From f7855c12c18bb97b7b9e49ab685ee558d8c0b47b Mon Sep 17 00:00:00 2001 From: Karl Berry Date: Mon, 23 Jul 2012 17:14:27 +0000 Subject: l3 3990 git-svn-id: svn://tug.org/texlive/trunk@27107 c570f23f-e606-0410-a88d-b1316a301751 --- .../doc/latex/l3experimental/l3dt/l3dt.pdf | Bin 520804 -> 553024 bytes .../doc/latex/l3experimental/l3sort/l3sort.pdf | Bin 501434 -> 515490 bytes .../doc/latex/l3experimental/l3str/l3flag.pdf | Bin 402815 -> 389876 bytes .../doc/latex/l3experimental/l3str/l3regex.pdf | Bin 713306 -> 983117 bytes .../doc/latex/l3experimental/l3str/l3str.pdf | Bin 776235 -> 850079 bytes .../latex/l3experimental/l3str/l3tl-analysis.pdf | Bin 507585 -> 519907 bytes .../doc/latex/l3experimental/l3str/l3tl-build.pdf | Bin 463230 -> 468383 bytes .../doc/latex/l3experimental/xcoffins/xcoffins.pdf | Bin 481218 -> 487843 bytes .../doc/latex/l3experimental/xgalley/l3galley.pdf | Bin 636687 -> 665514 bytes .../doc/latex/l3experimental/xgalley/xgalley.pdf | Bin 566960 -> 577252 bytes .../source/latex/l3experimental/l3dt/l3dt.dtx | 528 ++- .../source/latex/l3experimental/l3dt/l3dt.ins | 2 +- .../source/latex/l3experimental/l3sort/l3sort.dtx | 378 +- .../source/latex/l3experimental/l3sort/l3sort.ins | 2 +- .../source/latex/l3experimental/l3str/l3flag.dtx | 58 +- .../source/latex/l3experimental/l3str/l3regex.dtx | 3750 ++++++++++---------- .../source/latex/l3experimental/l3str/l3str.dtx | 2031 ++++++----- .../source/latex/l3experimental/l3str/l3str.ins | 2 +- .../latex/l3experimental/l3str/l3tl-analysis.dtx | 580 +-- .../latex/l3experimental/l3str/l3tl-build.dtx | 155 +- .../latex/l3experimental/xcoffins/xcoffins.dtx | 66 +- .../latex/l3experimental/xcoffins/xcoffins.ins | 2 +- .../latex/l3experimental/xgalley/l3galley.dtx | 629 ++-- .../latex/l3experimental/xgalley/xgalley.dtx | 244 +- .../latex/l3experimental/xgalley/xgalley.ins | 4 +- .../tex/latex/l3experimental/l3dt/l3dt.sty | 316 +- .../tex/latex/l3experimental/l3sort/l3sort.sty | 214 +- .../tex/latex/l3experimental/l3str/l3flag.sty | 40 +- .../latex/l3experimental/l3str/l3regex-trace.sty | 3010 ++++++++-------- .../tex/latex/l3experimental/l3str/l3regex.sty | 2960 +++++++-------- .../tex/latex/l3experimental/l3str/l3str.sty | 705 ++-- .../latex/l3experimental/l3str/l3tl-analysis.sty | 386 +- .../tex/latex/l3experimental/l3str/l3tl-build.sty | 90 +- .../tex/latex/l3experimental/xcoffins/xcoffins.sty | 44 +- .../tex/latex/l3experimental/xgalley/l3galley.sty | 491 ++- .../tex/latex/l3experimental/xgalley/xgalley.sty | 182 +- 36 files changed, 8521 insertions(+), 8348 deletions(-) (limited to 'Master/texmf-dist') diff --git a/Master/texmf-dist/doc/latex/l3experimental/l3dt/l3dt.pdf b/Master/texmf-dist/doc/latex/l3experimental/l3dt/l3dt.pdf index 5932347693b..ef1f6338685 100644 Binary files a/Master/texmf-dist/doc/latex/l3experimental/l3dt/l3dt.pdf and b/Master/texmf-dist/doc/latex/l3experimental/l3dt/l3dt.pdf differ diff --git a/Master/texmf-dist/doc/latex/l3experimental/l3sort/l3sort.pdf b/Master/texmf-dist/doc/latex/l3experimental/l3sort/l3sort.pdf index 9946455b071..643c77c2797 100644 Binary files a/Master/texmf-dist/doc/latex/l3experimental/l3sort/l3sort.pdf and b/Master/texmf-dist/doc/latex/l3experimental/l3sort/l3sort.pdf differ diff --git a/Master/texmf-dist/doc/latex/l3experimental/l3str/l3flag.pdf b/Master/texmf-dist/doc/latex/l3experimental/l3str/l3flag.pdf index 0c75fcf5818..6d452a1a251 100644 Binary files a/Master/texmf-dist/doc/latex/l3experimental/l3str/l3flag.pdf and b/Master/texmf-dist/doc/latex/l3experimental/l3str/l3flag.pdf differ diff --git a/Master/texmf-dist/doc/latex/l3experimental/l3str/l3regex.pdf b/Master/texmf-dist/doc/latex/l3experimental/l3str/l3regex.pdf index 9a2d148fd37..bf1c08b47e9 100644 Binary files a/Master/texmf-dist/doc/latex/l3experimental/l3str/l3regex.pdf and b/Master/texmf-dist/doc/latex/l3experimental/l3str/l3regex.pdf differ diff --git a/Master/texmf-dist/doc/latex/l3experimental/l3str/l3str.pdf b/Master/texmf-dist/doc/latex/l3experimental/l3str/l3str.pdf index ea0a481396d..dedf6013d1f 100644 Binary files a/Master/texmf-dist/doc/latex/l3experimental/l3str/l3str.pdf and b/Master/texmf-dist/doc/latex/l3experimental/l3str/l3str.pdf differ diff --git a/Master/texmf-dist/doc/latex/l3experimental/l3str/l3tl-analysis.pdf b/Master/texmf-dist/doc/latex/l3experimental/l3str/l3tl-analysis.pdf index 4286212f17a..0f6639fb947 100644 Binary files a/Master/texmf-dist/doc/latex/l3experimental/l3str/l3tl-analysis.pdf and b/Master/texmf-dist/doc/latex/l3experimental/l3str/l3tl-analysis.pdf differ diff --git a/Master/texmf-dist/doc/latex/l3experimental/l3str/l3tl-build.pdf b/Master/texmf-dist/doc/latex/l3experimental/l3str/l3tl-build.pdf index 1e9b60acacb..b5fe014e87e 100644 Binary files a/Master/texmf-dist/doc/latex/l3experimental/l3str/l3tl-build.pdf and b/Master/texmf-dist/doc/latex/l3experimental/l3str/l3tl-build.pdf differ diff --git a/Master/texmf-dist/doc/latex/l3experimental/xcoffins/xcoffins.pdf b/Master/texmf-dist/doc/latex/l3experimental/xcoffins/xcoffins.pdf index b8870804f7f..4c99ccfd767 100644 Binary files a/Master/texmf-dist/doc/latex/l3experimental/xcoffins/xcoffins.pdf and b/Master/texmf-dist/doc/latex/l3experimental/xcoffins/xcoffins.pdf differ diff --git a/Master/texmf-dist/doc/latex/l3experimental/xgalley/l3galley.pdf b/Master/texmf-dist/doc/latex/l3experimental/xgalley/l3galley.pdf index 459715894f1..821b441064c 100644 Binary files a/Master/texmf-dist/doc/latex/l3experimental/xgalley/l3galley.pdf and b/Master/texmf-dist/doc/latex/l3experimental/xgalley/l3galley.pdf differ diff --git a/Master/texmf-dist/doc/latex/l3experimental/xgalley/xgalley.pdf b/Master/texmf-dist/doc/latex/l3experimental/xgalley/xgalley.pdf index b9366016967..2489c2d4220 100644 Binary files a/Master/texmf-dist/doc/latex/l3experimental/xgalley/xgalley.pdf and b/Master/texmf-dist/doc/latex/l3experimental/xgalley/xgalley.pdf differ diff --git a/Master/texmf-dist/source/latex/l3experimental/l3dt/l3dt.dtx b/Master/texmf-dist/source/latex/l3experimental/l3dt/l3dt.dtx index 0fd090d9ef7..be3fc8eb1c1 100644 --- a/Master/texmf-dist/source/latex/l3experimental/l3dt/l3dt.dtx +++ b/Master/texmf-dist/source/latex/l3experimental/l3dt/l3dt.dtx @@ -36,7 +36,7 @@ % %<*driver|package> \RequirePackage{expl3} -\GetIdInfo$Id: l3dt.dtx 3459 2012-02-26 08:58:53Z joseph $ +\GetIdInfo$Id: l3dt.dtx 3877 2012-06-30 22:24:42Z joseph $ {L3 Experimental data tables} % %<*driver> @@ -165,9 +165,9 @@ % % \section{Removing data} % -% \begin{function}{\dt_del:Nn, \dt_gdel:Nn} +% \begin{function}{\dt_remove:Nn, \dt_gremove:Nn} % \begin{syntax} -% \cs{dt_del:Nn} \meta{dt} \Arg{key} +% \cs{dt_remove:Nn} \meta{dt} \Arg{key} % \end{syntax} % Deletes any entry from the current row of the \meta{data table} with % the \meta{key}. The \meta{key} is compared after processing with @@ -175,9 +175,9 @@ % all entries from a row does not delete the row itself. % \end{function} % -% \begin{function}{\dt_del:Nnn, \dt_gdel:Nnn} +% \begin{function}{\dt_remove:Nnn, \dt_gremove:Nnn} % \begin{syntax} -% \cs{dt_del:Nnn} \meta{dt} \Arg{row} \Arg{key} +% \cs{dt_remove:Nnn} \meta{dt} \Arg{row} \Arg{key} % \end{syntax} % Deletes any entry from the \meta{row} of the \meta{data table} with % the \meta{key}. The \meta{key} is compared after processing with @@ -384,88 +384,6 @@ % Scratch data tables for general use: these are never used by the kernel. % \end{variable} % -% \section{Internal function} -% -% \begin{variable}{\q_dt, \q_dt_header, \q_dt_row} -% Quarks used to construct the data table format. -% \end{variable} -% -% \begin{function}[EXP]{\dt_split_header:NT} -% \begin{syntax} -% \cs{dt_split_header:Nn} \meta{dt} \Arg{code} -% \end{syntax} -% Splits the \meta{data table} into the header part (containing the total -% number of rows present and the key list) and the body (containing the -% rows). The \meta{code} is then inserted, and should absorb the two parts -% of the split table as arguments. -% \end{function} -% -% \begin{function}{\dt_split_key:nnTF} -% \begin{syntax} -% \cs{dt_split_key:nnTF} \Arg{row} \Arg{key} \Arg{true code} \Arg{false code} -% \end{syntax} -% Searches the \meta{row} for the \meta{key}, using the -% comparison method as described for \cs{str_if_eq:nn}. If the \meta{key} is -% present, the \meta{true code} is left in the input stream followed by -% three \meta{balanced text} arguments -% \begin{enumerate} -% \item the partial \meta{row} for all keys before the \meta{key}, -% \item the \meta{value} for the \meta{key} and -% \item the partial \meta{row} for all keys after the \meta{key}. -% \end{enumerate} -% Thus the \meta{true code} must absorb three arguments. -% The two partial tables are structured such they may be recombined directly -% to produce a valid row lacking the entry for the \meta{key}. -% -% If the \meta{key} is not found in the \meta{row}, then the \meta{false code} -% is left in the input stream with no arguments. -% \end{function} -% -% \begin{function}{\dt_split_key_list:NnTF} -% \begin{syntax} -% \cs{dt_split_key_list:NnTF} \meta{dt} \Arg{key} \Arg{true code} \Arg{false code} -% \end{syntax} -% Searches the key list of the \meta{dt} for the \meta{key}, using the -% comparison method as described for \cs{str_if_eq:nn}. If the \meta{key} is -% present, the \meta{true code} is left in the input stream followed by -% five \meta{balanced text} arguments -% \begin{enumerate} -% \item the partial table for all keys before the \meta{key}, -% \item the \meta{id} for the \meta{key}, -% \item the \meta{type} for the \meta{key}, -% \item the \meta{header} for the \meta{key} and -% \item the partial table for all keys after the \meta{key}. -% \end{enumerate} -% Thus the \meta{true code} must absorb five arguments. -% The two partial tables are structured such they may be recombined directly -% to produce a valid key table lacking the entry for the \meta{key}. -% -% If the \meta{key} is not found in the key table, then the \meta{false code} -% is left in the input stream with no arguments. -% \end{function} -% -% \begin{function}{\dt_split_row:NnTF} -% \begin{syntax} -% \cs{dt_split_row:NnTF} \meta{dt} \Arg{row} \Arg{true code} \Arg{false code} -% \end{syntax} -% Searches the key list of the \meta{dt} for the \meta{row} -% (an \meta{integer expression}). If the -% \meta{row} is present, the \meta{true code} is left in the input stream -% followed by three \meta{balanced text} arguments -% \begin{enumerate} -% \item the partial table before the \meta{row}, -% \item the content of the \meta{row}, starting and ending with the -% row number, -% \item the partial table after the \meta{row}. -% \end{enumerate} -% Thus the \meta{true code} must absorb three arguments. -% The two partial tables are structured such they may be recombined directly -% to produce a valid key table lacking the entry for the \meta{row}. -% -% If the \meta{row} is not found in the key table, then the \meta{false code} -% is left in the input stream with no arguments. -% \end{function} -% % \end{documentation} % % \begin{implementation} @@ -475,12 +393,15 @@ % \begin{macrocode} %<*initex|package> % \end{macrocode} +% +% \begin{macrocode} +%<@@=dt> +% \end{macrocode} % % \begin{macrocode} %<*package> \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} -\package_check_loaded_expl: % % \end{macrocode} % @@ -495,34 +416,34 @@ % numbered: % \begin{quote} % \Arg{rows} \\ -% \cs{q_dt} \meta{key$_0$} \cs{q_dt} \meta{key$_1$} \cs{q_dt} \ldots \\ +% \cs{q_@@} \meta{key$_0$} \cs{q_@@} \meta{key$_1$} \cs{q_@@} \ldots \\ % \cs{q_nil} \\ -% \cs{q_dt_header} \\ -% \cs{q_dt_row} \\ +% \cs{q_@@_header} \\ +% \cs{q_@@_row} \\ % \meta{row$_0$} \\ -% \cs{q_dt} \meta{key$_0$} \cs{q_dt} \Arg{data$_{0,0}$} \\ -% \cs{q_dt} \meta{key$_1$} \cs{q_dt} \Arg{data$_{0,1}$} \\ +% \cs{q_@@} \meta{key$_0$} \cs{q_@@} \Arg{data$_{0,0}$} \\ +% \cs{q_@@} \meta{key$_1$} \cs{q_@@} \Arg{data$_{0,1}$} \\ % \ldots \\ -% \cs{q_dt} \\ +% \cs{q_@@} \\ % \cs{q_nil} \\ -% \cs{q_dt_row} \\ +% \cs{q_@@_row} \\ % \meta{row$_2$} \\ -% \cs{q_dt} \meta{key$_0$} \cs{q_dt} \Arg{data$_{1,0}$} \\ -% \cs{q_dt} \meta{key$_1$} \cs{q_dt} \Arg{data$_{1,1}$} \\ +% \cs{q_@@} \meta{key$_0$} \cs{q_@@} \Arg{data$_{1,0}$} \\ +% \cs{q_@@} \meta{key$_1$} \cs{q_@@} \Arg{data$_{1,1}$} \\ % \ldots \\ -% \cs{q_dt} \\ +% \cs{q_@@} \\ % \cs{q_nil} \\ -% \cs{q_dt_row} \\ +% \cs{q_@@_row} \\ % \ldots \\ -% \cs{q_dt_row} +% \cs{q_@@_row} % \end{quote} % -% \begin{variable}{\q_dt, \q_dt_row, \q_dt_header} +% \begin{variable}{\q_@@, \q_@@_row, \q_@@_header} % The quarks are set up. % \begin{macrocode} -\quark_new:N \q_dt -\quark_new:N \q_dt_row -\quark_new:N \q_dt_header +\quark_new:N \q_@@ +\quark_new:N \q_@@_row +\quark_new:N \q_@@_header % \end{macrocode} % \end{variable} % @@ -533,10 +454,10 @@ \tl_const:Nn \c_empty_dt { { 0 } - \q_dt + \q_@@ \q_nil - \q_dt_header - \q_dt_row + \q_@@_header + \q_@@_row } % \end{macrocode} % \end{variable} @@ -590,104 +511,104 @@ % % \subsection{Splitting functions} % -% \begin{macro}[aux]{\dt_split_aux:nnnn} -% \begin{macro}[aux]{\dt_split_aux:w} +% \begin{macro}[aux]{\@@_split:nnnn} +% \begin{macro}[aux]{\@@_split:w} % Two general auxiliaries. The \texttt{nnnn} function is used to apply the % \texttt{T} branch if a match is found and the \texttt{F} branch otherwise. % The \texttt{w} function is general purpose, and is used to define the % matching parameter set. % \begin{macrocode} -\cs_new_protected:Npn \dt_split_aux:nnnn #1#2#3#4 { #3 #2 } -\cs_new_protected:Npn \dt_split_aux:w { } +\cs_new_protected:Npn \@@_split:nnnn #1#2#3#4 { #3 #2 } +\cs_new_protected:Npn \@@_split:w { } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int,EXP]{\dt_split_header:NT} -% \begin{macro}[aux,EXP]{\dt_split_header_aux:wn} +% \begin{macro}[int,EXP]{\@@_split_header:NT} +% \begin{macro}[aux,EXP]{\@@_split_header:wn} % Splits the header from the table, inserting the code required to then % process the split table. The \cs{q_nil} is also removed from the end of the % header, as it is essentially a distraction here. % \begin{macrocode} -\cs_new:Npn \dt_split_header:NT #1#2 - { \exp_after:wN \dt_split_header_aux:wn #1 \q_stop {#2} } -\cs_new:Npn \dt_split_header_aux:wn #1 \q_nil \q_dt_header #2 \q_stop #3 - { #3 {#1} { \q_dt_header #2 } } +\cs_new:Npn \@@_split_header:NT #1#2 + { \exp_after:wN \@@_split_header:wn #1 \q_stop {#2} } +\cs_new:Npn \@@_split_header:wn #1 \q_nil \q_@@_header #2 \q_stop #3 + { #3 {#1} { \q_@@_header #2 } } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\dt_split_key:nnTF} -% \begin{macro}[aux]{\dt_split_key_aux:nnTF} +% \begin{macro}[int]{\@@_split_key:nnTF} +% \begin{macro}[aux]{\@@_split_key_aux:nnTF} % Here, the split is made for a partial list within a row. The row is % basically the same as a property list, so the split here is almost % identical to that in \cs{prop_split_aux:NnTF}. The row-end data is set up % such that it will not interfere with this process. % \begin{macrocode} -\cs_new_protected:Npn \dt_split_key:nnTF #1#2 - { \exp_args:No \dt_split_key_aux:nnTF { \tl_to_str:n {#2} } {#1} } -\cs_new_protected:Npn \dt_split_key_aux:nnTF #1#2 +\cs_new_protected:Npn \@@_split_key:nnTF #1#2 + { \exp_args:No \@@_split_key_aux:nnTF { \tl_to_str:n {#2} } {#1} } +\cs_new_protected:Npn \@@_split_key_aux:nnTF #1#2 { - \cs_set_protected:Npn \dt_split_aux:w - ##1 \q_dt #1 \q_dt ##2##3##4 \q_mark ##5 \q_stop - { \dt_split_aux:nnnn ##3 { { ##1 \q_dt } {##2} {##4} } } - \dt_split_aux:w #2 \q_mark - \q_dt #1 \q_dt { } { ? \use_ii:nn { } } \q_mark \q_stop + \cs_set_protected:Npn \@@_split:w + ##1 \q_@@ #1 \q_@@ ##2##3##4 \q_mark ##5 \q_stop + { \@@_split:nnnn ##3 { { ##1 \q_@@ } {##2} {##4} } } + \@@_split:w #2 \q_mark + \q_@@ #1 \q_@@ { } { ? \use_ii:nn { } } \q_mark \q_stop } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\dt_split_key_list:NnTF} -% \begin{macro}[aux]{\dt_split_key_list_aux:NnTF} +% \begin{macro}[int]{\@@_split_key_list:NnTF} +% \begin{macro}[aux]{\@@_split_key_list_aux:NnTF} % Finding a key in the header uses a similar approach to finding a key in % a property list. Here, if the key is found there will always be at least -% one token between \cs{q_dt} and \cs{q_dt_header} due to the \cs{q_nil} -% which is part of a new table. The use of |##1##2| in \cs{dt_split_aux:w} +% one token between \cs{q_@@} and \cs{q_@@_header} due to the \cs{q_nil} +% which is part of a new table. The use of |##1##2| in \cs{@@_split:w} % here is to deal with the overall number of rows. The set up here means % that this will always be unbraced then rebraced: simply grabbing |##1| % to include this and anything before the key of interest will give variable % results depending on whether the match is to the very first key or not. % \begin{macrocode} -\cs_new_protected:Npn \dt_split_key_list:NnTF #1#2 - { \exp_args:NNo \dt_split_key_list_aux:NnTF #1 { \tl_to_str:n {#2} } } -\cs_new_protected:Npn \dt_split_key_list_aux:NnTF #1#2 +\cs_new_protected:Npn \@@_split_key_list:NnTF #1#2 + { \exp_args:NNo \@@_split_key_list_aux:NnTF #1 { \tl_to_str:n {#2} } } +\cs_new_protected:Npn \@@_split_key_list_aux:NnTF #1#2 { - \cs_set_protected:Npn \dt_split_aux:w - ##1##2 \q_dt #2 \q_dt ##3##4 \q_dt_header ##5 \q_mark ##6 \q_stop + \cs_set_protected:Npn \@@_split:w + ##1##2 \q_@@ #2 \q_@@ ##3##4 \q_@@_header ##5 \q_mark ##6 \q_stop { - \dt_split_aux:nnnn ##3 - { { {##1} ##2 \q_dt } { ##3##4 \q_dt_header ##5 } } + \@@_split:nnnn ##3 + { { {##1} ##2 \q_@@ } { ##3##4 \q_@@_header ##5 } } } - \exp_after:wN \dt_split_aux:w #1 \q_mark - \q_dt #2 \q_dt { ? \use_ii:nn { } } \q_dt_header \q_mark \q_stop + \exp_after:wN \@@_split:w #1 \q_mark + \q_@@ #2 \q_@@ { ? \use_ii:nn { } } \q_@@_header \q_mark \q_stop } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\dt_split_row:NnTF} -% \begin{macro}[aux]{\dt_split_row_aux:NnTF, \dt_split_row_aux:NfTF} +% \begin{macro}[int]{\@@_split_row:NnTF} +% \begin{macro}[aux]{\@@_split_row_aux:NnTF, \@@_split_row_aux:NfTF} % The usual approach, here using the fact that each row start with row number % and ends with \cs{q_nil} so there will always be at least one token to be % absorbed as |##2|. The only odd thing to watch here is that the row % number is evaluated so that higher-level functions in the main do not % need to have an \texttt{f}-type variant. % \begin{macrocode} -\cs_new_protected:Npn \dt_split_row:NnTF #1#2 - { \dt_split_row_aux:NfTF #1 { \int_eval:n {#2} } } -\cs_new_protected:Npn \dt_split_row_aux:NnTF #1#2 +\cs_new_protected:Npn \@@_split_row:NnTF #1#2 + { \@@_split_row_aux:NfTF #1 { \int_eval:n {#2} } } +\cs_new_protected:Npn \@@_split_row_aux:NnTF #1#2 { - \cs_set_protected:Npn \dt_split_aux:w - ##1 \q_dt_row #2 \q_dt ##2##3 \q_dt_row ##4 \q_mark ##5 \q_stop + \cs_set_protected:Npn \@@_split:w + ##1 \q_@@_row #2 \q_@@ ##2##3 \q_@@_row ##4 \q_mark ##5 \q_stop { - \dt_split_aux:nnnn ##2 - { { ##1 \q_dt_row } { #2 \q_dt ##2##3 } {##4} } + \@@_split:nnnn ##2 + { { ##1 \q_@@_row } { #2 \q_@@ ##2##3 } {##4} } } - \exp_after:wN \dt_split_aux:w #1 \q_mark - \q_dt_row #2 \q_dt { ? \use_ii:nn { } } \q_dt_row \q_mark \q_stop + \exp_after:wN \@@_split:w #1 \q_mark + \q_@@_row #2 \q_@@ { ? \use_ii:nn { } } \q_@@_row \q_mark \q_stop } -\cs_generate_variant:Nn \dt_split_row_aux:NnTF { Nf } +\cs_generate_variant:Nn \@@_split_row_aux:NnTF { Nf } % \end{macrocode} % \end{macro} % \end{macro} @@ -695,33 +616,33 @@ % \subsection{Adding and removing data} % % \begin{macro}{\dt_add_key:Nn, \dt_gadd_key:Nn} -% \begin{macro}[aux]{\dt_add_key_aux:NNn} -% \begin{macro}[aux]{\dt_add_key_aux:NNnnn} +% \begin{macro}[aux]{\@@_add_key:NNn} +% \begin{macro}[aux]{\@@_add_key:NNnnn} % Here, there are two stages. If the key is already present in the list of % known keys then no action is taken, and the split list is thrown away. % On the other hand, if the key is not present then the header and body % are separated and the key is added to the end of the list of known keys -% (hence keys are ordered). The \cs{dt_split_header:Nn} function will have +% (hence keys are ordered). The \cs{@@_split_header:Nn} function will have % removed the \cs{q_nil} from the header, and so it is put back in here. % \begin{macrocode} -\cs_new_protected_nopar:Npn \dt_add_key:Nn { \dt_add_key_aux:NNn \tl_set:Nx } -\cs_new_protected_nopar:Npn \dt_gadd_key:Nn { \dt_add_key_aux:NNn \tl_gset:Nx } -\cs_new_protected:Npn \dt_add_key_aux:NNn #1#2#3 +\cs_new_protected_nopar:Npn \dt_add_key:Nn { \@@_add_key:NNn \tl_set:Nx } +\cs_new_protected_nopar:Npn \dt_gadd_key:Nn { \@@_add_key:NNn \tl_gset:Nx } +\cs_new_protected:Npn \@@_add_key:NNn #1#2#3 { - \dt_split_key_list:NnTF #2 {#3} + \@@_split_key_list:NnTF #2 {#3} { \use_none:nn } { - \dt_split_header:NT #2 - { \dt_add_key_aux:NNnnn #1 #2 {#3} } + \@@_split_header:NT #2 + { \@@_add_key:NNnnn #1 #2 {#3} } } } -\cs_new_protected:Npn \dt_add_key_aux:NNnnn #1#2#3#4#5 +\cs_new_protected:Npn \@@_add_key:NNnnn #1#2#3#4#5 { #1 #2 { \exp_not:n {#4} \tl_to_str:n {#3} - \exp_not:n { \q_dt \q_nil #5 } + \exp_not:n { \q_@@ \q_nil #5 } } } % \end{macrocode} @@ -730,28 +651,28 @@ % \end{macro} % % \begin{macro}{\dt_add_row:N, \dt_gadd_row:N} -% \begin{macro}[aux]{\dt_add_row_aux:NN} -% \begin{macro}[aux]{\dt_add_row_aux:NnN} -% \begin{macro}[aux,EXP]{\dt_add_row_aux:nw} +% \begin{macro}[aux]{\@@_add_row:NN} +% \begin{macro}[aux]{\@@_add_row:NnN} +% \begin{macro}[aux,EXP]{\@@_add_row:nw} % Adding a row means incrementing the total number and adding the structure % of an empty row. As finding the rows will get slow for large tables, this % is only done once. % \begin{macrocode} -\cs_new_protected_nopar:Npn \dt_add_row:N { \dt_add_row_aux:NN \tl_set:Nx } -\cs_new_protected_nopar:Npn \dt_gadd_row:N { \dt_add_row_aux:NN \tl_gset:Nx } -\cs_new_protected:Npn \dt_add_row_aux:NN #1#2 - { \exp_args:NNf \dt_add_row_aux:NnN #1 { \dt_rows:N #2 } #2 } -\cs_new_protected:Npn \dt_add_row_aux:NnN #1#2#3 +\cs_new_protected_nopar:Npn \dt_add_row:N { \@@_add_row:NN \tl_set:Nx } +\cs_new_protected_nopar:Npn \dt_gadd_row:N { \@@_add_row:NN \tl_gset:Nx } +\cs_new_protected:Npn \@@_add_row:NN #1#2 + { \exp_args:NNf \@@_add_row:NnN #1 { \dt_rows:N #2 } #2 } +\cs_new_protected:Npn \@@_add_row:NnN #1#2#3 { #1 #3 { { \int_eval:n { #2 + \c_one } } - \exp_after:wN \dt_add_row_aux:nw #3 \q_stop + \exp_after:wN \@@_add_row:nw #3 \q_stop #2 - \exp_not:n { \q_dt \q_nil \q_dt_row } + \exp_not:n { \q_@@ \q_nil \q_@@_row } } } -\cs_new:Npn \dt_add_row_aux:nw #1#2 \q_stop { \exp_not:n {#2} } +\cs_new:Npn \@@_add_row:nw #1#2 \q_stop { \exp_not:n {#2} } % \end{macrocode} % \end{macro} % \end{macro} @@ -770,29 +691,29 @@ % \end{macro} % % \begin{macro}{\dt_put:Nnnn, \dt_gput:Nnnn} -% \begin{macro}[aux]{\dt_put_aux:NNNnnn} -% \begin{macro}[aux]{\dt_put_aux:NNnnnnn} -% \begin{macro}[aux]{\dt_put_update:NNnnnnnnn} -% \begin{macro}[aux]{\dt_put_add_to_row:NNnnnnn} -% \begin{macro}[aux, EXP]{\dt_put_add_to_row_aux:w} +% \begin{macro}[aux]{\@@_put:NNNnnn} +% \begin{macro}[aux]{\@@_put:NNnnnnn} +% \begin{macro}[aux]{\@@_put_update:NNnnnnnnn} +% \begin{macro}[aux]{\@@_put_add_to_row:NNnnnnn} +% \begin{macro}[aux, EXP]{\@@_put_add_to_row_aux:w} % Adding to a row is a slightly complex procedure. The lead-off is the % standard combination across the local and global routes. % \begin{macrocode} \cs_new_protected_nopar:Npn \dt_put:Nnnn - { \dt_put_aux:NNNnnn \dt_add_key:Nn \tl_set:Nx } + { \@@_put:NNNnnn \dt_add_key:Nn \tl_set:Nx } \cs_new_protected_nopar:Npn \dt_gput:Nnnn - { \dt_put_aux:NNNnnn \dt_gadd_key:Nn \tl_gset:Nx } + { \@@_put:NNNnnn \dt_gadd_key:Nn \tl_gset:Nx } % \end{macrocode} % Add the key to the list those known, if necessary, then check that the % row requested makes sense. % \begin{macrocode} -\cs_new_protected:Npn \dt_put_aux:NNNnnn #1#2#3#4#5#6 +\cs_new_protected:Npn \@@_put:NNNnnn #1#2#3#4#5#6 { #1 #3 {#5} - \dt_split_row:NnTF #3 {#4} - { \dt_put_aux:NNnnnnn #2 #3 {#5} {#6} } + \@@_split_row:NnTF #3 {#4} + { \@@_put:NNnnnnn #2 #3 {#5} {#6} } { - \msg_kernel_error:nnxxx { dt } { unknown-row } + \__msg_kernel_error:nnxxx { dt } { unknown-row } { \token_to_str:N #3 } { \int_eval:n {#4} } { \dt_rows:N #3 } } } @@ -811,11 +732,11 @@ % stack if the key is already present. So there is some care needed sending % the parameters forward without running out of \TeX{} arguments. % \begin{macrocode} -\cs_new_protected:Npn \dt_put_aux:NNnnnnn #1#2#3#4#5#6#7 +\cs_new_protected:Npn \@@_put:NNnnnnn #1#2#3#4#5#6#7 { - \dt_split_key:nnTF {#6} {#3} - { \dt_put_update:NNnnnnnnn #1 #2 {#3} {#4} {#5} {#7} } - { \dt_put_add_to_row:NNnnnnn #1 #2 {#3} {#4} {#5} {#6} {#7} } + \@@_split_key:nnTF {#6} {#3} + { \@@_put_update:NNnnnnnnn #1 #2 {#3} {#4} {#5} {#7} } + { \@@_put_add_to_row:NNnnnnn #1 #2 {#3} {#4} {#5} {#6} {#7} } } % \end{macrocode} % The arguments here are @@ -834,32 +755,32 @@ % |#8| is needed. To try to keep things clear, there are a few more % \cs{exp_not:n} here than formally required. % \begin{macrocode} -\cs_new_protected:Npn \dt_put_update:NNnnnnnnn #1#2#3#4#5#6#7#8#9 +\cs_new_protected:Npn \@@_put_update:NNnnnnnnn #1#2#3#4#5#6#7#8#9 { #1 #2 { \exp_not:n { #5 #7 } \tl_to_str:n {#3} - \exp_not:n { \q_dt {#4} \q_dt #9 \q_dt_row #6 } + \exp_not:n { \q_@@ {#4} \q_@@ #9 \q_@@_row #6 } } } % \end{macrocode} % A slightly more complex case when adding an item. The arguments here are -% identical to those for \cs{dt_put_aux:NNnnnnnn}. The row has not been +% identical to those for \cs{@@_put:NNnnnnnn}. The row has not been % split, so the \cs{q_nil} there is removed and re-added to come after the % new content. % \begin{macrocode} -\cs_new_protected:Npn \dt_put_add_to_row:NNnnnnn #1#2#3#4#5#6#7 +\cs_new_protected:Npn \@@_put_add_to_row:NNnnnnn #1#2#3#4#5#6#7 { #1 #2 { \exp_not:n {#5} - \exp_not:o { \dt_put_add_row_aux:w #6 } + \exp_not:o { \@@_put_add_to_row_aux:w #6 } \tl_to_str:n {#3} - \exp_not:n { \q_dt {#4} \q_dt \q_nil \q_dt_row #7 } + \exp_not:n { \q_@@ {#4} \q_@@ \q_nil \q_@@_row #7 } } } -\cs_new:Npn \dt_put_add_row_aux:w #1 \q_nil {#1} +\cs_new:Npn \@@_put_add_to_row_aux:w #1 \q_nil {#1} % \end{macrocode} % \end{macro} % \end{macro} @@ -869,31 +790,29 @@ % \end{macro} % % \begin{macro}[EXP]{\dt_keys:N} -% \begin{macro}[aux, EXP]{\dt_keys_aux:nn} -% \begin{macro}[aux, EXP]{\dt_keys_aux:wN} +% \begin{macro}[aux, EXP]{\@@_keys:nn} +% \begin{macro}[aux, EXP]{\@@_keys:wN} % A quick mapping is needed to count keys. The \cs{use_none:nn} here -% is used to remove the number of rows and initial \cs{q_dt}. This could +% is used to remove the number of rows and initial \cs{q_@@}. This could % also be handled by starting from $-1$ rather than $0$, but this makes % the logic hopefully slightly clearer. % \begin{macrocode} \cs_new:Npn \dt_keys:N #1 - { \dt_split_header:NT #1 { \dt_keys_aux:nn } } -\cs_new:Npn \dt_keys_aux:nn #1#2 + { \@@_split_header:NT #1 { \@@_keys:nn } } +\cs_new:Npn \@@_keys:nn #1#2 { \int_eval:n { 0 - \exp_after:wN \dt_keys_aux:wN \use_none:nn #1 \q_recursion_tail \q_dt - \prg_break_point:n { } + \exp_after:wN \@@_keys:wN \use_none:nn #1 \q_recursion_tail \q_@@ + \__prg_break_point:Nn \dt_map_break: { } } } -\cs_new:Npn \dt_keys_aux:wN #1 \q_dt +\cs_new:Npn \@@_keys:wN #1 \q_@@ { - \if_meaning:w \q_recursion_tail #1 - \exp_after:wN \prg_map_break: - \fi: + \__quark_if_recursion_tail_break:nN {#1} \dt_map_break: +1 - \dt_keys_aux:wN + \@@_keys:wN } % \end{macrocode} % \end{macro} @@ -910,40 +829,40 @@ % % \subsection{Removing data} % -% \begin{macro}{\dt_del:Nn,\dt_gdel:Nn} +% \begin{macro}{\dt_remove:Nn,\dt_gremove:Nn} % Deleting to the current row is simply a special case of deleting to an % arbitrary row. % \begin{macrocode} -\cs_new_protected:Npn \dt_del:Nn #1 { \dt_del:Nnn #1 +\cs_new_protected:Npn \dt_remove:Nn #1 { \dt_remove:Nnn #1 { \dt_rows:N #1 - \c_one } } -\cs_new_protected:Npn \dt_gdel:Nn #1 { \dt_gdel:Nnn #1 +\cs_new_protected:Npn \dt_gremove:Nn #1 { \dt_gremove:Nnn #1 { \dt_rows:N #1 - \c_one } } % \end{macrocode} % \end{macro} % -% \begin{macro}{\dt_del:Nnn, \dt_gdel:Nnn} -% \begin{macro}[aux]{\dt_del_aux:NNnn} -% \begin{macro}[aux]{\dt_del_aux:NNnnnn} -% \begin{macro}[aux]{\dt_del_aux:NNnnnnn} +% \begin{macro}{\dt_remove:Nnn, \dt_gremove:Nnn} +% \begin{macro}[aux]{\dt_remove_aux:NNnn} +% \begin{macro}[aux]{\dt_remove_aux:NNnnnn} +% \begin{macro}[aux]{\dt_remove_aux:NNnnnnn} % Deleting a single entry from a single row means first splitting by row, % then splitting by key, and finally doing the assignment. If the row or the % key are not present then the entire function does nothing at all. % \begin{macrocode} -\cs_new_protected_nopar:Npn \dt_del:Nnn { \dt_del_aux:NNnn \tl_set:Nn } -\cs_new_protected_nopar:Npn \dt_gdel:Nnn { \dt_del_aux:NNnn \tl_gset:Nn } -\cs_new_protected:Npn \dt_del_aux:NNnn #1#2#3#4 +\cs_new_protected_nopar:Npn \dt_remove:Nnn { \dt_remove_aux:NNnn \tl_set:Nn } +\cs_new_protected_nopar:Npn \dt_gremove:Nnn { \dt_remove_aux:NNnn \tl_gset:Nn } +\cs_new_protected:Npn \dt_remove_aux:NNnn #1#2#3#4 { - \dt_split_row:NnTF #2 {#3} - { \dt_del_aux:NNnnnn #1 #2 {#4} } + \@@_split_row:NnTF #2 {#3} + { \dt_remove_aux:NNnnnn #1 #2 {#4} } { } } -\cs_new_protected:Npn \dt_del_aux:NNnnnn #1#2#3#4#5#6 +\cs_new_protected:Npn \dt_remove_aux:NNnnnn #1#2#3#4#5#6 { - \dt_split_key:nnTF {#5} {#3} - { \dt_del_aux:NNnnnnn #1 #2 {#4} {#6} } + \@@_split_key:nnTF {#5} {#3} + { \dt_remove_aux:NNnnnnn #1 #2 {#4} {#6} } { } } -\cs_new_protected:Npn \dt_del_aux:NNnnnnn #1#2#3#4#5#6#7 +\cs_new_protected:Npn \dt_remove_aux:NNnnnnn #1#2#3#4#5#6#7 { #1 #2 { #3 #5 #7 #4 } } % \end{macrocode} % \end{macro} @@ -968,25 +887,23 @@ { \dt_remove_key_aux:NNn \tl_gset:Nx } \cs_new_protected:Npn \dt_remove_key_aux:NNn #1#2#3 { - \dt_split_key_list:NnTF #2 {#3} + \@@_split_key_list:NnTF #2 {#3} { \exp_args:No \dt_remove_key_aux:nNNnn { \tl_to_str:n {#3} } #1 #2 } { } } \cs_new_protected:Npn \dt_remove_key_aux:nNNnn #1#2#3#4#5 { - \cs_set:Npn \dt_remove_key_aux:w ##1 \q_dt #1 \q_dt ##2 ##3 + \cs_set:Npn \dt_remove_key_aux:w ##1 \q_@@ #1 \q_@@ ##2 ##3 { \exp_not:n {##1} - \if_meaning:w \q_recursion_tail ##3 - \exp_after:wN \prg_map_break: - \fi: + \__quark_if_recursion_tail_break:nN {##3} \dt_map_break: \dt_remove_key_aux:w ##3 } #2 #3 { \exp_not:n {#4} - \dt_remove_key_aux:w #5 \q_dt #1 \q_dt { } \q_recursion_tail - \prg_break_point:n { } + \dt_remove_key_aux:w #5 \q_@@ #1 \q_@@ { } \q_recursion_tail + \__prg_break_point:Nn \dt_map_break: { } } } \cs_new:Npn \dt_remove_key_aux:w { } @@ -1011,7 +928,7 @@ { \dt_remove_row_aux:NNn \tl_gset:Nx } \cs_new_protected:Npn \dt_remove_row_aux:NNn #1#2#3 { - \dt_split_row:NnTF #2 {#3} + \@@_split_row:NnTF #2 {#3} { \dt_remove_row_aux:NNnnn #1 #2 } { } } @@ -1028,18 +945,16 @@ { { \int_eval:n { \dt_rows:N #2 - \c_one } } \dt_remove_row_aux:nw #3 \q_stop - \dt_remove_row_loop:nw #5 \q_recursion_tail \q_dt_row - \prg_break_point:n { } + \dt_remove_row_loop:nw #5 \q_recursion_tail \q_@@_row + \__prg_break_point:Nn \dt_map_break: { } } } -\cs_new_eq:NN \dt_remove_row_aux:nw \dt_add_row_aux:nw -\cs_new:Npn \dt_remove_row_loop:nw #1#2 \q_dt_row +\cs_new_eq:NN \dt_remove_row_aux:nw \@@_add_row:nw +\cs_new:Npn \dt_remove_row_loop:nw #1#2 \q_@@_row { - \if_meaning:w \q_recursion_tail #1 - \exp_after:wN \prg_map_break: - \fi: + \__quark_if_recursion_tail_break:nN {#1} \dt_map_break: \int_eval:n { #1 - \c_one } - \exp_not:n { #2 \q_dt_row } + \exp_not:n { #2 \q_@@_row } \dt_remove_row_loop:nw } % \end{macrocode} @@ -1060,13 +975,13 @@ % \begin{macrocode} \cs_new_protected:Npn \dt_get:NnnN #1#2#3#4 { - \dt_split_row:NnTF #1 {#2} + \@@_split_row:NnTF #1 {#2} { \dt_get_aux:nNnnn {#3} #4 } { \tl_set:Nn #4 { \q_no_value } } } \cs_new_protected:Npn \dt_get_aux:nNnnn #1#2#3#4#5 { - \dt_split_key:nnTF {#4} {#1} + \@@_split_key:nnTF {#4} {#1} { \dt_get_aux:Nnnn #2 } { \tl_set:Nn #2 { \q_no_value } } } @@ -1077,23 +992,23 @@ % \end{macro} % % \begin{macro}[TF]{\dt_get:NnnN} -% \begin{macro}[aux]{\dt_get_aux_true:nNnnn} -% \begin{macro}[aux]{\dt_get_aux_true:Nnnn} +% \begin{macro}[aux]{\@@_get_true:nNnnn} +% \begin{macro}[aux]{\@@_get_true:Nnnn} % The same idea as the standard method, but built as a conditional. % \begin{macrocode} \prg_new_protected_conditional:Npnn \dt_get:NnnN #1#2#3#4 { T , F , TF } { - \dt_split_row:NnTF #1 {#2} - { \dt_get_aux_true:nNnnn {#3} #4 } + \@@_split_row:NnTF #1 {#2} + { \@@_get_true:nNnnn {#3} #4 } { \prg_return_false: } } -\cs_new_protected:Npn \dt_get_aux_true:nNnnn #1#2#3#4#5 +\cs_new_protected:Npn \@@_get_true:nNnnn #1#2#3#4#5 { - \dt_split_key:nnTF {#4} {#1} - { \dt_get_aux_true:Nnnn #2 } + \@@_split_key:nnTF {#4} {#1} + { \@@_get_true:Nnnn #2 } { \prg_return_false: } } -\cs_new_protected:Npn \dt_get_aux_true:Nnnn #1#2#3#4 +\cs_new_protected:Npn \@@_get_true:Nnnn #1#2#3#4 { \tl_set:Nn #1 {#3} \prg_return_true: @@ -1130,10 +1045,10 @@ % \end{variable} % % \begin{macro}{\dt_map_variables:Nnn} -% \begin{macro}[aux]{\dt_map_variables_key:nn} -% \begin{macro}[aux]{\dt_map_variables_aux:nnn} -% \begin{macro}[aux]{\dt_map_variables_aux:nNNw} -% \begin{macro}[aux]{\dt_map_variables_aux:nnw} +% \begin{macro}[aux]{\@@_map_variables_key:nn} +% \begin{macro}[aux]{\@@_map_variables:nnn} +% \begin{macro}[aux]{\@@_map_variables:nNNw} +% \begin{macro}[aux]{\@@_map_variables:nnw} % Mapping across a data table is more complex than other cases as there % are two \enquote{dimensions} to worry about: the rows and the keys. % The first stage of the mapping is to convert the key--variable mapping @@ -1145,38 +1060,37 @@ { \int_gincr:N \g_dt_map_level_int \seq_gclear_new:c { g_dt_map_ \int_use:N \g_dt_map_level_int _seq } - \keyval_parse:NNn \use_none:n \dt_map_variables_key:nn {#2} - \dt_split_header:NT #1 { \dt_map_variables_aux:nnn {#3} } + \keyval_parse:NNn \use_none:n \@@_map_variables_key:nn {#2} + \@@_split_header:NT #1 { \@@_map_variables:nnn {#3} } } -\cs_new_protected:Npn \dt_map_variables_key:nn #1#2 +\cs_new_protected:Npn \@@_map_variables_key:nn #1#2 { \seq_gput_right:cn { g_dt_map_ \int_use:N \g_dt_map_level_int _seq } { {#1} #2 } } % \end{macrocode} -% As \cs{dt_split_header:NT} will leave a couple of tokens at the front +% As \cs{@@_split_header:NT} will leave a couple of tokens at the front % of the body part of the split, there is a quick piece of tidying up % to remove them. % \begin{macrocode} -\cs_new_protected:Npn \dt_map_variables_aux:nnn #1#2#3 - { \dt_map_variables_aux:nNNw {#1} #3 \q_stop } -\cs_new_protected:Npn \dt_map_variables_aux:nNNw - #1 \q_dt_header \q_dt_row #2 \q_stop +\cs_new_protected:Npn \@@_map_variables:nnn #1#2#3 + { \@@_map_variables:nNNw {#1} #3 \q_stop } +\cs_new_protected:Npn \@@_map_variables:nNNw + #1 \q_@@_header \q_@@_row #2 \q_stop { \int_zero_new:c { l_dt_map_ \int_use:N \g_dt_map_level_int _row_int } - \dt_map_variables_aux:nnw {#1} #2 { } \q_recursion_tail \q_dt_row - \prg_break_point:n { \int_gdecr:N \g_dt_map_level_int } + \@@_map_variables:nnw {#1} #2 { } \q_recursion_tail \q_@@_row + \__prg_break_point:Nn \dt_map_break: + { \int_gdecr:N \g_dt_map_level_int } } -\cs_new_protected:Npn \dt_map_variables_aux:nnw #1#2#3#4 \q_dt_row +\cs_new_protected:Npn \@@_map_variables:nnw #1#2#3#4 \q_@@_row { - \if_meaning:w \q_recursion_tail #3 - \exp_after:wN \dt_map_break: - \fi: + \__quark_if_recursion_tail_break:nN {#3} \dt_map_break: \seq_map_inline:cn { g_dt_map_ \int_use:N \g_dt_map_level_int _seq } { \dt_get_aux:nNnnn ##1 { } {#3#4} { } } #1 \int_incr:c { l_dt_map_ \int_use:N \g_dt_map_level_int _row_int } - \dt_map_variables_aux:nnw {#1} + \@@_map_variables:nnw {#1} } % \end{macrocode} % \end{macro} @@ -1187,10 +1101,12 @@ % % \begin{macro}[rEXP]{\dt_map_break:} % \begin{macro}[rEXP]{\dt_map_break:n} -% The break statements are simply copies. +% The break statements use the general \cs{__prg_map_break:Nn}. % \begin{macrocode} -\cs_new_eq:NN \dt_map_break: \prg_map_break: -\cs_new_eq:NN \dt_map_break:n \prg_map_break:n +\cs_new_nopar:Npn \dt_map_break: + { \__prg_map_break:Nn \dt_map_break: { } } +\cs_new_nopar:Npn \dt_map_break:n + { \__prg_map_break:Nn \dt_map_break: } % \end{macrocode} % \end{macro} % \end{macro} @@ -1213,33 +1129,33 @@ % \end{macro} % % \begin{macro}[pTF, EXP]{\dt_if_in:Nn} -% \begin{macro}[aux, EXP]{\dt_if_in_aux:nnn} -% \begin{macro}[aux, EXP]{\dt_if_in_aux:nwN} -% \begin{macro}[aux, EXP]{\dt_if_in_aux:n} +% \begin{macro}[aux, EXP]{\@@_if_in:nnn} +% \begin{macro}[aux, EXP]{\@@_if_in:nwN} +% \begin{macro}[aux, EXP]{\@@_if_in:n} % Expandably checking for the presence of a key in the table as a whole % requires a mapping to the header. The idea is the usual recursion set % up with a string-based comparison only after checking for the end of % the loop. % \begin{macrocode} \prg_new_conditional:Npnn \dt_if_in:Nn #1#2 { p , T , F , TF } - { \dt_split_header:NT #1 { \dt_if_in_aux:nnn {#2} } } -\cs_new:Npn \dt_if_in_aux:nnn #1#2#3 + { \@@_split_header:NT #1 { \@@_if_in:nnn {#2} } } +\cs_new:Npn \@@_if_in:nnn #1#2#3 { - \exp_last_unbraced:Nno \dt_if_in_aux:nwN {#1} { \use_none:nn #2 } - \q_recursion_tail \q_dt - \prg_break_point:n { } + \exp_last_unbraced:Nno \@@_if_in:nwN {#1} { \use_none:nn #2 } + \q_recursion_tail \q_@@ + \__prg_break_point: } -\cs_new:Npn \dt_if_in_aux:nwN #1#2 \q_dt +\cs_new:Npn \@@_if_in:nwN #1#2 \q_@@ { \if_meaning:w \q_recursion_tail #2 - \exp_after:wN \prg_map_break:n + \exp_after:wN \__prg_break:n \else: \exp_after:wN \use_none:n \fi: { \prg_return_false: } \str_if_eq:nnTF {#1} {#2} - { \prg_map_break:n { \prg_return_true: } } - { \dt_if_in_aux:nwN {#1} } + { \__prg_break:n { \prg_return_true: } } + { \@@_if_in:nwN {#1} } } % \end{macrocode} % \end{macro} @@ -1248,22 +1164,22 @@ % \end{macro} % % \begin{macro}[pTF, EXP]{\dt_if_in_row:Nnn} -% \begin{macro}[aux, EXP]{\dt_if_in_row_aux:nw} -% \begin{macro}[aux, EXP]{\dt_if_in_row_aux:nn} -% \begin{macro}[aux, EXP]{\dt_if_in_row_aux:nwn} -% \begin{macro}[aux, EXP]{\dt_if_in_row_aux:N} +% \begin{macro}[aux, EXP]{\@@_if_in_row:nw} +% \begin{macro}[aux, EXP]{\@@_if_in_row:nn} +% \begin{macro}[aux, EXP]{\@@_if_in_row:nwn} +% \begin{macro}[aux, EXP]{\@@_if_in_row:N} % Finding a key in a single row in an expandable way requires two mappings. % To start of with, there is a search for the row. This uses for termination -% the fact that each row starts \cs{q_dt_row} and ends \cs{q_nil}, and +% the fact that each row starts \cs{q_@@_row} and ends \cs{q_nil}, and % always contains at least the row number as the first \meta{balanced % text}. That can be replaced by the tail marker to terminate iteration: % all that is then needed is the correct placement of the clean-up code. % \begin{macrocode} \prg_new_conditional:Npnn \dt_if_in_row:Nnn #1#2#3 { p , T , F , TF } { - \exp_last_unbraced:Nno \dt_if_in_row_aux:nw {#2} #1 + \exp_last_unbraced:Nno \@@_if_in_row:nw {#2} #1 \q_recursion_tail \q_nil - \prg_break_point:n { } + \__prg_break_point: { \tl_to_str:n {#3} } } % \end{macrocode} @@ -1271,10 +1187,10 @@ % been found. That means that the row argument does not need to be converted % to a number earlier. % \begin{macrocode} -\cs_new:Npn \dt_if_in_row_aux:nw #1#2 \q_dt_row #3#4 \q_nil +\cs_new:Npn \@@_if_in_row:nw #1#2 \q_@@_row #3#4 \q_nil { \if_meaning:w \q_recursion_tail #3 - \exp_after:wN \prg_map_break:n + \exp_after:wN \__prg_break:n \else: \exp_after:wN \use_none:n \fi: @@ -1283,32 +1199,32 @@ \prg_return_false: } \int_compare:nNnTF {#1} = {#3} - { \prg_map_break:n { \exp_args:Nno \dt_if_in_row_aux:nn {#4} } } - { \dt_if_in_row_aux:nw {#1} } + { \__prg_break:n { \exp_args:Nno \@@_if_in_row:nn {#4} } } + { \@@_if_in_row:nw {#1} } } % \end{macrocode} % The second iteration is along the row. This is basically the same as -% \cs{prop_if_in:NnTF} with the \cs{q_dt} in place of \cs{q_prop}. +% \cs{prop_if_in:NnTF} with the \cs{q_@@} in place of \cs{q_prop}. % \begin{macrocode} -\cs_new:Npn \dt_if_in_row_aux:nn #1#2 +\cs_new:Npn \@@_if_in_row:nn #1#2 { - \dt_if_in_row_aux:nwn {#2} #1 {#2} \q_dt { } \q_recursion_tail - \prg_break_point:n { } + \@@_if_in_row:nwn {#2} #1 {#2} \q_@@ { } \q_recursion_tail + \__prg_break_point: } -\cs_new:Npn \dt_if_in_row_aux:nwn #1 \q_dt #2 \q_dt #3 +\cs_new:Npn \@@_if_in_row:nwn #1 \q_@@ #2 \q_@@ #3 { \str_if_eq:xxTF {#1} {#2} - { \dt_if_in_row_aux:N } - { \dt_if_in_row_aux:nwn {#1} } + { \@@_if_in_row:N } + { \@@_if_in_row:nwn {#1} } } -\cs_new:Npn \dt_if_in_row_aux:N #1 +\cs_new:Npn \@@_if_in_row:N #1 { - \if_meaning:w \q_dt #1 + \if_meaning:w \q_@@ #1 \prg_return_true: \else: \prg_return_false: \fi: - \prg_map_break: + \__prg_break: } % \end{macrocode} % \end{macro} @@ -1334,7 +1250,7 @@ % \subsection{Messages} % % \begin{macrocode} -\msg_kernel_new:nnnn { dt } { unknown-row } +\__msg_kernel_new:nnnn { dt } { unknown-row } { Data~table~#1~does~not~contain~a~row~'#2'. } { Data~table~#1~contains~#3~rows.~These~must~be~accessed~by~number:~row~ diff --git a/Master/texmf-dist/source/latex/l3experimental/l3dt/l3dt.ins b/Master/texmf-dist/source/latex/l3experimental/l3dt/l3dt.ins index 5b84dfca3dc..54db0f43b92 100644 --- a/Master/texmf-dist/source/latex/l3experimental/l3dt/l3dt.ins +++ b/Master/texmf-dist/source/latex/l3experimental/l3dt/l3dt.ins @@ -16,7 +16,7 @@ The released version of this bundle is available from CTAN. \fi -\input docstrip.tex +\input l3docstrip.tex \askforoverwritefalse \preamble diff --git a/Master/texmf-dist/source/latex/l3experimental/l3sort/l3sort.dtx b/Master/texmf-dist/source/latex/l3experimental/l3sort/l3sort.dtx index 0d80aebcc51..4d795c7c9f5 100644 --- a/Master/texmf-dist/source/latex/l3experimental/l3sort/l3sort.dtx +++ b/Master/texmf-dist/source/latex/l3experimental/l3sort/l3sort.dtx @@ -36,7 +36,7 @@ % %<*driver|package> \RequirePackage{expl3} -\GetIdInfo$Id: l3sort.dtx 3467 2012-02-26 15:39:35Z bruno $ +\GetIdInfo$Id: l3sort.dtx 3953 2012-07-11 16:15:09Z bruno $ {L3 Experimental sorting functions} % %<*driver> @@ -71,7 +71,7 @@ % % \section{\pkg{l3sort} documentation} % -% \LaTeX3 comes with a function to sort list variables (sequences, +% \LaTeX3 comes with a facility to sort list variables (sequences, % token lists, or comma-lists) according to some user-defined % comparison. For instance, % \begin{verbatim} @@ -133,6 +133,18 @@ % \meta{clist var}. % \end{function} % +% \begin{function}[EXP]{\tl_sort:nN} +% \begin{syntax} +% \cs{tl_sort:nN} \Arg{token list} \meta{conditional} +% \end{syntax} +% Sorts the items in the \meta{token list}, using the +% \meta{conditional} to compare items, and leaves the result in the +% input stream. The \meta{conditional} should have signature |:nnTF|, +% and return \texttt{true} if the two items being compared should be +% left in the same order, and \texttt{false} if the items should be +% swapped. +% \end{function} +% % \end{documentation} % % \begin{implementation} @@ -144,16 +156,19 @@ % \end{macrocode} % % \begin{macrocode} +%<@@=sort> +% \end{macrocode} +% +% \begin{macrocode} %<*package> \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} -\package_check_loaded_expl: % % \end{macrocode} % % \subsection{Variables} % -% \begin{variable}{\c_sort_max_length_int} +% \begin{variable}{\c_@@_max_length_int} % The maximum length of a sequence which will not overflow % the available registers depends on which engine is in use. % For $2^{N}$ registers, it is $3\cdot 2^{N-2}$: for that number @@ -164,63 +179,155 @@ % of size $2^{N-2}$, will fill up exactly the $2^{N-2}$ free % registers, totalling $2^{N-1}+2^{N-2}+2^{N-2}=2^{N}$ registers. % \begin{macrocode} -\int_const:Nn \c_sort_max_length_int +\int_const:Nn \c_@@_max_length_int { \luatex_if_engine:TF { 49152 } { 24576 } } % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_sort_length_int} +% \begin{variable}{\l_@@_length_int} % Length of the sequence which is being sorted. % \begin{macrocode} -\int_new:N \l_sort_length_int +\int_new:N \l_@@_length_int % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_sort_block_int} +% \begin{variable}{\l_@@_block_int} % Merge sort is done in several passes. In each pass, blocks of size -% \cs{l_sort_block_int} are merged in pairs. The block size starts +% \cs{l_@@_block_int} are merged in pairs. The block size starts % at $1$, and, for a length in the range $[2^k+1,2^{k+1}]$, reaches % $2^{k}$ in the last pass. % \begin{macrocode} -\int_new:N \l_sort_block_int +\int_new:N \l_@@_block_int % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_sort_begin_int} -% \begin{variable}{\l_sort_end_int} -% When merging two blocks, \cs{l_sort_begin_int} marks the lowest -% index in the two blocks, and \cs{l_sort_end_int} marks the +% \begin{variable}{\l_@@_begin_int} +% \begin{variable}{\l_@@_end_int} +% When merging two blocks, \cs{l_@@_begin_int} marks the lowest +% index in the two blocks, and \cs{l_@@_end_int} marks the % highest index, plus $1$. % \begin{macrocode} -\int_new:N \l_sort_begin_int -\int_new:N \l_sort_end_int +\int_new:N \l_@@_begin_int +\int_new:N \l_@@_end_int % \end{macrocode} % \end{variable} % \end{variable} % -% \begin{variable}{\l_sort_A_int} -% \begin{variable}{\l_sort_B_int} -% \begin{variable}{\l_sort_C_int} +% \begin{variable}{\l_@@_A_int} +% \begin{variable}{\l_@@_B_int} +% \begin{variable}{\l_@@_C_int} % When merging two blocks (whose end-points are \texttt{beg} % and \texttt{end}), $A$ starts from the high end of the low % block, and decreases until reaching \texttt{beg}. The index % $B$ starts from the top of the range and marks the register % in which a sorted item should be put. Finally, $C$ points % to the copy of the high block in the interval of registers -% starting at \cs{l_sort_length_int}, upwards. $C$ starts +% starting at \cs{l_@@_length_int}, upwards. $C$ starts % from the upper limit of that range. % \begin{macrocode} -\int_new:N \l_sort_A_int -\int_new:N \l_sort_B_int -\int_new:N \l_sort_C_int +\int_new:N \l_@@_A_int +\int_new:N \l_@@_B_int +\int_new:N \l_@@_C_int % \end{macrocode} % \end{variable} % \end{variable} % \end{variable} % -% \subsection{User commands} +% \subsection{Expandable sorting} +% +% Sorting expandably is very different from sorting and assigning to a +% variable. Since tokens cannot be stored, they must remain in the +% input stream, and be read through at every step. It is thus +% necessarily much slower (at best $O(n^2)$) than non-expandable sorting +% functions. +% +% A prototypical version of expandable quicksort is as follows. If the +% argument has no item, return nothing, otherwise partition, using the +% first item as a pivot (argument |#4| of |\__sort:nnNnn|). The +% arguments of \cs{__sort:nnNnn} are 1.~items less than |#4|, 2.~items +% greater than |#4|, 3.~comparison, 4.~pivot, 5.~next item to test. If +% |#5| is the tail of the list, call \cs{tl_sort:nN} on |#1| and on +% |#2|, placing |#4| in between; |\use:ff| expands the parts to make +% \cs{tl_sort:nN} \texttt{f}-expandable. Otherwise, compare |#4| and +% |#5| using |#3|. If they are ordered, place |#5| amongst the +% \enquote{greater} items, otherwise amongst the \enquote{lesser} items, +% and continue partitioning. +% \begin{verbatim} +% \cs_new:Npn \tl_sort:nN #1#2 +% { +% \tl_if_blank:nF {#1} +% { +% \__sort:nnNnn { } { } #2 +% #1 \q_recursion_tail \q_recursion_stop +% } +% } +% \cs_new:Npn \__sort:nnNnn #1#2#3#4#5 +% { +% \quark_if_recursion_tail_stop_do:nn {#5} +% { \use:ff { \tl_sort:nN {#1} #3 {#4} } { \tl_sort:nN {#2} #3 } } +% #3 {#4} {#5} +% { \__sort:nnNnn {#1} { #2 {#5} } #3 {#4} } +% { \__sort:nnNnn { #1 {#5} } {#2} #3 {#4} } +% } +% \cs_generate_variant:Nn \use:nn { ff } +% \end{verbatim} +% There are quite a few optimizations available here: the code below is +% less legible, but twice as fast. +% +% \begin{macro}[EXP]{\tl_sort:nN} +% \begin{macro}[EXP, aux] +% { +% \@@_quick_i:nNn, \@@_quick_ii:wn, +% \@@_quick_iii:nnNnn, \@@_quick_iv:nnNnwn +% } +% The \cs{@@_quick_i:nNn} function sorts |#1| using the comparison +% |#2|, then places its third argument (\meta{continuation}) before +% the result. This essentially avoids the rather slow \cs{use:ff}. +% First test if |#1| is blank: if it is, \cs{use_none:n} eats the +% break point, and \cs{@@_quick_ii:wn} removes all until the ending +% break point, unbracing the \meta{continuation}. If it is not blank, +% \cs{@@_quick_ii:wn} simply removes the end of that test until the +% break point. The third auxiliary plays the role of |\__sort:nnNnn| +% above, with an end-test replaced by \cs{__prg_break:}, which skips +% to the break point two lines later, unless |#5| is itself a break +% point. In that case, \cs{@@_quick_iv:nnNnwn} calls the first +% auxiliary, which sorts |#2| and places another call to itself before +% the result. +% \begin{macrocode} +\cs_new:Npn \tl_sort:nN #1#2 { \@@_quick_i:nNn {#1} #2 { } } +\cs_new:Npn \@@_quick_i:nNn #1#2 + { + \exp_after:wN \@@_quick_ii:wn + \use_none:n #1 \__prg_break_point: + \@@_quick_iii:nnNnn { } { } #2 + #1 + { \__prg_break_point: } + \__prg_break_point: + } +\cs_new:Npn \@@_quick_ii:wn #1 \__prg_break_point: #2 {#2} +\cs_new:Npn \@@_quick_iii:nnNnn #1#2#3#4#5 + { + \__prg_break: #5 + \@@_quick_iv:nnNnwn {#1} {#2} + \__prg_break_point: + #3 {#4} {#5} + { \@@_quick_iii:nnNnn {#1} { #2 {#5} } } + { \@@_quick_iii:nnNnn { #1 {#5} } {#2} } + #3 {#4} + } +\cs_new:Npn \@@_quick_iv:nnNnwn + #1#2 \__prg_break_point: #3#4 #5 \__prg_break_point: #6 + { + \@@_quick_i:nNn {#2} #3 + { \@@_quick_i:nNn {#1} #3 {#6} {#4} } + } +% \end{macrocode} +% \end{macro} +% \end{macro} +% +% \subsection{Protected user commands} % -% \begin{macro}[int]{\sort_aux:NNnNn} +% \begin{macro}[int]{\@@_main:NNNnNn} % Sorting happens in three steps. First store items % in \tn{toks} registers ranging from $0$ to the length % of the list, while checking that the list is not too @@ -232,32 +339,32 @@ % into a variable of the right type, by \texttt{x}-expanding % the code in |#3|, specific to each type of list. % \begin{macrocode} -\cs_new_protected:Npn \sort_aux:NNnNn #1#2#3#4#5 +\cs_new_protected:Npn \@@_main:NNNnNn #1#2#3#4#5#6 { \group_begin: - \l_sort_length_int \c_zero - #2 #4 + \l_@@_length_int \c_zero + #2 #5 { - \if_num:w \l_sort_length_int = \c_sort_max_length_int - \sort_too_long_error:Nw #4 + \if_int_compare:w \l_@@_length_int = \c_@@_max_length_int + \@@_too_long_error:NNw #3 #5 \fi: - \tex_toks:D \l_sort_length_int {##1} - \tex_advance:D \l_sort_length_int \c_one + \tex_toks:D \l_@@_length_int {##1} + \tex_advance:D \l_@@_length_int \c_one } - \cs_set:Npn \sort_compare:nn ##1 ##2 { #5 } - \l_sort_block_int \c_one - \sort_level: + \cs_set:Npn \sort_compare:nn ##1 ##2 { #6 } + \l_@@_block_int \c_one + \@@_level: \use:x { \group_end: - #1 \exp_not:N #4 {#3} + #1 \exp_not:N #5 {#4} } } % \end{macrocode} % \end{macro} % % \begin{macro}{\seq_sort:Nn, \seq_gsort:Nn} -% The first argument to \cs{sort_aux:NNnNn} is the final +% The first argument to \cs{@@_main:NNNnNn} is the final % assignment function used, either \cs{tl_set:Nn} or % \cs{tl_gset:Nn} to control local versus global results. % The second argument is what mapping function is used @@ -269,15 +376,15 @@ % \begin{macrocode} \cs_new_protected_nopar:Npn \seq_sort:Nn { - \sort_aux:NNnNn \tl_set:Nn - \seq_map_inline:Nn - { \sort_toks:NNw \exp_not:N \seq_item:n 0 ; } + \@@_main:NNNnNn \tl_set:Nn + \seq_map_inline:Nn \seq_map_break: + { \@@_toks:NNw \exp_not:N \__seq_item:n 0 ; } } \cs_new_protected_nopar:Npn \seq_gsort:Nn { - \sort_aux:NNnNn \tl_gset:Nn - \seq_map_inline:Nn - { \sort_toks:NNw \exp_not:N \seq_item:n 0 ; } + \@@_main:NNNnNn \tl_gset:Nn + \seq_map_inline:Nn \seq_map_break: + { \@@_toks:NNw \exp_not:N \__seq_item:n 0 ; } } % \end{macrocode} % \end{macro} @@ -291,21 +398,21 @@ % \begin{macrocode} \cs_new_protected_nopar:Npn \tl_sort:Nn { - \sort_aux:NNnNn \tl_set:Nn - \tl_map_inline:Nn - { \sort_toks:NNw \prg_do_nothing: \prg_do_nothing: 0 ; } + \@@_main:NNNnNn \tl_set:Nn + \tl_map_inline:Nn \tl_map_break: + { \@@_toks:NNw \prg_do_nothing: \prg_do_nothing: 0 ; } } \cs_new_protected_nopar:Npn \tl_gsort:Nn { - \sort_aux:NNnNn \tl_gset:Nn - \tl_map_inline:Nn - { \sort_toks:NNw \prg_do_nothing: \prg_do_nothing: 0 ; } + \@@_main:NNNnNn \tl_gset:Nn + \tl_map_inline:Nn \tl_map_break: + { \@@_toks:NNw \prg_do_nothing: \prg_do_nothing: 0 ; } } % \end{macrocode} % \end{macro} % % \begin{macro}{\clist_sort:Nn, \clist_gsort:Nn} -% \begin{macro}[aux]{\clist_sort_aux:NNn} +% \begin{macro}[aux]{\@@_sort:NNn} % The case of empty comma-lists is a little bit special as usual, % and filtered out: there is nothing to sort in that case. % Otherwise, the input is done with \cs{clist_map_inline:Nn}, @@ -313,24 +420,24 @@ % for sequences and token lists. The first comma must be removed. % An item must be wrapped in an extra set of braces if it contains % either the space or the comma characters. This is taken care of -% by \cs{clist_wrap_item:n}, but \cs{sort_toks:NNw} would simply +% by \cs{clist_wrap_item:n}, but \cs{@@_toks:NNw} would simply % feed \cs{tex_the:D} \cs{tex_toks:D} \meta{number} as an % argument to that function; hence we need to expand this argument % once to unpack the register. % \begin{macrocode} \cs_new_protected_nopar:Npn \clist_sort:Nn - { \clist_sort_aux:NNn \tl_set:Nn } + { \@@_sort:NNn \tl_set:Nn } \cs_new_protected_nopar:Npn \clist_gsort:Nn - { \clist_sort_aux:NNn \tl_gset:Nn } -\cs_new_protected:Npn \clist_sort_aux:NNn #1#2#3 + { \@@_sort:NNn \tl_gset:Nn } +\cs_new_protected:Npn \@@_sort:NNn #1#2#3 { \clist_if_empty:NF #2 { - \sort_aux:NNnNn #1 - \clist_map_inline:Nn + \@@_main:NNNnNn #1 + \clist_map_inline:Nn \clist_map_break: { \exp_last_unbraced:Nf \use_none:n - { \sort_toks:NNw \exp_args:No \clist_wrap_item:n 0 ; } + { \@@_toks:NNw \exp_args:No \__clist_wrap_item:n 0 ; } } #2 {#3} } @@ -339,13 +446,13 @@ % \end{macro} % \end{macro} % -% \begin{macro}{\sort_toks:NNw} +% \begin{macro}{\@@_toks:NNw} % Unpack the various \tn{toks} registers, from $0$ to the length % of the list. The functions |#1| and |#2| allow us to treat the % three data structures in a unified way: % \begin{itemize} -% \item for sequences, they are \cs{exp_not:N} \cs{seq_item:n}, -% expanding to the \cs{seq_item:n} separator, as expected; +% \item for sequences, they are \cs{exp_not:N} \cs{__seq_item:n}, +% expanding to the \cs{__seq_item:n} separator, as expected; % \item for token lists, they expand to nothing; % \item for comma lists, they expand to \cs{exp_args:No} % \cs{clist_wrap_item:n}, taking care of unpacking the register @@ -354,12 +461,12 @@ % braces. % \end{itemize} % \begin{macrocode} -\cs_new:Npn \sort_toks:NNw #1#2#3 ; +\cs_new:Npn \@@_toks:NNw #1#2#3 ; { - \if_num:w #3 < \l_sort_length_int + \if_int_compare:w #3 < \l_@@_length_int #1 #2 { \tex_the:D \tex_toks:D #3 } - \exp_after:wN \sort_toks:NNw \exp_after:wN #1 \exp_after:wN #2 - \int_use:N \int_eval:w #3 + \c_one \exp_after:wN ; + \exp_after:wN \@@_toks:NNw \exp_after:wN #1 \exp_after:wN #2 + \int_use:N \__int_eval:w #3 + \c_one \exp_after:wN ; \fi: } % \end{macrocode} @@ -367,28 +474,28 @@ % % \subsection{Sorting itself} % -% \begin{macro}[int]{\sort_level:} -% This function is called once blocks of size \cs{l_sort_block_int} +% \begin{macro}[int]{\@@_level:} +% This function is called once blocks of size \cs{l_@@_block_int} % (initially $1$) are each sorted. If the whole list fits in one % block, then we are done (this also takes care of the case of an % empty list or a list with one item). Otherwise, go through pairs % of blocks starting from $0$, then double the block size, and repeat. % \begin{macrocode} -\cs_new_protected_nopar:Npn \sort_level: +\cs_new_protected_nopar:Npn \@@_level: { - \if_num:w \l_sort_block_int < \l_sort_length_int - \l_sort_end_int \c_zero - \sort_merge_blocks: - \tex_multiply:D \l_sort_block_int \c_two - \exp_after:wN \sort_level: + \if_int_compare:w \l_@@_block_int < \l_@@_length_int + \l_@@_end_int \c_zero + \@@_merge_blocks: + \tex_multiply:D \l_@@_block_int \c_two + \exp_after:wN \@@_level: \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\sort_merge_blocks:} +% \begin{macro}[int]{\@@_merge_blocks:} % This function is called to merge a pair of blocks, starting at -% the last value of \cs{l_sort_end_int} (end-point of the previous +% the last value of \cs{l_@@_end_int} (end-point of the previous % pair of blocks). If shifting by one block to the right we reach % the end of the list, then this pass has ended: this end of the % list is sorted already. Store the result of that shift in $A$, @@ -398,30 +505,30 @@ % go beyond the length of the list. Copy this upper block of \tn{toks} % registers in registers above \texttt{length}, indexed by $C$: % this is covered by \cs{sort_copy_block:}. Once this is done we -% are ready to do the actual merger using \cs{sort_merge_blocks_aux:}, +% are ready to do the actual merger using \cs{@@_merge_blocks_aux:}, % after shifting $A$, $B$ and $C$ so that they point to the largest % index in their respective ranges rather than pointing just beyond % those ranges. Of course, once that pair of blocks is merged, % move on to the next pair. % \begin{macrocode} -\cs_new_protected_nopar:Npn \sort_merge_blocks: +\cs_new_protected_nopar:Npn \@@_merge_blocks: { - \l_sort_begin_int \l_sort_end_int - \tex_advance:D \l_sort_end_int \l_sort_block_int - \if_num:w \int_eval:w \l_sort_end_int < \l_sort_length_int - \l_sort_A_int \l_sort_end_int - \tex_advance:D \l_sort_end_int \l_sort_block_int - \if_num:w \l_sort_end_int > \l_sort_length_int - \l_sort_end_int \l_sort_length_int + \l_@@_begin_int \l_@@_end_int + \tex_advance:D \l_@@_end_int \l_@@_block_int + \if_int_compare:w \__int_eval:w \l_@@_end_int < \l_@@_length_int + \l_@@_A_int \l_@@_end_int + \tex_advance:D \l_@@_end_int \l_@@_block_int + \if_int_compare:w \l_@@_end_int > \l_@@_length_int + \l_@@_end_int \l_@@_length_int \fi: - \l_sort_B_int \l_sort_A_int - \l_sort_C_int \l_sort_length_int + \l_@@_B_int \l_@@_A_int + \l_@@_C_int \l_@@_length_int \sort_copy_block: - \tex_advance:D \l_sort_A_int \c_minus_one - \tex_advance:D \l_sort_B_int \c_minus_one - \tex_advance:D \l_sort_C_int \c_minus_one - \sort_merge_blocks_aux: - \exp_after:wN \sort_merge_blocks: + \tex_advance:D \l_@@_A_int \c_minus_one + \tex_advance:D \l_@@_B_int \c_minus_one + \tex_advance:D \l_@@_C_int \c_minus_one + \@@_merge_blocks_aux: + \exp_after:wN \@@_merge_blocks: \fi: } % \end{macrocode} @@ -430,16 +537,16 @@ % \begin{macro}[int]{\sort_copy_block:} % We wish to store a copy of the \enquote{upper} block of % \tn{toks} registers, ranging between the initial value of -% \cs{l_sort_B_int} (included) and \cs{l_sort_end_int} +% \cs{l_@@_B_int} (included) and \cs{l_@@_end_int} % (excluded) into a new range starting at the initial value -% of \cs{l_sort_C_int}, namely \cs{l_sort_length_int}. +% of \cs{l_@@_C_int}, namely \cs{l_@@_length_int}. % \begin{macrocode} \cs_new_protected_nopar:Npn \sort_copy_block: { - \tex_toks:D \l_sort_C_int \tex_toks:D \l_sort_B_int - \tex_advance:D \l_sort_C_int \c_one - \tex_advance:D \l_sort_B_int \c_one - \if_num:w \l_sort_B_int = \l_sort_end_int + \tex_toks:D \l_@@_C_int \tex_toks:D \l_@@_B_int + \tex_advance:D \l_@@_C_int \c_one + \tex_advance:D \l_@@_B_int \c_one + \if_int_compare:w \l_@@_B_int = \l_@@_end_int \use_i:nn \fi: \sort_copy_block: @@ -447,13 +554,13 @@ % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\sort_merge_blocks_aux:} -% At this stage, the first block starts at \cs{l_sort_begin_int}, -% and ends at \cs{l_sort_A_int}, and the second block starts at -% \cs{l_sort_length_int} and ends at \cs{l_sort_C_int}. The result -% of the merger is stored at positions indexed by \cs{l_sort_B_int}, -% which starts at $\cs{l_sort_end_int}-1$ and decreases down to -% \cs{l_sort_begin_int}, covering the full range of the two blocks. +% \begin{macro}[aux]{\@@_merge_blocks_aux:} +% At this stage, the first block starts at \cs{l_@@_begin_int}, +% and ends at \cs{l_@@_A_int}, and the second block starts at +% \cs{l_@@_length_int} and ends at \cs{l_@@_C_int}. The result +% of the merger is stored at positions indexed by \cs{l_@@_B_int}, +% which starts at $\cs{l_@@_end_int}-1$ and decreases down to +% \cs{l_@@_begin_int}, covering the full range of the two blocks. % In other words, we are building the merger starting with the % largest values. % The comparison function is defined to return either @@ -461,11 +568,11 @@ % means the arguments need to be given in the order they % appear originally in the list. % \begin{macrocode} -\cs_new_protected_nopar:Npn \sort_merge_blocks_aux: +\cs_new_protected_nopar:Npn \@@_merge_blocks_aux: { \exp_after:wN \sort_compare:nn \exp_after:wN - { \tex_the:D \tex_toks:D \exp_after:wN \l_sort_A_int \exp_after:wN } - \exp_after:wN { \tex_the:D \tex_toks:D \l_sort_C_int } + { \tex_the:D \tex_toks:D \exp_after:wN \l_@@_A_int \exp_after:wN } + \exp_after:wN { \tex_the:D \tex_toks:D \l_@@_C_int } } % \end{macrocode} % \end{macro} @@ -484,13 +591,13 @@ % \begin{macrocode} \cs_new_protected_nopar:Npn \sort_ordered: { - \tex_toks:D \l_sort_B_int \tex_toks:D \l_sort_C_int - \tex_advance:D \l_sort_B_int \c_minus_one - \tex_advance:D \l_sort_C_int \c_minus_one - \if_num:w \l_sort_C_int < \l_sort_length_int + \tex_toks:D \l_@@_B_int \tex_toks:D \l_@@_C_int + \tex_advance:D \l_@@_B_int \c_minus_one + \tex_advance:D \l_@@_C_int \c_minus_one + \if_int_compare:w \l_@@_C_int < \l_@@_length_int \use_i:nn \fi: - \sort_merge_blocks_aux: + \@@_merge_blocks_aux: } % \end{macrocode} % \end{macro} @@ -504,60 +611,59 @@ % block was exhausted ($A$ goes below the threshold). % In that case, all remaining \tn{toks} registers in % the second block, indexed by $C$, should be copied -% to the merger (see \cs{sort_merge_blocks_end:}). +% to the merger (see \cs{@@_merge_blocks_end:}). % \begin{macrocode} \cs_new_protected_nopar:Npn \sort_reversed: { - \tex_toks:D \l_sort_B_int \tex_toks:D \l_sort_A_int - \tex_advance:D \l_sort_B_int \c_minus_one - \tex_advance:D \l_sort_A_int \c_minus_one - \if_num:w \l_sort_A_int < \l_sort_begin_int - \sort_merge_blocks_end: \use_i:nn + \tex_toks:D \l_@@_B_int \tex_toks:D \l_@@_A_int + \tex_advance:D \l_@@_B_int \c_minus_one + \tex_advance:D \l_@@_A_int \c_minus_one + \if_int_compare:w \l_@@_A_int < \l_@@_begin_int + \@@_merge_blocks_end: \use_i:nn \fi: - \sort_merge_blocks_aux: + \@@_merge_blocks_aux: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\sort_merge_blocks_end:} +% \begin{macro}[aux]{\@@_merge_blocks_end:} % This function's task is to copy the \tn{toks} registers % in the block indexed by $C$ to the merger indexed by $B$. % The end can equally be detected by checking when $B$ reaches % the threshold \texttt{begin}, or when $C$ reaches % \texttt{length}. % \begin{macrocode} -\cs_new_protected_nopar:Npn \sort_merge_blocks_end: +\cs_new_protected_nopar:Npn \@@_merge_blocks_end: { - \tex_toks:D \l_sort_B_int \tex_toks:D \l_sort_C_int - \tex_advance:D \l_sort_B_int \c_minus_one - \tex_advance:D \l_sort_C_int \c_minus_one - \if_num:w \l_sort_B_int < \l_sort_begin_int + \tex_toks:D \l_@@_B_int \tex_toks:D \l_@@_C_int + \tex_advance:D \l_@@_B_int \c_minus_one + \tex_advance:D \l_@@_C_int \c_minus_one + \if_int_compare:w \l_@@_B_int < \l_@@_begin_int \use_i:nn \fi: - \sort_merge_blocks_end: + \@@_merge_blocks_end: } % \end{macrocode} % \end{macro} % % \subsection{Messages} % -% \begin{macro}{\sort_too_long_error:Nw} -% When there are too many items in a sequence, this is an error, -% and we clean up properly the mapping over items in the list. -% This relies on the fact that the breaking code for all mappings -% is unified. +% \begin{macro}{\@@_too_long_error:NNw} +% When there are too many items in a sequence, this is an error, and +% we clean up properly the mapping over items in the list: break using +% the type-specific breaking function |#1|. % \begin{macrocode} -\cs_new_protected:Npn \sort_too_long_error:Nw #1 \fi: +\cs_new_protected:Npn \@@_too_long_error:NNw #1#2 \fi: { \fi: - \msg_kernel_error:nnx { sort } { too-large } { \token_to_str:N #1 } - \prg_map_break: + \__msg_kernel_error:nnx { sort } { too-large } { \token_to_str:N #2 } + #1 } -\msg_kernel_new:nnnn { sort } { too-large } +\__msg_kernel_new:nnnn { sort } { too-large } { The~list~#1~is~too~long~to~be~sorted~by~TeX. } { - TeX~has~\int_use:N \c_max_register_int \ registers~available:~ - this~only~allows~to~sorts~with~up~to~\int_use:N \c_sort_max_length_int + TeX~has~\int_eval:n { \c_max_register_int + 1 }~registers~available:~ + this~only~allows~to~sorts~with~up~to~\int_use:N \c_@@_max_length_int \ items.~All~extra~items~will~be~ignored. } % \end{macrocode} diff --git a/Master/texmf-dist/source/latex/l3experimental/l3sort/l3sort.ins b/Master/texmf-dist/source/latex/l3experimental/l3sort/l3sort.ins index 0cc04d132fe..db6f7f2b5d0 100644 --- a/Master/texmf-dist/source/latex/l3experimental/l3sort/l3sort.ins +++ b/Master/texmf-dist/source/latex/l3experimental/l3sort/l3sort.ins @@ -16,7 +16,7 @@ The released version of this bundle is available from CTAN. \fi -\input docstrip.tex +\input l3docstrip.tex \askforoverwritefalse \preamble diff --git a/Master/texmf-dist/source/latex/l3experimental/l3str/l3flag.dtx b/Master/texmf-dist/source/latex/l3experimental/l3str/l3flag.dtx index 717065d7d27..d3b17183671 100644 --- a/Master/texmf-dist/source/latex/l3experimental/l3str/l3flag.dtx +++ b/Master/texmf-dist/source/latex/l3experimental/l3str/l3flag.dtx @@ -36,7 +36,7 @@ %<*driver|package> \RequirePackage{expl3} \GetIdInfo$Id: l3flag.dtx 3039 2011-12-08 09:22:35Z bruno $ - {L3 Experimental Flags} + {L3 Experimental flags} % %<*driver> \documentclass[full]{l3doc} @@ -175,6 +175,10 @@ % \end{macrocode} % % \begin{macrocode} +%<@@=flag> +% \end{macrocode} +% +% \begin{macrocode} \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} % \end{macrocode} @@ -187,28 +191,28 @@ % \begin{macrocode} \cs_new_protected:Npn \flag_new:n #1 { - \cs_new:cpn { flag_trap_#1:w } ##1 ; - { \exp_after:wN \use_none:n \cs:w l_#1_##1_flag \cs_end: } + \cs_new:cpn { @@_trap_#1:w } ##1 ; + { \exp_after:wN \use_none:n \cs:w @@_#1_##1: \cs_end: } } % \end{macrocode} % \end{macro} % % \begin{macro}{\flag_clear:n} -% \begin{macro}[aux]{\flag_clear_aux:ww} +% \begin{macro}[aux]{\@@_clear:ww} % Undefine control sequences, starting from the |_0| flag, upwards, % until reaching an undefined control sequence. % \begin{macrocode} \cs_new_protected:Npn \flag_clear:n #1 - { \flag_clear_aux:ww 0 ; #1 \q_stop } -\cs_new_protected:Npn \flag_clear_aux:ww #1 ; #2 \q_stop + { \@@_clear:ww 0 ; #1 \q_stop } +\cs_new_protected:Npn \@@_clear:ww #1 ; #2 \q_stop { - \if_cs_exist:w l_#2_#1_flag \cs_end: + \if_cs_exist:w @@_#2_#1: \cs_end: \else: \exp_after:wN \use_none_delimit_by_q_stop:w \fi: - \cs_set_eq:cN { l_#2_#1_flag } \c_undefined:D - \exp_after:wN \flag_clear_aux:ww - \int_use:N \int_eval:w \c_one + #1 ; + \cs_set_eq:cN { @@_#2_#1: } \c_undefined:D + \exp_after:wN \@@_clear:ww + \int_use:N \__int_eval:w \c_one + #1 ; #2 \q_stop } % \end{macrocode} @@ -216,7 +220,8 @@ % \end{macro} % % \begin{macro}{\flag_clear_new:n} -% A flag exist if \cs{flag_trap_\meta{flag name}:n} exists. +% As for other datatypes, clear the \meta{flag} or create a new one, +% as appropriate. % \begin{macrocode} \cs_new_protected:Npn \flag_clear_new:n #1 { \flag_if_exist:nTF {#1} { \flag_clear:n } { \flag_new:n } {#1} } @@ -224,22 +229,23 @@ % \end{macro} % % \begin{macro}{\flag_set_trap:nn} -% Should that \pkg{l3flag} function check whether the flag exists? +% ^^A todo: check that the flag exists. +% Redefine the trap. % \begin{macrocode} \cs_new_protected:Npn \flag_set_trap:nn #1#2 - { \cs_set:cpn { flag_trap_#1:w } ##1 ; {#2} } + { \cs_set:cpn { @@_trap_#1:w } ##1 ; {#2} } % \end{macrocode} % \end{macro} % % \subsection{Expandable flag commands} % % \begin{macro}[EXP, pTF]{\flag_if_exist:n} -% A \meta{flag} is defined if the corresponding \enquote{trap} is -% defined. +% A flag exist if the corresponding trap \cs{@@_trap_\meta{flag +% name}:n} is defined. % \begin{macrocode} \prg_new_conditional:Npnn \flag_if_exist:n #1 { p , T , F , TF } { - \cs_if_exist:cTF { flag_trap_#1:w } + \cs_if_exist:cTF { @@_trap_#1:w } { \prg_return_true: } { \prg_return_false: } } % \end{macrocode} @@ -250,7 +256,7 @@ % \begin{macrocode} \prg_new_conditional:Npnn \flag_if_raised:n #1 { p , T , F , TF } { - \if_cs_exist:w l_#1_0_flag \cs_end: + \if_cs_exist:w @@_#1_0: \cs_end: \prg_return_true: \else: \prg_return_false: @@ -260,21 +266,21 @@ % \end{macro} % % \begin{macro}[EXP]{\flag_height:n} -% \begin{macro}[EXP, aux]{\flag_height_loop:ww, \flag_height_end:ww} +% \begin{macro}[EXP, aux]{\@@_height_loop:ww, \@@_height_end:ww} % Extract the value of the flag by going through all of the % |_|\meta{integer} control sequences starting from $0$. % \begin{macrocode} -\cs_new:Npn \flag_height:n #1 { \flag_height_loop:ww 0; #1 \q_stop } -\cs_new:Npn \flag_height_loop:ww #1 ; #2 \q_stop +\cs_new:Npn \flag_height:n #1 { \@@_height_loop:ww 0; #1 \q_stop } +\cs_new:Npn \@@_height_loop:ww #1 ; #2 \q_stop { - \if_cs_exist:w l_#2_#1_flag \cs_end: - \exp_after:wN \flag_height_loop:ww \int_use:N \int_eval:w \c_one + + \if_cs_exist:w @@_#2_#1: \cs_end: + \exp_after:wN \@@_height_loop:ww \int_use:N \__int_eval:w \c_one + \else: - \exp_after:wN \flag_height_end:ww + \exp_after:wN \@@_height_end:ww \fi: #1 ; #2 \q_stop } -\cs_new:Npn \flag_height_end:ww #1 ; #2 \q_stop { #1 } +\cs_new:Npn \@@_height_end:ww #1 ; #2 \q_stop { #1 } % \end{macrocode} % \end{macro} % \end{macro} @@ -284,8 +290,8 @@ % \begin{macrocode} \cs_new:Npn \flag_raise:n #1 { - \cs:w flag_trap_#1:w \exp_after:wN \cs_end: - \int_value:w \flag_height:n {#1} ; + \cs:w @@_trap_#1:w \exp_after:wN \cs_end: + \__int_value:w \flag_height:n {#1} ; } % \end{macrocode} % \end{macro} diff --git a/Master/texmf-dist/source/latex/l3experimental/l3str/l3regex.dtx b/Master/texmf-dist/source/latex/l3experimental/l3str/l3regex.dtx index bb290aaa276..498036bdaa7 100644 --- a/Master/texmf-dist/source/latex/l3experimental/l3str/l3regex.dtx +++ b/Master/texmf-dist/source/latex/l3experimental/l3str/l3regex.dtx @@ -35,8 +35,8 @@ % %<*driver|package> \RequirePackage{expl3} -\GetIdInfo$Id: l3regex.dtx 3488 2012-03-03 19:49:03Z bruno $ - {L3 Experimental Regular Expressions} +\GetIdInfo$Id: l3regex.dtx 3940 2012-07-09 00:41:44Z bruno $ + {L3 Experimental regular expressions} % %<*driver> \documentclass[full]{l3doc} @@ -592,7 +592,7 @@ % \item Detect attempts to use back-references. % \item Test for the maximum register \cs{c_max_register_int}. % \item Find out whether the fact that |\W| and friends match the -% end-marker leads to bugs. Possibly update \cs{regex_item_reverse:n}. +% end-marker leads to bugs. Possibly update \cs{__regex_item_reverse:n}. % \item Enforce that |\cC| can only be followed by a match-all dot. % \end{itemize} % @@ -605,7 +605,7 @@ % \item Only build \c{...} once. % \item Use \tn{skip} for the left and right state stacks when % compiling a regex. -% \item Should \cs{regex_action_free_group:n} only be used for greedy +% \item Should \cs{__regex_action_free_group:n} only be used for greedy % |{n,}| quantifier? (I think not.) % \item Quantifiers for |\u| and assertions. % \item Improve digit grabbing for the |\g| escape in replacement. @@ -614,16 +614,16 @@ % \texttt{current_state} and \texttt{current_submatches}. % \item If possible, when a state is reused by the same thread, kill % other subthreads. -% \item Use \tn{dimen} registers rather than \cs{l_regex_balance_tl} -% to build \cs{regex_replacement_balance_one_match:n}. +% \item Use \tn{dimen} registers rather than \cs{l__regex_balance_tl} +% to build \cs{__regex_replacement_balance_one_match:n}. % \item Reduce the number of epsilon-transitions in alternatives. % \item Optimize simple strings: use less states (|abcade| should give % two states, for |abc| and |ade|). [Does that really make sense?] % \item Optimize groups with no alternative. -% \item Optimize states with a single \cs{regex_action_free:n}. -% \item Optimize the use of \cs{regex_action_success:} by inserting it +% \item Optimize states with a single \cs{__regex_action_free:n}. +% \item Optimize the use of \cs{__regex_action_success:} by inserting it % in state $2$ directly instead of having an extra transition. -% \item Optimize the use of \cs{prg_stepwise_...} functions. +% \item Optimize the use of \cs{int_step_...} functions. % \item Groups don't capture within regexes for csnames; optimize and % document. % \item Decide and document what |\c{\c{...}}| should do in the @@ -700,11 +700,20 @@ % % \section{\pkg{l3regex} implementation} % -%<*package> % \begin{macrocode} +%<*initex|package> +% \end{macrocode} +% +% \begin{macrocode} +%<@@=regex> +% \end{macrocode} +% +% \begin{macrocode} +%<*package> \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} \RequirePackage{l3tl-build, l3tl-analysis, l3flag, l3str} +% % \end{macrocode} % % \subsection{Plan of attack} @@ -750,7 +759,7 @@ % are ordered according to the greediness of quantifiers. % \item \emph{Step}: used when matching, starts at $0$, incremented % every time a character is read, and is not reset when searching -% for repeated matches. The integer \cs{l_regex_step_int} is a +% for repeated matches. The integer \cs{l_@@_step_int} is a % unique id for all the steps of the matching algorithm. % \end{itemize} % @@ -791,7 +800,7 @@ % submatches as would be extracted by the \cs{regex_extract} % functions. Since smaller \tn{skip} registers are used, the minimum % index is twice \texttt{max_state}, and the used registers go up to -% \cs{l_regex_submatch_int}. They are organized in blocks of +% \cs{l_@@_submatch_int}. They are organized in blocks of % \texttt{capturing_group}, each block corresponding to one match % with all its submatches stored in consecutive \tn{skip}s. % \end{itemize} @@ -820,114 +829,114 @@ % % \subsubsection{Constants and variables} % -% \begin{macro}[aux]{\regex_tmp:w} +% \begin{macro}[aux]{\@@_tmp:w} % Temporary function used for various short-term purposes. % \begin{macrocode} -\cs_new:Npn \regex_tmp:w { } +\cs_new:Npn \@@_tmp:w { } % \end{macrocode} % \end{macro} % % \begin{variable} % { -% \l_regex_internal_a_tl, \l_regex_internal_b_tl, -% \l_regex_internal_a_int, \l_regex_internal_b_int, -% \l_regex_internal_c_int, \l_regex_internal_bool, -% \l_regex_internal_seq, \g_regex_internal_tl, +% \l_@@_internal_a_tl, \l_@@_internal_b_tl, +% \l_@@_internal_a_int, \l_@@_internal_b_int, +% \l_@@_internal_c_int, \l_@@_internal_bool, +% \l_@@_internal_seq, \g_@@_internal_tl, % } % Temporary variables used for various purposes. % \begin{macrocode} -\tl_new:N \l_regex_internal_a_tl -\tl_new:N \l_regex_internal_b_tl -\int_new:N \l_regex_internal_a_int -\int_new:N \l_regex_internal_b_int -\int_new:N \l_regex_internal_c_int -\bool_new:N \l_regex_internal_bool -\seq_new:N \l_regex_internal_seq -\tl_new:N \g_regex_internal_tl +\tl_new:N \l_@@_internal_a_tl +\tl_new:N \l_@@_internal_b_tl +\int_new:N \l_@@_internal_a_int +\int_new:N \l_@@_internal_b_int +\int_new:N \l_@@_internal_c_int +\bool_new:N \l_@@_internal_bool +\seq_new:N \l_@@_internal_seq +\tl_new:N \g_@@_internal_tl % \end{macrocode} % \end{variable} % -% \begin{variable}{\c_regex_no_match_regex} +% \begin{variable}{\c_@@_no_match_regex} % This regular expression matches nothing, but is still a valid % regular expression. We could use a failing assertion, but I went for % an empty class. It is used as the initial value for regular % expressions declared using \cs{regex_new:N}. % \begin{macrocode} -\tl_const:Nn \c_regex_no_match_regex +\tl_const:Nn \c_@@_no_match_regex { - \regex_branch:n - { \regex_class:NnnnN \c_true_bool { } { 1 } { 0 } \c_true_bool } + \@@_branch:n + { \@@_class:NnnnN \c_true_bool { } { 1 } { 0 } \c_true_bool } } % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_regex_balance_int} +% \begin{variable}{\l_@@_balance_int} % The first thing we do when matching is to go once through the query % token list and store the information for each token as \tn{muskip} -% and \tn{toks} registers. During this phase, \cs{l_regex_balance_int} +% and \tn{toks} registers. During this phase, \cs{l_@@_balance_int} % counts the balance of begin-group and end-group character tokens % which appear before a given point in the token list, and we store it % as the shrink component of each \tn{muskip} register. This variable % is also used to keep track of the balance in the replacement text. % \begin{macrocode} -\int_new:N \l_regex_balance_int +\int_new:N \l_@@_balance_int % \end{macrocode} % \end{variable} % % \subsubsection{Testing characters} % -% \begin{macro}[int]{\regex_break_point:TF} -% \begin{macro}[int]{\regex_break_true:w} +% \begin{macro}[int]{\@@_break_point:TF} +% \begin{macro}[int]{\@@_break_true:w} % When testing whether a character of the query token list matches % a given character class in the regular expression, we often % have to test it against several ranges of characters, checking % if any one of those matches. This is done with a structure like % \begin{quote} % \meta{test1} \ldots{} \meta{test$\sb{n}$} \\ -% \cs{regex_break_point:TF} \Arg{true code} \Arg{false code} +% \cs{@@_break_point:TF} \Arg{true code} \Arg{false code} % \end{quote} -% If any of the tests succeeds, it calls \cs{regex_break_true:w}, +% If any of the tests succeeds, it calls \cs{@@_break_true:w}, % which cleans up and leaves \meta{true code} in the input stream. -% Otherwise, \cs{regex_break_point:TF} leaves the \meta{false code} +% Otherwise, \cs{@@_break_point:TF} leaves the \meta{false code} % in the input stream. % \begin{macrocode} -\cs_new_protected:Npn \regex_break_true:w - #1 \regex_break_point:TF #2 #3 {#2} -\cs_new_protected:Npn \regex_break_point:TF #1 #2 { #2 } +\cs_new_protected:Npn \@@_break_true:w + #1 \@@_break_point:TF #2 #3 {#2} +\cs_new_protected:Npn \@@_break_point:TF #1 #2 { #2 } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\regex_item_reverse:n} +% \begin{macro}[int]{\@@_item_reverse:n} % This function makes showing regular expressions easier, and lets us % define |\D| in terms of |\d| for instance. There is a subtelty: the % end of the query is marked by $-2$, and will thus match |\D| and % other negated properties; this case is catched by another part of % the code. % \begin{macrocode} -\cs_new_protected:Npn \regex_item_reverse:n #1 +\cs_new_protected:Npn \@@_item_reverse:n #1 { #1 - \regex_break_point:TF { } \regex_break_true:w + \@@_break_point:TF { } \@@_break_true:w } % \end{macrocode} % \end{macro} % % \begin{macro}[int] -% {\regex_item_caseful_equal:n, \regex_item_caseful_range:nn} -% Simple comparisons triggering \cs{regex_break_true:w} when true. +% {\@@_item_caseful_equal:n, \@@_item_caseful_range:nn} +% Simple comparisons triggering \cs{@@_break_true:w} when true. % \begin{macrocode} -\cs_new_protected:Npn \regex_item_caseful_equal:n #1 +\cs_new_protected:Npn \@@_item_caseful_equal:n #1 { - \if_num:w #1 = \l_regex_current_char_int - \exp_after:wN \regex_break_true:w + \if_int_compare:w #1 = \l_@@_current_char_int + \exp_after:wN \@@_break_true:w \fi: } -\cs_new_protected:Npn \regex_item_caseful_range:nn #1 #2 +\cs_new_protected:Npn \@@_item_caseful_range:nn #1 #2 { - \reverse_if:N \if_num:w #1 > \l_regex_current_char_int - \reverse_if:N \if_num:w #2 < \l_regex_current_char_int - \exp_after:wN \exp_after:wN \exp_after:wN \regex_break_true:w + \reverse_if:N \if_int_compare:w #1 > \l_@@_current_char_int + \reverse_if:N \if_int_compare:w #2 < \l_@@_current_char_int + \exp_after:wN \exp_after:wN \exp_after:wN \@@_break_true:w \fi: \fi: } @@ -935,63 +944,63 @@ % \end{macro} % % \begin{macro}[int] -% {\regex_item_caseless_equal:n, \regex_item_caseless_range:nn} +% {\@@_item_caseless_equal:n, \@@_item_caseless_range:nn} % For caseless matching, we perform the test both on the % \texttt{current_char} and on the \texttt{case_changed_char}. Before % doing the second set of tests, we make sure that % \texttt{case_changed_char} has been computed. % \begin{macrocode} -\cs_new_protected:Npn \regex_item_caseless_equal:n #1 +\cs_new_protected:Npn \@@_item_caseless_equal:n #1 { - \if_num:w #1 = \l_regex_current_char_int - \exp_after:wN \regex_break_true:w + \if_int_compare:w #1 = \l_@@_current_char_int + \exp_after:wN \@@_break_true:w \fi: - \if_num:w \l_regex_case_changed_char_int = \c_max_int - \regex_compute_case_changed_char: + \if_int_compare:w \l_@@_case_changed_char_int = \c_max_int + \@@_compute_case_changed_char: \fi: - \if_num:w #1 = \l_regex_case_changed_char_int - \exp_after:wN \regex_break_true:w + \if_int_compare:w #1 = \l_@@_case_changed_char_int + \exp_after:wN \@@_break_true:w \fi: } -\cs_new_protected:Npn \regex_item_caseless_range:nn #1 #2 +\cs_new_protected:Npn \@@_item_caseless_range:nn #1 #2 { - \reverse_if:N \if_num:w #1 > \l_regex_current_char_int - \reverse_if:N \if_num:w #2 < \l_regex_current_char_int - \exp_after:wN \exp_after:wN \exp_after:wN \regex_break_true:w + \reverse_if:N \if_int_compare:w #1 > \l_@@_current_char_int + \reverse_if:N \if_int_compare:w #2 < \l_@@_current_char_int + \exp_after:wN \exp_after:wN \exp_after:wN \@@_break_true:w \fi: \fi: - \if_num:w \l_regex_case_changed_char_int = \c_max_int - \regex_compute_case_changed_char: + \if_int_compare:w \l_@@_case_changed_char_int = \c_max_int + \@@_compute_case_changed_char: \fi: - \reverse_if:N \if_num:w #1 > \l_regex_case_changed_char_int - \reverse_if:N \if_num:w #2 < \l_regex_case_changed_char_int - \exp_after:wN \exp_after:wN \exp_after:wN \regex_break_true:w + \reverse_if:N \if_int_compare:w #1 > \l_@@_case_changed_char_int + \reverse_if:N \if_int_compare:w #2 < \l_@@_case_changed_char_int + \exp_after:wN \exp_after:wN \exp_after:wN \@@_break_true:w \fi: \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_compute_case_changed_char:} -% This function is called when \cs{l_regex_case_changed_char_int} has +% \begin{macro}[int]{\@@_compute_case_changed_char:} +% This function is called when \cs{l_@@_case_changed_char_int} has % not yet been computed (or rather, when it is set to the marker value % \cs{c_max_int}). If the current character code is in the range % $[65,90]$ (upper-case), then add $32$, making it lowercase. If it is % in the lower-case letter range $[97,122]$, subtract $32$. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_compute_case_changed_char: +\cs_new_protected_nopar:Npn \@@_compute_case_changed_char: { - \int_set_eq:NN \l_regex_case_changed_char_int \l_regex_current_char_int - \if_num:w \l_regex_current_char_int < \c_ninety_one - \if_num:w \l_regex_current_char_int < \c_sixty_five + \int_set_eq:NN \l_@@_case_changed_char_int \l_@@_current_char_int + \if_int_compare:w \l_@@_current_char_int < \c_ninety_one + \if_int_compare:w \l_@@_current_char_int < \c_sixty_five \else: - \int_add:Nn \l_regex_case_changed_char_int { \c_thirty_two } + \int_add:Nn \l_@@_case_changed_char_int { \c_thirty_two } \fi: \else: - \if_num:w \l_regex_current_char_int < \c_one_hundred_twenty_three - \if_num:w \l_regex_current_char_int < \c_ninety_seven + \if_int_compare:w \l_@@_current_char_int < \c_one_hundred_twenty_three + \if_int_compare:w \l_@@_current_char_int < \c_ninety_seven \else: - \int_sub:Nn \l_regex_case_changed_char_int { \c_thirty_two } + \int_sub:Nn \l_@@_case_changed_char_int { \c_thirty_two } \fi: \fi: \fi: @@ -999,79 +1008,79 @@ % \end{macrocode} % \end{macro} % -% \begin{macro}[int, EXP]{\regex_item_equal:n, \regex_item_range:nn} +% \begin{macro}[int, EXP]{\@@_item_equal:n, \@@_item_range:nn} % Those must always be defined to expand to a \texttt{caseful} % (default) or \texttt{caseless} version, and not be protected: they % must expand when compiling, to hard-code which tests are caseless or % caseful. % \begin{macrocode} -\cs_new_eq:NN \regex_item_equal:n ? -\cs_new_eq:NN \regex_item_range:nn ? +\cs_new_eq:NN \@@_item_equal:n ? +\cs_new_eq:NN \@@_item_range:nn ? % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_item_catcode:nT, \regex_item_catcode_reverse:nT} -% \begin{macro}[aux]{\regex_item_catcode_aux:} +% \begin{macro}[int]{\@@_item_catcode:nT, \@@_item_catcode_reverse:nT} +% \begin{macro}[aux]{\@@_item_catcode:} % The argument is a sum of powers of $4$ with exponents given by the % allowed category codes (between $0$ and $13$). Dividing by a given % power of $4$ gives an odd result if and only if that category code % is allowed. If the catcode does not match, then skip the character % code tests which follow. % \begin{macrocode} -\cs_new_protected:Npn \regex_item_catcode_aux: +\cs_new_protected:Npn \@@_item_catcode: { " - \if_case:w \l_regex_current_catcode_int + \if_case:w \l_@@_current_catcode_int 1 \or: 4 \or: 10 \or: 40 \or: 100 \or: \or: 1000 \or: 4000 \or: 10000 \or: \or: 100000 \or: 400000 \or: 1000000 \or: 4000000 \else: 1*\c_zero \fi: } -\cs_new_protected:Npn \regex_item_catcode:nT #1 +\cs_new_protected:Npn \@@_item_catcode:nT #1 { - \if_int_odd:w \int_eval:w #1 / \regex_item_catcode_aux: \int_eval_end: + \if_int_odd:w \__int_eval:w #1 / \@@_item_catcode: \__int_eval_end: \exp_after:wN \use:n \else: \exp_after:wN \use_none:n \fi: } -\cs_new_protected:Npn \regex_item_catcode_reverse:nT #1#2 - { \regex_item_catcode:nT {#1} { \regex_item_reverse:n {#2} } } +\cs_new_protected:Npn \@@_item_catcode_reverse:nT #1#2 + { \@@_item_catcode:nT {#1} { \@@_item_reverse:n {#2} } } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\regex_item_exact:nn, \regex_item_exact_cs:c} +% \begin{macro}[int]{\@@_item_exact:nn, \@@_item_exact_cs:c} % This matches an exact \meta{category}-\meta{character code} pair, or % an exact control sequence. % \begin{macrocode} -\cs_new_protected:Npn \regex_item_exact:nn #1#2 +\cs_new_protected:Npn \@@_item_exact:nn #1#2 { - \if_num:w #1 = \l_regex_current_catcode_int - \if_num:w #2 = \l_regex_current_char_int - \exp_after:wN \exp_after:wN \exp_after:wN \regex_break_true:w + \if_int_compare:w #1 = \l_@@_current_catcode_int + \if_int_compare:w #2 = \l_@@_current_char_int + \exp_after:wN \exp_after:wN \exp_after:wN \@@_break_true:w \fi: \fi: } -\cs_new_protected:Npn \regex_item_exact_cs:c #1 +\cs_new_protected:Npn \@@_item_exact_cs:c #1 { - \int_compare:nNnTF \l_regex_current_catcode_int = \c_zero + \int_compare:nNnTF \l_@@_current_catcode_int = \c_zero { - \str_if_eq:xxTF + \str_if_eq_x:nnTF { \exp_after:wN \exp_after:wN \exp_after:wN \cs_to_str:N - \tex_the:D \tex_toks:D \l_regex_current_pos_int + \tex_the:D \tex_toks:D \l_@@_current_pos_int } { #1 } - { \regex_break_true:w } { } + { \@@_break_true:w } { } } { } } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_item_cs:n} +% \begin{macro}[int]{\@@_item_cs:n} % Match a control sequence (the argument is a compiled regex). % First test the catcode of the current token to be zero. % Then perform the matching test, and break if the csname @@ -1079,25 +1088,25 @@ % of the \tn{toks}\meta{current position} (of the form \cs{exp_not:n} % \Arg{control sequence}) to \meta{control sequence}. % \begin{macrocode} -\cs_new_protected:Npn \regex_item_cs:n #1 +\cs_new_protected:Npn \@@_item_cs:n #1 { - \int_compare:nNnT \l_regex_current_catcode_int = \c_zero + \int_compare:nNnT \l_@@_current_catcode_int = \c_zero { \group_begin: - \regex_single_match: - \regex_disable_submatches: - \regex_build_for_cs:n {#1} - \bool_set_eq:NN \l_regex_saved_success_bool \g_regex_success_bool - \exp_args:Nx \regex_match:n + \@@_single_match: + \@@_disable_submatches: + \@@_build_for_cs:n {#1} + \bool_set_eq:NN \l_@@_saved_success_bool \g_@@_success_bool + \exp_args:Nx \@@_match:n { \exp_after:wN \exp_after:wN \exp_after:wN \cs_to_str:N - \tex_the:D \tex_toks:D \l_regex_current_pos_int + \tex_the:D \tex_toks:D \l_@@_current_pos_int } - \if_meaning:w \c_true_bool \g_regex_success_bool - \group_insert_after:N \regex_break_true:w + \if_meaning:w \c_true_bool \g_@@_success_bool + \group_insert_after:N \@@_break_true:w \fi: - \bool_gset_eq:NN \g_regex_success_bool \l_regex_saved_success_bool + \bool_gset_eq:NN \g_@@_success_bool \l_@@_saved_success_bool \group_end: } } @@ -1108,8 +1117,8 @@ % % \begin{macro}[aux] % { -% \regex_prop_d:, \regex_prop_h:, \regex_prop_s:, -% \regex_prop_v:, \regex_prop_w:, \regex_prop_N: +% \@@_prop_d:, \@@_prop_h:, \@@_prop_s:, +% \@@_prop_v:, \@@_prop_w:, \@@_prop_N: % } % Character property tests for |\d|, |\W|, \emph{etc.} These character % properties are not affected by the |(?i)| option. The characters @@ -1120,84 +1129,84 @@ % order in which the various tests appear is optimized for usual % mostly lower case letter text. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_prop_d: - { \regex_item_caseful_range:nn \c_forty_eight { 57 } } % 0--9 -\cs_new_protected_nopar:Npn \regex_prop_h: +\cs_new_protected_nopar:Npn \@@_prop_d: + { \@@_item_caseful_range:nn \c_forty_eight { 57 } } % 0--9 +\cs_new_protected_nopar:Npn \@@_prop_h: { - \regex_item_caseful_equal:n \c_thirty_two % space - \regex_item_caseful_equal:n \c_nine % tab + \@@_item_caseful_equal:n \c_thirty_two % space + \@@_item_caseful_equal:n \c_nine % tab } -\cs_new_protected_nopar:Npn \regex_prop_s: +\cs_new_protected_nopar:Npn \@@_prop_s: { - \regex_item_caseful_equal:n \c_thirty_two % space - \regex_item_caseful_equal:n \c_nine % tab - \regex_item_caseful_equal:n \c_ten % lf - \regex_item_caseful_equal:n \c_twelve % ff - \regex_item_caseful_equal:n \c_thirteen % cr + \@@_item_caseful_equal:n \c_thirty_two % space + \@@_item_caseful_equal:n \c_nine % tab + \@@_item_caseful_equal:n \c_ten % lf + \@@_item_caseful_equal:n \c_twelve % ff + \@@_item_caseful_equal:n \c_thirteen % cr } -\cs_new_protected_nopar:Npn \regex_prop_v: - { \regex_item_caseful_range:nn \c_ten \c_thirteen } % lf, vtab, ff, cr -\cs_new_protected_nopar:Npn \regex_prop_w: +\cs_new_protected_nopar:Npn \@@_prop_v: + { \@@_item_caseful_range:nn \c_ten \c_thirteen } % lf, vtab, ff, cr +\cs_new_protected_nopar:Npn \@@_prop_w: { - \regex_item_caseful_range:nn \c_ninety_seven { 122 } % a--z - \regex_item_caseful_range:nn \c_sixty_five { 90 } % A--Z - \regex_item_caseful_range:nn \c_forty_eight { 57 } % 0--9 - \regex_item_caseful_equal:n { 95 } % _ + \@@_item_caseful_range:nn \c_ninety_seven { 122 } % a--z + \@@_item_caseful_range:nn \c_sixty_five { 90 } % A--Z + \@@_item_caseful_range:nn \c_forty_eight { 57 } % 0--9 + \@@_item_caseful_equal:n { 95 } % _ } -\cs_new_protected_nopar:Npn \regex_prop_N: - { \regex_item_reverse:n { \regex_item_caseful_equal:n \c_ten } } +\cs_new_protected_nopar:Npn \@@_prop_N: + { \@@_item_reverse:n { \@@_item_caseful_equal:n \c_ten } } % \end{macrocode} % \end{macro} % % \begin{macro}[aux] % { -% \regex_posix_alnum:, \regex_posix_alpha:, \regex_posix_ascii:, -% \regex_posix_blank:, \regex_posix_cntrl:, \regex_posix_digit:, -% \regex_posix_graph:, \regex_posix_lower:, \regex_posix_print:, -% \regex_posix_punct:, \regex_posix_space:, \regex_posix_upper:, -% \regex_posix_word: , \regex_posix_xdigit: +% \@@_posix_alnum:, \@@_posix_alpha:, \@@_posix_ascii:, +% \@@_posix_blank:, \@@_posix_cntrl:, \@@_posix_digit:, +% \@@_posix_graph:, \@@_posix_lower:, \@@_posix_print:, +% \@@_posix_punct:, \@@_posix_space:, \@@_posix_upper:, +% \@@_posix_word: , \@@_posix_xdigit: % } % \textsc{posix} properties. No surprise. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_posix_alnum: - { \regex_posix_alpha: \regex_posix_digit: } -\cs_new_protected_nopar:Npn \regex_posix_alpha: - { \regex_posix_lower: \regex_posix_upper: } -\cs_new_protected_nopar:Npn \regex_posix_ascii: - { \regex_item_caseful_range:nn \c_zero \c_one_hundred_twenty_seven } -\cs_new_eq:NN \regex_posix_blank: \regex_prop_h: -\cs_new_protected_nopar:Npn \regex_posix_cntrl: +\cs_new_protected_nopar:Npn \@@_posix_alnum: + { \@@_posix_alpha: \@@_posix_digit: } +\cs_new_protected_nopar:Npn \@@_posix_alpha: + { \@@_posix_lower: \@@_posix_upper: } +\cs_new_protected_nopar:Npn \@@_posix_ascii: + { \@@_item_caseful_range:nn \c_zero \c_one_hundred_twenty_seven } +\cs_new_eq:NN \@@_posix_blank: \@@_prop_h: +\cs_new_protected_nopar:Npn \@@_posix_cntrl: { - \regex_item_caseful_range:nn \c_zero { 31 } - \regex_item_caseful_equal:n \c_one_hundred_twenty_seven + \@@_item_caseful_range:nn \c_zero { 31 } + \@@_item_caseful_equal:n \c_one_hundred_twenty_seven } -\cs_new_eq:NN \regex_posix_digit: \regex_prop_d: -\cs_new_protected_nopar:Npn \regex_posix_graph: - { \regex_item_caseful_range:nn { 33 } { 126 } } -\cs_new_protected_nopar:Npn \regex_posix_lower: - { \regex_item_caseful_range:nn \c_ninety_seven { 122 } } -\cs_new_protected_nopar:Npn \regex_posix_print: - { \regex_item_caseful_range:nn \c_thirty_two { 126 } } -\cs_new_protected_nopar:Npn \regex_posix_punct: +\cs_new_eq:NN \@@_posix_digit: \@@_prop_d: +\cs_new_protected_nopar:Npn \@@_posix_graph: + { \@@_item_caseful_range:nn { 33 } { 126 } } +\cs_new_protected_nopar:Npn \@@_posix_lower: + { \@@_item_caseful_range:nn \c_ninety_seven { 122 } } +\cs_new_protected_nopar:Npn \@@_posix_print: + { \@@_item_caseful_range:nn \c_thirty_two { 126 } } +\cs_new_protected_nopar:Npn \@@_posix_punct: { - \regex_item_caseful_range:nn { 33 } { 47 } - \regex_item_caseful_range:nn { 58 } { 64 } - \regex_item_caseful_range:nn { 91 } { 96 } - \regex_item_caseful_range:nn { 123 } { 126 } + \@@_item_caseful_range:nn { 33 } { 47 } + \@@_item_caseful_range:nn { 58 } { 64 } + \@@_item_caseful_range:nn { 91 } { 96 } + \@@_item_caseful_range:nn { 123 } { 126 } } -\cs_new_protected_nopar:Npn \regex_posix_space: +\cs_new_protected_nopar:Npn \@@_posix_space: { - \regex_item_caseful_equal:n \c_thirty_two - \regex_item_caseful_range:nn \c_nine \c_thirteen + \@@_item_caseful_equal:n \c_thirty_two + \@@_item_caseful_range:nn \c_nine \c_thirteen } -\cs_new_protected_nopar:Npn \regex_posix_upper: - { \regex_item_caseful_range:nn \c_sixty_five { 90 } } -\cs_new_eq:NN \regex_posix_word: \regex_prop_w: -\cs_new_protected_nopar:Npn \regex_posix_xdigit: +\cs_new_protected_nopar:Npn \@@_posix_upper: + { \@@_item_caseful_range:nn \c_sixty_five { 90 } } +\cs_new_eq:NN \@@_posix_word: \@@_prop_w: +\cs_new_protected_nopar:Npn \@@_posix_xdigit: { - \regex_posix_digit: - \regex_item_caseful_range:nn \c_sixty_five { 70 } - \regex_item_caseful_range:nn \c_ninety_seven { 102 } + \@@_posix_digit: + \@@_item_caseful_range:nn \c_sixty_five { 70 } + \@@_item_caseful_range:nn \c_ninety_seven { 102 } } % \end{macrocode} % \end{macro} @@ -1215,7 +1224,7 @@ % issues (those can become quite complex to handle in combination with % ranges in character classes). % -% Usage: \cs{regex_escape_use:nnnn} \meta{inline~1} \meta{inline~2} +% Usage: \cs{@@_escape_use:nnnn} \meta{inline~1} \meta{inline~2} % \meta{inline~3} \Arg{token list} The \meta{token list} is converted to % a string, then read from left to right, interpreting backslashes as % escaping the next character. Unescaped characters are fed to the @@ -1230,152 +1239,152 @@ % The conversion is mostly done within an \texttt{x}-expanding % assignment, except for the |\x| escape sequence, which is not amenable % to that in general. For this, we use the general framework of -% \cs{tl_set_build:Nw}. +% \cs{__tl_build:Nw}. % -% \begin{macro}[int]{\regex_escape_use:nnnn} -% The result is built in \cs{l_regex_internal_a_tl}, which is then +% \begin{macro}[int]{\@@_escape_use:nnnn} +% The result is built in \cs{l_@@_internal_a_tl}, which is then % left in the input stream. Go through |#4| once, applying |#1|, % |#2|, or |#3| as relevant to each character (after de-escaping -% it). Note that we cannot replace \cs{tl_se:Nx} and -% \cs{tl_build_one:o} by a single call to \cs{tl_build_one:x}, because -% the \texttt{x}-expanding assignment is interrupted by |\x|. +% it). Note that we cannot replace \cs{tl_set:Nx} and +% \cs{__tl_build_one:o} by a single call to \cs{__tl_build_one:x}, because +% the \texttt{x}-expanding assignment may be interrupted by |\x|. % \begin{macrocode} -\cs_new_protected:Npn \regex_escape_use:nnnn #1#2#3#4 +\cs_new_protected:Npn \@@_escape_use:nnnn #1#2#3#4 { -% \trace_push:nnn { regex } { 1 } { regex_escape_use:nnnn } - \tl_set_build:Nw \l_regex_internal_a_tl - \cs_set_nopar:Npn \regex_escape_unescaped:N ##1 { #1 } - \cs_set_nopar:Npn \regex_escape_escaped:N ##1 { #2 } - \cs_set_nopar:Npn \regex_escape_raw:N ##1 { #3 } +% \trace_push:nnn { regex } { 1 } { @@_escape_use:nnnn } + \__tl_build:Nw \l_@@_internal_a_tl + \cs_set_nopar:Npn \@@_escape_unescaped:N ##1 { #1 } + \cs_set_nopar:Npn \@@_escape_escaped:N ##1 { #2 } + \cs_set_nopar:Npn \@@_escape_raw:N ##1 { #3 } \int_set:Nn \tex_escapechar:D { 92 } - \str_gset_other:Nn \g_regex_internal_tl { #4 } - \tl_set:Nx \l_regex_internal_b_tl + \__str_gset_other:Nn \g_@@_internal_tl { #4 } + \tl_set:Nx \l_@@_internal_b_tl { - \exp_after:wN \regex_escape_loop:N \g_regex_internal_tl - { break } \prg_break_point:n { } + \exp_after:wN \@@_escape_loop:N \g_@@_internal_tl + { break } \__prg_break_point: } - \tl_build_one:o \l_regex_internal_b_tl - \tl_build_end: -% \trace_pop:nnn { regex } { 1 } { regex_escape_use:nnnn } - \l_regex_internal_a_tl + \__tl_build_one:o \l_@@_internal_b_tl + \__tl_build_end: +% \trace_pop:nnn { regex } { 1 } { @@_escape_use:nnnn } + \l_@@_internal_a_tl } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_escape_loop:N} -% \begin{macro}[aux]+\regex_escape_\:w+ -% \cs{regex_escape_loop:N} reads one character: if it is special +% \begin{macro}[aux]{\@@_escape_loop:N} +% \begin{macro}[aux]+\@@_escape_\:w+ +% \cs{@@_escape_loop:N} reads one character: if it is special % (space, backslash, or end-marker), perform the associated action, % otherwise it is simply an unescaped character. After a backslash, % the same is done, but unknown characters are \enquote{escaped}. % \begin{macrocode} -\cs_new:Npn \regex_escape_loop:N #1 +\cs_new:Npn \@@_escape_loop:N #1 { - \cs_if_exist_use:cF { regex_escape_\token_to_str:N #1:w } - { \regex_escape_unescaped:N #1 } - \regex_escape_loop:N + \cs_if_exist_use:cF { @@_escape_\token_to_str:N #1:w } + { \@@_escape_unescaped:N #1 } + \@@_escape_loop:N } -\cs_new_nopar:cpn { regex_escape_ \c_backslash_str :w } - \regex_escape_loop:N #1 +\cs_new_nopar:cpn { @@_escape_ \c_backslash_str :w } + \@@_escape_loop:N #1 { - \cs_if_exist_use:cF { regex_escape_/\token_to_str:N #1:w } - { \regex_escape_escaped:N #1 } - \regex_escape_loop:N + \cs_if_exist_use:cF { @@_escape_/\token_to_str:N #1:w } + { \@@_escape_escaped:N #1 } + \@@_escape_loop:N } % \end{macrocode} % \end{macro} % \end{macro} % % \begin{macro}[aux] -% {\regex_escape_unescaped:N, \regex_escape_escaped:N, \regex_escape_raw:N} +% {\@@_escape_unescaped:N, \@@_escape_escaped:N, \@@_escape_raw:N} % Those functions are never called before being given a new meaning, % so their definitions here don't matter. % \begin{macrocode} -\cs_new_eq:NN \regex_escape_unescaped:N ? -\cs_new_eq:NN \regex_escape_escaped:N ? -\cs_new_eq:NN \regex_escape_raw:N ? +\cs_new_eq:NN \@@_escape_unescaped:N ? +\cs_new_eq:NN \@@_escape_escaped:N ? +\cs_new_eq:NN \@@_escape_raw:N ? % \end{macrocode} % \end{macro} % % \begin{macro}[aux] % { -% \regex_escape_break:w, \regex_escape_/break:w, -% \regex_escape_/a:w, \regex_escape_/e:w, \regex_escape_/f:w, -% \regex_escape_/n:w, \regex_escape_/r:w, \regex_escape_/t:w +% \@@_escape_break:w, \@@_escape_/break:w, +% \@@_escape_/a:w, \@@_escape_/e:w, \@@_escape_/f:w, +% \@@_escape_/n:w, \@@_escape_/r:w, \@@_escape_/t:w % } -% \begin{macro}[aux]+\regex_escape_ :w+ +% \begin{macro}[aux]+\@@_escape_ :w+ % The loop is ended upon seeing the end-marker % \enquote{\texttt{break}}, with an error if the string ended in a % backslash. Spaces are ignored, and |\a|, |\e|, |\f|, |\n|, |\r|, % |\t| take their meaning here. % \begin{macrocode} -\cs_new_eq:NN \regex_escape_break:w \prg_map_break: -\cs_new_nopar:cpn { regex_escape_/break:w } +\cs_new_eq:NN \@@_escape_break:w \__prg_break: +\cs_new_nopar:cpn { @@_escape_/break:w } { \if_false: { \fi: } - \msg_kernel_error:nn { regex } { trailing-backslash } + \__msg_kernel_error:nn { regex } { trailing-backslash } \exp_after:wN \use_none:n \exp_after:wN { \if_false: } \fi: } -\cs_new_nopar:cpn { regex_escape_~:w } { } -\cs_new_nopar:cpx { regex_escape_/a:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^G } -\cs_new_nopar:cpx { regex_escape_/t:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^I } -\cs_new_nopar:cpx { regex_escape_/n:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^J } -\cs_new_nopar:cpx { regex_escape_/f:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^L } -\cs_new_nopar:cpx { regex_escape_/r:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^M } -\cs_new_nopar:cpx { regex_escape_/e:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^[ } +\cs_new_nopar:cpn { @@_escape_~:w } { } +\cs_new_nopar:cpx { @@_escape_/a:w } + { \exp_not:N \@@_escape_raw:N \iow_char:N \^^G } +\cs_new_nopar:cpx { @@_escape_/t:w } + { \exp_not:N \@@_escape_raw:N \iow_char:N \^^I } +\cs_new_nopar:cpx { @@_escape_/n:w } + { \exp_not:N \@@_escape_raw:N \iow_char:N \^^J } +\cs_new_nopar:cpx { @@_escape_/f:w } + { \exp_not:N \@@_escape_raw:N \iow_char:N \^^L } +\cs_new_nopar:cpx { @@_escape_/r:w } + { \exp_not:N \@@_escape_raw:N \iow_char:N \^^M } +\cs_new_nopar:cpx { @@_escape_/e:w } + { \exp_not:N \@@_escape_raw:N \iow_char:N \^^[ } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[aux]{\regex_escape_/x:w} -% \begin{macro}[aux]{\regex_escape_x_end:w, \regex_escape_x_large:n} -% When |\x| is encountered, \cs{regex_escape_x_test:N} is responsible +% \begin{macro}[aux]{\@@_escape_/x:w} +% \begin{macro}[aux]{\@@_escape_x_end:w, \@@_escape_x_large:n} +% When |\x| is encountered, \cs{@@_escape_x_test:N} is responsible % for grabbing some hexadecimal digits, and feeding the result to -% \cs{regex_escape_x_end:w}. If the number is $<256$, then it is -% turned into a byte and fed to \cs{regex_escape_raw:N}. Otherwise, +% \cs{@@_escape_x_end:w}. If the number is $<256$, then it is +% turned into a byte and fed to \cs{@@_escape_raw:N}. Otherwise, % interrupt the assignment, and either produce an error, or use a % standard \tn{lowercase} trick depending on the precise value. % \begin{macrocode} -\cs_new:cpn { regex_escape_/x:w } \regex_escape_loop:N +\cs_new:cpn { @@_escape_/x:w } \@@_escape_loop:N { - \exp_after:wN \regex_escape_x_end:w - \int_value:w "0 \regex_escape_x_test:N + \exp_after:wN \@@_escape_x_end:w + \__int_value:w "0 \@@_escape_x_test:N } -\cs_new:Npn \regex_escape_x_end:w #1 ; +\cs_new:Npn \@@_escape_x_end:w #1 ; { \int_compare:nNnTF {#1} < \c_two_hundred_fifty_six { - \exp_last_unbraced:Nf \regex_escape_raw:N - { \str_output_byte:n {#1} } + \exp_last_unbraced:Nf \@@_escape_raw:N + { \__str_output_byte:n {#1} } } - { \regex_escape_x_large:n {#1} } + { \@@_escape_x_large:n {#1} } } \group_begin: \char_set_catcode_other:n { 0 } - \cs_new:Npn \regex_escape_x_large:n #1 + \cs_new:Npn \@@_escape_x_large:n #1 { \if_false: { \fi: } - \tl_build_one:o \l_regex_internal_b_tl + \__tl_build_one:o \l_@@_internal_b_tl \int_compare:nNnTF {#1} > \c_max_char_int { - \msg_kernel_error:nnx { regex } { x-overflow } {#1} - \tl_set:Nx \l_regex_internal_b_tl - { \if_false: } \fi: \regex_escape_loop:N + \__msg_kernel_error:nnx { regex } { x-overflow } {#1} + \tl_set:Nx \l_@@_internal_b_tl + { \if_false: } \fi: \@@_escape_loop:N } { \char_set_lccode:nn { \c_zero } {#1} \tl_to_lowercase:n { - \tl_set:Nx \l_regex_internal_b_tl + \tl_set:Nx \l_@@_internal_b_tl { \if_false: } \fi: - \regex_escape_raw:N ^^@ - \regex_escape_loop:N + \@@_escape_raw:N ^^@ + \@@_escape_loop:N } } } @@ -1384,75 +1393,75 @@ % \end{macro} % \end{macro} % -% \begin{macro}[aux]{\regex_escape_x_test:N} +% \begin{macro}[aux]{\@@_escape_x_test:N} % Find out whether the first character is a left brace (allowing any % number of hexadecimal digits), or not (allowing up to two % hexadecimal digits). We need to check for the end-of-string marker. -% Eventually, call either \cs{regex_escape_x_loop:N} or -% \cs{regex_escape_x_ii:N}. +% Eventually, call either \cs{@@_escape_x_loop:N} or +% \cs{@@_escape_x_ii:N}. % \begin{macrocode} -\cs_new:Npn \regex_escape_x_test:N #1 +\cs_new:Npn \@@_escape_x_test:N #1 { - \str_if_eq:xxTF {#1} { break } { ; } + \str_if_eq_x:nnTF {#1} { break } { ; } { \if_charcode:w \c_space_token #1 - \exp_after:wN \regex_escape_x_test:N + \exp_after:wN \@@_escape_x_test:N \else: - \exp_after:wN \regex_escape_x_test_ii:N + \exp_after:wN \@@_escape_x_test_ii:N \exp_after:wN #1 \fi: } } -\cs_new:Npn \regex_escape_x_test_ii:N #1 +\cs_new:Npn \@@_escape_x_test_ii:N #1 { \if_charcode:w \c_lbrace_str #1 - \exp_after:wN \regex_escape_x_loop:N + \exp_after:wN \@@_escape_x_loop:N \else: - \str_aux_hexadecimal_use:NTF #1 - { \exp_after:wN \regex_escape_x_ii:N } - { ; \exp_after:wN \regex_escape_loop:N \exp_after:wN #1 } + \__str_hexadecimal_use:NTF #1 + { \exp_after:wN \@@_escape_x_ii:N } + { ; \exp_after:wN \@@_escape_loop:N \exp_after:wN #1 } \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_escape_x_ii:N} +% \begin{macro}[aux]{\@@_escape_x_ii:N} % This looks for the second digit in the unbraced case. % \begin{macrocode} -\cs_new:Npn \regex_escape_x_ii:N #1 +\cs_new:Npn \@@_escape_x_ii:N #1 { - \str_if_eq:xxTF {#1} { break } { ; } + \str_if_eq_x:nnTF {#1} { break } { ; } { - \str_aux_hexadecimal_use:NTF #1 - { ; \regex_escape_loop:N } - { ; \regex_escape_loop:N #1 } + \__str_hexadecimal_use:NTF #1 + { ; \@@_escape_loop:N } + { ; \@@_escape_loop:N #1 } } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_escape_x_loop:N} +% \begin{macro}[aux]{\@@_escape_x_loop:N} % Grab hexadecimal digits, skip spaces, and at the end, check that % there is a right brace, otherwise raise an error outside the % assignment. % \begin{macrocode} -\cs_new:Npn \regex_escape_x_loop:N #1 +\cs_new:Npn \@@_escape_x_loop:N #1 { - \str_aux_hexadecimal_use:NTF #1 - { \regex_escape_x_loop:N } + \__str_hexadecimal_use:NTF #1 + { \@@_escape_x_loop:N } { \token_if_eq_charcode:NNTF \c_space_token #1 - { \regex_escape_x_loop:N } + { \@@_escape_x_loop:N } { ; \exp_after:wN \token_if_eq_charcode:NNTF \c_rbrace_str #1 - { \regex_escape_loop:N } + { \@@_escape_loop:N } { \if_false: { \fi: } - \tl_build_one:o \l_regex_internal_b_tl - \msg_kernel_error:nn { regex } { x-missing-rbrace } {#1} - \tl_set:Nx \l_regex_internal_b_tl - { \if_false: } \fi: \regex_escape_loop:N #1 + \__tl_build_one:o \l_@@_internal_b_tl + \__msg_kernel_error:nn { regex } { x-missing-rbrace } {#1} + \tl_set:Nx \l_@@_internal_b_tl + { \if_false: } \fi: \@@_escape_loop:N #1 } } } @@ -1461,7 +1470,7 @@ % \end{macro} % % \begin{macro}[EXP, aux] -% {\regex_char_if_alphanumeric:NTF, \regex_char_if_special:NTF} +% {\@@_char_if_alphanumeric:NTF, \@@_char_if_special:NTF} % These two tests are used in the first pass when parsing a regular % expression. That pass is responsible for finding escaped and % non-escaped characters, and recognizing which ones have special @@ -1478,7 +1487,7 @@ % \end{itemize} % The code is ugly, and highly based on magic numbers and the ascii % codes of characters. This is mostly unavoidable for performance -% reasons: testing for instance with \cs{str_if_contains_char:nN} +% reasons: testing for instance with \cs{__str_if_contains_char:nN} % would be much slower. Maybe the tests can be optimized a little % bit more. % Here, \enquote{alphanumeric} means \texttt{0}--\texttt{9}, @@ -1487,41 +1496,41 @@ % but printable ascii, from space (hex \texttt{20}) to % \texttt{del} (hex \texttt{7E}). % \begin{macrocode} -\prg_new_conditional:Npnn \regex_char_if_special:N #1 { TF } +\prg_new_conditional:Npnn \@@_char_if_special:N #1 { TF } { - \if_num:w `#1 < \c_ninety_one - \if_num:w `#1 < \c_fifty_eight - \if_num:w `#1 < \c_forty_eight - \if_num:w `#1 < \c_thirty_two + \if_int_compare:w `#1 < \c_ninety_one + \if_int_compare:w `#1 < \c_fifty_eight + \if_int_compare:w `#1 < \c_forty_eight + \if_int_compare:w `#1 < \c_thirty_two \prg_return_false: \else: \prg_return_true: \fi: \else: \prg_return_false: \fi: \else: - \if_num:w `#1 < \c_sixty_five + \if_int_compare:w `#1 < \c_sixty_five \prg_return_true: \else: \prg_return_false: \fi: \fi: \else: - \if_num:w `#1 < \c_one_hundred_twenty_three - \if_num:w `#1 < \c_ninety_seven + \if_int_compare:w `#1 < \c_one_hundred_twenty_three + \if_int_compare:w `#1 < \c_ninety_seven \prg_return_true: \else: \prg_return_false: \fi: \else: - \if_num:w `#1 < \c_one_hundred_twenty_seven + \if_int_compare:w `#1 < \c_one_hundred_twenty_seven \prg_return_true: \else: \prg_return_false: \fi: \fi: \fi: } -\prg_new_conditional:Npnn \regex_char_if_alphanumeric:N #1 { TF } +\prg_new_conditional:Npnn \@@_char_if_alphanumeric:N #1 { TF } { - \if_num:w `#1 < \c_ninety_one - \if_num:w `#1 < \c_fifty_eight - \if_num:w `#1 < \c_forty_eight + \if_int_compare:w `#1 < \c_ninety_one + \if_int_compare:w `#1 < \c_fifty_eight + \if_int_compare:w `#1 < \c_forty_eight \prg_return_false: \else: \prg_return_true: \fi: \else: - \if_num:w `#1 < \c_sixty_five + \if_int_compare:w `#1 < \c_sixty_five \prg_return_false: \else: \prg_return_true: \fi: \fi: \else: - \if_num:w `#1 < \c_one_hundred_twenty_three - \if_num:w `#1 < \c_ninety_seven + \if_int_compare:w `#1 < \c_one_hundred_twenty_three + \if_int_compare:w `#1 < \c_ninety_seven \prg_return_false: \else: \prg_return_true: \fi: \else: \prg_return_false: @@ -1539,152 +1548,152 @@ % then turned into states of an automaton in the building % phase. Compiled regular expressions consist of the following: % \begin{itemize} -% \item \cs{regex_class:NnnnN} \meta{boolean} \Arg{tests} \Arg{min} +% \item \cs{@@_class:NnnnN} \meta{boolean} \Arg{tests} \Arg{min} % \Arg{more} \meta{lazyness} -% \item \cs{regex_group:nnnN} \Arg{branches} \Arg{min} \Arg{more} -% \meta{lazyness}, also \cs{regex_group_no_capture:nnnN} and -% \cs{regex_group_resetting:nnnN} with the same syntax. -% \item \cs{regex_branch:n} \Arg{contents} -% \item \cs{regex_command_K:} -% \item \cs{regex_assertion:Nn} \meta{boolean} \Arg{assertion test}, -% where the \meta{assertion test} is \cs{regex_b_test:} or -% |{|\cs{regex_anchor:N} \meta{integer}|}| +% \item \cs{@@_group:nnnN} \Arg{branches} \Arg{min} \Arg{more} +% \meta{lazyness}, also \cs{@@_group_no_capture:nnnN} and +% \cs{@@_group_resetting:nnnN} with the same syntax. +% \item \cs{@@_branch:n} \Arg{contents} +% \item \cs{@@_command_K:} +% \item \cs{@@_assertion:Nn} \meta{boolean} \Arg{assertion test}, +% where the \meta{assertion test} is \cs{@@_b_test:} or +% |{|\cs{@@_anchor:N} \meta{integer}|}| % \end{itemize} % Tests can be the following: % \begin{itemize} -% \item \cs{regex_item_caseful_equal:n} \Arg{char code} -% \item \cs{regex_item_caseless_equal:n} \Arg{char code} -% \item \cs{regex_item_caseful_range:nn} \Arg{min} \Arg{max} -% \item \cs{regex_item_caseless_range:nn} \Arg{min} \Arg{max} -% \item \cs{regex_item_catcode:nT} \Arg{catcode bitmap} \Arg{tests} -% \item \cs{regex_item_catcode_reverse:nT} \Arg{catcode bitmap} \Arg{tests} -% \item \cs{regex_item_reverse:n} \Arg{tests} -% \item \cs{regex_item_exact:nn} \Arg{catcode} \Arg{char code} -% \item \cs{regex_item_exact_cs:c} \Arg{csname} -% \item \cs{regex_item_cs:n} \Arg{compiled regex} +% \item \cs{@@_item_caseful_equal:n} \Arg{char code} +% \item \cs{@@_item_caseless_equal:n} \Arg{char code} +% \item \cs{@@_item_caseful_range:nn} \Arg{min} \Arg{max} +% \item \cs{@@_item_caseless_range:nn} \Arg{min} \Arg{max} +% \item \cs{@@_item_catcode:nT} \Arg{catcode bitmap} \Arg{tests} +% \item \cs{@@_item_catcode_reverse:nT} \Arg{catcode bitmap} \Arg{tests} +% \item \cs{@@_item_reverse:n} \Arg{tests} +% \item \cs{@@_item_exact:nn} \Arg{catcode} \Arg{char code} +% \item \cs{@@_item_exact_cs:c} \Arg{csname} +% \item \cs{@@_item_cs:n} \Arg{compiled regex} % \end{itemize} % % \subsubsection{Variables used when compiling} % -% \begin{variable}{\l_regex_group_level_int} +% \begin{variable}{\l_@@_group_level_int} % We make sure to open the same number of groups as we close. % \begin{macrocode} -\int_new:N \l_regex_group_level_int +\int_new:N \l_@@_group_level_int % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_regex_mode_int} +% \begin{variable}{\l_@@_mode_int} % While compiling, ten modes are recognized, labelled $-63$, $-23$, % $-6$, $-2$, $0$, $2$, $3$, $6$, $23$, $63$. See % section~\ref{sec:regex-modes}. % \begin{macrocode} -\int_new:N \l_regex_mode_int +\int_new:N \l_@@_mode_int % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_regex_catcodes_int, \l_regex_default_catcodes_int} -% \begin{variable}{\l_regex_catcodes_bool} +% \begin{variable}{\l_@@_catcodes_int, \l_@@_default_catcodes_int} +% \begin{variable}{\l_@@_catcodes_bool} % We wish to allow constructions such as |\c[^BE](..\cL[a-z]..)|, % where the outer catcode test applies to the whole group, but is % superseeded by the inner catcode test. For this to work, we need to % keep track of lists of allowed category codes: -% \cs{l_regex_catcodes_int} and \cs{l_regex_default_catcodes_int} are +% \cs{l_@@_catcodes_int} and \cs{l_@@_default_catcodes_int} are % bitmaps, sums of $4^c$, for all allowed catcodes $c$. The latter is % local to each capturing group, and we reset -% \cs{l_regex_catcodes_int} to that value after each character or +% \cs{l_@@_catcodes_int} to that value after each character or % class, changing it only when encountering a |\c| escape. The boolean % records whether the list of categories of a catcode test has to be % inverted: compare |\c[^BE]| and |\c[BE]|. % \begin{macrocode} -\int_new:N \l_regex_catcodes_int -\int_new:N \l_regex_default_catcodes_int -\bool_new:N \l_regex_catcodes_bool +\int_new:N \l_@@_catcodes_int +\int_new:N \l_@@_default_catcodes_int +\bool_new:N \l_@@_catcodes_bool % \end{macrocode} % \end{variable} % \end{variable} % % \begin{variable} % { -% \c_regex_catcode_C_int, \c_regex_catcode_B_int, \c_regex_catcode_E_int, -% \c_regex_catcode_M_int, \c_regex_catcode_T_int, \c_regex_catcode_P_int, -% \c_regex_catcode_U_int, \c_regex_catcode_D_int, \c_regex_catcode_S_int, -% \c_regex_catcode_L_int, \c_regex_catcode_O_int, \c_regex_catcode_A_int +% \c_@@_catcode_C_int, \c_@@_catcode_B_int, \c_@@_catcode_E_int, +% \c_@@_catcode_M_int, \c_@@_catcode_T_int, \c_@@_catcode_P_int, +% \c_@@_catcode_U_int, \c_@@_catcode_D_int, \c_@@_catcode_S_int, +% \c_@@_catcode_L_int, \c_@@_catcode_O_int, \c_@@_catcode_A_int % } -% \begin{variable}{\c_regex_all_catcodes_int} +% \begin{variable}{\c_@@_all_catcodes_int} % Constants: $4^c$ for each category, and the sum of all powers of $4$. % \begin{macrocode} -\int_const:Nn \c_regex_catcode_C_int { "1 } -\int_const:Nn \c_regex_catcode_B_int { "4 } -\int_const:Nn \c_regex_catcode_E_int { "10 } -\int_const:Nn \c_regex_catcode_M_int { "40 } -\int_const:Nn \c_regex_catcode_T_int { "100 } -\int_const:Nn \c_regex_catcode_P_int { "1000 } -\int_const:Nn \c_regex_catcode_U_int { "4000 } -\int_const:Nn \c_regex_catcode_D_int { "10000 } -\int_const:Nn \c_regex_catcode_S_int { "100000 } -\int_const:Nn \c_regex_catcode_L_int { "400000 } -\int_const:Nn \c_regex_catcode_O_int { "1000000 } -\int_const:Nn \c_regex_catcode_A_int { "4000000 } -\int_const:Nn \c_regex_all_catcodes_int { "5515155 } +\int_const:Nn \c_@@_catcode_C_int { "1 } +\int_const:Nn \c_@@_catcode_B_int { "4 } +\int_const:Nn \c_@@_catcode_E_int { "10 } +\int_const:Nn \c_@@_catcode_M_int { "40 } +\int_const:Nn \c_@@_catcode_T_int { "100 } +\int_const:Nn \c_@@_catcode_P_int { "1000 } +\int_const:Nn \c_@@_catcode_U_int { "4000 } +\int_const:Nn \c_@@_catcode_D_int { "10000 } +\int_const:Nn \c_@@_catcode_S_int { "100000 } +\int_const:Nn \c_@@_catcode_L_int { "400000 } +\int_const:Nn \c_@@_catcode_O_int { "1000000 } +\int_const:Nn \c_@@_catcode_A_int { "4000000 } +\int_const:Nn \c_@@_all_catcodes_int { "5515155 } % \end{macrocode} % \end{variable} % \end{variable} % -% \begin{variable}{\l_regex_internal_regex} +% \begin{variable}{\l_@@_internal_regex} % The compilation step stores its result in this variable. % \begin{macrocode} -\cs_new_eq:NN \l_regex_internal_regex \c_regex_no_match_regex +\cs_new_eq:NN \l_@@_internal_regex \c_@@_no_match_regex % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_regex_show_prefix_seq} +% \begin{variable}{\l_@@_show_prefix_seq} % This sequence holds the prefix that makes up the line displayed to % the user. The various items must be removed from the right, which is % tricky with a token list, hence we use a sequence. % \begin{macrocode} -\seq_new:N \l_regex_show_prefix_seq +\seq_new:N \l_@@_show_prefix_seq % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_regex_show_lines_int} +% \begin{variable}{\l_@@_show_lines_int} % A hack. To know whether a given class has a single item in it or % not, we count the number of lines when showing the class. % \begin{macrocode} -\int_new:N \l_regex_show_lines_int +\int_new:N \l_@@_show_lines_int % \end{macrocode} % \end{variable} % % \subsubsection{Generic helpers used when compiling} % -% \begin{macro}[int]{\regex_get_digits:NTFw} -% \begin{macro}[aux, rEXP]{\regex_get_digits_loop:w} +% \begin{macro}[int]{\@@_get_digits:NTFw} +% \begin{macro}[aux, rEXP]{\@@_get_digits_loop:w} % If followed by some raw digits, collect them one by one in the % integer variable |#1|, and take the \texttt{true} branch. Otherwise, % take the \texttt{false} branch. % \begin{macrocode} -\cs_new_protected:Npn \regex_get_digits:NTFw #1#2#3#4#5 +\cs_new_protected:Npn \@@_get_digits:NTFw #1#2#3#4#5 { - \regex_if_raw_digit:NNTF #4 #5 - { #1 = #5 \regex_get_digits_loop:nw {#2} } + \@@_if_raw_digit:NNTF #4 #5 + { #1 = #5 \@@_get_digits_loop:nw {#2} } { #3 #4 #5 } } -\cs_new:Npn \regex_get_digits_loop:nw #1#2#3 +\cs_new:Npn \@@_get_digits_loop:nw #1#2#3 { - \regex_if_raw_digit:NNTF #2 #3 - { #3 \regex_get_digits_loop:nw {#1} } + \@@_if_raw_digit:NNTF #2 #3 + { #3 \@@_get_digits_loop:nw {#1} } { \scan_stop: #1 #2 #3 } } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[aux, EXP]{\regex_if_raw_digit:NNTF} +% \begin{macro}[aux, EXP]{\@@_if_raw_digit:NNTF} % Test used when grabbing digits for the |{m,n}| quantifier. % It only accepts non-escaped digits. % \begin{macrocode} -\prg_new_conditional:Npnn \regex_if_raw_digit:NN #1#2 { TF } +\prg_new_conditional:Npnn \@@_if_raw_digit:NN #1#2 { TF } { - \if_meaning:w \regex_compile_raw:N #1 + \if_meaning:w \@@_compile_raw:N #1 \if_int_compare:w \c_one < 1 #2 \exp_stop_f: \prg_return_true: \else: @@ -1740,15 +1749,15 @@ % any mode. % \end{itemize} % -% \begin{macro}[int, EXP]{\regex_if_in_class:TF} +% \begin{macro}[int, EXP]{\@@_if_in_class:TF} % Test whether we are directly in a character class (at the innermost % level of nesting). There, many escape sequences are not recognized, % and special characters are normal. Also, for every raw character, we % must look ahead for a possible raw dash. % \begin{macrocode} -\cs_new_nopar:Npn \regex_if_in_class:TF +\cs_new_nopar:Npn \@@_if_in_class:TF { - \if_int_odd:w \l_regex_mode_int + \if_int_odd:w \l_@@_mode_int \exp_after:wN \use_i:nn \else: \exp_after:wN \use_ii:nn @@ -1757,16 +1766,16 @@ % \end{macrocode} % \end{macro} % -% \begin{macro}[int, EXP]{\regex_if_in_cs:TF} +% \begin{macro}[int, EXP]{\@@_if_in_cs:TF} % Right braces are special only directly inside control sequences (at % the inner-most level of nesting, not counting groups). % \begin{macrocode} -\cs_new_nopar:Npn \regex_if_in_cs:TF +\cs_new_nopar:Npn \@@_if_in_cs:TF { - \if_int_odd:w \l_regex_mode_int + \if_int_odd:w \l_@@_mode_int \exp_after:wN \use_ii:nn \else: - \if_int_compare:w \l_regex_mode_int < \c_zero + \if_int_compare:w \l_@@_mode_int < \c_zero \exp_after:wN \exp_after:wN \exp_after:wN \use_i:nn \else: \exp_after:wN \exp_after:wN \exp_after:wN \use_ii:nn @@ -1776,16 +1785,16 @@ % \end{macrocode} % \end{macro} % -% \begin{macro}[int, EXP]{\regex_if_in_class_or_catcode:TF} +% \begin{macro}[int, EXP]{\@@_if_in_class_or_catcode:TF} % Assertions are only allowed in modes $0$, $-2$, and $-6$, % \emph{i.e.}, even, non-positive modes. % \begin{macrocode} -\cs_new_nopar:Npn \regex_if_in_class_or_catcode:TF +\cs_new_nopar:Npn \@@_if_in_class_or_catcode:TF { - \if_int_odd:w \l_regex_mode_int + \if_int_odd:w \l_@@_mode_int \exp_after:wN \use_i:nn \else: - \if_int_compare:w \l_regex_mode_int > \c_zero + \if_int_compare:w \l_@@_mode_int > \c_zero \exp_after:wN \exp_after:wN \exp_after:wN \use_i:nn \else: \exp_after:wN \exp_after:wN \exp_after:wN \use_ii:nn @@ -1795,15 +1804,15 @@ % \end{macrocode} % \end{macro} % -% \begin{macro}[int, EXP]{\regex_if_within_catcode:TF} +% \begin{macro}[int, EXP]{\@@_if_within_catcode:TF} % This test takes the true branch if we are in a catcode test, either % immediately following it (modes $2$ and $6$) or in a class on which % it applies (modes $23$ and $63$). This is used to tweak how left % brackets behave in modes $2$ and $6$. % \begin{macrocode} -\cs_new_nopar:Npn \regex_if_within_catcode:TF +\cs_new_nopar:Npn \@@_if_within_catcode:TF { - \if_int_compare:w \l_regex_mode_int > \c_zero + \if_int_compare:w \l_@@_mode_int > \c_zero \exp_after:wN \use_i:nn \else: \exp_after:wN \use_ii:nn @@ -1812,19 +1821,19 @@ % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_chk_c_allowed:T} +% \begin{macro}[aux]{\@@_chk_c_allowed:T} % The |\c| escape sequence is only allowed in modes $0$ and $3$, % \emph{i.e.}, not within any other |\c| escape sequence. % \begin{macrocode} -\cs_new_protected:Npn \regex_chk_c_allowed:T +\cs_new_protected:Npn \@@_chk_c_allowed:T { - \if_num:w \l_regex_mode_int = \c_zero + \if_int_compare:w \l_@@_mode_int = \c_zero \exp_after:wN \use:n \else: - \if_num:w \l_regex_mode_int = \c_three + \if_int_compare:w \l_@@_mode_int = \c_three \exp_after:wN \exp_after:wN \exp_after:wN \use:n \else: - \msg_kernel_error:nn { regex } { c-bad-mode } + \__msg_kernel_error:nn { regex } { c-bad-mode } \exp_after:wN \exp_after:wN \exp_after:wN \use_none:n \fi: \fi: @@ -1832,17 +1841,17 @@ % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_mode_quit_c:} +% \begin{macro}[aux]{\@@_mode_quit_c:} % This function changes the mode as it is needed just after a catcode % test. % \begin{macrocode} -\cs_new_protected:Npn \regex_mode_quit_c: +\cs_new_protected:Npn \@@_mode_quit_c: { - \if_num:w \l_regex_mode_int = \c_two - \l_regex_mode_int = \c_zero + \if_int_compare:w \l_@@_mode_int = \c_two + \l_@@_mode_int = \c_zero \else: - \if_num:w \l_regex_mode_int = \c_six - \l_regex_mode_int = \c_three + \if_int_compare:w \l_@@_mode_int = \c_six + \l_@@_mode_int = \c_three \fi: \fi: } @@ -1851,214 +1860,214 @@ % % \subsubsection{Framework} % -% \begin{macro}[int]{\regex_compile:w, \regex_compile_end:} +% \begin{macro}[int]{\@@_compile:w, \@@_compile_end:} % Used when compiling a user regex or a regex for the |\c{...}| escape % sequence within another regex. Start building a token list within a % group (with \texttt{x}-expansion at the outset), and set a few % variables (group level, catcodes), then start the first branch. At % the end, make sure there are no dangling classes nor groups, close -% the last branch: we are done building \cs{l_regex_internal_regex}. +% the last branch: we are done building \cs{l_@@_internal_regex}. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_compile:w +\cs_new_protected_nopar:Npn \@@_compile:w { - \tl_set_build_x:Nw \l_regex_internal_regex - \int_zero:N \l_regex_group_level_int - \int_set_eq:NN \l_regex_default_catcodes_int \c_regex_all_catcodes_int - \int_set_eq:NN \l_regex_catcodes_int \l_regex_default_catcodes_int - \cs_set_nopar:Npn \regex_item_equal:n { \regex_item_caseful_equal:n } - \cs_set_nopar:Npn \regex_item_range:nn { \regex_item_caseful_range:nn } - \tl_build_one:n { \regex_branch:n { \if_false: } \fi: } + \__tl_build_x:Nw \l_@@_internal_regex + \int_zero:N \l_@@_group_level_int + \int_set_eq:NN \l_@@_default_catcodes_int \c_@@_all_catcodes_int + \int_set_eq:NN \l_@@_catcodes_int \l_@@_default_catcodes_int + \cs_set_nopar:Npn \@@_item_equal:n { \@@_item_caseful_equal:n } + \cs_set_nopar:Npn \@@_item_range:nn { \@@_item_caseful_range:nn } + \__tl_build_one:n { \@@_branch:n { \if_false: } \fi: } } -\cs_new_protected_nopar:Npn \regex_compile_end: +\cs_new_protected_nopar:Npn \@@_compile_end: { - \regex_if_in_class:TF + \@@_if_in_class:TF { - \msg_kernel_error:nn { regex } { missing-rbrack } - \use:c { regex_compile_]: } + \__msg_kernel_error:nn { regex } { missing-rbrack } + \use:c { @@_compile_]: } \prg_do_nothing: \prg_do_nothing: } { } - \if_num:w \l_regex_group_level_int > \c_zero - \msg_kernel_error:nnx { regex } { missing-rparen } - { \int_use:N \l_regex_group_level_int } + \if_int_compare:w \l_@@_group_level_int > \c_zero + \__msg_kernel_error:nnx { regex } { missing-rparen } + { \int_use:N \l_@@_group_level_int } \prg_replicate:nn - { \l_regex_group_level_int } + { \l_@@_group_level_int } { - \tl_build_one:n + \__tl_build_one:n { \if_false: { \fi: } \if_false: { \fi: } { 1 } { 0 } \c_true_bool } - \tl_build_end: - \tl_build_one:o \l_regex_internal_regex + \__tl_build_end: + \__tl_build_one:o \l_@@_internal_regex } \fi: - \tl_build_one:n { \if_false: { \fi: } } - \tl_build_end: + \__tl_build_one:n { \if_false: { \fi: } } + \__tl_build_end: } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_compile:n} -% The compilation is done between \cs{regex_compile:w} and -% \cs{regex_compile_end:}, starting in mode~$0$. Then -% \cs{regex_escape_use:nnnn} distinguishes special characters, escaped +% \begin{macro}[int]{\@@_compile:n} +% The compilation is done between \cs{@@_compile:w} and +% \cs{@@_compile_end:}, starting in mode~$0$. Then +% \cs{@@_escape_use:nnnn} distinguishes special characters, escaped % alphanumerics, and raw characters, interpreting |\a|, |\x| and other % sequences. The $4$ trailing \cs{prg_do_nothing:} are needed because % some functions defined later look up to $4$ tokens ahead. Before % ending, make sure that any |\c{...}| is properly closed. % \begin{macrocode} -\cs_new_protected:Npn \regex_compile:n #1 +\cs_new_protected:Npn \@@_compile:n #1 { - \regex_compile:w + \@@_compile:w \int_set:Nn \tex_escapechar:D { 92 } - \int_set_eq:NN \l_regex_mode_int \c_zero - \regex_escape_use:nnnn + \int_set_eq:NN \l_@@_mode_int \c_zero + \@@_escape_use:nnnn { - \regex_char_if_special:NTF ##1 - \regex_compile_special:N \regex_compile_raw:N ##1 + \@@_char_if_special:NTF ##1 + \@@_compile_special:N \@@_compile_raw:N ##1 } { - \regex_char_if_alphanumeric:NTF ##1 - \regex_compile_escaped:N \regex_compile_raw:N ##1 + \@@_char_if_alphanumeric:NTF ##1 + \@@_compile_escaped:N \@@_compile_raw:N ##1 } - { \regex_compile_raw:N ##1 } + { \@@_compile_raw:N ##1 } { #1 } \prg_do_nothing: \prg_do_nothing: \prg_do_nothing: \prg_do_nothing: - \int_compare:nNnT \l_regex_mode_int < \c_zero + \int_compare:nNnT \l_@@_mode_int < \c_zero { - \msg_kernel_error:nn { regex } { c-missing-rbrace } - \regex_compile_end: - \regex_compile_one:x - { \regex_item_cs:n { \exp_not:o \l_regex_internal_regex } } + \__msg_kernel_error:nn { regex } { c-missing-rbrace } + \@@_compile_end: + \@@_compile_one:x + { \@@_item_cs:n { \exp_not:o \l_@@_internal_regex } } \prg_do_nothing: \prg_do_nothing: \prg_do_nothing: \prg_do_nothing: } - \regex_compile_end: + \@@_compile_end: } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_compile_escaped:N, \regex_compile_special:N} +% \begin{macro}[int]{\@@_compile_escaped:N, \@@_compile_special:N} % If the special character or escaped alphanumeric has a particular % meaning in regexes, the corresponding function is used. Otherwise, % it is interpreted as a raw character. We distinguish special % characters from escaped alphanumeric characters because they behave % differently when appearing as an end-point of a range. % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_special:N #1 +\cs_new_protected:Npn \@@_compile_special:N #1 { - \cs_if_exist_use:cF { regex_compile_#1: } - { \regex_compile_raw:N #1 } + \cs_if_exist_use:cF { @@_compile_#1: } + { \@@_compile_raw:N #1 } } -\cs_new_protected:Npn \regex_compile_escaped:N #1 +\cs_new_protected:Npn \@@_compile_escaped:N #1 { - \cs_if_exist_use:cF { regex_compile_/#1: } - { \regex_compile_raw:N #1 } + \cs_if_exist_use:cF { @@_compile_/#1: } + { \@@_compile_raw:N #1 } } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_compile_one:x} +% \begin{macro}[int]{\@@_compile_one:x} % This is used after finding one \enquote{test}, such as |\d|, or a % raw character. If that followed a catcode test (\emph{e.g.}, |\cL|), % then restore the mode. If we are not in a class, then the test is -% \enquote{standalone}, and we need to add \cs{regex_class:NnnnN} and +% \enquote{standalone}, and we need to add \cs{@@_class:NnnnN} and % search for quantifiers. In any case, insert the test, possibly % together with a catcode test if appropriate. % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_one:x #1 +\cs_new_protected:Npn \@@_compile_one:x #1 { - \regex_mode_quit_c: - \regex_if_in_class:TF { } + \@@_mode_quit_c: + \@@_if_in_class:TF { } { - \tl_build_one:n - { \regex_class:NnnnN \c_true_bool { \if_false: } \fi: } + \__tl_build_one:n + { \@@_class:NnnnN \c_true_bool { \if_false: } \fi: } } - \tl_build_one:x + \__tl_build_one:x { - \if_num:w \l_regex_catcodes_int < \c_regex_all_catcodes_int - \regex_item_catcode:nT { \int_use:N \l_regex_catcodes_int } + \if_int_compare:w \l_@@_catcodes_int < \c_@@_all_catcodes_int + \@@_item_catcode:nT { \int_use:N \l_@@_catcodes_int } { \exp_not:N \exp_not:n {#1} } \else: \exp_not:N \exp_not:n {#1} \fi: } - \int_set_eq:NN \l_regex_catcodes_int \l_regex_default_catcodes_int - \regex_if_in_class:TF { } { \regex_compile_quantifier:w } + \int_set_eq:NN \l_@@_catcodes_int \l_@@_default_catcodes_int + \@@_if_in_class:TF { } { \@@_compile_quantifier:w } } % \end{macrocode} % \end{macro} % % \begin{macro}[int] -% {\regex_compile_abort_tokens:n, \regex_compile_abort_tokens:x} +% {\@@_compile_abort_tokens:n, \@@_compile_abort_tokens:x} % This function places the collected tokens back in the input stream, % each as a raw character. Spaces are not preserved. % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_abort_tokens:n #1 +\cs_new_protected:Npn \@@_compile_abort_tokens:n #1 { \use:x { \exp_args:No \tl_map_function:nN { \tl_to_str:n {#1} } - \regex_compile_raw:N + \@@_compile_raw:N } } -\cs_generate_variant:Nn \regex_compile_abort_tokens:n { x } +\cs_generate_variant:Nn \@@_compile_abort_tokens:n { x } % \end{macrocode} % \end{macro} % % \subsubsection{Quantifers} % -% \begin{macro}[int]{\regex_compile_quantifier:w} +% \begin{macro}[int]{\@@_compile_quantifier:w} % This looks ahead and finds any quantifier (special character equal % to either of \texttt{?+*\{}). % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_quantifier:w #1#2 +\cs_new_protected:Npn \@@_compile_quantifier:w #1#2 { - \token_if_eq_meaning:NNTF #1 \regex_compile_special:N + \token_if_eq_meaning:NNTF #1 \@@_compile_special:N { - \cs_if_exist_use:cF { regex_compile_quantifier_#2:w } - { \regex_compile_quantifier_none: #1 #2 } + \cs_if_exist_use:cF { @@_compile_quantifier_#2:w } + { \@@_compile_quantifier_none: #1 #2 } } - { \regex_compile_quantifier_none: #1 #2 } + { \@@_compile_quantifier_none: #1 #2 } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_compile_quantifier_none:} -% \begin{macro}[aux]{\regex_compile_quantifier_abort:xNN} +% \begin{macro}[aux]{\@@_compile_quantifier_none:} +% \begin{macro}[aux]{\@@_compile_quantifier_abort:xNN} % Those functions are called whenever there is no quantifier, or a % braced construction is invalid (equivalent to no quantifier, and % whatever characters were grabbed are left raw). % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_quantifier_none: - { \tl_build_one:n { \if_false: { \fi: } { 1 } { 0 } \c_false_bool } } -\cs_new_protected:Npn \regex_compile_quantifier_abort:xNN #1#2#3 +\cs_new_protected:Npn \@@_compile_quantifier_none: + { \__tl_build_one:n { \if_false: { \fi: } { 1 } { 0 } \c_false_bool } } +\cs_new_protected:Npn \@@_compile_quantifier_abort:xNN #1#2#3 { - \regex_compile_quantifier_none: - \msg_kernel_warning:nnxx { regex } { invalid-quantifier } {#1} {#3} - \regex_compile_abort_tokens:x {#1} + \@@_compile_quantifier_none: + \__msg_kernel_warning:nnxx { regex } { invalid-quantifier } {#1} {#3} + \@@_compile_abort_tokens:x {#1} #2 #3 } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[aux]{\regex_compile_quantifier_lazyness:nnNN} +% \begin{macro}[aux]{\@@_compile_quantifier_lazyness:nnNN} % Once the \enquote{main} quantifier (\texttt{?}, \texttt{*}, % \texttt{+} or a braced construction) is found, we check whether it % is lazy (followed by a question mark). We then add to the compiled -% regex a closing brace (ending \cs{regex_class:NnnnN} and friends), +% regex a closing brace (ending \cs{@@_class:NnnnN} and friends), % the start-point of the range, its end-point, and a boolean, % \texttt{true} for lazy and \texttt{false} for greedy operators. % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_quantifier_lazyness:nnNN #1#2#3#4 +\cs_new_protected:Npn \@@_compile_quantifier_lazyness:nnNN #1#2#3#4 { - \str_if_eq:nnTF { #3 #4 } { \regex_compile_special:N ? } - { \tl_build_one:n { \if_false: { \fi: } { #1 } { #2 } \c_true_bool } } + \str_if_eq:nnTF { #3 #4 } { \@@_compile_special:N ? } + { \__tl_build_one:n { \if_false: { \fi: } { #1 } { #2 } \c_true_bool } } { - \tl_build_one:n { \if_false: { \fi: } { #1 } { #2 } \c_false_bool } + \__tl_build_one:n { \if_false: { \fi: } { #1 } { #2 } \c_false_bool } #3 #4 } } @@ -2067,106 +2076,106 @@ % % \begin{macro}[aux] % { -% \regex_compile_quantifier_?:w, -% \regex_compile_quantifier_*:w, -% \regex_compile_quantifier_+:w +% \@@_compile_quantifier_?:w, +% \@@_compile_quantifier_*:w, +% \@@_compile_quantifier_+:w % } % For each \enquote{basic} quantifier, |?|, |*|, |+|, feed the correct -% arguments to \cs{regex_compile_quantifier_lazyness:nnNN}, $-1$ means +% arguments to \cs{@@_compile_quantifier_lazyness:nnNN}, $-1$ means % that there is no upper bound on the number of repetitions. % \begin{macrocode} -\cs_new_protected_nopar:cpn { regex_compile_quantifier_?:w } - { \regex_compile_quantifier_lazyness:nnNN { 0 } { 1 } } -\cs_new_protected_nopar:cpn { regex_compile_quantifier_*:w } - { \regex_compile_quantifier_lazyness:nnNN { 0 } { -1 } } -\cs_new_protected_nopar:cpn { regex_compile_quantifier_+:w } - { \regex_compile_quantifier_lazyness:nnNN { 1 } { -1 } } +\cs_new_protected_nopar:cpn { @@_compile_quantifier_?:w } + { \@@_compile_quantifier_lazyness:nnNN { 0 } { 1 } } +\cs_new_protected_nopar:cpn { @@_compile_quantifier_*:w } + { \@@_compile_quantifier_lazyness:nnNN { 0 } { -1 } } +\cs_new_protected_nopar:cpn { @@_compile_quantifier_+:w } + { \@@_compile_quantifier_lazyness:nnNN { 1 } { -1 } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]+\regex_compile_quantifier_{:w+ ^^A} +% \begin{macro}[aux]+\@@_compile_quantifier_{:w+ ^^A} % \begin{macro}[aux] % { -% \regex_compile_quantifier_braced_i:w, -% \regex_compile_quantifier_braced_ii:w, -% \regex_compile_quantifier_braced_iii:w, +% \@@_compile_quantifier_braced_i:w, +% \@@_compile_quantifier_braced_ii:w, +% \@@_compile_quantifier_braced_iii:w, % } % Three possible syntaxes: \texttt{\{\meta{int}\}}, % \texttt{\{\meta{int},\}}, or \texttt{\{\meta{int},\meta{int}\}}. Any % other syntax causes us to abort and put whatever we collected back % in the input stream, as \texttt{raw} characters, including the -% opening brace. Grab a number into \cs{l_regex_internal_a_int}. If +% opening brace. Grab a number into \cs{l_@@_internal_a_int}. If % the number is followed by a right brace, the range is $[a,a]$. If % followed by a comma, grab one more number, and call the \texttt{_ii} % or \texttt{_iii} auxiliary. Those auxiliaries check for a closing % brace, leading to the range $[a,\infty]$ or $[a,b]$, encoded as % $\{a\}\{-1\}$ and $\{a\}\{b-a\}$. % \begin{macrocode} -\cs_new_protected:cpn { regex_compile_quantifier_ \c_lbrace_str :w } +\cs_new_protected:cpn { @@_compile_quantifier_ \c_lbrace_str :w } { - \regex_get_digits:NTFw \l_regex_internal_a_int - { \regex_compile_quantifier_braced_i:w } - { \regex_compile_quantifier_abort:xNN { \c_lbrace_str } } + \@@_get_digits:NTFw \l_@@_internal_a_int + { \@@_compile_quantifier_braced_i:w } + { \@@_compile_quantifier_abort:xNN { \c_lbrace_str } } } -\cs_new_protected:Npn \regex_compile_quantifier_braced_i:w #1#2 +\cs_new_protected:Npn \@@_compile_quantifier_braced_i:w #1#2 { - \prg_case_str:xxn { #1 #2 } + \str_case_x:nnn { #1 #2 } { - { \regex_compile_special:N \c_rbrace_str } + { \@@_compile_special:N \c_rbrace_str } { - \exp_args:No \regex_compile_quantifier_lazyness:nnNN - { \int_use:N \l_regex_internal_a_int } { 0 } + \exp_args:No \@@_compile_quantifier_lazyness:nnNN + { \int_use:N \l_@@_internal_a_int } { 0 } } - { \regex_compile_special:N , } + { \@@_compile_special:N , } { - \regex_get_digits:NTFw \l_regex_internal_b_int - { \regex_compile_quantifier_braced_iii:w } - { \regex_compile_quantifier_braced_ii:w } + \@@_get_digits:NTFw \l_@@_internal_b_int + { \@@_compile_quantifier_braced_iii:w } + { \@@_compile_quantifier_braced_ii:w } } } { - \regex_compile_quantifier_abort:xNN - { \c_lbrace_str \int_use:N \l_regex_internal_a_int } + \@@_compile_quantifier_abort:xNN + { \c_lbrace_str \int_use:N \l_@@_internal_a_int } #1 #2 } } -\cs_new_protected:Npn \regex_compile_quantifier_braced_ii:w #1#2 +\cs_new_protected:Npn \@@_compile_quantifier_braced_ii:w #1#2 { - \str_if_eq:xxTF - { #1 #2 } { \regex_compile_special:N \c_rbrace_str } + \str_if_eq_x:nnTF + { #1 #2 } { \@@_compile_special:N \c_rbrace_str } { - \exp_args:No \regex_compile_quantifier_lazyness:nnNN - { \int_use:N \l_regex_internal_a_int } { -1 } + \exp_args:No \@@_compile_quantifier_lazyness:nnNN + { \int_use:N \l_@@_internal_a_int } { -1 } } { - \regex_compile_quantifier_abort:xNN - { \c_lbrace_str \int_use:N \l_regex_internal_a_int , } + \@@_compile_quantifier_abort:xNN + { \c_lbrace_str \int_use:N \l_@@_internal_a_int , } #1 #2 } } -\cs_new_protected:Npn \regex_compile_quantifier_braced_iii:w #1#2 +\cs_new_protected:Npn \@@_compile_quantifier_braced_iii:w #1#2 { - \str_if_eq:xxTF - { #1 #2 } { \regex_compile_special:N \c_rbrace_str } + \str_if_eq_x:nnTF + { #1 #2 } { \@@_compile_special:N \c_rbrace_str } { - \if_num:w \l_regex_internal_a_int > \l_regex_internal_b_int - \msg_kernel_error:nnxx { regex } { backwards-quantifier } - { \int_use:N \l_regex_internal_a_int } - { \int_use:N \l_regex_internal_b_int } - \int_zero:N \l_regex_internal_b_int + \if_int_compare:w \l_@@_internal_a_int > \l_@@_internal_b_int + \__msg_kernel_error:nnxx { regex } { backwards-quantifier } + { \int_use:N \l_@@_internal_a_int } + { \int_use:N \l_@@_internal_b_int } + \int_zero:N \l_@@_internal_b_int \else: - \int_sub:Nn \l_regex_internal_b_int \l_regex_internal_a_int + \int_sub:Nn \l_@@_internal_b_int \l_@@_internal_a_int \fi: - \exp_args:Noo \regex_compile_quantifier_lazyness:nnNN - { \int_use:N \l_regex_internal_a_int } - { \int_use:N \l_regex_internal_b_int } + \exp_args:Noo \@@_compile_quantifier_lazyness:nnNN + { \int_use:N \l_@@_internal_a_int } + { \int_use:N \l_@@_internal_b_int } } { - \regex_compile_quantifier_abort:xNN + \@@_compile_quantifier_abort:xNN { \c_lbrace_str - \int_use:N \l_regex_internal_a_int , - \int_use:N \l_regex_internal_b_int + \int_use:N \l_@@_internal_a_int , + \int_use:N \l_@@_internal_b_int } #1 #2 } @@ -2177,57 +2186,57 @@ % % \subsubsection{Raw characters} % -% \begin{macro}[int]{\regex_compile_raw_error:N} +% \begin{macro}[int]{\@@_compile_raw_error:N} % Within character classes, and following catcode tests, some escaped % alphanumeric sequences such as |\b| do not have any meaning. They % are replaced by a raw character, after spitting out an error. % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_raw_error:N #1 +\cs_new_protected:Npn \@@_compile_raw_error:N #1 { - \msg_kernel_error:nnx { regex } { bad-escape } {#1} - \regex_compile_raw:N #1 + \__msg_kernel_error:nnx { regex } { bad-escape } {#1} + \@@_compile_raw:N #1 } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_compile_raw:N} +% \begin{macro}[int]{\@@_compile_raw:N} % If we are in a character class and the next character is an % unescaped dash, this denotes a range. Otherwise, the current % character |#1| matches itself. % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_raw:N #1#2#3 +\cs_new_protected:Npn \@@_compile_raw:N #1#2#3 { - \regex_if_in_class:TF + \@@_if_in_class:TF { - \str_if_eq:nnTF {#2#3} { \regex_compile_special:N - } - { \regex_compile_range:Nw #1 } + \str_if_eq:nnTF {#2#3} { \@@_compile_special:N - } + { \@@_compile_range:Nw #1 } { - \regex_compile_one:x - { \regex_item_equal:n { \int_value:w `#1 ~ } } + \@@_compile_one:x + { \@@_item_equal:n { \__int_value:w `#1 ~ } } #2 #3 } } { - \regex_compile_one:x - { \regex_item_equal:n { \int_value:w `#1 ~ } } + \@@_compile_one:x + { \@@_item_equal:n { \__int_value:w `#1 ~ } } #2 #3 } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_compile_range:Nw, \regex_if_end_range:NNTF} +% \begin{macro}[aux]{\@@_compile_range:Nw, \@@_if_end_range:NNTF} % We have just read a raw character followed by a dash; this should be % followed by an end-point for the range. Valid end-points are: any % raw character; any special character, except a right bracket. In % particular, escaped characters are forbidden. % \begin{macrocode} -\prg_new_protected_conditional:Npnn \regex_if_end_range:NN #1#2 { TF } +\prg_new_protected_conditional:Npnn \@@_if_end_range:NN #1#2 { TF } { - \if_meaning:w \regex_compile_raw:N #1 + \if_meaning:w \@@_compile_raw:N #1 \prg_return_true: \else: - \if_meaning:w \regex_compile_special:N #1 + \if_meaning:w \@@_compile_special:N #1 \if_charcode:w ] #2 \prg_return_false: \else: @@ -2238,31 +2247,31 @@ \fi: \fi: } -\cs_new_protected:Npn \regex_compile_range:Nw #1#2#3 +\cs_new_protected:Npn \@@_compile_range:Nw #1#2#3 { - \regex_if_end_range:NNTF #2 #3 + \@@_if_end_range:NNTF #2 #3 { - \if_num:w `#1 > `#3 \exp_stop_f: - \msg_kernel_error:nnxx { regex } { range-backwards } {#1} {#3} + \if_int_compare:w `#1 > `#3 \exp_stop_f: + \__msg_kernel_error:nnxx { regex } { range-backwards } {#1} {#3} \else: - \tl_build_one:x + \__tl_build_one:x { - \if_num:w `#1 = `#3 \exp_stop_f: - \regex_item_equal:n + \if_int_compare:w `#1 = `#3 \exp_stop_f: + \@@_item_equal:n \else: - \regex_item_range:nn { \int_value:w `#1 ~ } + \@@_item_range:nn { \__int_value:w `#1 ~ } \fi: - { \int_value:w `#3 ~ } + { \__int_value:w `#3 ~ } } \fi: } { - \msg_kernel_warning:nnxx { regex } { range-missing-end } + \__msg_kernel_warning:nnxx { regex } { range-missing-end } {#1} { \c_backslash_str #3 } - \tl_build_one:x + \__tl_build_one:x { - \regex_item_equal:n { \int_value:w `#1 ~ } - \regex_item_equal:n { \int_value:w `- ~ } + \@@_item_equal:n { \__int_value:w `#1 ~ } + \@@_item_equal:n { \__int_value:w `- ~ } } #2#3 } @@ -2272,21 +2281,21 @@ % % \subsubsection{Character properties} % -% \begin{macro}[aux]{\regex_compile_.:, \regex_prop_.:} +% \begin{macro}[aux]{\@@_compile_.:, \@@_prop_.:} % In a class, the dot has no special meaning. Outside, insert -% \cs{regex_prop_.:}, which matches any character or control +% \cs{@@_prop_.:}, which matches any character or control % sequence, and refuses $-2$ (end-marker). % \begin{macrocode} -\cs_new_protected_nopar:cpx { regex_compile_.: } +\cs_new_protected_nopar:cpx { @@_compile_.: } { - \exp_not:N \regex_if_in_class:TF - { \regex_compile_raw:N . } - { \regex_compile_one:x \exp_not:c { regex_prop_.: } } + \exp_not:N \@@_if_in_class:TF + { \@@_compile_raw:N . } + { \@@_compile_one:x \exp_not:c { @@_prop_.: } } } -\cs_new_protected_nopar:cpn { regex_prop_.: } +\cs_new_protected_nopar:cpn { @@_prop_.: } { - \if_num:w \l_regex_current_char_int > - \c_two - \exp_after:wN \regex_break_true:w + \if_int_compare:w \l_@@_current_char_int > - \c_two + \exp_after:wN \@@_break_true:w \fi: } % \end{macrocode} @@ -2294,75 +2303,75 @@ % % \begin{macro}[aux] % { -% \regex_compile_/d:, \regex_compile_/D:, -% \regex_compile_/h:, \regex_compile_/H:, -% \regex_compile_/s:, \regex_compile_/S:, -% \regex_compile_/v:, \regex_compile_/V:, -% \regex_compile_/w:, \regex_compile_/W:, -% \regex_compile_/N:, +% \@@_compile_/d:, \@@_compile_/D:, +% \@@_compile_/h:, \@@_compile_/H:, +% \@@_compile_/s:, \@@_compile_/S:, +% \@@_compile_/v:, \@@_compile_/V:, +% \@@_compile_/w:, \@@_compile_/W:, +% \@@_compile_/N:, % } -% The constants \cs{regex_prop_d:}, \emph{etc.} hold +% The constants \cs{@@_prop_d:}, \emph{etc.} hold % a list of tests which match the corresponding character -% class, and jump to the \cs{regex_break_point:TF} marker. +% class, and jump to the \cs{@@_break_point:TF} marker. % As for a normal character, we check for quantifiers. % \begin{macrocode} -\cs_set_protected:Npn \regex_tmp:w #1#2 +\cs_set_protected:Npn \@@_tmp:w #1#2 { - \cs_new_protected_nopar:cpx { regex_compile_/#1: } - { \regex_compile_one:x \exp_not:c { regex_prop_#1: } } - \cs_new_protected_nopar:cpx { regex_compile_/#2: } + \cs_new_protected_nopar:cpx { @@_compile_/#1: } + { \@@_compile_one:x \exp_not:c { @@_prop_#1: } } + \cs_new_protected_nopar:cpx { @@_compile_/#2: } { - \regex_compile_one:x - { \regex_item_reverse:n \exp_not:c { regex_prop_#1: } } + \@@_compile_one:x + { \@@_item_reverse:n \exp_not:c { @@_prop_#1: } } } } -\regex_tmp:w d D -\regex_tmp:w h H -\regex_tmp:w s S -\regex_tmp:w v V -\regex_tmp:w w W -\cs_new_protected_nopar:cpn { regex_compile_/N: } - { \regex_compile_one:x \regex_prop_N: } +\@@_tmp:w d D +\@@_tmp:w h H +\@@_tmp:w s S +\@@_tmp:w v V +\@@_tmp:w w W +\cs_new_protected_nopar:cpn { @@_compile_/N: } + { \@@_compile_one:x \@@_prop_N: } % \end{macrocode} % \end{macro} % % \subsubsection{Anchoring and simple assertions} % -% \begin{macro}[aux]{\regex_compile_anchor:NF} -% \begin{macro}[aux]+\regex_compile_^:+ -% \begin{macro}[aux]{\regex_compile_/A:, \regex_compile_/G:} -% \begin{macro}[aux]+\regex_compile_$:+ -% \begin{macro}[aux]{\regex_compile_/Z:, \regex_compile_/z:} +% \begin{macro}[aux]{\@@_compile_anchor:NF} +% \begin{macro}[aux]+\@@_compile_^:+ +% \begin{macro}[aux]{\@@_compile_/A:, \@@_compile_/G:} +% \begin{macro}[aux]+\@@_compile_$:+ +% \begin{macro}[aux]{\@@_compile_/Z:, \@@_compile_/z:} % In modes where assertions are allowed, anchor to the start of the % query, the start of the match, or the end of the query, depending on % the integer |#1|. In other modes, |#2| treats the character as raw, % with an error for escaped letters (|$| is valid in a class, but |\A| % is definitely a mistake on the user's part). % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_anchor:NF #1#2 +\cs_new_protected:Npn \@@_compile_anchor:NF #1#2 { - \regex_if_in_class_or_catcode:TF {#2} + \@@_if_in_class_or_catcode:TF {#2} { - \tl_build_one:n - { \regex_assertion:Nn \c_true_bool { \regex_anchor:N #1 } } + \__tl_build_one:n + { \@@_assertion:Nn \c_true_bool { \@@_anchor:N #1 } } } } -\cs_set_protected:Npn \regex_tmp:w #1#2 +\cs_set_protected:Npn \@@_tmp:w #1#2 { - \cs_new_protected_nopar:cpn { regex_compile_/#1: } - { \regex_compile_anchor:NF #2 { \regex_compile_raw_error:N #1 } } + \cs_new_protected_nopar:cpn { @@_compile_/#1: } + { \@@_compile_anchor:NF #2 { \@@_compile_raw_error:N #1 } } } -\regex_tmp:w A \l_regex_min_pos_int -\regex_tmp:w G \l_regex_start_pos_int -\regex_tmp:w Z \l_regex_max_pos_int -\regex_tmp:w z \l_regex_max_pos_int -\cs_set_protected:Npn \regex_tmp:w #1#2 +\@@_tmp:w A \l_@@_min_pos_int +\@@_tmp:w G \l_@@_start_pos_int +\@@_tmp:w Z \l_@@_max_pos_int +\@@_tmp:w z \l_@@_max_pos_int +\cs_set_protected:Npn \@@_tmp:w #1#2 { - \cs_new_protected_nopar:cpn { regex_compile_#1: } - { \regex_compile_anchor:NF #2 { \regex_compile_raw:N #1 } } + \cs_new_protected_nopar:cpn { @@_compile_#1: } + { \@@_compile_anchor:NF #2 { \@@_compile_raw:N #1 } } } -\exp_args:Nx \regex_tmp:w { \iow_char:N \^ } \l_regex_min_pos_int -\exp_args:Nx \regex_tmp:w { \iow_char:N \$ } \l_regex_max_pos_int +\exp_args:Nx \@@_tmp:w { \iow_char:N \^ } \l_@@_min_pos_int +\exp_args:Nx \@@_tmp:w { \iow_char:N \$ } \l_@@_max_pos_int % \end{macrocode} % \end{macro} % \end{macro} @@ -2370,27 +2379,27 @@ % \end{macro} % \end{macro} % -% \begin{macro}[aux]{\regex_compile_/b:, \regex_compile_/B:} +% \begin{macro}[aux]{\@@_compile_/b:, \@@_compile_/B:} % Contrarily to |^| and |$|, which could be implemented without really % knowing what precedes in the token list, this requires more % information, namely, the knowledge of the last character code. % \begin{macrocode} -\cs_new_protected_nopar:cpn { regex_compile_/b: } +\cs_new_protected_nopar:cpn { @@_compile_/b: } { - \regex_if_in_class_or_catcode:TF - { \regex_compile_raw_error:N b } + \@@_if_in_class_or_catcode:TF + { \@@_compile_raw_error:N b } { - \tl_build_one:n - { \regex_assertion:Nn \c_true_bool { \regex_b_test: } } + \__tl_build_one:n + { \@@_assertion:Nn \c_true_bool { \@@_b_test: } } } } -\cs_new_protected_nopar:cpn { regex_compile_/B: } +\cs_new_protected_nopar:cpn { @@_compile_/B: } { - \regex_if_in_class_or_catcode:TF - { \regex_compile_raw_error:N B } + \@@_if_in_class_or_catcode:TF + { \@@_compile_raw_error:N B } { - \tl_build_one:n - { \regex_assertion:Nn \c_false_bool { \regex_b_test: } } + \__tl_build_one:n + { \@@_assertion:Nn \c_false_bool { \@@_b_test: } } } } % \end{macrocode} @@ -2398,116 +2407,116 @@ % % \subsubsection{Character classes} % -% \begin{macro}[aux]{\regex_compile_]:} +% \begin{macro}[aux]{\@@_compile_]:} % Outside a class, right brackets have no meaning. In a class, change % the mode ($m\to (m-15)/13$, truncated) to reflect the fact that we % are leaving the class. Look for quantifiers, unless we are still in % a class after leaving one (the case of |[...\cL[...]...]|). % quantifiers. % \begin{macrocode} -\cs_new_protected:cpn { regex_compile_]: } +\cs_new_protected:cpn { @@_compile_]: } { - \regex_if_in_class:TF + \@@_if_in_class:TF { - \if_num:w \l_regex_mode_int > \c_sixteen - \tl_build_one:n { \if_false: { \fi: } } + \if_int_compare:w \l_@@_mode_int > \c_sixteen + \__tl_build_one:n { \if_false: { \fi: } } \fi: - \tex_advance:D \l_regex_mode_int - \c_fifteen - \tex_divide:D \l_regex_mode_int \c_thirteen - \if_int_odd:w \l_regex_mode_int \else: - \exp_after:wN \regex_compile_quantifier:w + \tex_advance:D \l_@@_mode_int - \c_fifteen + \tex_divide:D \l_@@_mode_int \c_thirteen + \if_int_odd:w \l_@@_mode_int \else: + \exp_after:wN \@@_compile_quantifier:w \fi: } - { \regex_compile_raw:N ] } + { \@@_compile_raw:N ] } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_compile_[:} +% \begin{macro}[aux]{\@@_compile_[:} % In a class, left brackets might introduce a \textsc{posix} character % class, or mean nothing. Immediately following |\c|\meta{category}, % we must insert the apropriate catcode test, then parse the class; we % pre-expand the catcode as an optimization. Otherwise (modes $0$, % $-2$ and $-6$) just parse the class. The mode is updated later. % \begin{macrocode} -\cs_new_protected_nopar:cpn { regex_compile_[: } +\cs_new_protected_nopar:cpn { @@_compile_[: } { - \regex_if_in_class:TF - { \regex_compile_class_posix_test:w } + \@@_if_in_class:TF + { \@@_compile_class_posix_test:w } { - \regex_if_within_catcode:TF + \@@_if_within_catcode:TF { - \exp_after:wN \regex_compile_class_catcode:w - \int_use:N \l_regex_catcodes_int ; + \exp_after:wN \@@_compile_class_catcode:w + \int_use:N \l_@@_catcodes_int ; } - { \regex_compile_class_normal:w } + { \@@_compile_class_normal:w } } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_compile_class_normal:w} -% In the \enquote{normal} case, we will insert \cs{regex_class:NnnnN} +% \begin{macro}[aux]{\@@_compile_class_normal:w} +% In the \enquote{normal} case, we will insert \cs{@@_class:NnnnN} % \meta{boolean} in the compiled code. The \meta{boolean} is true for % positive classes, and false for negative classes, characterized by a -% leading |^|. The auxiliary \cs{regex_compile_class:TFNN} also +% leading |^|. The auxiliary \cs{@@_compile_class:TFNN} also % checks for a leading |]| which has a special meaning. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_compile_class_normal:w +\cs_new_protected_nopar:Npn \@@_compile_class_normal:w { - \regex_compile_class:TFNN - { \regex_class:NnnnN \c_true_bool } - { \regex_class:NnnnN \c_false_bool } + \@@_compile_class:TFNN + { \@@_class:NnnnN \c_true_bool } + { \@@_class:NnnnN \c_false_bool } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_compile_class_catcode:w} +% \begin{macro}[aux]{\@@_compile_class_catcode:w} % This function is called for a left bracket in modes $2$ or $6$ % (catcode test, and catcode test within a class). In mode $2$ the % whole construction needs to be put in a class (like single % character). Then determine if the class is positive or negative, -% inserting \cs{regex_item_catcode:nT} or the \texttt{reverse} variant +% inserting \cs{@@_item_catcode:nT} or the \texttt{reverse} variant % as appropriate, each with the current catcodes bitmap |#1| as an % argument, and reset the catcodes. % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_class_catcode:w #1; +\cs_new_protected:Npn \@@_compile_class_catcode:w #1; { - \if_int_compare:w \l_regex_mode_int = \c_two - \tl_build_one:n - { \regex_class:NnnnN \c_true_bool { \if_false: } \fi: } + \if_int_compare:w \l_@@_mode_int = \c_two + \__tl_build_one:n + { \@@_class:NnnnN \c_true_bool { \if_false: } \fi: } \fi: - \int_set_eq:NN \l_regex_catcodes_int \l_regex_default_catcodes_int - \regex_compile_class:TFNN - { \regex_item_catcode:nT {#1} } - { \regex_item_catcode_reverse:nT {#1} } + \int_set_eq:NN \l_@@_catcodes_int \l_@@_default_catcodes_int + \@@_compile_class:TFNN + { \@@_item_catcode:nT {#1} } + { \@@_item_catcode_reverse:nT {#1} } } % \end{macrocode} % \end{macro} % % \begin{macro}[aux] -% {\regex_compile_class:TFNN, \regex_compile_class_ii:NN} +% {\@@_compile_class:TFNN, \@@_compile_class_ii:NN} % If the first character is |^|, then the class is negative (use % |#2|), otherwise it is positive (use |#1|). If the next character % is a right bracket, then it should be changed to a raw one. % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_class:TFNN #1#2#3#4 +\cs_new_protected:Npn \@@_compile_class:TFNN #1#2#3#4 { - \l_regex_mode_int = \int_value:w \l_regex_mode_int 3 \exp_stop_f: - \str_if_eq:nnTF { #3 #4 } { \regex_compile_special:N ^ } + \l_@@_mode_int = \__int_value:w \l_@@_mode_int 3 \exp_stop_f: + \str_if_eq:nnTF { #3 #4 } { \@@_compile_special:N ^ } { - \tl_build_one:n { #2 { \if_false: } \fi: } - \regex_compile_class_ii:NN + \__tl_build_one:n { #2 { \if_false: } \fi: } + \@@_compile_class_ii:NN } { - \tl_build_one:n { #1 { \if_false: } \fi: } - \regex_compile_class_ii:NN #3 #4 + \__tl_build_one:n { #1 { \if_false: } \fi: } + \@@_compile_class_ii:NN #3 #4 } } -\cs_new_protected:Npn \regex_compile_class_ii:NN #1#2 +\cs_new_protected:Npn \@@_compile_class_ii:NN #1#2 { \token_if_eq_charcode:NNTF #2 ] - { \regex_compile_raw:N #2 } + { \@@_compile_raw:N #2 } { #1 #2 } } % \end{macrocode} @@ -2515,80 +2524,80 @@ % % \begin{macro}[aux] % { -% \regex_compile_class_posix_test:w, -% \regex_compile_class_posix:NNNNw, -% \regex_compile_class_posix_loop:w, -% \regex_compile_class_posix_end:w +% \@@_compile_class_posix_test:w, +% \@@_compile_class_posix:NNNNw, +% \@@_compile_class_posix_loop:w, +% \@@_compile_class_posix_end:w % } % Here we check for a syntax such as |[:alpha:]|. We also detect |[=| % and |[.| which have a meaning in \textsc{posix} regular expressions, % but are not implemented in \pkg{l3regex}. In case we see |[:|, grab % raw characters until hopefully reaching |:]|. If that's missing, or % the \textsc{posix} class is unknown, abort. If all is right, add the -% test to the current class, with an extra \cs{regex_item_reverse:n} +% test to the current class, with an extra \cs{@@_item_reverse:n} % for negative classes. % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_class_posix_test:w #1#2 +\cs_new_protected:Npn \@@_compile_class_posix_test:w #1#2 { - \token_if_eq_meaning:NNT \regex_compile_special:N #1 + \token_if_eq_meaning:NNT \@@_compile_special:N #1 { - \prg_case_str:nnn { #2 } + \str_case:nnn { #2 } { - : { \regex_compile_class_posix:NNNNw } - = { \msg_kernel_warning:nnx { regex } { posix-unsupported } { = } } - . { \msg_kernel_warning:nnx { regex } { posix-unsupported } { . } } + : { \@@_compile_class_posix:NNNNw } + = { \__msg_kernel_warning:nnx { regex } { posix-unsupported } { = } } + . { \__msg_kernel_warning:nnx { regex } { posix-unsupported } { . } } } { } } - \regex_compile_raw:N [ #1 #2 + \@@_compile_raw:N [ #1 #2 } -\cs_new_protected:Npn \regex_compile_class_posix:NNNNw #1#2#3#4#5#6 +\cs_new_protected:Npn \@@_compile_class_posix:NNNNw #1#2#3#4#5#6 { - \str_if_eq:nnTF { #5 #6 } { \regex_compile_special:N ^ } + \str_if_eq:nnTF { #5 #6 } { \@@_compile_special:N ^ } { - \bool_set_false:N \l_regex_internal_bool - \tl_set:Nx \l_regex_internal_a_tl { \if_false: } \fi: - \regex_compile_class_posix_loop:w + \bool_set_false:N \l_@@_internal_bool + \tl_set:Nx \l_@@_internal_a_tl { \if_false: } \fi: + \@@_compile_class_posix_loop:w } { - \bool_set_true:N \l_regex_internal_bool - \tl_set:Nx \l_regex_internal_a_tl { \if_false: } \fi: - \regex_compile_class_posix_loop:w #5 #6 + \bool_set_true:N \l_@@_internal_bool + \tl_set:Nx \l_@@_internal_a_tl { \if_false: } \fi: + \@@_compile_class_posix_loop:w #5 #6 } } -\cs_new:Npn \regex_compile_class_posix_loop:w #1#2 +\cs_new:Npn \@@_compile_class_posix_loop:w #1#2 { - \token_if_eq_meaning:NNTF \regex_compile_raw:N #1 - { #2 \regex_compile_class_posix_loop:w } - { \if_false: { \fi: } \regex_compile_class_posix_end:w #1 #2 } + \token_if_eq_meaning:NNTF \@@_compile_raw:N #1 + { #2 \@@_compile_class_posix_loop:w } + { \if_false: { \fi: } \@@_compile_class_posix_end:w #1 #2 } } -\cs_new_protected:Npn \regex_compile_class_posix_end:w #1#2#3#4 +\cs_new_protected:Npn \@@_compile_class_posix_end:w #1#2#3#4 { \str_if_eq:nnTF { #1 #2 #3 #4 } - { \regex_compile_special:N : \regex_compile_special:N ] } + { \@@_compile_special:N : \@@_compile_special:N ] } { - \cs_if_exist:cTF { regex_posix_ \l_regex_internal_a_tl : } + \cs_if_exist:cTF { @@_posix_ \l_@@_internal_a_tl : } { - \regex_compile_one:x + \@@_compile_one:x { - \bool_if:NF \l_regex_internal_bool \regex_item_reverse:n - \exp_not:c { regex_posix_ \l_regex_internal_a_tl : } + \bool_if:NF \l_@@_internal_bool \@@_item_reverse:n + \exp_not:c { @@_posix_ \l_@@_internal_a_tl : } } } { - \msg_kernel_warning:nnx { regex } { posix-unknown } - { \l_regex_internal_a_tl } - \regex_compile_abort_tokens:x + \__msg_kernel_warning:nnx { regex } { posix-unknown } + { \l_@@_internal_a_tl } + \@@_compile_abort_tokens:x { - [: \bool_if:NF \l_regex_internal_bool { ^ } - \l_regex_internal_a_tl :] + [: \bool_if:NF \l_@@_internal_bool { ^ } + \l_@@_internal_a_tl :] } } } { - \msg_kernel_error:nnxx { regex } { posix-missing-close } - { [: \l_regex_internal_a_tl } { #2 #4 } - \regex_compile_abort_tokens:x { [: \l_regex_internal_a_tl } + \__msg_kernel_error:nnxx { regex } { posix-missing-close } + { [: \l_@@_internal_a_tl } { #2 #4 } + \@@_compile_abort_tokens:x { [: \l_@@_internal_a_tl } #1 #2 #3 #4 } } @@ -2597,149 +2606,149 @@ % % \subsubsection{Groups and alternations} % -% \begin{macro}[aux]{\regex_compile_group_begin:N, \regex_compile_group_end:} +% \begin{macro}[aux]{\@@_compile_group_begin:N, \@@_compile_group_end:} % The contents of a regex group are turned into compiled code in -% \cs{l_regex_internal_regex}, which ends up with items of the form -% \cs{regex_branch:n} \Arg{concatenation}. This construction is done +% \cs{l_@@_internal_regex}, which ends up with items of the form +% \cs{@@_branch:n} \Arg{concatenation}. This construction is done % using \pkg{l3tl-build} within a \TeX{} group, which automatically % makes sure that options (case-sensitivity and default catcode) are % reset at the end of the group. The argument |#1| is -% \cs{regex_group:nnnN} or a variant thereof. A small subtelty to +% \cs{@@_group:nnnN} or a variant thereof. A small subtelty to % support |\cL(abc)| as a shorthand for |(\cLa\cLb\cLc)|: exit any % pending catcode test, save the category code at the start of the % group as the default catcode for that group, and make sure that the % catcode is restored to the default outside the group. % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_group_begin:N #1 +\cs_new_protected:Npn \@@_compile_group_begin:N #1 { - \tl_build_one:n { #1 { \if_false: } \fi: } - \regex_mode_quit_c: - \tl_set_build:Nw \l_regex_internal_regex - \int_set_eq:NN \l_regex_default_catcodes_int \l_regex_catcodes_int - \int_incr:N \l_regex_group_level_int - \tl_build_one:n { \regex_branch:n { \if_false: } \fi: } + \__tl_build_one:n { #1 { \if_false: } \fi: } + \@@_mode_quit_c: + \__tl_build:Nw \l_@@_internal_regex + \int_set_eq:NN \l_@@_default_catcodes_int \l_@@_catcodes_int + \int_incr:N \l_@@_group_level_int + \__tl_build_one:n { \@@_branch:n { \if_false: } \fi: } } -\cs_new_protected:Npn \regex_compile_group_end: +\cs_new_protected:Npn \@@_compile_group_end: { - \if_num:w \l_regex_group_level_int > \c_zero - \tl_build_one:n { \if_false: { \fi: } } - \tl_build_end: - \int_set_eq:NN \l_regex_catcodes_int \l_regex_default_catcodes_int - \tl_build_one:o \l_regex_internal_regex - \exp_after:wN \regex_compile_quantifier:w + \if_int_compare:w \l_@@_group_level_int > \c_zero + \__tl_build_one:n { \if_false: { \fi: } } + \__tl_build_end: + \int_set_eq:NN \l_@@_catcodes_int \l_@@_default_catcodes_int + \__tl_build_one:o \l_@@_internal_regex + \exp_after:wN \@@_compile_quantifier:w \else: - \msg_kernel_warning:nn { regex } { extra-rparen } - \exp_after:wN \regex_compile_raw:N \exp_after:wN ) + \__msg_kernel_warning:nn { regex } { extra-rparen } + \exp_after:wN \@@_compile_raw:N \exp_after:wN ) \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_compile_(:} +% \begin{macro}[aux]{\@@_compile_(:} % In a class, parentheses are not special. Outside, check for a |?|, % denoting special groups, and run the code for the corresponding % special group. % \begin{macrocode} -\cs_new_protected_nopar:cpn { regex_compile_(: } +\cs_new_protected_nopar:cpn { @@_compile_(: } { - \regex_if_in_class:TF { \regex_compile_raw:N ( } - { \regex_compile_lparen:w } + \@@_if_in_class:TF { \@@_compile_raw:N ( } + { \@@_compile_lparen:w } } -\cs_new_protected:Npn \regex_compile_lparen:w #1#2#3#4 +\cs_new_protected:Npn \@@_compile_lparen:w #1#2#3#4 { - \str_if_eq:nnTF { #1 #2 } { \regex_compile_special:N ? } + \str_if_eq:nnTF { #1 #2 } { \@@_compile_special:N ? } { \cs_if_exist_use:cF - { regex_compile_special_group_\token_to_str:N #4 :w } + { @@_compile_special_group_\token_to_str:N #4 :w } { - \msg_kernel_warning:nnx { regex } { special-group-unknown } + \__msg_kernel_warning:nnx { regex } { special-group-unknown } { (? \token_to_str:N #4 } - \regex_compile_group_begin:N \regex_group:nnnN - \regex_compile_raw:N ? #3 #4 + \@@_compile_group_begin:N \@@_group:nnnN + \@@_compile_raw:N ? #3 #4 } } { - \regex_compile_group_begin:N \regex_group:nnnN + \@@_compile_group_begin:N \@@_group:nnnN #1 #2 #3 #4 } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]+\regex_compile_|:+ +% \begin{macro}[aux]+\@@_compile_|:+ % In a class, the pipe is not special. Otherwise, end the current % branch and open another one. % \begin{macrocode} -\cs_new_protected_nopar:cpn { regex_compile_|: } +\cs_new_protected_nopar:cpn { @@_compile_|: } { - \regex_if_in_class:TF { \regex_compile_raw:N | } + \@@_if_in_class:TF { \@@_compile_raw:N | } { - \tl_build_one:n - { \if_false: { \fi: } \regex_branch:n { \if_false: } \fi: } + \__tl_build_one:n + { \if_false: { \fi: } \@@_branch:n { \if_false: } \fi: } } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_compile_):} +% \begin{macro}[aux]{\@@_compile_):} % Within a class, parentheses are not special. Outside, close a group. % \begin{macrocode} -\cs_new_protected_nopar:cpn { regex_compile_): } +\cs_new_protected_nopar:cpn { @@_compile_): } { - \regex_if_in_class:TF { \regex_compile_raw:N ) } - { \regex_compile_group_end: } + \@@_if_in_class:TF { \@@_compile_raw:N ) } + { \@@_compile_group_end: } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_compile_special_group_::w} -% \begin{macro}[aux]+\regex_compile_special_group_|:w+ +% \begin{macro}[aux]{\@@_compile_special_group_::w} +% \begin{macro}[aux]+\@@_compile_special_group_|:w+ % Non-capturing, and resetting groups are easy to take care of during % compilation; for those groups, the harder parts will come when % building. % \begin{macrocode} -\cs_new_protected_nopar:cpn { regex_compile_special_group_::w } - { \regex_compile_group_begin:N \regex_group_no_capture:nnnN } -\cs_new_protected_nopar:cpn { regex_compile_special_group_|:w } - { \regex_compile_group_begin:N \regex_group_resetting:nnnN } +\cs_new_protected_nopar:cpn { @@_compile_special_group_::w } + { \@@_compile_group_begin:N \@@_group_no_capture:nnnN } +\cs_new_protected_nopar:cpn { @@_compile_special_group_|:w } + { \@@_compile_group_begin:N \@@_group_resetting:nnnN } % \end{macrocode} % \end{macro} % \end{macro} % % \begin{macro}[aux] -% {\regex_compile_special_group_i:w, \regex_compile_special_group_-:w} +% {\@@_compile_special_group_i:w, \@@_compile_special_group_-:w} % The match can be made case-insensitive by setting the option with % \texttt{(?i)}; the original behaviour is restored by \texttt{(?-i)}. % This is the only supported option. % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_special_group_i:w #1#2 +\cs_new_protected:Npn \@@_compile_special_group_i:w #1#2 { - \str_if_eq:nnTF { #1 #2 } { \regex_compile_special:N ) } + \str_if_eq:nnTF { #1 #2 } { \@@_compile_special:N ) } { - \cs_set_nopar:Npn \regex_item_equal:n { \regex_item_caseless_equal:n } - \cs_set_nopar:Npn \regex_item_range:nn { \regex_item_caseless_range:nn } + \cs_set_nopar:Npn \@@_item_equal:n { \@@_item_caseless_equal:n } + \cs_set_nopar:Npn \@@_item_range:nn { \@@_item_caseless_range:nn } } { - \msg_kernel_warning:nnx { regex } { unknown-option } { (?i #2 } - \regex_compile_raw:N ( - \regex_compile_raw:N ? - \regex_compile_raw:N i + \__msg_kernel_warning:nnx { regex } { unknown-option } { (?i #2 } + \@@_compile_raw:N ( + \@@_compile_raw:N ? + \@@_compile_raw:N i #1 #2 } } -\cs_new_protected_nopar:cpn { regex_compile_special_group_-:w } #1#2#3#4 +\cs_new_protected_nopar:cpn { @@_compile_special_group_-:w } #1#2#3#4 { \str_if_eq:nnTF { #1 #2 #3 #4 } - { \regex_compile_raw:N i \regex_compile_special:N ) } + { \@@_compile_raw:N i \@@_compile_special:N ) } { - \cs_set_nopar:Npn \regex_item_equal:n { \regex_item_caseful_equal:n } - \cs_set_nopar:Npn \regex_item_range:nn { \regex_item_caseful_range:nn } + \cs_set_nopar:Npn \@@_item_equal:n { \@@_item_caseful_equal:n } + \cs_set_nopar:Npn \@@_item_range:nn { \@@_item_caseful_range:nn } } { - \msg_kernel_warning:nnx { regex } { unknown-option } { (?-#2#4 } - \regex_compile_raw:N ( - \regex_compile_raw:N ? - \regex_compile_raw:N - + \__msg_kernel_warning:nnx { regex } { unknown-option } { (?-#2#4 } + \@@_compile_raw:N ( + \@@_compile_raw:N ? + \@@_compile_raw:N - #1 #2 #3 #4 } } @@ -2748,28 +2757,28 @@ % % \subsubsection{Catcodes and csnames} % -% \begin{macro}[aux]{\regex_compile_/c:, \regex_compile_c_test:NN} +% \begin{macro}[aux]{\@@_compile_/c:, \@@_compile_c_test:NN} % The |\c| escape sequence can be followed by a capital letter % representing a character category, by a left bracket which starts a % list of categories, or by a brace group holding a regular expression % for a control sequence name. Otherwise, raise an error. % \begin{macrocode} -\cs_new_protected:cpn { regex_compile_/c: } - { \regex_chk_c_allowed:T { \regex_compile_c_test:NN } } -\cs_new_protected:Npn \regex_compile_c_test:NN #1#2 +\cs_new_protected:cpn { @@_compile_/c: } + { \@@_chk_c_allowed:T { \@@_compile_c_test:NN } } +\cs_new_protected:Npn \@@_compile_c_test:NN #1#2 { - \token_if_eq_meaning:NNTF #1 \regex_compile_raw:N + \token_if_eq_meaning:NNTF #1 \@@_compile_raw:N { - \int_if_exist:cTF { c_regex_catcode_#2_int } + \int_if_exist:cTF { c_@@_catcode_#2_int } { - \int_set_eq:Nc \l_regex_catcodes_int { c_regex_catcode_#2_int } - \l_regex_mode_int - = \if_case:w \l_regex_mode_int \c_two \else: \c_six \fi: + \int_set_eq:Nc \l_@@_catcodes_int { c_@@_catcode_#2_int } + \l_@@_mode_int + = \if_case:w \l_@@_mode_int \c_two \else: \c_six \fi: } } - { \cs_if_exist_use:cF { regex_compile_c_#2:w } } + { \cs_if_exist_use:cF { @@_compile_c_#2:w } } { - \msg_kernel_error:nnx { regex } { c-missing-category } {#2} + \__msg_kernel_error:nnx { regex } { c-missing-category } {#2} #1 #2 } } @@ -2778,86 +2787,86 @@ % % \begin{macro}[aux] % { -% \regex_compile_c_[:w, -% \regex_compile_c_lbrack_loop:NN, -% \regex_compile_c_lbrack_add:N, -% \regex_compile_c_lbrack_end:, +% \@@_compile_c_[:w, +% \@@_compile_c_lbrack_loop:NN, +% \@@_compile_c_lbrack_add:N, +% \@@_compile_c_lbrack_end:, % } % When encountering |\c[|, the task is to collect uppercase letters % representing character categories. First check for |^| which negates % the list of category codes. % \begin{macrocode} -\cs_new_protected:cpn { regex_compile_c_[:w } #1#2 +\cs_new_protected:cpn { @@_compile_c_[:w } #1#2 { - \l_regex_mode_int - = \if_case:w \l_regex_mode_int \c_two \else: \c_six \fi: - \int_zero:N \l_regex_catcodes_int - \str_if_eq:nnTF { #1 #2 } { \regex_compile_special:N ^ } + \l_@@_mode_int + = \if_case:w \l_@@_mode_int \c_two \else: \c_six \fi: + \int_zero:N \l_@@_catcodes_int + \str_if_eq:nnTF { #1 #2 } { \@@_compile_special:N ^ } { - \bool_set_false:N \l_regex_catcodes_bool - \regex_compile_c_lbrack_loop:NN + \bool_set_false:N \l_@@_catcodes_bool + \@@_compile_c_lbrack_loop:NN } { - \bool_set_true:N \l_regex_catcodes_bool - \regex_compile_c_lbrack_loop:NN + \bool_set_true:N \l_@@_catcodes_bool + \@@_compile_c_lbrack_loop:NN #1 #2 } } -\cs_new_protected:Npn \regex_compile_c_lbrack_loop:NN #1#2 +\cs_new_protected:Npn \@@_compile_c_lbrack_loop:NN #1#2 { - \token_if_eq_meaning:NNTF #1 \regex_compile_raw:N + \token_if_eq_meaning:NNTF #1 \@@_compile_raw:N { - \int_if_exist:cTF { c_regex_catcode_#2_int } + \int_if_exist:cTF { c_@@_catcode_#2_int } { - \exp_args:Nc \regex_compile_c_lbrack_add:N - { c_regex_catcode_#2_int } - \regex_compile_c_lbrack_loop:NN + \exp_args:Nc \@@_compile_c_lbrack_add:N + { c_@@_catcode_#2_int } + \@@_compile_c_lbrack_loop:NN } } { \token_if_eq_charcode:NNTF #2 ] - { \regex_compile_c_lbrack_end: } + { \@@_compile_c_lbrack_end: } } { - \msg_kernel_error:nnx { regex } { c-missing-rbrack } {#2} - \regex_compile_c_lbrack_end: + \__msg_kernel_error:nnx { regex } { c-missing-rbrack } {#2} + \@@_compile_c_lbrack_end: #1 #2 } } -\cs_new_protected:Npn \regex_compile_c_lbrack_add:N #1 +\cs_new_protected:Npn \@@_compile_c_lbrack_add:N #1 { - \if_int_odd:w \int_eval:w \l_regex_catcodes_int / #1 \int_eval_end: + \if_int_odd:w \__int_eval:w \l_@@_catcodes_int / #1 \__int_eval_end: \else: - \tex_advance:D \l_regex_catcodes_int #1 + \tex_advance:D \l_@@_catcodes_int #1 \fi: } -\cs_new_protected_nopar:Npn \regex_compile_c_lbrack_end: +\cs_new_protected_nopar:Npn \@@_compile_c_lbrack_end: { - \if_meaning:w \c_false_bool \l_regex_catcodes_bool - \int_set:Nn \l_regex_catcodes_int - { \c_regex_all_catcodes_int - \l_regex_catcodes_int } + \if_meaning:w \c_false_bool \l_@@_catcodes_bool + \int_set:Nn \l_@@_catcodes_int + { \c_@@_all_catcodes_int - \l_@@_catcodes_int } \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}+\regex_compile_c_{:+ +% \begin{macro}+\@@_compile_c_{:+ % The case of a left brace is easy, based on what we have done so far: % in a group, compile the regular expression, after changing the mode % to forbid nesting |\c|. Additionally, disable submatch tracking % since groups don't escape the scope of |\c{...}|. % \begin{macrocode} -\cs_new_protected_nopar:cpn { regex_compile_c_ \c_lbrace_str :w } +\cs_new_protected_nopar:cpn { @@_compile_c_ \c_lbrace_str :w } { - \regex_compile:w - \regex_disable_submatches: - \l_regex_mode_int - = - \if_case:w \l_regex_mode_int \c_two \else: \c_six \fi: + \@@_compile:w + \@@_disable_submatches: + \l_@@_mode_int + = - \if_case:w \l_@@_mode_int \c_two \else: \c_six \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}+\regex_compile_}:+ +% \begin{macro}+\@@_compile_}:+ % Non-escaped right braces are only special if they appear when % compiling the regular expression for a csname, but not within a % class: |\c{[}{]}| matches the control sequences |\}| and @@ -2866,23 +2875,23 @@ % dangling class or group). Then insert the corresponding test in the % outer regex. % \begin{macrocode} -\cs_new_protected:cpn { regex_compile_ \c_rbrace_str : } +\cs_new_protected:cpn { @@_compile_ \c_rbrace_str : } { - \regex_if_in_cs:TF + \@@_if_in_cs:TF { - \regex_compile_end: - \regex_compile_one:x - { \regex_item_cs:n { \exp_not:o \l_regex_internal_regex } } + \@@_compile_end: + \@@_compile_one:x + { \@@_item_cs:n { \exp_not:o \l_@@_internal_regex } } } - { \exp_after:wN \regex_compile_raw:N \c_rbrace_str } + { \exp_after:wN \@@_compile_raw:N \c_rbrace_str } } % \end{macrocode} % \end{macro} % % \subsubsection{Raw token lists with \cs{u}} % -% \begin{macro}[aux]{\regex_compile_/u:} -% \begin{macro}[aux, EXP]{\regex_compile_u_loop:NN} +% \begin{macro}[aux]{\@@_compile_/u:} +% \begin{macro}[aux, EXP]{\@@_compile_u_loop:NN} % The |\u| escape is invalid in classes and directly following a % catcode test. Otherwise, it must be followed by a left brace. We % then collect the characters for the argument of |\u| within an @@ -2893,37 +2902,37 @@ % raw and special characters, and stop when encountering a special % right brace, any escaped character, or the end-marker. % \begin{macrocode} -\cs_new_protected:cpn { regex_compile_/u: } #1#2 +\cs_new_protected:cpn { @@_compile_/u: } #1#2 { - \regex_if_in_class_or_catcode:TF - { \regex_compile_raw_error:N u #1 #2 } + \@@_if_in_class_or_catcode:TF + { \@@_compile_raw_error:N u #1 #2 } { - \str_if_eq:xxTF {#1#2} { \regex_compile_special:N \c_lbrace_str } + \str_if_eq_x:nnTF {#1#2} { \@@_compile_special:N \c_lbrace_str } { - \tl_set:Nx \l_regex_internal_a_tl { \if_false: } \fi: - \regex_compile_u_loop:NN + \tl_set:Nx \l_@@_internal_a_tl { \if_false: } \fi: + \@@_compile_u_loop:NN } { - \msg_kernel_error:nn { regex } { u-missing-lbrace } - \regex_compile_raw:N u #1 #2 + \__msg_kernel_error:nn { regex } { u-missing-lbrace } + \@@_compile_raw:N u #1 #2 } } } -\cs_new:Npn \regex_compile_u_loop:NN #1#2 +\cs_new:Npn \@@_compile_u_loop:NN #1#2 { - \token_if_eq_meaning:NNTF #1 \regex_compile_raw:N - { #2 \regex_compile_u_loop:NN } + \token_if_eq_meaning:NNTF #1 \@@_compile_raw:N + { #2 \@@_compile_u_loop:NN } { - \token_if_eq_meaning:NNTF #1 \regex_compile_special:N + \token_if_eq_meaning:NNTF #1 \@@_compile_special:N { \exp_after:wN \token_if_eq_charcode:NNTF \c_rbrace_str #2 - { \if_false: { \fi: } \regex_compile_u_end: } - { #2 \regex_compile_u_loop:NN } + { \if_false: { \fi: } \@@_compile_u_end: } + { #2 \@@_compile_u_loop:NN } } { \if_false: { \fi: } - \msg_kernel_error:nnx { regex } { u-missing-rbrace } {#2} - \regex_compile_u_end: + \__msg_kernel_error:nnx { regex } { u-missing-rbrace } {#2} + \@@_compile_u_end: #1 #2 } } @@ -2932,66 +2941,66 @@ % \end{macro} % \end{macro} % -% \begin{macro}[aux]{\regex_compile_u_end:} +% \begin{macro}[aux]{\@@_compile_u_end:} % Once we have extracted the variable's name, we store the contents of -% that variable in \cs{l_regex_internal_a_tl}. The behaviour of |\u| +% that variable in \cs{l_@@_internal_a_tl}. The behaviour of |\u| % then depends on whether we are within a |\c{...}| escape (in this % case, the variable is turned to a string), or not. % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_u_end: +\cs_new_protected:Npn \@@_compile_u_end: { - \tl_set:Nv \l_regex_internal_a_tl { \l_regex_internal_a_tl } - \if_num:w \l_regex_mode_int = \c_zero - \regex_compile_u_not_cs: + \tl_set:Nv \l_@@_internal_a_tl { \l_@@_internal_a_tl } + \if_int_compare:w \l_@@_mode_int = \c_zero + \@@_compile_u_not_cs: \else: - \regex_compile_u_in_cs: + \@@_compile_u_in_cs: \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_compile_u_in_cs:} +% \begin{macro}[aux]{\@@_compile_u_in_cs:} % When |\u| appears within a control sequence, we convert the variable % to a string with escaped spaces. Then for each character insert a % class matching exactly that character, once. % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_u_in_cs: +\cs_new_protected:Npn \@@_compile_u_in_cs: { - \exp_args:NNo \str_gset_other:Nn \g_regex_internal_tl - { \l_regex_internal_a_tl } - \tl_build_one:x + \exp_args:NNo \__str_gset_other:Nn \g_@@_internal_tl + { \l_@@_internal_a_tl } + \__tl_build_one:x { - \tl_map_function:NN \g_regex_internal_tl - \regex_compile_u_in_cs_aux:n + \tl_map_function:NN \g_@@_internal_tl + \@@_compile_u_in_cs_aux:n } } -\cs_new:Npn \regex_compile_u_in_cs_aux:n #1 +\cs_new:Npn \@@_compile_u_in_cs_aux:n #1 { - \regex_class:NnnnN \c_true_bool - { \regex_item_caseful_equal:n { \int_value:w `#1 } } + \@@_class:NnnnN \c_true_bool + { \@@_item_caseful_equal:n { \__int_value:w `#1 } } { 1 } { 0 } \c_false_bool } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_compile_u_not_cs:} +% \begin{macro}[aux]{\@@_compile_u_not_cs:} % In mode $0$, the |\u| escape adds one state to the NFA for each -% token in \cs{l_regex_internal_a_tl}. If a given \meta{token} is a +% token in \cs{l_@@_internal_a_tl}. If a given \meta{token} is a % control sequence, then insert a string comparison test, otherwise, -% \cs{regex_item_exact:nn} which compares catcode and character code. +% \cs{@@_item_exact:nn} which compares catcode and character code. % \begin{macrocode} -\cs_new_protected:Npn \regex_compile_u_not_cs: +\cs_new_protected:Npn \@@_compile_u_not_cs: { - \exp_args:No \tl_analysis_map_inline:nn { \l_regex_internal_a_tl } + \exp_args:No \__tl_analysis_map_inline:nn { \l_@@_internal_a_tl } { - \tl_build_one:n + \__tl_build_one:n { - \regex_class:NnnnN \c_true_bool + \@@_class:NnnnN \c_true_bool { - \if_num:w "##2 = \c_zero - \regex_item_exact_cs:c { \exp_after:wN \cs_to_str:N ##1 } + \if_int_compare:w "##2 = \c_zero + \@@_item_exact_cs:c { \exp_after:wN \cs_to_str:N ##1 } \else: - \regex_item_exact:nn { \int_value:w "##2 } { ##3 } + \@@_item_exact:nn { \__int_value:w "##2 } { ##3 } \fi: } { 1 } { 0 } \c_false_bool @@ -3003,136 +3012,140 @@ % % \subsubsection{Other} % -% \begin{macro}[aux]{\regex_compile_/K:} +% \begin{macro}[aux]{\@@_compile_/K:} % The |\K| control sequence is currently the only \enquote{command}, % which performs some action, rather than matching something. It is % allowed in the same contexts as |\b|. At the compilation stage, we % leave it as a single control sequence, defined later. % \begin{macrocode} -\cs_new_protected_nopar:cpn { regex_compile_/K: } +\cs_new_protected_nopar:cpn { @@_compile_/K: } { - \int_compare:nNnTF \l_regex_mode_int = \c_zero - { \tl_build_one:n { \regex_command_K: } } - { \regex_compile_raw_error:N K } + \int_compare:nNnTF \l_@@_mode_int = \c_zero + { \__tl_build_one:n { \@@_command_K: } } + { \@@_compile_raw_error:N K } } % \end{macrocode} % \end{macro} % % \subsubsection{Showing regexes} % -% \begin{macro}[aux]{\regex_show_aux:Nx} -% Within a \cs{tl_set_build:Nw} \ldots{} \cs{tl_build_end:} group, we +% \begin{macro}[aux]{\@@_show:Nx} +% Within a \cs{__tl_build:Nw} \ldots{} \cs{__tl_build_end:} group, we % redefine all the function that can appear in a compiled regex, then % run the regex. The result is then shown. % \begin{macrocode} -\cs_new_protected:Npn \regex_show_aux:Nx #1#2 +\cs_new_protected:Npn \@@_show:Nx #1#2 { - \tl_set_build:Nw \l_regex_internal_a_tl - \cs_set_protected_nopar:Npn \regex_branch:n + \__tl_build:Nw \l_@@_internal_a_tl + \cs_set_protected_nopar:Npn \@@_branch:n { - \seq_pop_right:NN \l_regex_show_prefix_seq \l_regex_internal_a_tl - \regex_show_one:n { +-branch } - \seq_put_right:No \l_regex_show_prefix_seq \l_regex_internal_a_tl + \seq_pop_right:NN \l_@@_show_prefix_seq \l_@@_internal_a_tl + \@@_show_one:n { +-branch } + \seq_put_right:No \l_@@_show_prefix_seq \l_@@_internal_a_tl \use:n } - \cs_set_protected_nopar:Npn \regex_group:nnnN - { \regex_show_group_aux:nnnnN { } } - \cs_set_protected_nopar:Npn \regex_group_no_capture:nnnN - { \regex_show_group_aux:nnnnN { ~(no~capture) } } - \cs_set_protected_nopar:Npn \regex_group_resetting:nnnN - { \regex_show_group_aux:nnnnN { ~(resetting) } } - \cs_set_eq:NN \regex_class:NnnnN \regex_show_class:NnnnN - \cs_set_protected_nopar:Npn \regex_command_K: - { \regex_show_one:n { reset~match~start~(\iow_char:N\\K) } } - \cs_set_protected:Npn \regex_assertion:Nn ##1##2 - { \regex_show_one:n { \bool_if:NF ##1 { negative~ } assertion:~##2 } } - \cs_set_nopar:Npn \regex_b_test: { word~boundary } - \cs_set_eq:NN \regex_anchor:N \regex_show_anchor_to_str:N - \cs_set_protected:Npn \regex_item_caseful_equal:n ##1 - { \regex_show_one:n { char~code~\int_eval:n{##1} } } - \cs_set_protected:Npn \regex_item_caseful_range:nn ##1##2 - { \regex_show_one:n { range~[\int_eval:n{##1}, \int_eval:n{##2}] } } - \cs_set_protected:Npn \regex_item_caseless_equal:n ##1 - { \regex_show_one:n { char~code~\int_eval:n{##1}~(caseless) } } - \cs_set_protected:Npn \regex_item_caseless_range:nn ##1##2 + \cs_set_protected_nopar:Npn \@@_group:nnnN + { \@@_show_group_aux:nnnnN { } } + \cs_set_protected_nopar:Npn \@@_group_no_capture:nnnN + { \@@_show_group_aux:nnnnN { ~(no~capture) } } + \cs_set_protected_nopar:Npn \@@_group_resetting:nnnN + { \@@_show_group_aux:nnnnN { ~(resetting) } } + \cs_set_eq:NN \@@_class:NnnnN \@@_show_class:NnnnN + \cs_set_protected_nopar:Npn \@@_command_K: + { \@@_show_one:n { reset~match~start~(\iow_char:N\\K) } } + \cs_set_protected:Npn \@@_assertion:Nn ##1##2 + { \@@_show_one:n { \bool_if:NF ##1 { negative~ } assertion:~##2 } } + \cs_set_nopar:Npn \@@_b_test: { word~boundary } + \cs_set_eq:NN \@@_anchor:N \@@_show_anchor_to_str:N + \cs_set_protected:Npn \@@_item_caseful_equal:n ##1 + { \@@_show_one:n { char~code~\int_eval:n{##1} } } + \cs_set_protected:Npn \@@_item_caseful_range:nn ##1##2 + { \@@_show_one:n { range~[\int_eval:n{##1}, \int_eval:n{##2}] } } + \cs_set_protected:Npn \@@_item_caseless_equal:n ##1 + { \@@_show_one:n { char~code~\int_eval:n{##1}~(caseless) } } + \cs_set_protected:Npn \@@_item_caseless_range:nn ##1##2 { - \regex_show_one:n + \@@_show_one:n { Range~[\int_eval:n{##1}, \int_eval:n{##2}]~(caseless) } } - \cs_set_protected:Npn \regex_item_catcode:nT - { \regex_show_item_catcode:NnT \c_true_bool } - \cs_set_protected:Npn \regex_item_catcode_reverse:nT - { \regex_show_item_catcode:NnT \c_false_bool } - \cs_set_protected:Npn \regex_item_reverse:n - { \regex_show_scope:nn { Reversed~match } } - \cs_set_protected:Npn \regex_item_exact:nn ##1##2 - { \regex_show_one:n { char~##2,~catcode~##1 } } - \cs_set_protected:Npn \regex_item_exact_cs:c ##1 - { \regex_show_one:n { control~sequence~\iow_char:N\\##1 } } - \cs_set_protected:Npn \regex_item_cs:n - { \regex_show_scope:nn { control~sequence } } - \cs_set:cpn { regex_prop_.: } { \regex_show_one:n { any~token } } - \seq_clear:N \l_regex_show_prefix_seq - \regex_show_push:n { ~ } + \cs_set_protected:Npn \@@_item_catcode:nT + { \@@_show_item_catcode:NnT \c_true_bool } + \cs_set_protected:Npn \@@_item_catcode_reverse:nT + { \@@_show_item_catcode:NnT \c_false_bool } + \cs_set_protected:Npn \@@_item_reverse:n + { \@@_show_scope:nn { Reversed~match } } + \cs_set_protected:Npn \@@_item_exact:nn ##1##2 + { \@@_show_one:n { char~##2,~catcode~##1 } } + \cs_set_protected:Npn \@@_item_exact_cs:c ##1 + { \@@_show_one:n { control~sequence~\iow_char:N\\##1 } } + \cs_set_protected:Npn \@@_item_cs:n + { \@@_show_scope:nn { control~sequence } } + \cs_set:cpn { @@_prop_.: } { \@@_show_one:n { any~token } } + \seq_clear:N \l_@@_show_prefix_seq + \@@_show_push:n { ~ } #1 - \tl_build_end: - \msg_aux_show:x { > Compiled~regex~#2: \l_regex_internal_a_tl } + \__tl_build_end: + \__msg_show_variable:x { > Compiled~regex~#2: \l_@@_internal_a_tl } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_show_one:n} +% \begin{macro}[aux]{\@@_show_one:n} % Every part of the final message go through this function, which adds % one line to the output, with the apropriate prefix. % \begin{macrocode} -\cs_new_protected:Npn \regex_show_one:n #1 +\cs_new_protected:Npn \@@_show_one:n #1 { - \int_incr:N \l_regex_show_lines_int - \tl_build_one:x - { \iow_newline: \seq_use:N \l_regex_show_prefix_seq #1 } + \int_incr:N \l_@@_show_lines_int + \__tl_build_one:x + { + \iow_newline: + \seq_map_function:NN \l_@@_show_prefix_seq \use:n + #1 + } } % \end{macrocode} % \end{macro} % % \begin{macro}[aux] -% {\regex_show_push:n, \regex_show_pop:, \regex_show_scope:nn} +% {\@@_show_push:n, \@@_show_pop:, \@@_show_scope:nn} % Enter and exit levels of nesting. The \texttt{scope} function prints % its first argument as an \enquote{introduction}, then performs its % second argument in a deeper level of nesting. % \begin{macrocode} -\cs_new_protected:Npn \regex_show_push:n #1 - { \seq_put_right:Nx \l_regex_show_prefix_seq { #1 ~ } } -\cs_new_protected:Npn \regex_show_pop: - { \seq_pop_right:NN \l_regex_show_prefix_seq \l_regex_internal_a_tl } -\cs_new_protected:Npn \regex_show_scope:nn #1#2 +\cs_new_protected:Npn \@@_show_push:n #1 + { \seq_put_right:Nx \l_@@_show_prefix_seq { #1 ~ } } +\cs_new_protected:Npn \@@_show_pop: + { \seq_pop_right:NN \l_@@_show_prefix_seq \l_@@_internal_a_tl } +\cs_new_protected:Npn \@@_show_scope:nn #1#2 { - \regex_show_one:n {#1} - \regex_show_push:n { ~ } + \@@_show_one:n {#1} + \@@_show_push:n { ~ } #2 - \regex_show_pop: + \@@_show_pop: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_show_group_aux:nnnnN} +% \begin{macro}[aux]{\@@_show_group_aux:nnnnN} % We display all groups in the same way, simply adding a message, % \texttt{(no capture)} or \texttt{(resetting)}, to special groups. % The odd \cs{use_ii:nn} avoids printing a spurious \texttt{+-branch} % for the first branch. % \begin{macrocode} -\cs_new_protected:Npn \regex_show_group_aux:nnnnN #1#2#3#4#5 +\cs_new_protected:Npn \@@_show_group_aux:nnnnN #1#2#3#4#5 { - \regex_show_one:n { ,-group~begin #1 } - \regex_show_push:n { | } + \@@_show_one:n { ,-group~begin #1 } + \@@_show_push:n { | } \use_ii:nn #2 - \regex_show_pop: - \regex_show_one:n - { `-group~end \regex_msg_repeated:nnN {#3} {#4} #5 } + \@@_show_pop: + \@@_show_one:n + { `-group~end \@@_msg_repeated:nnN {#3} {#4} #5 } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_show_class:NnnnN} +% \begin{macro}[aux]{\@@_show_class:NnnnN} % I'm entirely unhappy about this function: I couldn't find a way to % test if a class is a single test. Instead, collect the % representation of the tests in the class. If that had more than one @@ -3143,79 +3156,81 @@ % incorrect). That's clunky, but not too expensive, since it's only % one test. % \begin{macrocode} -\cs_set:Npn \regex_show_class:NnnnN #1#2#3#4#5 +\cs_set:Npn \@@_show_class:NnnnN #1#2#3#4#5 { - \tl_set_build:Nw \l_regex_internal_a_tl - \int_zero:N \l_regex_show_lines_int - \regex_show_push:n {~} + \__tl_build:Nw \l_@@_internal_a_tl + \int_zero:N \l_@@_show_lines_int + \@@_show_push:n {~} #2 \exp_last_unbraced:Nf - \prg_case_int:nnn { \l_regex_show_lines_int } + \int_case:nnn { \l_@@_show_lines_int } { {0} { - \tl_build_end: - \regex_show_one:n { \bool_if:NTF #1 { Fail } { Pass } } + \__tl_build_end: + \@@_show_one:n { \bool_if:NTF #1 { Fail } { Pass } } } {1} { - \tl_build_end: + \__tl_build_end: \bool_if:NTF #1 { #2 - \tl_build_one:n { \regex_msg_repeated:nnN {#3} {#4} #5 } + \__tl_build_one:n { \@@_msg_repeated:nnN {#3} {#4} #5 } } { - \regex_show_one:n - { Don't~match~\regex_msg_repeated:nnN {#3} {#4} #5 } - \tl_build_one:o \l_regex_internal_a_tl + \@@_show_one:n + { Don't~match~\@@_msg_repeated:nnN {#3} {#4} #5 } + \__tl_build_one:o \l_@@_internal_a_tl } } } { - \tl_build_end: - \regex_show_one:n + \__tl_build_end: + \@@_show_one:n { \bool_if:NTF #1 { M } { Don't~m } atch - \regex_msg_repeated:nnN {#3} {#4} #5 + \@@_msg_repeated:nnN {#3} {#4} #5 } - \tl_build_one:o \l_regex_internal_a_tl + \__tl_build_one:o \l_@@_internal_a_tl } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux, rEXP]{\regex_show_anchor_to_str:N} +% \begin{macro}[aux, rEXP]{\@@_show_anchor_to_str:N} % The argument is an integer telling us where the anchor is. We % convert that to the relevant info. % \begin{macrocode} -\cs_new:Npn \regex_show_anchor_to_str:N #1 +\cs_new:Npn \@@_show_anchor_to_str:N #1 { anchor~at~ - \prg_case_str:nnn { #1 } + \str_case:nnn { #1 } { - { \l_regex_min_pos_int } { start~(\iow_char:N\\A) } - { \l_regex_start_pos_int } { start~of~match~(\iow_char:N\\G) } - { \l_regex_max_pos_int } { end~(\iow_char:N\\Z) } + { \l_@@_min_pos_int } { start~(\iow_char:N\\A) } + { \l_@@_start_pos_int } { start~of~match~(\iow_char:N\\G) } + { \l_@@_max_pos_int } { end~(\iow_char:N\\Z) } } { } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_show_item_catcode:NnT} +% \begin{macro}[aux]{\@@_show_item_catcode:NnT} % Produce a sequence of categories which the catcode bitmap |#2| % contains, and show it, indenting the tests on which this catcode % constraint applies. % \begin{macrocode} -\cs_new_protected:Npn \regex_show_item_catcode:NnT #1#2 +\cs_new_protected:Npn \@@_show_item_catcode:NnT #1#2 { - \seq_set_split:Nnn \l_regex_internal_seq { } { CBEMTPUDSLOA } - \seq_set_filter:NNn \l_regex_internal_seq \l_regex_internal_seq - { \int_if_odd_p:n { #2 / \int_use:c { c_regex_catcode_##1_int } } } - \regex_show_scope:nn + \seq_set_split:Nnn \l_@@_internal_seq { } { CBEMTPUDSLOA } + \seq_set_filter:NNn \l_@@_internal_seq \l_@@_internal_seq + { \int_if_odd_p:n { #2 / \int_use:c { c_@@_catcode_##1_int } } } + \@@_show_scope:nn { - categories~\seq_use:N \l_regex_internal_seq, ~ + categories~ + \seq_map_function:NN \l_@@_internal_seq \use:n + , ~ \bool_if:NF #1 { negative~ } class } } @@ -3226,41 +3241,41 @@ % % \subsubsection{Variables used while building} % -% \begin{variable}{\l_regex_min_state_int, \l_regex_max_state_int} -% The last state that was allocated is $\cs{l_regex_max_state_int}-1$, -% so that \cs{l_regex_max_state_int} always points to a free state. +% \begin{variable}{\l_@@_min_state_int, \l_@@_max_state_int} +% The last state that was allocated is $\cs{l_@@_max_state_int}-1$, +% so that \cs{l_@@_max_state_int} always points to a free state. % The \texttt{min_state} variable is always $0$, but is included to % avoid hard-coding this value. % \begin{macrocode} -\int_new:N \l_regex_min_state_int -\int_new:N \l_regex_max_state_int +\int_new:N \l_@@_min_state_int +\int_new:N \l_@@_max_state_int % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_regex_left_state_int, \l_regex_right_state_int} -% \begin{variable}{\l_regex_left_state_seq, \l_regex_right_state_seq} +% \begin{variable}{\l_@@_left_state_int, \l_@@_right_state_int} +% \begin{variable}{\l_@@_left_state_seq, \l_@@_right_state_seq} % Alternatives are implemented by branching from a \texttt{left} state % into the various choices, then merging those into a \texttt{right} % state. We store information about those states in two sequences. % Those states are also used to implement group quantifiers. Most % often, the left and right pointers only differ by~$1$. % \begin{macrocode} -\int_new:N \l_regex_left_state_int -\int_new:N \l_regex_right_state_int -\seq_new:N \l_regex_left_state_seq -\seq_new:N \l_regex_right_state_seq +\int_new:N \l_@@_left_state_int +\int_new:N \l_@@_right_state_int +\seq_new:N \l_@@_left_state_seq +\seq_new:N \l_@@_right_state_seq % \end{macrocode} % \end{variable} % \end{variable} % -% \begin{variable}{\l_regex_capturing_group_int} -% \cs{l_regex_capturing_group_int} is the \textsc{id} number that will +% \begin{variable}{\l_@@_capturing_group_int} +% \cs{l_@@_capturing_group_int} is the \textsc{id} number that will % be assigned to a capturing group if one was opened now. This starts % at $0$ for the group enclosing the full regular expression, and % groups are counted in the order of their left parenthesis, except % when encountering \texttt{resetting} groups. % \begin{macrocode} -\int_new:N \l_regex_capturing_group_int +\int_new:N \l_@@_capturing_group_int % \end{macrocode} % \end{variable} % @@ -3270,22 +3285,22 @@ % Each state of the \textsc{nfa} is stored in a \tn{toks}. The % operations which can appear in the \tn{toks} are % \begin{itemize} -% \item \cs{regex_action_start_wildcard:} inserted at the start +% \item \cs{@@_action_start_wildcard:} inserted at the start % of the regular expression to make it unanchored. -% \item \cs{regex_action_success:} marks the exit state of the +% \item \cs{@@_action_success:} marks the exit state of the % \textsc{nfa}. -% \item \cs{regex_action_cost:n} \Arg{shift} is a transition from the +% \item \cs{@@_action_cost:n} \Arg{shift} is a transition from the % current \meta{state} to $\meta{state}+\meta{shift}$, which % consumes the current character: the target state is saved and will % be considered again when matching at the next position. -% \item \cs{regex_action_free:n} \Arg{shift}, and -% \cs{regex_action_free_group:n} \Arg{shift} are free transitions, +% \item \cs{@@_action_free:n} \Arg{shift}, and +% \cs{@@_action_free_group:n} \Arg{shift} are free transitions, % which immediately perform the actions for the state % $\meta{state}+\meta{shift}$ of the \textsc{nfa}. They differ in % how they detect and avoid infinite loops. For now, we just need to % know that the \texttt{group} variant must be used for transitions % back to the start of a group. -% \item \cs{regex_action_submatch:n} \Arg{key} where the \meta{key} is +% \item \cs{@@_action_submatch:n} \Arg{key} where the \meta{key} is % a group number followed by |<| or |>| for the beginning or end of % group. This causes the current position in the query to be stored % as the \meta{key} submatch boundary. @@ -3306,7 +3321,7 @@ % corresponding end-points of nested groups. % \end{itemize} % -% \begin{macro}[int]{\regex_build:n, \regex_build:N} +% \begin{macro}[int]{\@@_build:n, \@@_build:N} % The \texttt{n}-type function first compiles its argument. Reset some % variables. Allocate two states, and put a wildcard in state $0$ % (transitions to state $1$ and $0$ state). Then build the regex @@ -3314,104 +3329,104 @@ % value of \texttt{capturing_group}). Finally, if the match reaches the % last state, it is successful. % \begin{macrocode} -\cs_new_protected:Npn \regex_build:n #1 +\cs_new_protected:Npn \@@_build:n #1 { - \regex_compile:n {#1} - \regex_build:N \l_regex_internal_regex + \@@_compile:n {#1} + \@@_build:N \l_@@_internal_regex } -\cs_new_protected:Npn \regex_build:N #1 +\cs_new_protected:Npn \@@_build:N #1 { -% \trace_push:nnn { regex } { 1 } { regex_build } +% \trace_push:nnn { regex } { 1 } { @@_build } \int_set:Nn \tex_escapechar:D { 92 } - \int_zero:N \l_regex_capturing_group_int - \int_set_eq:NN \l_regex_max_state_int \l_regex_min_state_int - \regex_build_new_state: - \regex_build_new_state: - \regex_toks_put_right:Nn \l_regex_left_state_int - { \regex_action_start_wildcard: } - \regex_group:nnnN {#1} { 1 } { 0 } \c_false_bool - \regex_toks_put_right:Nn \l_regex_right_state_int - { \regex_action_success: } -% \regex_trace_states:n { 2 } -% \trace_pop:nnn { regex } { 1 } { regex_build } + \int_zero:N \l_@@_capturing_group_int + \int_set_eq:NN \l_@@_max_state_int \l_@@_min_state_int + \@@_build_new_state: + \@@_build_new_state: + \@@_toks_put_right:Nn \l_@@_left_state_int + { \@@_action_start_wildcard: } + \@@_group:nnnN {#1} { 1 } { 0 } \c_false_bool + \@@_toks_put_right:Nn \l_@@_right_state_int + { \@@_action_success: } +% \@@_trace_states:n { 2 } +% \trace_pop:nnn { regex } { 1 } { @@_build } } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_build_for_cs:n} +% \begin{macro}[int]{\@@_build_for_cs:n} % When using a regex to match a cs, we don't insert a wildcard, we % anchor at the end, and since we ignore submatches, there is no need % to surround the expression with a group. However, for branches to % work properly at the outer level, we need to put the appropriate % \texttt{left} and \texttt{right} states in their sequence. % \begin{macrocode} -\cs_new_protected:Npn \regex_build_for_cs:n #1 +\cs_new_protected:Npn \@@_build_for_cs:n #1 { -% \trace_push:nnn { regex } { 1 } { regex_build_for_cs } - \int_set_eq:NN \l_regex_max_state_int \l_regex_min_state_int - \regex_build_new_state: - \regex_build_new_state: - \regex_push_lr_states: +% \trace_push:nnn { regex } { 1 } { @@_build_for_cs } + \int_set_eq:NN \l_@@_max_state_int \l_@@_min_state_int + \@@_build_new_state: + \@@_build_new_state: + \@@_push_lr_states: #1 - \regex_pop_lr_states: - \regex_toks_put_right:Nn \l_regex_right_state_int + \@@_pop_lr_states: + \@@_toks_put_right:Nn \l_@@_right_state_int { - \if_num:w \l_regex_current_pos_int = \l_regex_max_pos_int - \exp_after:wN \regex_action_success: + \if_int_compare:w \l_@@_current_pos_int = \l_@@_max_pos_int + \exp_after:wN \@@_action_success: \fi: } -% \regex_trace_states:n { 2 } -% \trace_pop:nnn { regex } { 1 } { regex_build_for_cs } +% \@@_trace_states:n { 2 } +% \trace_pop:nnn { regex } { 1 } { @@_build_for_cs } } % \end{macrocode} % \end{macro} % % \subsubsection{Helpers for building an \textsc{nfa}} % -% \begin{macro}[int]{\regex_push_lr_states:, \regex_pop_lr_states:} +% \begin{macro}[int]{\@@_push_lr_states:, \@@_pop_lr_states:} % When building the regular expression, we keep track of pointers to % the left-end and right-end of each group without help from \TeX{}'s % grouping. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_push_lr_states: +\cs_new_protected_nopar:Npn \@@_push_lr_states: { - \seq_push:No \l_regex_left_state_seq - { \int_use:N \l_regex_left_state_int } - \seq_push:No \l_regex_right_state_seq - { \int_use:N \l_regex_right_state_int } + \seq_push:No \l_@@_left_state_seq + { \int_use:N \l_@@_left_state_int } + \seq_push:No \l_@@_right_state_seq + { \int_use:N \l_@@_right_state_int } } -\cs_new_protected_nopar:Npn \regex_pop_lr_states: +\cs_new_protected_nopar:Npn \@@_pop_lr_states: { - \seq_pop:NN \l_regex_left_state_seq \l_regex_internal_a_tl - \int_set:Nn \l_regex_left_state_int \l_regex_internal_a_tl - \seq_pop:NN \l_regex_right_state_seq \l_regex_internal_a_tl - \int_set:Nn \l_regex_right_state_int \l_regex_internal_a_tl + \seq_pop:NN \l_@@_left_state_seq \l_@@_internal_a_tl + \int_set:Nn \l_@@_left_state_int \l_@@_internal_a_tl + \seq_pop:NN \l_@@_right_state_seq \l_@@_internal_a_tl + \int_set:Nn \l_@@_right_state_int \l_@@_internal_a_tl } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_toks_put_left:Nx} -% \begin{macro}[int]{\regex_toks_put_right:Nx, \regex_toks_put_right:Nn} +% \begin{macro}[int]{\@@_toks_put_left:Nx} +% \begin{macro}[int]{\@@_toks_put_right:Nx, \@@_toks_put_right:Nn} % During the building phase we wish to add \texttt{x}-expanded % material to \tn{toks}, either to the left or to the right. The % expansion is done \enquote{by hand} for optimization (these % operations are used quite a lot). The \texttt{Nn} version of -% \cs{regex_toks_put_right:Nx} is provided because it is more +% \cs{@@_toks_put_right:Nx} is provided because it is more % efficient than \texttt{x}-expanding with \cs{exp_not:n}. % \begin{macrocode} -\cs_new_protected:Npn \regex_toks_put_left:Nx #1#2 +\cs_new_protected:Npn \@@_toks_put_left:Nx #1#2 { - \cs_set_nopar:Npx \regex_tmp:w { #2 } + \cs_set_nopar:Npx \@@_tmp:w { #2 } \tex_toks:D #1 \exp_after:wN \exp_after:wN \exp_after:wN - { \exp_after:wN \regex_tmp:w \tex_the:D \tex_toks:D #1 } + { \exp_after:wN \@@_tmp:w \tex_the:D \tex_toks:D #1 } } -\cs_new_protected:Npn \regex_toks_put_right:Nx #1#2 +\cs_new_protected:Npn \@@_toks_put_right:Nx #1#2 { - \cs_set_nopar:Npx \regex_tmp:w {#2} + \cs_set_nopar:Npx \@@_tmp:w {#2} \tex_toks:D #1 \exp_after:wN - { \tex_the:D \tex_toks:D \exp_after:wN #1 \regex_tmp:w } + { \tex_the:D \tex_toks:D \exp_after:wN #1 \@@_tmp:w } } -\cs_new_protected:Npn \regex_toks_put_right:Nn #1#2 +\cs_new_protected:Npn \@@_toks_put_right:Nn #1#2 { \tex_toks:D #1 \exp_after:wN { \tex_the:D \tex_toks:D #1 #2 } } % \end{macrocode} % \end{macro} @@ -3419,8 +3434,8 @@ % % \begin{macro}[int] % { -% \regex_build_transition_left:NNN, -% \regex_build_transition_right:nNn +% \@@_build_transition_left:NNN, +% \@@_build_transition_right:nNn % } % Add a transition from |#2| to |#3| using the function |#1|. The % \texttt{left} function is used for higher priority transitions, and @@ -3428,56 +3443,56 @@ % should be performed later). The signatures differ to reflect the % differing usage later on. Both functions could be optimized. % \begin{macrocode} -\cs_new_protected:Npn \regex_build_transition_left:NNN #1#2#3 - { \regex_toks_put_left:Nx #2 { #1 { \int_eval:n { #3 - #2 } } } } -\cs_new_protected:Npn \regex_build_transition_right:nNn #1#2#3 - { \regex_toks_put_right:Nx #2 { #1 { \int_eval:n { #3 - #2 } } } } +\cs_new_protected:Npn \@@_build_transition_left:NNN #1#2#3 + { \@@_toks_put_left:Nx #2 { #1 { \int_eval:n { #3 - #2 } } } } +\cs_new_protected:Npn \@@_build_transition_right:nNn #1#2#3 + { \@@_toks_put_right:Nx #2 { #1 { \int_eval:n { #3 - #2 } } } } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_build_new_state:} +% \begin{macro}[int]{\@@_build_new_state:} % Add a new empty state to the \textsc{nfa}. Then update the % \texttt{left}, \texttt{right}, and \texttt{max} states, so that the % \texttt{right} state is the new empty state, and the \texttt{left} % state points to the previously \enquote{current} state. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_build_new_state: +\cs_new_protected_nopar:Npn \@@_build_new_state: { %<*trace> \trace:nnx { regex } { 2 } { regex~new~state~ - L=\int_use:N \l_regex_left_state_int ~ -> ~ - R=\int_use:N \l_regex_right_state_int ~ -> ~ - M=\int_use:N \l_regex_max_state_int ~ -> ~ - \int_eval:n { \l_regex_max_state_int + \c_one } + L=\int_use:N \l_@@_left_state_int ~ -> ~ + R=\int_use:N \l_@@_right_state_int ~ -> ~ + M=\int_use:N \l_@@_max_state_int ~ -> ~ + \int_eval:n { \l_@@_max_state_int + \c_one } } % - \tex_toks:D \l_regex_max_state_int { } - \int_set_eq:NN \l_regex_left_state_int \l_regex_right_state_int - \int_set_eq:NN \l_regex_right_state_int \l_regex_max_state_int - \int_incr:N \l_regex_max_state_int + \tex_toks:D \l_@@_max_state_int { } + \int_set_eq:NN \l_@@_left_state_int \l_@@_right_state_int + \int_set_eq:NN \l_@@_right_state_int \l_@@_max_state_int + \int_incr:N \l_@@_max_state_int } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_build_transitions_lazyness:NNNNN} +% \begin{macro}[int]{\@@_build_transitions_lazyness:NNNNN} % This function creates a new state, and puts two transitions starting % from the old current state. The order of the transitions is % controlled by |#1|, true for lazy quantifiers, and false for greedy % quantifiers. % \begin{macrocode} -\cs_new_protected:Npn \regex_build_transitions_lazyness:NNNNN #1#2#3#4#5 +\cs_new_protected:Npn \@@_build_transitions_lazyness:NNNNN #1#2#3#4#5 { - \regex_build_new_state: - \regex_toks_put_right:Nx \l_regex_left_state_int + \@@_build_new_state: + \@@_toks_put_right:Nx \l_@@_left_state_int { \if_meaning:w \c_true_bool #1 - #2 { \int_eval:n { #3 - \l_regex_left_state_int } } - #4 { \int_eval:n { #5 - \l_regex_left_state_int } } + #2 { \int_eval:n { #3 - \l_@@_left_state_int } } + #4 { \int_eval:n { #5 - \l_@@_left_state_int } } \else: - #4 { \int_eval:n { #5 - \l_regex_left_state_int } } - #2 { \int_eval:n { #3 - \l_regex_left_state_int } } + #4 { \int_eval:n { #5 - \l_@@_left_state_int } } + #2 { \int_eval:n { #3 - \l_@@_left_state_int } } \fi: } } @@ -3486,83 +3501,83 @@ % % \subsubsection{Building classes} % -% \begin{macro}[int]{\regex_class:NnnnN} -% \begin{macro}[int, rEXP]{\regex_tests_action_cost:n} +% \begin{macro}[int]{\@@_class:NnnnN} +% \begin{macro}[int, rEXP]{\@@_tests_action_cost:n} % The arguments are: \meta{boolean} \Arg{tests} \Arg{min} \Arg{more} % \meta{lazyness}. First store the tests with a trailing -% \cs{regex_action_cost:n}, in the true branch of -% \cs{regex_break_point:TF} for positive classes, or the false branch +% \cs{@@_action_cost:n}, in the true branch of +% \cs{@@_break_point:TF} for positive classes, or the false branch % for negative classes. The integer \meta{more} is $0$ for fixed % repetitions, $-1$ for unbounded repetitions, and % $\meta{max}-\meta{min}$ for a range of repetitions. % \begin{macrocode} -\cs_new_protected:Npn \regex_class:NnnnN #1#2#3#4#5 +\cs_new_protected:Npn \@@_class:NnnnN #1#2#3#4#5 { - \cs_set_nopar:Npx \regex_tests_action_cost:n ##1 + \cs_set_nopar:Npx \@@_tests_action_cost:n ##1 { \exp_not:n { \exp_not:n {#2} } \bool_if:NTF #1 - { \regex_break_point:TF { \regex_action_cost:n {##1} } { } } - { \regex_break_point:TF { } { \regex_action_cost:n {##1} } } + { \@@_break_point:TF { \@@_action_cost:n {##1} } { } } + { \@@_break_point:TF { } { \@@_action_cost:n {##1} } } } \if_case:w - #4 \exp_stop_f: - \regex_class_repeat:n {#3} - \or: \regex_class_repeat:nN {#3} #5 - \else: \regex_class_repeat:nnN {#3} {#4} #5 + \@@_class_repeat:n {#3} + \or: \@@_class_repeat:nN {#3} #5 + \else: \@@_class_repeat:nnN {#3} {#4} #5 \fi: } -\cs_new:Npn \regex_tests_action_cost:n { \regex_action_cost:n } +\cs_new:Npn \@@_tests_action_cost:n { \@@_action_cost:n } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[aux]{\regex_class_repeat:n} +% \begin{macro}[aux]{\@@_class_repeat:n} % This is used for a fixed number of repetitions. Build one state for % each repetition, with a transition controlled by the tests that we % have collected. That works just fine for |#1|${}=0$ repetitions: % nothing is built. % \begin{macrocode} -\cs_new_protected:Npn \regex_class_repeat:n #1 +\cs_new_protected:Npn \@@_class_repeat:n #1 { \prg_replicate:nn {#1} { - \regex_build_new_state: - \regex_build_transition_right:nNn \regex_tests_action_cost:n - \l_regex_left_state_int \l_regex_right_state_int + \@@_build_new_state: + \@@_build_transition_right:nNn \@@_tests_action_cost:n + \l_@@_left_state_int \l_@@_right_state_int } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_class_repeat:nN} +% \begin{macro}[aux]{\@@_class_repeat:nN} % This implements unbounded repetitions of a single class (\emph{e.g.} % the |*| and |+| quantifiers). If the minimum number |#1| of % repetitions is $0$, then build a transition from the current state % to itself governed by the tests, and a free transition to a new % state (hence skipping the tests). Otherwise, call -% \cs{regex_class_repeat:n} for the code to match |#1| repetitions, +% \cs{@@_class_repeat:n} for the code to match |#1| repetitions, % and add free transitions from the last state to the previous one, % and to a new one. In both cases, the order of transitions is % controlled by the lazyness boolean |#2|. % \begin{macrocode} -\cs_new_protected:Npn \regex_class_repeat:nN #1#2 +\cs_new_protected:Npn \@@_class_repeat:nN #1#2 { - \if_num:w #1 = \c_zero - \regex_build_transitions_lazyness:NNNNN #2 - \regex_action_free:n \l_regex_right_state_int - \regex_tests_action_cost:n \l_regex_left_state_int + \if_int_compare:w #1 = \c_zero + \@@_build_transitions_lazyness:NNNNN #2 + \@@_action_free:n \l_@@_right_state_int + \@@_tests_action_cost:n \l_@@_left_state_int \else: - \regex_class_repeat:n {#1} - \int_set_eq:NN \l_regex_internal_a_int \l_regex_left_state_int - \regex_build_transitions_lazyness:NNNNN #2 - \regex_action_free:n \l_regex_right_state_int - \regex_action_free:n \l_regex_internal_a_int + \@@_class_repeat:n {#1} + \int_set_eq:NN \l_@@_internal_a_int \l_@@_left_state_int + \@@_build_transitions_lazyness:NNNNN #2 + \@@_action_free:n \l_@@_right_state_int + \@@_action_free:n \l_@@_internal_a_int \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_class_repeat:nnN} +% \begin{macro}[aux]{\@@_class_repeat:nnN} % We want to build the code to match from |#1| to $|#1|+|#2|$ % repetitions. Match |#1| repetitions (can be $0$). Compute the final % state of the next construction as \texttt{a}. Build $|#2|>0$ states, @@ -3571,16 +3586,16 @@ % \texttt{a} is safe because states are allocated in order, starting % from \texttt{max_state}. % \begin{macrocode} -\cs_new_protected:Npn \regex_class_repeat:nnN #1#2#3 +\cs_new_protected:Npn \@@_class_repeat:nnN #1#2#3 { - \regex_class_repeat:n {#1} - \int_set:Nn \l_regex_internal_a_int - { \l_regex_max_state_int + #2 - \c_one } + \@@_class_repeat:n {#1} + \int_set:Nn \l_@@_internal_a_int + { \l_@@_max_state_int + #2 - \c_one } \prg_replicate:nn { #2 } { - \regex_build_transitions_lazyness:NNNNN #3 - \regex_action_free:n \l_regex_internal_a_int - \regex_tests_action_cost:n \l_regex_right_state_int + \@@_build_transitions_lazyness:NNNNN #3 + \@@_action_free:n \l_@@_internal_a_int + \@@_tests_action_cost:n \l_@@_right_state_int } } % \end{macrocode} @@ -3588,7 +3603,7 @@ % % \subsubsection{Building groups} % -% \begin{macro}[aux]{\regex_group_aux:nnnnN} +% \begin{macro}[aux]{\@@_group_aux:nnnnN} % Arguments: \Arg{label} \Arg{contents} \Arg{min} \Arg{more} % \meta{lazyness}. If \meta{min} is $0$, we need to add a state before % building the group, so that the thread which skips the group does @@ -3605,146 +3620,146 @@ % auxiliaries expects \texttt{left_state} and \texttt{right_state} to % be set properly. % \begin{macrocode} -\cs_new_protected:Npn \regex_group_aux:nnnnN #1#2#3#4#5 +\cs_new_protected:Npn \@@_group_aux:nnnnN #1#2#3#4#5 { -% \trace_push:nnn { regex } { 1 } { regex_group } - \if_num:w #3 = \c_zero - \regex_build_new_state: -%\assert_int:n { \l_regex_max_state_int = \l_regex_right_state_int + 1 } - \regex_build_transition_right:nNn \regex_action_free_group:n - \l_regex_left_state_int \l_regex_right_state_int +% \trace_push:nnn { regex } { 1 } { @@_group } + \if_int_compare:w #3 = \c_zero + \@@_build_new_state: +%\assert_int:n { \l_@@_max_state_int = \l_@@_right_state_int + 1 } + \@@_build_transition_right:nNn \@@_action_free_group:n + \l_@@_left_state_int \l_@@_right_state_int \fi: - \regex_build_new_state: - \regex_push_lr_states: + \@@_build_new_state: + \@@_push_lr_states: #2 - \regex_pop_lr_states: + \@@_pop_lr_states: \if_case:w - #4 \exp_stop_f: - \regex_group_repeat:nn {#1} {#3} - \or: \regex_group_repeat:nnN {#1} {#3} #5 - \else: \regex_group_repeat:nnnN {#1} {#3} {#4} #5 + \@@_group_repeat:nn {#1} {#3} + \or: \@@_group_repeat:nnN {#1} {#3} #5 + \else: \@@_group_repeat:nnnN {#1} {#3} {#4} #5 \fi: -% \trace_pop:nnn { regex } { 1 } { regex_group } +% \trace_pop:nnn { regex } { 1 } { @@_group } } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_group:nnnN, \regex_group_no_capture:nnnN} -% Hand to \cs{regex_group_aux:nnnnnN} the label of that group +% \begin{macro}[int]{\@@_group:nnnN, \@@_group_no_capture:nnnN} +% Hand to \cs{@@_group_aux:nnnnnN} the label of that group % (expanded), and the group itself, with some extra commands to % perform. % \begin{macrocode} -\cs_new_protected:Npn \regex_group:nnnN #1 +\cs_new_protected:Npn \@@_group:nnnN #1 { - \exp_args:No \regex_group_aux:nnnnN - { \int_use:N \l_regex_capturing_group_int } + \exp_args:No \@@_group_aux:nnnnN + { \int_use:N \l_@@_capturing_group_int } { - \int_incr:N \l_regex_capturing_group_int + \int_incr:N \l_@@_capturing_group_int #1 } } -\cs_new_protected_nopar:Npn \regex_group_no_capture:nnnN - { \regex_group_aux:nnnnN { -1 } } +\cs_new_protected_nopar:Npn \@@_group_no_capture:nnnN + { \@@_group_aux:nnnnN { -1 } } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_group_resetting:nnnN} -% \begin{macro}[aux]{\regex_group_resetting_loop:nnNn} -% Again, hand the label $-1$ to \cs{regex_group_aux:nnnnN}, but this +% \begin{macro}[int]{\@@_group_resetting:nnnN} +% \begin{macro}[aux]{\@@_group_resetting_loop:nnNn} +% Again, hand the label $-1$ to \cs{@@_group_aux:nnnnN}, but this % time we work a little bit harder to keep track of the maximum group % label at the end of any branch, and to reset the group number at % each branch. This relies on the fact that a compiled regex always is -% a sequence of items of the form \cs{regex_branch:n} \Arg{branch}. +% a sequence of items of the form \cs{@@_branch:n} \Arg{branch}. % \begin{macrocode} -\cs_new_protected:Npn \regex_group_resetting:nnnN #1 +\cs_new_protected:Npn \@@_group_resetting:nnnN #1 { - \regex_group_aux:nnnnN { -1 } + \@@_group_aux:nnnnN { -1 } { - \exp_args:Noo \regex_group_resetting_loop:nnNn - { \int_use:N \l_regex_capturing_group_int } - { \int_use:N \l_regex_capturing_group_int } + \exp_args:Noo \@@_group_resetting_loop:nnNn + { \int_use:N \l_@@_capturing_group_int } + { \int_use:N \l_@@_capturing_group_int } #1 - { ?? \prg_map_break:n } { } - \prg_break_point:n { } + { ?? \__prg_break:n } { } + \__prg_break_point: } } -\cs_new_protected:Npn \regex_group_resetting_loop:nnNn #1#2#3#4 +\cs_new_protected:Npn \@@_group_resetting_loop:nnNn #1#2#3#4 { - \use_none:nn #3 { \int_set:Nn \l_regex_capturing_group_int {#1} } - \int_set:Nn \l_regex_capturing_group_int {#2} + \use_none:nn #3 { \int_set:Nn \l_@@_capturing_group_int {#1} } + \int_set:Nn \l_@@_capturing_group_int {#2} #3 {#4} - \exp_args:Nf \regex_group_resetting_loop:nnNn - { \int_max:nn {#1} { \l_regex_capturing_group_int } } + \exp_args:Nf \@@_group_resetting_loop:nnNn + { \int_max:nn {#1} { \l_@@_capturing_group_int } } {#2} } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\regex_branch:n} +% \begin{macro}[int]{\@@_branch:n} % Add a free transition from the left state of the current group to a % brand new state, starting point of this branch. Once the branch is % built, add a transition from its last state to the right state of % the group. The left and right states of the group are extracted from % the relevant sequences. % \begin{macrocode} -\cs_new_protected:Npn \regex_branch:n #1 +\cs_new_protected:Npn \@@_branch:n #1 { -% \trace_push:nnn { regex } { 1 } { regex_branch } - \regex_build_new_state: - \seq_get:NN \l_regex_left_state_seq \l_regex_internal_a_tl - \int_set:Nn \l_regex_left_state_int \l_regex_internal_a_tl - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_left_state_int \l_regex_right_state_int +% \trace_push:nnn { regex } { 1 } { @@_branch } + \@@_build_new_state: + \seq_get:NN \l_@@_left_state_seq \l_@@_internal_a_tl + \int_set:Nn \l_@@_left_state_int \l_@@_internal_a_tl + \@@_build_transition_right:nNn \@@_action_free:n + \l_@@_left_state_int \l_@@_right_state_int #1 - \seq_get:NN \l_regex_right_state_seq \l_regex_internal_a_tl - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_right_state_int \l_regex_internal_a_tl -% \trace_pop:nnn { regex } { 1 } { regex_branch } + \seq_get:NN \l_@@_right_state_seq \l_@@_internal_a_tl + \@@_build_transition_right:nNn \@@_action_free:n + \l_@@_right_state_int \l_@@_internal_a_tl +% \trace_pop:nnn { regex } { 1 } { @@_branch } } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_group_repeat:nn} +% \begin{macro}[int]{\@@_group_repeat:nn} % This function is called to repeat a group a fixed number of times % |#2|; if this is $0$ we remove the group altogether (but don't reset % the \texttt{capturing_group} label). Otherwise, the auxiliary -% \cs{regex_group_repeat_aux:n} copies |#2| times the \tn{toks} for +% \cs{@@_group_repeat_aux:n} copies |#2| times the \tn{toks} for % the group, and leaves \texttt{internal_a} pointing to the left end % of the last repetition. We only record the submatch information at % the last repetition. Finally, add a state at the end (the transition % to it has been taken care of by the replicating auxiliary. % \begin{macrocode} -\cs_new_protected:Npn \regex_group_repeat:nn #1#2 +\cs_new_protected:Npn \@@_group_repeat:nn #1#2 { - \if_num:w #2 = \c_zero - \int_set:Nn \l_regex_max_state_int - { \l_regex_left_state_int - \c_one } - \regex_build_new_state: + \if_int_compare:w #2 = \c_zero + \int_set:Nn \l_@@_max_state_int + { \l_@@_left_state_int - \c_one } + \@@_build_new_state: \else: - \regex_group_repeat_aux:n {#2} - \regex_group_submatches:nNN {#1} - \l_regex_internal_a_int \l_regex_right_state_int - \regex_build_new_state: + \@@_group_repeat_aux:n {#2} + \@@_group_submatches:nNN {#1} + \l_@@_internal_a_int \l_@@_right_state_int + \@@_build_new_state: \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_group_submatches:nNN} +% \begin{macro}[aux]{\@@_group_submatches:nNN} % This inserts in states |#2| and |#3| the code for tracking % submatches of the group |#1|, unless inhibited by a label of $-1$. % \begin{macrocode} -\cs_new_protected:Npn \regex_group_submatches:nNN #1#2#3 +\cs_new_protected:Npn \@@_group_submatches:nNN #1#2#3 { - \if_num:w #1 > \c_minus_one - \regex_toks_put_left:Nx #2 { \regex_action_submatch:n { #1 < } } - \regex_toks_put_left:Nx #3 { \regex_action_submatch:n { #1 > } } + \if_int_compare:w #1 > \c_minus_one + \@@_toks_put_left:Nx #2 { \@@_action_submatch:n { #1 < } } + \@@_toks_put_left:Nx #3 { \@@_action_submatch:n { #1 > } } \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_group_repeat_aux:n} +% \begin{macro}[aux]{\@@_group_repeat_aux:n} % Here we repeat \tn{toks} ranging from \texttt{left_state} to % \texttt{max_state}, $|#1|>0$ times. First add a transition so that % the copies will \enquote{chain} properly. Compute the shift @@ -3754,33 +3769,33 @@ % the end, \texttt{b} is equal to the \texttt{max_state}, and % \texttt{a} points to the left of the last copy of the group. % \begin{macrocode} -\cs_new_protected:Npn \regex_group_repeat_aux:n #1 +\cs_new_protected:Npn \@@_group_repeat_aux:n #1 { - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_right_state_int \l_regex_max_state_int - \int_set_eq:NN \l_regex_internal_a_int \l_regex_left_state_int - \int_set_eq:NN \l_regex_internal_b_int \l_regex_max_state_int - \if_num:w \int_eval:w #1 > \c_one - \int_set:Nn \l_regex_internal_c_int + \@@_build_transition_right:nNn \@@_action_free:n + \l_@@_right_state_int \l_@@_max_state_int + \int_set_eq:NN \l_@@_internal_a_int \l_@@_left_state_int + \int_set_eq:NN \l_@@_internal_b_int \l_@@_max_state_int + \if_int_compare:w \__int_eval:w #1 > \c_one + \int_set:Nn \l_@@_internal_c_int { ( #1 - \c_one ) - * ( \l_regex_internal_b_int - \l_regex_internal_a_int ) + * ( \l_@@_internal_b_int - \l_@@_internal_a_int ) } - \tex_advance:D \l_regex_right_state_int \l_regex_internal_c_int - \tex_advance:D \l_regex_max_state_int \l_regex_internal_c_int - \prg_replicate:nn \l_regex_internal_c_int + \tex_advance:D \l_@@_right_state_int \l_@@_internal_c_int + \tex_advance:D \l_@@_max_state_int \l_@@_internal_c_int + \prg_replicate:nn \l_@@_internal_c_int { - \tex_toks:D \l_regex_internal_b_int - = \tex_toks:D \l_regex_internal_a_int - \tex_advance:D \l_regex_internal_a_int \c_one - \tex_advance:D \l_regex_internal_b_int \c_one + \tex_toks:D \l_@@_internal_b_int + = \tex_toks:D \l_@@_internal_a_int + \tex_advance:D \l_@@_internal_a_int \c_one + \tex_advance:D \l_@@_internal_b_int \c_one } \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_group_repeat:nnN} +% \begin{macro}[aux]{\@@_group_repeat:nnN} % This function is called to repeat a group at least $n$ times; the % case $n=0$ is very different from $n>0$. Assume first that $n=0$. % Insert submatch tracking information at the start and end of the @@ -3798,43 +3813,43 @@ % end up before submatch tracking, but that is irrelevant since it % only does so when going again through the group, recording new % matches. Finally, add a state; we already have a transition pointing -% to it from \cs{regex_group_repeat_aux:n}. -% \begin{macrocode} -\cs_new_protected:Npn \regex_group_repeat:nnN #1#2#3 - { - \if_num:w #2 = \c_zero - \regex_group_submatches:nNN {#1} - \l_regex_left_state_int \l_regex_right_state_int - \int_set:Nn \l_regex_internal_a_int - { \l_regex_left_state_int - \c_one } - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_right_state_int \l_regex_internal_a_int - \regex_build_new_state: +% to it from \cs{@@_group_repeat_aux:n}. +% \begin{macrocode} +\cs_new_protected:Npn \@@_group_repeat:nnN #1#2#3 + { + \if_int_compare:w #2 = \c_zero + \@@_group_submatches:nNN {#1} + \l_@@_left_state_int \l_@@_right_state_int + \int_set:Nn \l_@@_internal_a_int + { \l_@@_left_state_int - \c_one } + \@@_build_transition_right:nNn \@@_action_free:n + \l_@@_right_state_int \l_@@_internal_a_int + \@@_build_new_state: \if_meaning:w \c_true_bool #3 - \regex_build_transition_left:NNN \regex_action_free:n - \l_regex_internal_a_int \l_regex_right_state_int + \@@_build_transition_left:NNN \@@_action_free:n + \l_@@_internal_a_int \l_@@_right_state_int \else: - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_internal_a_int \l_regex_right_state_int + \@@_build_transition_right:nNn \@@_action_free:n + \l_@@_internal_a_int \l_@@_right_state_int \fi: \else: - \regex_group_repeat_aux:n {#2} - \regex_group_submatches:nNN {#1} - \l_regex_internal_a_int \l_regex_right_state_int + \@@_group_repeat_aux:n {#2} + \@@_group_submatches:nNN {#1} + \l_@@_internal_a_int \l_@@_right_state_int \if_meaning:w \c_true_bool #3 - \regex_build_transition_right:nNn \regex_action_free_group:n - \l_regex_right_state_int \l_regex_internal_a_int + \@@_build_transition_right:nNn \@@_action_free_group:n + \l_@@_right_state_int \l_@@_internal_a_int \else: - \regex_build_transition_left:NNN \regex_action_free_group:n - \l_regex_right_state_int \l_regex_internal_a_int + \@@_build_transition_left:NNN \@@_action_free_group:n + \l_@@_right_state_int \l_@@_internal_a_int \fi: - \regex_build_new_state: + \@@_build_new_state: \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_group_repeat:nnnN} +% \begin{macro}[aux]{\@@_group_repeat:nnnN} % We wish to repeat the group between |#2| and $|#2|+|#3|$ times, with % a lazyness controlled by |#4|. We insert submatch tracking up front: % in principle, we could avoid recording submatches for the first |#2| @@ -3851,106 +3866,106 @@ % not follow the normal pattern: we had to add it \enquote{by hand} % earlier. % \begin{macrocode} -\cs_new_protected:Npn \regex_group_repeat:nnnN #1#2#3#4 +\cs_new_protected:Npn \@@_group_repeat:nnnN #1#2#3#4 { - \regex_group_submatches:nNN {#1} - \l_regex_left_state_int \l_regex_right_state_int - \regex_group_repeat_aux:n { #2 + #3 } + \@@_group_submatches:nNN {#1} + \l_@@_left_state_int \l_@@_right_state_int + \@@_group_repeat_aux:n { #2 + #3 } \if_meaning:w \c_true_bool #4 - \int_set_eq:NN \l_regex_left_state_int \l_regex_max_state_int + \int_set_eq:NN \l_@@_left_state_int \l_@@_max_state_int \prg_replicate:nn { #3 } { - \int_sub:Nn \l_regex_left_state_int - { \l_regex_internal_b_int - \l_regex_internal_a_int } - \regex_build_transition_left:NNN \regex_action_free:n - \l_regex_left_state_int \l_regex_max_state_int + \int_sub:Nn \l_@@_left_state_int + { \l_@@_internal_b_int - \l_@@_internal_a_int } + \@@_build_transition_left:NNN \@@_action_free:n + \l_@@_left_state_int \l_@@_max_state_int } \else: \prg_replicate:nn { #3 - \c_one } { - \int_sub:Nn \l_regex_right_state_int - { \l_regex_internal_b_int - \l_regex_internal_a_int } - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_right_state_int \l_regex_max_state_int + \int_sub:Nn \l_@@_right_state_int + { \l_@@_internal_b_int - \l_@@_internal_a_int } + \@@_build_transition_right:nNn \@@_action_free:n + \l_@@_right_state_int \l_@@_max_state_int } - \if_num:w #2 = \c_zero - \int_set:Nn \l_regex_right_state_int - { \l_regex_left_state_int - \c_one } + \if_int_compare:w #2 = \c_zero + \int_set:Nn \l_@@_right_state_int + { \l_@@_left_state_int - \c_one } \else: - \int_sub:Nn \l_regex_right_state_int - { \l_regex_internal_b_int - \l_regex_internal_a_int } + \int_sub:Nn \l_@@_right_state_int + { \l_@@_internal_b_int - \l_@@_internal_a_int } \fi: - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_right_state_int \l_regex_max_state_int + \@@_build_transition_right:nNn \@@_action_free:n + \l_@@_right_state_int \l_@@_max_state_int \fi: - \regex_build_new_state: + \@@_build_new_state: } % \end{macrocode} % \end{macro} % % \subsubsection{Others} % -% \begin{macro}[int]{\regex_assertion:Nn, \regex_b_test:, \regex_anchor:N} -% Usage: \cs{regex_assertion:Nn} \meta{boolean} \Arg{test}, where the +% \begin{macro}[int]{\@@_assertion:Nn, \@@_b_test:, \@@_anchor:N} +% Usage: \cs{@@_assertion:Nn} \meta{boolean} \Arg{test}, where the % \meta{test} is either of the two other functions. Add a free % transition to a new state, conditionally to the assertion test. The -% \cs{regex_b_test:} test is used by the |\b| and |\B| escape: check +% \cs{@@_b_test:} test is used by the |\b| and |\B| escape: check % if the last character was a word character or not, and do the same % to the current character. The boundary-markers of the string are % non-word characters for this purpose. Anchors at the start or end -% of match use \cs{regex_anchor:N}, with a position controlled by the +% of match use \cs{@@_anchor:N}, with a position controlled by the % integer |#1|. % \begin{macrocode} -\cs_new_protected:Npn \regex_assertion:Nn #1#2 +\cs_new_protected:Npn \@@_assertion:Nn #1#2 { - \regex_build_new_state: - \regex_toks_put_right:Nx \l_regex_left_state_int + \@@_build_new_state: + \@@_toks_put_right:Nx \l_@@_left_state_int { \exp_not:n {#2} - \regex_break_point:TF + \@@_break_point:TF \bool_if:NF #1 { { } } { - \regex_action_free:n + \@@_action_free:n { \int_eval:n - { \l_regex_right_state_int - \l_regex_left_state_int } + { \l_@@_right_state_int - \l_@@_left_state_int } } } \bool_if:NT #1 { { } } } } -\cs_new_protected:Npn \regex_anchor:N #1 +\cs_new_protected:Npn \@@_anchor:N #1 { - \if_num:w #1 = \l_regex_current_pos_int - \exp_after:wN \regex_break_true:w + \if_int_compare:w #1 = \l_@@_current_pos_int + \exp_after:wN \@@_break_true:w \fi: } -\cs_new_protected_nopar:Npn \regex_b_test: +\cs_new_protected_nopar:Npn \@@_b_test: { \group_begin: - \int_set_eq:NN \l_regex_current_char_int \l_regex_last_char_int - \regex_prop_w: - \regex_break_point:TF - { \group_end: \regex_item_reverse:n \regex_prop_w: } - { \group_end: \regex_prop_w: } + \int_set_eq:NN \l_@@_current_char_int \l_@@_last_char_int + \@@_prop_w: + \@@_break_point:TF + { \group_end: \@@_item_reverse:n \@@_prop_w: } + { \group_end: \@@_prop_w: } } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_command_K:} +% \begin{macro}[int]{\@@_command_K:} % Change the starting point of the $0$-th submatch (full match), and % transition to a new state, pretending that this is a fresh thread. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_command_K: +\cs_new_protected_nopar:Npn \@@_command_K: { - \regex_build_new_state: - \regex_toks_put_right:Nx \l_regex_left_state_int + \@@_build_new_state: + \@@_toks_put_right:Nx \l_@@_left_state_int { - \regex_action_submatch:n { 0< } - \bool_set_true:N \l_regex_fresh_thread_bool - \regex_action_free:n - { \int_eval:n { \l_regex_right_state_int - \l_regex_left_state_int } } - \bool_set_false:N \l_regex_fresh_thread_bool + \@@_action_submatch:n { 0< } + \bool_set_true:N \l_@@_fresh_thread_bool + \@@_action_free:n + { \int_eval:n { \l_@@_right_state_int - \l_@@_left_state_int } } + \bool_set_false:N \l_@@_fresh_thread_bool } } % \end{macrocode} @@ -3991,8 +4006,8 @@ % match, should have a higher priority than not going through the group. % % We solve this by distinguishing \enquote{normal} free transitions -% \cs{regex_action_free:n} from transitions -% \cs{regex_action_free_group:n} which go back to the start of the +% \cs{@@_action_free:n} from transitions +% \cs{@@_action_free_group:n} which go back to the start of the % group. The former will keep threads unless they have been visited by a % \enquote{completed} thread, while the latter kind of transition also % prevents going back to a state visited by the current thread. @@ -4001,11 +4016,11 @@ % % \begin{variable} % { -% \l_regex_min_pos_int, -% \l_regex_max_pos_int, -% \l_regex_current_pos_int, -% \l_regex_start_pos_int, -% \l_regex_success_pos_int, +% \l_@@_min_pos_int, +% \l_@@_max_pos_int, +% \l_@@_current_pos_int, +% \l_@@_start_pos_int, +% \l_@@_success_pos_int, % } % The tokens in the query are indexed from \texttt{min_pos} for the % first to $\texttt{max_pos}-1$ for the last, and their information is @@ -4018,20 +4033,20 @@ % updated whenever a thread succeeds, is used as the next starting % position. % \begin{macrocode} -\int_new:N \l_regex_min_pos_int -\int_new:N \l_regex_max_pos_int -\int_new:N \l_regex_current_pos_int -\int_new:N \l_regex_start_pos_int -\int_new:N \l_regex_success_pos_int +\int_new:N \l_@@_min_pos_int +\int_new:N \l_@@_max_pos_int +\int_new:N \l_@@_current_pos_int +\int_new:N \l_@@_start_pos_int +\int_new:N \l_@@_success_pos_int % \end{macrocode} % \end{variable} % % \begin{variable} % { -% \l_regex_current_char_int, -% \l_regex_current_catcode_int, -% \l_regex_last_char_int, -% \l_regex_case_changed_char_int +% \l_@@_current_char_int, +% \l_@@_current_catcode_int, +% \l_@@_last_char_int, +% \l_@@_case_changed_char_int % } % The character and category codes of the token at the current % position; the character code of the token at the previous position; @@ -4041,40 +4056,40 @@ % \texttt{current_char} variable is also used in various other phases % to hold a character code. % \begin{macrocode} -\int_new:N \l_regex_current_char_int -\int_new:N \l_regex_current_catcode_int -\int_new:N \l_regex_last_char_int -\int_new:N \l_regex_case_changed_char_int +\int_new:N \l_@@_current_char_int +\int_new:N \l_@@_current_catcode_int +\int_new:N \l_@@_last_char_int +\int_new:N \l_@@_case_changed_char_int % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_regex_current_state_int} +% \begin{variable}{\l_@@_current_state_int} % For every character in the token list, each of the active states is -% considered in turn. The variable \cs{l_regex_current_state_int} +% considered in turn. The variable \cs{l_@@_current_state_int} % holds the state of the \textsc{nfa} which is currently considered: % transitions are then given as shifts relative to the current state. % \begin{macrocode} -\int_new:N \l_regex_current_state_int +\int_new:N \l_@@_current_state_int % \end{macrocode} % \end{variable} % % \begin{variable} -% {\l_regex_current_submatches_prop, \l_regex_success_submatches_prop} +% {\l_@@_current_submatches_prop, \l_@@_success_submatches_prop} % The submatches for the thread which is currently active are stored % in the \texttt{current_submatches} property list variable. This -% property list is stored by \cs{regex_action_cost:n} into the +% property list is stored by \cs{@@_action_cost:n} into the % \tn{toks} register for the target state of the transition, to be % retrieved when matching at the next position. When a thread % succeeds, this property list is copied to -% \cs{l_regex_success_submatches_prop}: only the last sucessful thread +% \cs{l_@@_success_submatches_prop}: only the last sucessful thread % will remain there. % \begin{macrocode} -\prop_new:N \l_regex_current_submatches_prop -\prop_new:N \l_regex_success_submatches_prop +\prop_new:N \l_@@_current_submatches_prop +\prop_new:N \l_@@_success_submatches_prop % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_regex_step_int} +% \begin{variable}{\l_@@_step_int} % This integer, always even, is increased every time a character in % the query is read, and not reset when doing multiple matches. For % each \meta{state} in the \textsc{nfa} we store in @@ -4089,11 +4104,11 @@ % attach each set of submatch information to a given iteration (and % automatically discard it when it corresponds to a past step). % \begin{macrocode} -\int_new:N \l_regex_step_int +\int_new:N \l_@@_step_int % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_regex_min_active_int, \l_regex_max_active_int} +% \begin{variable}{\l_@@_min_active_int, \l_@@_max_active_int} % All the currently active states are kept in order of precedence in % the \tn{skip} registers, and the corresponding submatches in the % \tn{toks}. For our purposes, those serve as an array, indexed from @@ -4102,71 +4117,71 @@ % space can immediately be reused, and \texttt{max_active} is reset to % \texttt{min_active}, effectively clearing the array. % \begin{macrocode} -\int_new:N \l_regex_min_active_int -\int_new:N \l_regex_max_active_int +\int_new:N \l_@@_min_active_int +\int_new:N \l_@@_max_active_int % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_regex_every_match_tl} +% \begin{variable}{\l_@@_every_match_tl} % Every time a match is found, this token list is used. For single % matching, the token list is empty. For multiple matching, the token % list is set to repeat the matching, after performing some operation -% which depends on the user function. See \cs{regex_single_match:} and -% \cs{regex_multi_match:n}. +% which depends on the user function. See \cs{@@_single_match:} and +% \cs{@@_multi_match:n}. % \begin{macrocode} -\tl_new:N \l_regex_every_match_tl +\tl_new:N \l_@@_every_match_tl % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_regex_fresh_thread_bool, \l_regex_empty_success_bool} -% \begin{macro}[aux]{\regex_if_two_empty_matches:F} +% \begin{variable}{\l_@@_fresh_thread_bool, \l_@@_empty_success_bool} +% \begin{macro}[aux]{\@@_if_two_empty_matches:F} % When doing multiple matches, we need to avoid infinite loops where % each iteration matches the same empty token list. When an empty % token list is matched, the next successful match of the same empty % token list is suppressed. We detect empty matches by setting -% \cs{l_regex_fresh_thread_bool} to \texttt{true} for threads which +% \cs{l_@@_fresh_thread_bool} to \texttt{true} for threads which % directly come from the start of the regex or from the |\K| command, % and testing that boolean whenever a thread succeeds. The function -% \cs{regex_if_two_empty_matches:F} is redefined at every match +% \cs{@@_if_two_empty_matches:F} is redefined at every match % attempt, depending on whether the previous match was empty or not: % if it was, then the function must cancel a purported success if it % is empty and at the same spot as the previous match; otherwise, we % definitely don't have two identical empty matches, so the function % is \cs{use:n}. % \begin{macrocode} -\bool_new:N \l_regex_fresh_thread_bool -\bool_new:N \l_regex_empty_success_bool -\cs_new_eq:NN \regex_if_two_empty_matches:F \use:n +\bool_new:N \l_@@_fresh_thread_bool +\bool_new:N \l_@@_empty_success_bool +\cs_new_eq:NN \@@_if_two_empty_matches:F \use:n % \end{macrocode} % \end{macro} % \end{variable} % % \begin{variable} % { -% \g_regex_success_bool, -% \l_regex_saved_success_bool, -% \l_regex_match_success_bool +% \g_@@_success_bool, +% \l_@@_saved_success_bool, +% \l_@@_match_success_bool % } -% The boolean \cs{l_regex_match_success_bool} is true if the current -% match attempt was successful, and \cs{g_regex_success_bool} is true +% The boolean \cs{l_@@_match_success_bool} is true if the current +% match attempt was successful, and \cs{g_@@_success_bool} is true % if there was at least one successful match. This is the only global % variable in this whole module, but we would need it to be local when % matching a control sequence with |\c{...}|. This is done by saving -% the global variable into \cs{l_regex_saved_success_bool}, which is +% the global variable into \cs{l_@@_saved_success_bool}, which is % local, hence not affected by the changes due to inner regex % functions. % \begin{macrocode} -\bool_new:N \g_regex_success_bool -\bool_new:N \l_regex_saved_success_bool -\bool_new:N \l_regex_match_success_bool +\bool_new:N \g_@@_success_bool +\bool_new:N \l_@@_saved_success_bool +\bool_new:N \l_@@_match_success_bool % \end{macrocode} % \end{variable} % % \subsubsection{Matching: framework} % -% \begin{macro}[int]{\regex_match:n} +% \begin{macro}[int]{\@@_match:n} % First store the query into \tn{toks} and \tn{muskip} registers (see -% \cs{regex_query_set:nnn}). Then initialize the variables that should +% \cs{@@_query_set:nnn}). Then initialize the variables that should % be set once for each user function (even for multiple % matches). Namely, the overall matching is not yet sucessful; none of % the states should be marked as visited (\tn{dimen} registers), and @@ -4175,39 +4190,39 @@ % smothering an empty match at the start). Once all this is set up, we % are ready for the ride. Find the first match. % \begin{macrocode} -\cs_new_protected:Npn \regex_match:n #1 +\cs_new_protected:Npn \@@_match:n #1 { -% \trace_push:nnx { regex } { 1 } { regex_match } +% \trace_push:nnx { regex } { 1 } { @@_match } % \trace:nnx { regex } { 1 } { analyzing~query~token~list } - \int_zero:N \l_regex_balance_int - \int_set:Nn \l_regex_current_pos_int { \c_two * \l_regex_max_state_int } - \regex_query_set:nnn { } { -1 } { -2 } - \int_set_eq:NN \l_regex_min_pos_int \l_regex_current_pos_int - \tl_analysis_map_inline:nn {#1} - { \regex_query_set:nnn {##1} {"##2} {##3} } - \int_set_eq:NN \l_regex_max_pos_int \l_regex_current_pos_int - \regex_query_set:nnn { } { -1 } { -2 } + \int_zero:N \l_@@_balance_int + \int_set:Nn \l_@@_current_pos_int { \c_two * \l_@@_max_state_int } + \@@_query_set:nnn { } { -1 } { -2 } + \int_set_eq:NN \l_@@_min_pos_int \l_@@_current_pos_int + \__tl_analysis_map_inline:nn {#1} + { \@@_query_set:nnn {##1} {"##2} {##3} } + \int_set_eq:NN \l_@@_max_pos_int \l_@@_current_pos_int + \@@_query_set:nnn { } { -1 } { -2 } % \trace:nnx { regex } { 1 } { initializing } - \bool_gset_false:N \g_regex_success_bool - \prg_stepwise_inline:nnnn - \l_regex_min_state_int \c_one { \l_regex_max_state_int - \c_one } + \bool_gset_false:N \g_@@_success_bool + \int_step_inline:nnnn + \l_@@_min_state_int \c_one { \l_@@_max_state_int - \c_one } { \tex_dimen:D ##1 \c_one sp \scan_stop: } - \int_set_eq:NN \l_regex_min_active_int \l_regex_max_state_int - \int_set_eq:NN \l_regex_step_int \c_zero - \int_set_eq:NN \l_regex_success_pos_int \l_regex_min_pos_int - \int_set:Nn \l_regex_submatch_int - { \c_two * \l_regex_max_state_int } - \bool_set_false:N \l_regex_empty_success_bool - \regex_match_once: -% \trace_pop:nnx { regex } { 1 } { regex_match } + \int_set_eq:NN \l_@@_min_active_int \l_@@_max_state_int + \int_set_eq:NN \l_@@_step_int \c_zero + \int_set_eq:NN \l_@@_success_pos_int \l_@@_min_pos_int + \int_set:Nn \l_@@_submatch_int + { \c_two * \l_@@_max_state_int } + \bool_set_false:N \l_@@_empty_success_bool + \@@_match_once: +% \trace_pop:nnx { regex } { 1 } { @@_match } } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_match_once:} +% \begin{macro}[int]{\@@_match_once:} % This function finds one match, then does some action defined by the % \texttt{every_match} token list, which may recursively call -% \cs{regex_match_once:}. First initialize some variables: set the +% \cs{@@_match_once:}. First initialize some variables: set the % conditional which detects identical empty matches; this match % attempt starts at the previous \texttt{success_pos}, is not yet % sucessful, and has no submatches yet; clear the array of active @@ -4215,148 +4230,149 @@ % ready to read our first token in the query, but we actually start % one position earlier than the start, and \texttt{get} that token, so % that the \texttt{last_char} will be set properly for word -% boundaries. Then call \cs{regex_match_loop:}, which runs through the +% boundaries. Then call \cs{@@_match_loop:}, which runs through the % query until the end or until a successful match breaks early. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_match_once: +\cs_new_protected_nopar:Npn \@@_match_once: { - \if_meaning:w \c_true_bool \l_regex_empty_success_bool - \cs_set_nopar:Npn \regex_if_two_empty_matches:F - { \int_compare:nNnF \l_regex_start_pos_int = \l_regex_current_pos_int } + \if_meaning:w \c_true_bool \l_@@_empty_success_bool + \cs_set_nopar:Npn \@@_if_two_empty_matches:F + { \int_compare:nNnF \l_@@_start_pos_int = \l_@@_current_pos_int } \else: - \cs_set_eq:NN \regex_if_two_empty_matches:F \use:n + \cs_set_eq:NN \@@_if_two_empty_matches:F \use:n \fi: - \int_set_eq:NN \l_regex_start_pos_int \l_regex_success_pos_int - \bool_set_false:N \l_regex_match_success_bool - \prop_clear:N \l_regex_current_submatches_prop - \int_set_eq:NN \l_regex_max_active_int \l_regex_min_active_int - \regex_store_state:n { \l_regex_min_state_int } - \int_set:Nn \l_regex_current_pos_int - { \l_regex_start_pos_int - \c_one } - \regex_query_get: - \regex_match_loop: - \l_regex_every_match_tl + \int_set_eq:NN \l_@@_start_pos_int \l_@@_success_pos_int + \bool_set_false:N \l_@@_match_success_bool + \prop_clear:N \l_@@_current_submatches_prop + \int_set_eq:NN \l_@@_max_active_int \l_@@_min_active_int + \@@_store_state:n { \l_@@_min_state_int } + \int_set:Nn \l_@@_current_pos_int + { \l_@@_start_pos_int - \c_one } + \@@_query_get: + \@@_match_loop: + \l_@@_every_match_tl } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_single_match:, \regex_multi_match:n} +% \begin{macro}[int]{\@@_single_match:, \@@_multi_match:n} % For a single match, the overall success is determined by whether the % only match attempt is a success. When doing multiple matches, the % overall matching is successful as soon as any match % succeeds. Perform the action |#1|, then find the next match. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_single_match: +\cs_new_protected_nopar:Npn \@@_single_match: { - \tl_set:Nn \l_regex_every_match_tl - { \bool_gset_eq:NN \g_regex_success_bool \l_regex_match_success_bool } + \tl_set:Nn \l_@@_every_match_tl + { \bool_gset_eq:NN \g_@@_success_bool \l_@@_match_success_bool } } -\cs_new_protected:Npn \regex_multi_match:n #1 +\cs_new_protected:Npn \@@_multi_match:n #1 { - \tl_set:Nn \l_regex_every_match_tl + \tl_set:Nn \l_@@_every_match_tl { - \if_meaning:w \c_true_bool \l_regex_match_success_bool - \bool_gset_true:N \g_regex_success_bool + \if_meaning:w \c_true_bool \l_@@_match_success_bool + \bool_gset_true:N \g_@@_success_bool #1 - \exp_after:wN \regex_match_once: + \exp_after:wN \@@_match_once: \fi: } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_match_loop:} -% \begin{macro}[aux, rEXP]{\regex_match_one_active:w} +% \begin{macro}[aux]{\@@_match_loop:} +% \begin{macro}[aux, rEXP]{\@@_match_one_active:w} % At each new position, set some variables and get the new character % and category from the query. Then unpack the array of active % threads, and clear it by resetting its length % (\texttt{max_active}). This results in a sequence of -% \cs{regex_use_state_and_submatches:nn} \Arg{state} \Arg{prop}, and +% \cs{@@_use_state_and_submatches:nn} \Arg{state} \Arg{prop}, and % we consider those states one by one in order. As soon as a thread % succeeds, exit the step, and, if there are threads to consider at the % next position, and we have not reached the end of the string, % repeat the loop. Otherwise, the last thread that succeeded is what -% \cs{regex_match_once:} matches. We explain the \texttt{fresh_thread} -% business when describing \cs{regex_action_wildcard:}. +% \cs{@@_match_once:} matches. We explain the \texttt{fresh_thread} +% business when describing \cs{@@_action_wildcard:}. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_match_loop: +\cs_new_protected_nopar:Npn \@@_match_loop: { - \tex_advance:D \l_regex_step_int \c_two - \int_incr:N \l_regex_current_pos_int - \int_set_eq:NN \l_regex_last_char_int \l_regex_current_char_int - \int_set_eq:NN \l_regex_case_changed_char_int \c_max_int - \regex_query_get: + \tex_advance:D \l_@@_step_int \c_two + \int_incr:N \l_@@_current_pos_int + \int_set_eq:NN \l_@@_last_char_int \l_@@_current_char_int + \int_set_eq:NN \l_@@_case_changed_char_int \c_max_int + \@@_query_get: \use:x { - \int_set_eq:NN \l_regex_max_active_int \l_regex_min_active_int - \exp_after:wN \regex_match_one_active:w - \int_use:N \l_regex_min_active_int ; + \int_set_eq:NN \l_@@_max_active_int \l_@@_min_active_int + \exp_after:wN \@@_match_one_active:w + \int_use:N \l_@@_min_active_int ; } - \prg_break_point:n { \bool_set_false:N \l_regex_fresh_thread_bool } - \if_num:w \l_regex_max_active_int > \l_regex_min_active_int - \if_num:w \l_regex_current_pos_int < \l_regex_max_pos_int - \exp_after:wN \exp_after:wN \exp_after:wN \regex_match_loop: + \__prg_break_point: + \bool_set_false:N \l_@@_fresh_thread_bool %^^A was arg of break_point:n + \if_int_compare:w \l_@@_max_active_int > \l_@@_min_active_int + \if_int_compare:w \l_@@_current_pos_int < \l_@@_max_pos_int + \exp_after:wN \exp_after:wN \exp_after:wN \@@_match_loop: \fi: \fi: } -\cs_new:Npn \regex_match_one_active:w #1; +\cs_new:Npn \@@_match_one_active:w #1; { - \if_num:w #1 < \l_regex_max_active_int - \regex_use_state_and_submatches:nn - { \int_value:w \tex_skip:D #1 } + \if_int_compare:w #1 < \l_@@_max_active_int + \@@_use_state_and_submatches:nn + { \__int_value:w \tex_skip:D #1 } { \tex_the:D \tex_toks:D #1 } - \exp_after:wN \regex_match_one_active:w - \int_use:N \int_eval:w #1 + \c_one \exp_after:wN ; + \exp_after:wN \@@_match_one_active:w + \int_use:N \__int_eval:w #1 + \c_one \exp_after:wN ; \fi: } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[aux]{\regex_query_set:nnn} +% \begin{macro}[aux]{\@@_query_set:nnn} % The arguments are: tokens that \texttt{o} and \texttt{x} expand to % one token of the query, the catcode, and the character code. Store % those, and the current brace balance (used later to check for % overall brace balance) in a \tn{muskip} register and a \tn{toks}, % then update the \texttt{balance}. % \begin{macrocode} -\cs_new_protected:Npn \regex_query_set:nnn #1#2#3 +\cs_new_protected:Npn \@@_query_set:nnn #1#2#3 { - \tex_muskip:D \l_regex_current_pos_int + \tex_muskip:D \l_@@_current_pos_int = \etex_gluetomu:D #3 sp plus #2 sp - minus \l_regex_balance_int sp + minus \l_@@_balance_int sp \scan_stop: - \tex_toks:D \l_regex_current_pos_int {#1} - \int_incr:N \l_regex_current_pos_int + \tex_toks:D \l_@@_current_pos_int {#1} + \int_incr:N \l_@@_current_pos_int \if_case:w #2 \exp_stop_f: - \or: \int_incr:N \l_regex_balance_int - \or: \int_decr:N \l_regex_balance_int + \or: \int_incr:N \l_@@_balance_int + \or: \int_decr:N \l_@@_balance_int \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_query_get:} +% \begin{macro}[aux]{\@@_query_get:} % Extract the current character and category codes from the % \tn{muskip} register of the current position: those are the main and % the stretch components, and we need a conversion to avoid \TeX{}'s % \enquote{incompatible glue units} error. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_query_get: +\cs_new_protected_nopar:Npn \@@_query_get: { - \l_regex_current_char_int - = \etex_mutoglue:D \tex_muskip:D \l_regex_current_pos_int - \l_regex_current_catcode_int = \etex_gluestretch:D - \etex_mutoglue:D \tex_muskip:D \l_regex_current_pos_int + \l_@@_current_char_int + = \etex_mutoglue:D \tex_muskip:D \l_@@_current_pos_int + \l_@@_current_catcode_int = \etex_gluestretch:D + \etex_mutoglue:D \tex_muskip:D \l_@@_current_pos_int } % \end{macrocode} % \end{macro} % % \subsubsection{Using states of the \textsc{nfa}} % -% \begin{macro}[int]{\regex_use_state:} +% \begin{macro}[int]{\@@_use_state:} % Use the current \textsc{nfa} instruction. The state is initially % marked as belonging to the current \texttt{step}: this allows normal % free transition to repeat, but group-repeating transitions @@ -4364,32 +4380,33 @@ % state is marked as $\texttt{step}+1$: any thread hitting it at that % point will be terminated. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_use_state: +\cs_new_protected_nopar:Npn \@@_use_state: { %<*trace> - \trace:nnx { regex } { 2 } { state~\int_use:N \l_regex_current_state_int } + \trace:nnx { regex } { 2 } { state~\int_use:N \l_@@_current_state_int } % - \tex_dimen:D \l_regex_current_state_int - = \l_regex_step_int sp \scan_stop: - \tex_the:D \tex_toks:D \l_regex_current_state_int - \tex_dimen:D \l_regex_current_state_int - = \int_eval:w \l_regex_step_int + \c_one \int_eval_end: sp \scan_stop: + \tex_dimen:D \l_@@_current_state_int + = \l_@@_step_int sp \scan_stop: + \tex_the:D \tex_toks:D \l_@@_current_state_int + \tex_dimen:D \l_@@_current_state_int + = \__int_eval:w \l_@@_step_int + \c_one \__int_eval_end: sp \scan_stop: } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_use_state_and_submatches:nn} +% \begin{macro}[int]{\@@_use_state_and_submatches:nn} % This function is called as one item in the array of active threads % after that array has been unpacked for a new step. Update the % \texttt{current_state} and \texttt{current_submatches} and use the % state if it has not yet been encountered at this step. % \begin{macrocode} -\cs_new_protected:Npn \regex_use_state_and_submatches:nn #1 #2 +\cs_new_protected:Npn \@@_use_state_and_submatches:nn #1 #2 { - \int_set:Nn \l_regex_current_state_int {#1} - \if_num:w \tex_dimen:D \l_regex_current_state_int < \l_regex_step_int - \tl_set:Nn \l_regex_current_submatches_prop {#2} - \exp_after:wN \regex_use_state: + \int_set:Nn \l_@@_current_state_int {#1} + \if_int_compare:w \tex_dimen:D \l_@@_current_state_int + < \l_@@_step_int + \tl_set:Nn \l_@@_current_submatches_prop {#2} + \exp_after:wN \@@_use_state: \fi: \scan_stop: } @@ -4398,31 +4415,31 @@ % % \subsubsection{Actions when matching} % -% \begin{macro}[int]{\regex_action_start_wildcard:} +% \begin{macro}[int]{\@@_action_start_wildcard:} % For an unanchored match, state $0$ has a free transition to the next % and a costly one to itself, to repeat at the next position. To catch % repeated identical empty matches, we need to know if a successful % thread corresponds to an empty match. The instruction resetting -% \cs{l_regex_fresh_thread_bool} may be skipped by a successful -% thread, hence we had to add it to \cs{regex_match_loop:} too. +% \cs{l_@@_fresh_thread_bool} may be skipped by a successful +% thread, hence we had to add it to \cs{@@_match_loop:} too. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_action_start_wildcard: +\cs_new_protected_nopar:Npn \@@_action_start_wildcard: { - \bool_set_true:N \l_regex_fresh_thread_bool - \regex_action_free:n {1} - \bool_set_false:N \l_regex_fresh_thread_bool - \regex_action_cost:n {0} + \bool_set_true:N \l_@@_fresh_thread_bool + \@@_action_free:n {1} + \bool_set_false:N \l_@@_fresh_thread_bool + \@@_action_cost:n {0} } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_action_free:n, \regex_action_free_group:n} -% \begin{macro}[aux]{\regex_action_free_aux:nn} +% \begin{macro}[int]{\@@_action_free:n, \@@_action_free_group:n} +% \begin{macro}[aux]{\@@_action_free_aux:nn} % These functions copy a thread after checking that the \textsc{nfa} % state has not already been used at this position. If not, store % submatches in the new state, and insert the instructions for that % state in the input stream. Then restore the old value of -% \cs{l_regex_current_state_int} and of the current submatches. The +% \cs{l_@@_current_state_int} and of the current submatches. The % two types of free transitions differ by how they test that the state % has not been encountered yet: the \texttt{group} version is % stricter, and will not use a state if it was used earlier in the @@ -4430,116 +4447,116 @@ % \enquote{normal} version will revisit a state when within the thread % itself. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_action_free:n - { \regex_action_free_aux:nn { > \l_regex_step_int \else: } } -\cs_new_protected_nopar:Npn \regex_action_free_group:n - { \regex_action_free_aux:nn { < \l_regex_step_int } } -\cs_new_protected:Npn \regex_action_free_aux:nn #1#2 +\cs_new_protected_nopar:Npn \@@_action_free:n + { \@@_action_free_aux:nn { > \l_@@_step_int \else: } } +\cs_new_protected_nopar:Npn \@@_action_free_group:n + { \@@_action_free_aux:nn { < \l_@@_step_int } } +\cs_new_protected:Npn \@@_action_free_aux:nn #1#2 { \use:x { - \int_add:Nn \l_regex_current_state_int {#2} + \int_add:Nn \l_@@_current_state_int {#2} \exp_not:n { - \if_num:w \tex_dimen:D \l_regex_current_state_int #1 - \exp_after:wN \regex_use_state: + \if_int_compare:w \tex_dimen:D \l_@@_current_state_int #1 + \exp_after:wN \@@_use_state: \fi: } - \int_set:Nn \l_regex_current_state_int - { \int_use:N \l_regex_current_state_int } - \tl_set:Nn \exp_not:N \l_regex_current_submatches_prop - { \exp_not:o \l_regex_current_submatches_prop } + \int_set:Nn \l_@@_current_state_int + { \int_use:N \l_@@_current_state_int } + \tl_set:Nn \exp_not:N \l_@@_current_submatches_prop + { \exp_not:o \l_@@_current_submatches_prop } } } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\regex_action_cost:n} +% \begin{macro}[int]{\@@_action_cost:n} % A transition which consumes the current character and shifts the % state by |#1|. The resulting state is stored in the \tn{skip} array % for use at the next position, and we also store the current % submatches. % \begin{macrocode} -\cs_new_protected:Npn \regex_action_cost:n #1 +\cs_new_protected:Npn \@@_action_cost:n #1 { - \exp_args:No \regex_store_state:n - { \int_use:N \int_eval:w \l_regex_current_state_int + #1 } + \exp_args:No \@@_store_state:n + { \int_use:N \__int_eval:w \l_@@_current_state_int + #1 } } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_store_state:n} -% \begin{macro}[aux]{\regex_store_submatches:} +% \begin{macro}[int]{\@@_store_state:n} +% \begin{macro}[aux]{\@@_store_submatches:} % Put the given state in the array of \tn{skip} registers (converted % to a dimension in scaled points), and increment the length of the % array. Then store the current submatch in the This is done by -% increasing the pointer \cs{l_regex_max_active_int}, and converting +% increasing the pointer \cs{l_@@_max_active_int}, and converting % the integer to a dimension (suitable for a \tn{skip} assignment) in % scaled points. % \begin{macrocode} -\cs_new_protected:Npn \regex_store_state:n #1 +\cs_new_protected:Npn \@@_store_state:n #1 { - \regex_store_submatches: - \tex_skip:D \l_regex_max_active_int = #1 sp \scan_stop: - \int_incr:N \l_regex_max_active_int + \@@_store_submatches: + \tex_skip:D \l_@@_max_active_int = #1 sp \scan_stop: + \int_incr:N \l_@@_max_active_int } -\cs_new_protected_nopar:Npn \regex_store_submatches: +\cs_new_protected_nopar:Npn \@@_store_submatches: { - \tex_toks:D \l_regex_max_active_int \exp_after:wN - { \l_regex_current_submatches_prop } + \tex_toks:D \l_@@_max_active_int \exp_after:wN + { \l_@@_current_submatches_prop } } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\regex_disable_submatches:} +% \begin{macro}[int]{\@@_disable_submatches:} % Some user functions don't require tracking submatches. % We get a performance improvement by simply defining the % relevant functions to remove their argument and do nothing % with it. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_disable_submatches: +\cs_new_protected_nopar:Npn \@@_disable_submatches: { - \cs_set_protected_nopar:Npn \regex_store_submatches: { } - \cs_set_protected:Npn \regex_action_submatch:n ##1 { } + \cs_set_protected_nopar:Npn \@@_store_submatches: { } + \cs_set_protected:Npn \@@_action_submatch:n ##1 { } } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_action_submatch:n} +% \begin{macro}[int]{\@@_action_submatch:n} % Update the current submatches with the information from the current % position. Maybe a bottleneck. % \begin{macrocode} -\cs_new_protected:Npn \regex_action_submatch:n #1 +\cs_new_protected:Npn \@@_action_submatch:n #1 { - \prop_put:Nno \l_regex_current_submatches_prop {#1} - { \int_use:N \l_regex_current_pos_int } + \prop_put:Nno \l_@@_current_submatches_prop {#1} + { \int_use:N \l_@@_current_pos_int } } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\regex_action_success:} +% \begin{macro}[int]{\@@_action_success:} % There is a successful match when an execution path reaches the last % state in the \textsc{nfa}, unless this marks a second identical % empty match. Then mark that there was a successful match; it is % empty if it is \enquote{fresh}; and we store the current position % and submatches. The current step is then interrupted with -% \cs{prg_map_break:}, and only paths with higher precedence are +% \cs{__prg_break:}, and only paths with higher precedence are % pursued further. The values stored here may be overwritten by a % later success of a path with higher precedence. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_action_success: +\cs_new_protected_nopar:Npn \@@_action_success: { - \regex_if_two_empty_matches:F + \@@_if_two_empty_matches:F { - \bool_set_true:N \l_regex_match_success_bool - \bool_set_eq:NN \l_regex_empty_success_bool - \l_regex_fresh_thread_bool - \int_set_eq:NN \l_regex_success_pos_int \l_regex_current_pos_int - \prop_set_eq:NN \l_regex_success_submatches_prop - \l_regex_current_submatches_prop - \prg_map_break: + \bool_set_true:N \l_@@_match_success_bool + \bool_set_eq:NN \l_@@_empty_success_bool + \l_@@_fresh_thread_bool + \int_set_eq:NN \l_@@_success_pos_int \l_@@_current_pos_int + \prop_set_eq:NN \l_@@_success_submatches_prop + \l_@@_current_submatches_prop + \__prg_break: } } % \end{macrocode} @@ -4549,27 +4566,27 @@ % % \subsubsection{Variables and helpers used in replacement} % -% \begin{variable}{\l_regex_replacement_csnames_int} +% \begin{variable}{\l_@@_replacement_csnames_int} % The behaviour of closing braces inside a replacement text depends on % whether a sequences |\c{| or |\u{| has been encountered. The number % of \enquote{open} such sequences that should be closed by |}| is -% stored in \cs{l_regex_replacement_csnames_int}, and decreased by +% stored in \cs{l_@@_replacement_csnames_int}, and decreased by % $1$ by each |}|. % \begin{macrocode} -\int_new:N \l_regex_replacement_csnames_int +\int_new:N \l_@@_replacement_csnames_int % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_regex_balance_tl} +% \begin{variable}{\l_@@_balance_tl} % This token list holds the replacement text for -% \cs{regex_replacement_balance_one_match:n} while it is being built +% \cs{@@_replacement_balance_one_match:n} while it is being built % incrementally. % \begin{macrocode} -\tl_new:N \l_regex_balance_tl +\tl_new:N \l_@@_balance_tl % \end{macrocode} % \end{variable} % -% \begin{macro}[aux, rEXP]{\regex_replacement_balance_one_match:n} +% \begin{macro}[aux, rEXP]{\@@_replacement_balance_one_match:n} % This expects as an argument the first index of a range of \tn{skip} % registers which hold the submatch information for a given match. It % can be used within an integer expression to obtain the brace balance @@ -4582,13 +4599,13 @@ % function must result in a valid integer expression (hence a leading % |+| in the actual definition). % \begin{macrocode} -\cs_new:Npn \regex_replacement_balance_one_match:n #1 - { - \regex_submatch_balance:n {#1} } +\cs_new:Npn \@@_replacement_balance_one_match:n #1 + { - \@@_submatch_balance:n {#1} } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux, rEXP]{\regex_replacement_do_one_match:n} -% The input is the same as \cs{regex_replacement_balance_one_match:n}. +% \begin{macro}[aux, rEXP]{\@@_replacement_do_one_match:n} +% The input is the same as \cs{@@_replacement_balance_one_match:n}. % This function is redefined to expand to the part of the token list % from the end of the previous match to a given match, followed by the % replacement text. Hence concatenating the result of this function @@ -4597,16 +4614,16 @@ % will produce the fully replaced token list. The initialization does % not matter, but we set it as for an empty replacement. % \begin{macrocode} -\cs_new:Npn \regex_replacement_do_one_match:n #1 +\cs_new:Npn \@@_replacement_do_one_match:n #1 { - \regex_query_range:nn + \@@_query_range:nn { \etex_glueshrink:D \tex_skip:D #1 } { \tex_skip:D #1 } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_exp_not:N} +% \begin{macro}[aux]{\@@_replacement_exp_not:N} % This function lets us navigate around the fact that the primitive % \cs{exp_not:n} requires a braced argument. As far as I can tell, it % is only needed if the user tries to include in the replacement text @@ -4615,18 +4632,18 @@ % assignment, \cs{exp_not:N}~|#| behaves as a single |#|, whereas % \cs{exp_not:n}~|{#}| behaves as a doubled |##|. % \begin{macrocode} -\cs_new:Npn \regex_replacement_exp_not:N #1 { \exp_not:n {#1} } +\cs_new:Npn \@@_replacement_exp_not:N #1 { \exp_not:n {#1} } % \end{macrocode} % \end{macro} % % \subsubsection{Query and brace balance} % -% \begin{macro}[int, rEXP]{\regex_query_range:nn} -% \begin{macro}[aux, rEXP]{\regex_query_range_loop:ww} +% \begin{macro}[int, rEXP]{\@@_query_range:nn} +% \begin{macro}[aux, rEXP]{\@@_query_range_loop:ww} % When it is time to extract submatches from the token list, the % various tokens are stored in \tn{toks} registers numbered from -% \cs{l_regex_min_pos_int} inclusive to \cs{l_regex_max_pos_int} -% exclusive. The function \cs{regex_query_range:nn} \Arg{min} +% \cs{l_@@_min_pos_int} inclusive to \cs{l_@@_max_pos_int} +% exclusive. The function \cs{@@_query_range:nn} \Arg{min} % \Arg{max} unpacks registers from the position \meta{min} to the % position $\meta{max}-1$ included. Once this is expanded, a second % \texttt{x}-expansion will result in the actual tokens from the @@ -4634,44 +4651,44 @@ % very end of their operation, after checking (and correcting) the % brace balance first. % \begin{macrocode} -\cs_new:Npn \regex_query_range:nn #1#2 +\cs_new:Npn \@@_query_range:nn #1#2 { - \exp_after:wN \regex_query_range_loop:ww - \int_use:N \int_eval:w #1 \exp_after:wN ; - \int_use:N \int_eval:w #2 ; - \prg_break_point:n { } + \exp_after:wN \@@_query_range_loop:ww + \int_use:N \__int_eval:w #1 \exp_after:wN ; + \int_use:N \__int_eval:w #2 ; + \__prg_break_point: } -\cs_new:Npn \regex_query_range_loop:ww #1 ; #2 ; +\cs_new:Npn \@@_query_range_loop:ww #1 ; #2 ; { - \if_num:w #1 < #2 \exp_stop_f: + \if_int_compare:w #1 < #2 \exp_stop_f: \else: - \exp_after:wN \prg_map_break: + \exp_after:wN \__prg_break: \fi: \tex_the:D \tex_toks:D #1 \exp_stop_f: - \exp_after:wN \regex_query_range_loop:ww - \int_use:N \int_eval:w #1 + \c_one ; #2 ; + \exp_after:wN \@@_query_range_loop:ww + \int_use:N \__int_eval:w #1 + \c_one ; #2 ; } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\regex_query_submatch:n} +% \begin{macro}[int]{\@@_query_submatch:n} % When this function is called, \tn{skip}$i$ holds the start and end % positions for the $i$-th overall submatch as its main and stretch % components. In the case of repeated matches, submatches from all the % matches are put one after the other in blocks of -% \cs{l_regex_capturing_group_int} \tn{skip} registers. +% \cs{l_@@_capturing_group_int} \tn{skip} registers. % \begin{macrocode} -\cs_new:Npn \regex_query_submatch:n #1 +\cs_new:Npn \@@_query_submatch:n #1 { - \regex_query_range:nn - { \tex_skip:D \int_eval:w #1 } - { \etex_gluestretch:D \tex_skip:D \int_eval:w #1 } + \@@_query_range:nn + { \tex_skip:D \__int_eval:w #1 } + { \etex_gluestretch:D \tex_skip:D \__int_eval:w #1 } } % \end{macrocode} % \end{macro} % -% \begin{macro}[rEXP]{\regex_submatch_balance:n} +% \begin{macro}[rEXP]{\@@_submatch_balance:n} % Every user function must result in a balanced token list (unbalanced % token lists cannot be stored by TeX). When we unpacked the query, we % kept track of the brace balance as the shrink component of @@ -4683,7 +4700,7 @@ % gobbled by \cs{etex_muexpr:D}, and the whole expression can be cast % safely to an integer (no trailing expansion). % \begin{macrocode} -\cs_new_protected:Npn \regex_submatch_balance:n #1 +\cs_new_protected:Npn \@@_submatch_balance:n #1 { \etex_glueshrink:D \etex_mutoglue:D \etex_muexpr:D \tex_muskip:D \etex_gluestretch:D \tex_skip:D #1 @@ -4695,12 +4712,12 @@ % % \subsubsection{Framework} % -% \begin{macro}[int]{\regex_replacement:n} -% \begin{macro}[aux]{\regex_replacement_aux:n} +% \begin{macro}[int]{\@@_replacement:n} +% \begin{macro}[aux]{\@@_replacement_aux:n} % The replacement text is built incrementally by abusing \tn{toks} % within a group (see \pkg{l3tl-build}). We keep track in -% \cs{l_regex_balance_int} of the balance of explicit begin- and -% end-group tokens and \cs{l_regex_balance_tl} will consist of some +% \cs{l_@@_balance_int} of the balance of explicit begin- and +% end-group tokens and \cs{l_@@_balance_tl} will consist of some % code to compute the brace balance from submatches (see its % description). Detect unescaped right braces, and escaped characters, % with trailing \cs{prg_do_nothing:} because some of the later @@ -4708,42 +4725,42 @@ % parsed, make sure that there is no open csname. Finally, define the % \texttt{balance_one_match} and \texttt{do_one_match} functions. % \begin{macrocode} -\cs_new_protected:Npn \regex_replacement:n #1 +\cs_new_protected:Npn \@@_replacement:n #1 { -% \trace_push:nnn { regex } { 1 } { regex_replacement:n } - \tl_set_build:Nw \l_regex_internal_a_tl - \int_zero:N \l_regex_balance_int - \tl_clear:N \l_regex_balance_tl - \regex_escape_use:nnnn +% \trace_push:nnn { regex } { 1 } { @@_replacement:n } + \__tl_build:Nw \l_@@_internal_a_tl + \int_zero:N \l_@@_balance_int + \tl_clear:N \l_@@_balance_tl + \@@_escape_use:nnnn { \if_charcode:w \c_rbrace_str ##1 - \regex_replacement_rbrace:N \else: \tl_build_one:n \fi: ##1 + \@@_replacement_rbrace:N \else: \__tl_build_one:n \fi: ##1 } - { \regex_replacement_escaped:N ##1 } - { \tl_build_one:n ##1 } + { \@@_replacement_escaped:N ##1 } + { \__tl_build_one:n ##1 } {#1} \prg_do_nothing: \prg_do_nothing: - \if_int_compare:w \l_regex_replacement_csnames_int > \c_zero - \msg_kernel_error:nnx { regex } { replacement-missing-rbrace } - { \int_use:N \l_regex_replacement_csnames_int } - \tl_build_one:x - { \prg_replicate:nn \l_regex_replacement_csnames_int \cs_end: } + \if_int_compare:w \l_@@_replacement_csnames_int > \c_zero + \__msg_kernel_error:nnx { regex } { replacement-missing-rbrace } + { \int_use:N \l_@@_replacement_csnames_int } + \__tl_build_one:x + { \prg_replicate:nn \l_@@_replacement_csnames_int \cs_end: } \fi: - \cs_gset:Npx \regex_replacement_balance_one_match:n ##1 + \cs_gset:Npx \@@_replacement_balance_one_match:n ##1 { - + \int_use:N \l_regex_balance_int - \l_regex_balance_tl - - \regex_submatch_balance:n {##1} + + \int_use:N \l_@@_balance_int + \l_@@_balance_tl + - \@@_submatch_balance:n {##1} } - \tl_build_end: - \exp_args:No \regex_replacement_aux:n \l_regex_internal_a_tl -% \trace_pop:nnn { regex } { 1 } { regex_replacement:n } + \__tl_build_end: + \exp_args:No \@@_replacement_aux:n \l_@@_internal_a_tl +% \trace_pop:nnn { regex } { 1 } { @@_replacement:n } } -\cs_new_protected:Npn \regex_replacement_aux:n #1 +\cs_new_protected:Npn \@@_replacement_aux:n #1 { - \cs_set:Npn \regex_replacement_do_one_match:n ##1 + \cs_set:Npn \@@_replacement_do_one_match:n ##1 { - \regex_query_range:nn + \@@_query_range:nn { \etex_glueshrink:D \tex_skip:D ##1 } { \tex_skip:D ##1 } #1 @@ -4753,19 +4770,19 @@ % \end{macro} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_escaped:N} +% \begin{macro}[aux]{\@@_replacement_escaped:N} % As in parsing a regular expression, we use an auxiliary built from % |#1| if defined. Otherwise, check for escaped digits (standing from % submatches from $0$ to $9$): anything else is a raw character. % \begin{macrocode} -\cs_new_protected:Npn \regex_replacement_escaped:N #1 +\cs_new_protected:Npn \@@_replacement_escaped:N #1 { - \cs_if_exist_use:cF { regex_replacement_#1:w } + \cs_if_exist_use:cF { @@_replacement_#1:w } { - \if_num:w \c_one < 1#1 \exp_stop_f: - \regex_replacement_put_submatch:n {#1} + \if_int_compare:w \c_one < 1#1 \exp_stop_f: + \@@_replacement_put_submatch:n {#1} \else: - \tl_build_one:n #1 + \__tl_build_one:n #1 \fi: } } @@ -4774,7 +4791,7 @@ % % \subsubsection{Submatches} % -% \begin{macro}[aux]{\regex_replacement_put_submatch:n} +% \begin{macro}[aux]{\@@_replacement_put_submatch:n} % Insert a submatch in the replacement text. This is dropped if the % submatch number is larger than the number of capturing groups. % Unless the submatch appears inside a |\c{...}| or |\u{...}| @@ -4784,53 +4801,53 @@ % expandable, and would be expanded too early (short of adding % \cs{exp_not:N}, making the code messy again). % \begin{macrocode} -\cs_new_protected:Npn \regex_replacement_put_submatch:n #1 +\cs_new_protected:Npn \@@_replacement_put_submatch:n #1 { - \if_num:w #1 < \l_regex_capturing_group_int - \tl_build_one:n { \regex_query_submatch:n { #1 + ##1 } } - \if_num:w \l_regex_replacement_csnames_int = \c_zero - \tl_put_right:Nn \l_regex_balance_tl - { + \regex_submatch_balance:n { \int_eval:w #1+##1 \int_eval_end: } } + \if_int_compare:w #1 < \l_@@_capturing_group_int + \__tl_build_one:n { \@@_query_submatch:n { #1 + ##1 } } + \if_int_compare:w \l_@@_replacement_csnames_int = \c_zero + \tl_put_right:Nn \l_@@_balance_tl + { + \@@_submatch_balance:n { \__int_eval:w #1+##1 \__int_eval_end: } } \fi: \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_g:w, \regex_replacement_g_digits:NN} +% \begin{macro}[aux]{\@@_replacement_g:w, \@@_replacement_g_digits:NN} % An ugly method to grab digits for the |\g| escape sequence. At the % end of the run of digits, check that it ends with a right brace. % \begin{macrocode} -\cs_new_protected:Npn \regex_replacement_g:w #1#2 +\cs_new_protected:Npn \@@_replacement_g:w #1#2 { - \str_if_eq:xxTF { #1#2 } { \tl_build_one:n \c_lbrace_str } + \str_if_eq_x:nnTF { #1#2 } { \__tl_build_one:n \c_lbrace_str } { - \int_zero:N \l_regex_internal_a_int - \regex_replacement_g_digits:NN + \int_zero:N \l_@@_internal_a_int + \@@_replacement_g_digits:NN } - { \regex_replacement_error:NNN g #1 #2 } + { \@@_replacement_error:NNN g #1 #2 } } -\cs_new_protected:Npn \regex_replacement_g_digits:NN #1#2 +\cs_new_protected:Npn \@@_replacement_g_digits:NN #1#2 { - \token_if_eq_meaning:NNTF #1 \tl_build_one:n + \token_if_eq_meaning:NNTF #1 \__tl_build_one:n { - \if_num:w \c_one < 1#2 \exp_stop_f: - \int_set:Nn \l_regex_internal_a_int - { \c_ten * \l_regex_internal_a_int + #2 } + \if_int_compare:w \c_one < 1#2 \exp_stop_f: + \int_set:Nn \l_@@_internal_a_int + { \c_ten * \l_@@_internal_a_int + #2 } \exp_after:wN \use_i:nnn - \exp_after:wN \regex_replacement_g_digits:NN + \exp_after:wN \@@_replacement_g_digits:NN \else: - \exp_after:wN \regex_replacement_error:NNN + \exp_after:wN \@@_replacement_error:NNN \exp_after:wN g \fi: } { - \if_meaning:w \regex_replacement_rbrace:N #1 - \exp_args:No \regex_replacement_put_submatch:n - { \int_use:N \l_regex_internal_a_int } + \if_meaning:w \@@_replacement_rbrace:N #1 + \exp_args:No \@@_replacement_put_submatch:n + { \int_use:N \l_@@_internal_a_int } \exp_after:wN \use_none:nn \else: - \exp_after:wN \regex_replacement_error:NNN + \exp_after:wN \@@_replacement_error:NNN \exp_after:wN g \fi: } @@ -4841,8 +4858,8 @@ % % \subsubsection{Csnames in replacement} % -% \begin{macro}[aux]{\regex_replacement_c:w} -% \begin{macro}[aux]+\regex_replacement_c_{:w+ +% \begin{macro}[aux]{\@@_replacement_c:w} +% \begin{macro}[aux]+\@@_replacement_c_{:w+ % |\c| can be followed by a left brace, or by a letter for which we % have defined a way to produce that category of characters. The % appropriate definitions for catcodes are introduced later. For @@ -4851,62 +4868,62 @@ % with a weird cross-over between \cs{exp_not:n} and \cs{exp_not:N} % (see this helper's description for an explanation). % \begin{macrocode} -\cs_new_protected:Npn \regex_replacement_c:w #1#2 +\cs_new_protected:Npn \@@_replacement_c:w #1#2 { - \token_if_eq_meaning:NNTF #1 \tl_build_one:n + \token_if_eq_meaning:NNTF #1 \__tl_build_one:n { - \cs_if_exist_use:cF { regex_replacement_c_#2:w } - { \regex_replacement_error:NNN c #1#2 } + \cs_if_exist_use:cF { @@_replacement_c_#2:w } + { \@@_replacement_error:NNN c #1#2 } } - { \regex_replacement_error:NNN c #1#2 } + { \@@_replacement_error:NNN c #1#2 } } -\cs_new_protected_nopar:cpn { regex_replacement_c_ \c_lbrace_str :w } +\cs_new_protected_nopar:cpn { @@_replacement_c_ \c_lbrace_str :w } { - \if_case:w \l_regex_replacement_csnames_int - \tl_build_one:n - { \exp_not:n { \exp_after:wN \regex_replacement_exp_not:N \cs:w } } + \if_case:w \l_@@_replacement_csnames_int + \__tl_build_one:n + { \exp_not:n { \exp_after:wN \@@_replacement_exp_not:N \cs:w } } \else: - \tl_build_one:n { \exp_not:n { \exp_after:wN \tl_to_str:N \cs:w } } + \__tl_build_one:n { \exp_not:n { \exp_after:wN \tl_to_str:N \cs:w } } \fi: - \int_incr:N \l_regex_replacement_csnames_int + \int_incr:N \l_@@_replacement_csnames_int } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_u:w} +% \begin{macro}[aux]{\@@_replacement_u:w} % Check that |\u| is followed by a left brace. If so, start a control % sequence with \cs{cs:w}, which is then unpacked either with % \cs{exp_not:V} or \cs{tl_to_str:V} depending on the current context. % \begin{macrocode} -\cs_new_protected:Npn \regex_replacement_u:w #1#2 +\cs_new_protected:Npn \@@_replacement_u:w #1#2 { - \str_if_eq:xxTF { #1#2 } { \tl_build_one:n \c_lbrace_str } + \str_if_eq_x:nnTF { #1#2 } { \__tl_build_one:n \c_lbrace_str } { - \if_case:w \l_regex_replacement_csnames_int - \tl_build_one:n { \exp_not:n { \exp_after:wN \exp_not:V \cs:w } } + \if_case:w \l_@@_replacement_csnames_int + \__tl_build_one:n { \exp_not:n { \exp_after:wN \exp_not:V \cs:w } } \else: - \tl_build_one:n { \exp_not:n { \exp_after:wN \tl_to_str:V \cs:w } } + \__tl_build_one:n { \exp_not:n { \exp_after:wN \tl_to_str:V \cs:w } } \fi: - \int_incr:N \l_regex_replacement_csnames_int + \int_incr:N \l_@@_replacement_csnames_int } - { \regex_replacement_error:NNN u #1#2 } + { \@@_replacement_error:NNN u #1#2 } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_rbrace:N} +% \begin{macro}[aux]{\@@_replacement_rbrace:N} % Within a |\c{...}| or |\u{...}| construction, end the control % sequence, and decrease the brace count. Otherwise, this is a raw % right brace. % \begin{macrocode} -\cs_new_protected:Npn \regex_replacement_rbrace:N #1 +\cs_new_protected:Npn \@@_replacement_rbrace:N #1 { - \if_int_compare:w \l_regex_replacement_csnames_int > \c_zero - \tl_build_one:n \cs_end: - \int_decr:N \l_regex_replacement_csnames_int + \if_int_compare:w \l_@@_replacement_csnames_int > \c_zero + \__tl_build_one:n \cs_end: + \int_decr:N \l_@@_replacement_csnames_int \else: - \tl_build_one:n #1 + \__tl_build_one:n #1 \fi: } % \end{macrocode} @@ -4923,7 +4940,7 @@ \group_begin: % \end{macrocode} % -% \begin{macro}[aux]{\regex_replacement_char:nNN} +% \begin{macro}[aux]{\@@_replacement_char:nNN} % The only way to produce an arbitrary character--catcode pair is to % use the \tn{lowercase} or \tn{uppercase} primitives. This is a % wrapper for our purposes. The first argument is the null character @@ -4931,30 +4948,30 @@ % from the input stream: |#3| is the character whose character code to % reproduce. % \begin{macrocode} - \cs_new_protected:Npn \regex_replacement_char:nNN #1#2#3 + \cs_new_protected:Npn \@@_replacement_char:nNN #1#2#3 { \if_meaning:w \prg_do_nothing: #3 - \msg_kernel_error:nn { regex } { replacement-catcode-end } + \__msg_kernel_error:nn { regex } { replacement-catcode-end } \else: \tex_lccode:D \c_zero = `#3 \scan_stop: - \tl_to_lowercase:n { \tl_build_one:n {#1} } + \tl_to_lowercase:n { \__tl_build_one:n {#1} } \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_c_A:w} +% \begin{macro}[aux]{\@@_replacement_c_A:w} % For an active character, expansion must be avoided, twice because we % later do two \texttt{x}-expansions, to unpack \tn{toks} for the % query, and to expand their contents to tokens of the query. % \begin{macrocode} \char_set_catcode_active:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_A:w - { \regex_replacement_char:nNN { \exp_not:n { \exp_not:N ^^@ } } } + \cs_new_protected_nopar:Npn \@@_replacement_c_A:w + { \@@_replacement_char:nNN { \exp_not:n { \exp_not:N ^^@ } } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_c_B:w} +% \begin{macro}[aux]{\@@_replacement_c_B:w} % An explicit begin-group token increases the balance, unless within a % |\c{...}| or |\u{...}| construction. Add the desired begin-group % character, using the standard \cs{if_false:} trick. We eventually @@ -4964,82 +4981,82 @@ % with \pkg{l3tl-analysis}. % \begin{macrocode} \char_set_catcode_group_begin:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_B:w + \cs_new_protected_nopar:Npn \@@_replacement_c_B:w { - \if_num:w \l_regex_replacement_csnames_int = \c_zero - \int_incr:N \l_regex_balance_int + \if_int_compare:w \l_@@_replacement_csnames_int = \c_zero + \int_incr:N \l_@@_balance_int \fi: - \regex_replacement_char:nNN + \@@_replacement_char:nNN { \exp_not:n { \exp_after:wN ^^@ \if_false: } \fi: } } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_c_C:w} +% \begin{macro}[aux]{\@@_replacement_c_C:w} % This is not quite catcode-related: when the user requests a % character with category \enquote{control sequence}, the % one-character control symbol is returned. As for the active % character, we prepare for two \texttt{x}-expansions. % \begin{macrocode} - \cs_new_protected:Npn \regex_replacement_c_C:w #1#2 - { \tl_build_one:n { \exp_not:N \exp_not:N \exp_not:c {#2} } } + \cs_new_protected:Npn \@@_replacement_c_C:w #1#2 + { \__tl_build_one:n { \exp_not:N \exp_not:N \exp_not:c {#2} } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_c_D:w} +% \begin{macro}[aux]{\@@_replacement_c_D:w} % Subscripts fit the mould: \tn{lowercase} the null byte with the % correct category. % \begin{macrocode} \char_set_catcode_math_subscript:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_D:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \@@_replacement_c_D:w + { \@@_replacement_char:nNN { ^^@ } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_c_E:w} +% \begin{macro}[aux]{\@@_replacement_c_E:w} % Similar to the begin-group case, the second \texttt{x}-expansion % produces the bare end-group token. % \begin{macrocode} \char_set_catcode_group_end:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_E:w + \cs_new_protected_nopar:Npn \@@_replacement_c_E:w { - \if_num:w \l_regex_replacement_csnames_int = \c_zero - \int_decr:N \l_regex_balance_int + \if_int_compare:w \l_@@_replacement_csnames_int = \c_zero + \int_decr:N \l_@@_balance_int \fi: - \regex_replacement_char:nNN + \@@_replacement_char:nNN { \exp_not:n { \if_false: { \fi: ^^@ } } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_c_L:w} +% \begin{macro}[aux]{\@@_replacement_c_L:w} % Simply \tn{lowercase} a letter null byte to produce an arbitrary letter. % \begin{macrocode} \char_set_catcode_letter:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_L:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \@@_replacement_c_L:w + { \@@_replacement_char:nNN { ^^@ } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_c_M:w} +% \begin{macro}[aux]{\@@_replacement_c_M:w} % No surprise here, we lowercase the null math toggle. % \begin{macrocode} \char_set_catcode_math_toggle:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_M:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \@@_replacement_c_M:w + { \@@_replacement_char:nNN { ^^@ } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_c_O:w} +% \begin{macro}[aux]{\@@_replacement_c_O:w} % Lowercase an other null byte. % \begin{macrocode} \char_set_catcode_other:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_O:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \@@_replacement_c_O:w + { \@@_replacement_char:nNN { ^^@ } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_c_P:w} +% \begin{macro}[aux]{\@@_replacement_c_P:w} % For macro parameters, expansion is a tricky issue. We need to % prepare for two \texttt{x}-expansions and passing through various % macro definitions. Note that we cannot replace one \cs{exp_not:n} by @@ -5048,53 +5065,53 @@ % parameter character would be doubled. % \begin{macrocode} \char_set_catcode_parameter:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_P:w + \cs_new_protected_nopar:Npn \@@_replacement_c_P:w { - \regex_replacement_char:nNN + \@@_replacement_char:nNN { \exp_not:n { \exp_not:n { ^^@^^@^^@^^@ } } } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_c_S:w} +% \begin{macro}[aux]{\@@_replacement_c_S:w} % Spaces are normalized on input by \TeX{} to have character code % $32$. It is in fact impossible to get a token with character code % $0$ and category code $10$. Hence we use $32$ instead of $0$ as our % base character. % \begin{macrocode} - \cs_new_protected:Npn \regex_replacement_c_S:w #1#2 + \cs_new_protected:Npn \@@_replacement_c_S:w #1#2 { \if_meaning:w \prg_do_nothing: #2 - \msg_kernel_error:nn { regex } { replacement-catcode-end } + \__msg_kernel_error:nn { regex } { replacement-catcode-end } \else: - \if_num:w `#2 = \c_zero - \msg_kernel_error:nn { regex } { replacement-null-space } + \if_int_compare:w `#2 = \c_zero + \__msg_kernel_error:nn { regex } { replacement-null-space } \fi: \tex_lccode:D 32 = `#2 \scan_stop: - \tl_to_lowercase:n { \tl_build_one:n {~} } + \tl_to_lowercase:n { \__tl_build_one:n {~} } \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_c_T:w} +% \begin{macro}[aux]{\@@_replacement_c_T:w} % No surprise for alignment tabs here. Those are surrounded by the % appropriate braces whenever necessary, hence they don't cause % trouble in alignment settings. % \begin{macrocode} \char_set_catcode_alignment:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_T:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \@@_replacement_c_T:w + { \@@_replacement_char:nNN { ^^@ } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replacement_c_U:w} -% Simple call to \cs{regex_replacement_char:nNN} which lowercases the +% \begin{macro}[aux]{\@@_replacement_c_U:w} +% Simple call to \cs{@@_replacement_char:nNN} which lowercases the % math superscript |^^@|. % \begin{macrocode} \char_set_catcode_math_superscript:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_U:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \@@_replacement_c_U:w + { \@@_replacement_char:nNN { ^^@ } } % \end{macrocode} % \end{macro} % @@ -5105,14 +5122,14 @@ % % \subsubsection{An error} % -% \begin{macro}[aux]{\regex_replacement_error:NNN} +% \begin{macro}[aux]{\@@_replacement_error:NNN} % Simple error reporting by calling one of the messages % \texttt{replacement-c}, \texttt{replacement-g}, or % \texttt{replacement-u}. % \begin{macrocode} -\cs_new_protected:Npn \regex_replacement_error:NNN #1#2#3 +\cs_new_protected:Npn \@@_replacement_error:NNN #1#2#3 { - \msg_kernel_error:nnx { regex } { replacement-#1 } {#3} + \__msg_kernel_error:nnx { regex } { replacement-#1 } {#3} #2 #3 } % \end{macrocode} @@ -5125,7 +5142,7 @@ % nothing. % \begin{macrocode} \cs_new_protected:Npn \regex_new:N #1 - { \cs_new_eq:NN #1 \c_regex_no_match_regex } + { \cs_new_eq:NN #1 \c_@@_no_match_regex } % \end{macrocode} % \end{macro} % @@ -5135,18 +5152,18 @@ % \begin{macrocode} \cs_new_protected_nopar:Npn \regex_set:Nn #1#2 { - \regex_compile:n {#2} - \tl_set_eq:NN #1 \l_regex_internal_regex + \@@_compile:n {#2} + \tl_set_eq:NN #1 \l_@@_internal_regex } \cs_new_protected_nopar:Npn \regex_gset:Nn #1#2 { - \regex_compile:n {#2} - \tl_gset_eq:NN #1 \l_regex_internal_regex + \@@_compile:n {#2} + \tl_gset_eq:NN #1 \l_@@_internal_regex } \cs_new_protected_nopar:Npn \regex_const:Nn #1#2 { - \regex_compile:n {#2} - \tl_const:Nx #1 { \exp_not:o \l_regex_internal_regex } + \@@_compile:n {#2} + \tl_const:Nx #1 { \exp_not:o \l_@@_internal_regex } } % \end{macrocode} % \end{macro} @@ -5154,16 +5171,16 @@ % \begin{macro}{\regex_show:N, \regex_show:n} % User functions: the \texttt{n} variant requires compilation first. % Then show the variable with some appropriate text. The auxiliary -% \cs{regex_show_aux:Nx} is defined in a different section. +% \cs{@@_show:Nx} is defined in a different section. % \begin{macrocode} \cs_new_protected:Npn \regex_show:n #1 { - \regex_compile:n {#1} - \regex_show_aux:Nx \l_regex_internal_regex + \@@_compile:n {#1} + \@@_show:Nx \l_@@_internal_regex { { \tl_to_str:n {#1} } } } \cs_new_protected:Npn \regex_show:N #1 - { \regex_show_aux:Nx #1 { variable~\token_to_str:N #1 } } + { \@@_show:Nx #1 { variable~\token_to_str:N #1 } } % \end{macrocode} % \end{macro} % @@ -5176,13 +5193,13 @@ % \begin{macrocode} \prg_new_protected_conditional:Npnn \regex_match:nn #1#2 { T , F , TF } { - \regex_match_aux:nn { \regex_build:n {#1} } {#2} - \regex_aux_return: + \@@_if_match:nn { \@@_build:n {#1} } {#2} + \@@_return: } \prg_new_protected_conditional:Npnn \regex_match:Nn #1#2 { T , F , TF } { - \regex_match_aux:nn { \regex_build:N #1 } {#2} - \regex_aux_return: + \@@_if_match:nn { \@@_build:N #1 } {#2} + \@@_return: } % \end{macrocode} % \end{macro} @@ -5191,9 +5208,9 @@ % Again, use an auxiliary whose first argument builds the \textsc{nfa}. % \begin{macrocode} \cs_new_protected:Npn \regex_count:nnN #1 - { \regex_count_aux:nnN { \regex_build:n {#1} } } + { \@@_count:nnN { \@@_build:n {#1} } } \cs_new_protected:Npn \regex_count:NnN #1 - { \regex_count_aux:nnN { \regex_build:N #1 } } + { \@@_count:nnN { \@@_build:N #1 } } % \end{macrocode} % \end{macro} % @@ -5214,60 +5231,60 @@ % \regex_split:nnN, \regex_split:NnN % } % We define here $40$ user functions, following a common pattern in -% terms of \texttt{_aux:nnN} auxiliaries, defined in the coming -% subsections. The auxiliary is handed \cs{regex_build:n} or -% \cs{regex_build:N} with the appropriate regex argument, then all +% terms of \texttt{:nnN} auxiliaries, defined in the coming +% subsections. The auxiliary is handed \cs{@@_build:n} or +% \cs{@@_build:N} with the appropriate regex argument, then all % other necessary arguments (replacement text, token list, \emph{etc.} -% The conditionals call \cs{regex_aux_return:} to return either +% The conditionals call \cs{@@_return:} to return either % \texttt{true} or \texttt{false} once matching has been performed. % \begin{macrocode} -\cs_set_protected:Npn \regex_tmp:w #1#2#3 +\cs_set_protected:Npn \@@_tmp:w #1#2#3 { - \cs_new_protected:Npn #2 ##1 { #1 { \regex_build:n {##1} } } - \cs_new_protected:Npn #3 ##1 { #1 { \regex_build:N ##1 } } + \cs_new_protected:Npn #2 ##1 { #1 { \@@_build:n {##1} } } + \cs_new_protected:Npn #3 ##1 { #1 { \@@_build:N ##1 } } \prg_new_protected_conditional:Npnn #2 ##1##2##3 { T , F , TF } - { #1 { \regex_build:n {##1} } {##2} ##3 \regex_aux_return: } + { #1 { \@@_build:n {##1} } {##2} ##3 \@@_return: } \prg_new_protected_conditional:Npnn #3 ##1##2##3 { T , F , TF } - { #1 { \regex_build:N ##1 } {##2} ##3 \regex_aux_return: } + { #1 { \@@_build:N ##1 } {##2} ##3 \@@_return: } } -\regex_tmp:w \regex_extract_once_aux:nnN +\@@_tmp:w \@@_extract_once:nnN \regex_extract_once:nnN \regex_extract_once:NnN -\regex_tmp:w \regex_extract_all_aux:nnN +\@@_tmp:w \@@_extract_all:nnN \regex_extract_all:nnN \regex_extract_all:NnN -\regex_tmp:w \regex_replace_once_aux:nnN +\@@_tmp:w \@@_replace_once:nnN \regex_replace_once:nnN \regex_replace_once:NnN -\regex_tmp:w \regex_replace_all_aux:nnN +\@@_tmp:w \@@_replace_all:nnN \regex_replace_all:nnN \regex_replace_all:NnN -\regex_tmp:w \regex_split_aux:nnN \regex_split:nnN \regex_split:NnN +\@@_tmp:w \@@_split:nnN \regex_split:nnN \regex_split:NnN % \end{macrocode} % \end{macro} % \end{macro} % % \subsubsection{Variables and helpers for user functions} % -% \begin{variable}{\l_regex_match_count_int} +% \begin{variable}{\l_@@_match_count_int} % The number of matches found so far is stored -% in \cs{l_regex_match_count_int}. This is only used +% in \cs{l_@@_match_count_int}. This is only used % in the \cs{regex_count:nnN} functions. % \begin{macrocode} -\int_new:N \l_regex_match_count_int +\int_new:N \l_@@_match_count_int % \end{macrocode} % \end{variable} % -% \begin{variable}{regex_begin, regex_end} +% \begin{variable}{@@_begin, @@_end} % Those flags are raised to indicate extra begin-group % or end-group tokens when extracting submatches. % \begin{macrocode} -\flag_new:n { regex_begin } -\flag_new:n { regex_end } +\flag_new:n { @@_begin } +\flag_new:n { @@_end } % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_regex_submatch_int, \l_regex_zeroth_submatch_int} +% \begin{variable}{\l_@@_submatch_int, \l_@@_zeroth_submatch_int} % The end-points of each submatch are stored as main and stretch % components of \tn{skip}\meta{submatch}, where \meta{submatch} ranges -% from \cs{l_regex_max_state_int} (inclusive) to -% \cs{l_regex_submatch_int} (exclusive). Each successful match comes +% from \cs{l_@@_max_state_int} (inclusive) to +% \cs{l_@@_submatch_int} (exclusive). Each successful match comes % with a $0$-th submatch (the full match), and one match for each % capturing group: submatches corresponding to the last successful % match are labelled starting at @@ -5275,19 +5292,19 @@ % $0$-th submatch is the position at which that match attempt started: % this is used for splitting and replacements. % \begin{macrocode} -\int_new:N \l_regex_submatch_int -\int_new:N \l_regex_zeroth_submatch_int +\int_new:N \l_@@_submatch_int +\int_new:N \l_@@_zeroth_submatch_int % \end{macrocode} % \end{variable} % -% \begin{macro}[aux]{\regex_aux_return:} +% \begin{macro}[aux]{\@@_return:} % This function triggers either \cs{prg_return_false:} or % \cs{prg_return_true:} as appropriate to whether a match was found or % not. It is used by all user conditionals. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_aux_return: +\cs_new_protected_nopar:Npn \@@_return: { - \if_meaning:w \c_true_bool \g_regex_success_bool + \if_meaning:w \c_true_bool \g_@@_success_bool \prg_return_true: \else: \prg_return_false: @@ -5298,73 +5315,73 @@ % % \subsubsection{Matching} % -% \begin{macro}[aux]{\regex_match_aux:nn} +% \begin{macro}[aux]{\@@_if_match:nn} % We don't track submatches, and stop after a single match. Build the % \textsc{nfa} with |#1|, and perform the match on the query |#2|. % \begin{macrocode} -\cs_new_protected:Npn \regex_match_aux:nn #1#2 +\cs_new_protected:Npn \@@_if_match:nn #1#2 { \group_begin: - \regex_disable_submatches: - \regex_single_match: + \@@_disable_submatches: + \@@_single_match: #1 - \regex_match:n {#2} + \@@_match:n {#2} \group_end: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_count_aux:nnN} +% \begin{macro}[aux]{\@@_count:nnN} % Again, we don't care about submatches. Instead of aborting after the % first \enquote{longest match} is found, we search for multiple -% matches, incrementing \cs{l_regex_match_count_int} every time to +% matches, incrementing \cs{l_@@_match_count_int} every time to % record the number of matches. Build the \textsc{nfa} and match. At % the end, store the result in the user's variable. % \begin{macrocode} -\cs_new_protected:Npn \regex_count_aux:nnN #1#2#3 +\cs_new_protected:Npn \@@_count:nnN #1#2#3 { \group_begin: - \regex_disable_submatches: - \int_zero:N \l_regex_match_count_int - \regex_multi_match:n { \int_incr:N \l_regex_match_count_int } + \@@_disable_submatches: + \int_zero:N \l_@@_match_count_int + \@@_multi_match:n { \int_incr:N \l_@@_match_count_int } #1 - \regex_match:n {#2} + \@@_match:n {#2} \exp_args:NNNo \group_end: - \int_set:Nn #3 { \int_use:N \l_regex_match_count_int } + \int_set:Nn #3 { \int_use:N \l_@@_match_count_int } } % \end{macrocode} % \end{macro} % % \subsubsection{Extracting submatches} % -% \begin{macro}[aux]{\regex_extract_once_aux:nnN, \regex_extract_all_aux:nnN} +% \begin{macro}[aux]{\@@_extract_once:nnN, \@@_extract_all:nnN} % Match once or multiple times. After each match (or after the only -% match), extract the submatches using \cs{regex_extract:}. At the +% match), extract the submatches using \cs{@@_extract:}. At the % end, store the sequence containing all the submatches into the user % variable |#3| after closing the group. % \begin{macrocode} -\cs_new_protected:Npn \regex_extract_once_aux:nnN #1#2#3 +\cs_new_protected:Npn \@@_extract_once:nnN #1#2#3 { \group_begin: - \regex_single_match: + \@@_single_match: #1 - \regex_match:n {#2} - \regex_extract: - \regex_group_end_extract_seq:N #3 + \@@_match:n {#2} + \@@_extract: + \@@_group_end_extract_seq:N #3 } -\cs_new_protected:Npn \regex_extract_all_aux:nnN #1#2#3 +\cs_new_protected:Npn \@@_extract_all:nnN #1#2#3 { \group_begin: - \regex_multi_match:n { \regex_extract: } + \@@_multi_match:n { \@@_extract: } #1 - \regex_match:n {#2} - \regex_group_end_extract_seq:N #3 + \@@_match:n {#2} + \@@_group_end_extract_seq:N #3 } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_split_aux:nnN} +% \begin{macro}[aux]{\@@_split:nnN} % Splitting at submatches is a bit more tricky. For each match, % extract all submatches, and replace the zeroth submatch by the part % of the query between the start of the match attempt and the start of @@ -5373,101 +5390,101 @@ % match, store the last part of the token list, which ranges from the % start of the match attempt to the end of the query. This step is % inhibited if the last match was empty and at the very end: decrement -% \cs{l_regex_submatch_int}, which controls which \tn{skip} registers +% \cs{l_@@_submatch_int}, which controls which \tn{skip} registers % will be used. % \begin{macrocode} -\cs_new_protected:Npn \regex_split_aux:nnN #1#2#3 +\cs_new_protected:Npn \@@_split:nnN #1#2#3 { \group_begin: - \regex_multi_match:n + \@@_multi_match:n { - \if_num:w \l_regex_start_pos_int < \l_regex_success_pos_int - \regex_extract: - \tex_skip:D \l_regex_zeroth_submatch_int - = \l_regex_start_pos_int sp - plus \tex_skip:D \l_regex_zeroth_submatch_int \scan_stop: + \if_int_compare:w \l_@@_start_pos_int < \l_@@_success_pos_int + \@@_extract: + \tex_skip:D \l_@@_zeroth_submatch_int + = \l_@@_start_pos_int sp + plus \tex_skip:D \l_@@_zeroth_submatch_int \scan_stop: \fi: } #1 - \regex_match:n {#2} -%\assert_int:n { \l_regex_current_pos_int = \l_regex_max_pos_int } - \tex_skip:D \l_regex_submatch_int - = \l_regex_start_pos_int sp plus \l_regex_max_pos_int sp \scan_stop: - \int_incr:N \l_regex_submatch_int - \if_meaning:w \c_true_bool \l_regex_empty_success_bool - \if_num:w \l_regex_start_pos_int = \l_regex_max_pos_int - \int_decr:N \l_regex_submatch_int + \@@_match:n {#2} +%\assert_int:n { \l_@@_current_pos_int = \l_@@_max_pos_int } + \tex_skip:D \l_@@_submatch_int + = \l_@@_start_pos_int sp plus \l_@@_max_pos_int sp \scan_stop: + \int_incr:N \l_@@_submatch_int + \if_meaning:w \c_true_bool \l_@@_empty_success_bool + \if_int_compare:w \l_@@_start_pos_int = \l_@@_max_pos_int + \int_decr:N \l_@@_submatch_int \fi: \fi: - \regex_group_end_extract_seq:N #3 + \@@_group_end_extract_seq:N #3 } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_group_end_extract_seq:N} +% \begin{macro}[aux]{\@@_group_end_extract_seq:N} % The end-points of submatches are stored as the main and stretch -% components of \tn{skip} registers from \cs{l_regex_max_state_int} to -% \cs{l_regex_submatch_int} (exclusive). Extract the relevant ranges -% into \cs{l_regex_internal_a_tl}. We detect unbalanced results using -% the two flags \texttt{regex_begin} and \texttt{regex_end}, raised +% components of \tn{skip} registers from \cs{l_@@_max_state_int} to +% \cs{l_@@_submatch_int} (exclusive). Extract the relevant ranges +% into \cs{l_@@_internal_a_tl}. We detect unbalanced results using +% the two flags \texttt{@@_begin} and \texttt{@@_end}, raised % whenever we see too many begin-group or end-group tokens in a -% submatch. We disable \cs{seq_item:n} to prevent two +% submatch. We disable \cs{__seq_item:n} to prevent two % \texttt{x}-expansions. % \begin{macrocode} -\cs_new_protected:Npn \regex_group_end_extract_seq:N #1 +\cs_new_protected:Npn \@@_group_end_extract_seq:N #1 { - \cs_set_eq:NN \seq_item:n \scan_stop: - \flag_clear:n { regex_begin } - \flag_clear:n { regex_end } - \tl_set:Nx \l_regex_internal_a_tl + \cs_set_eq:NN \__seq_item:n \scan_stop: + \flag_clear:n { @@_begin } + \flag_clear:n { @@_end } + \tl_set:Nx \l_@@_internal_a_tl { - \prg_stepwise_function:nnnN - { \c_two * \l_regex_max_state_int } + \int_step_function:nnnN + { \c_two * \l_@@_max_state_int } \c_one - { \l_regex_submatch_int - \c_one } - \regex_extract_seq_aux:n + { \l_@@_submatch_int - \c_one } + \@@_extract_seq_aux:n } \int_compare:nNnF - { \flag_height:n { regex_begin } + \flag_height:n { regex_end } } + { \flag_height:n { @@_begin } + \flag_height:n { @@_end } } = \c_zero { - \msg_kernel_error:nnxxx { regex } { result-unbalanced } + \__msg_kernel_error:nnxxx { regex } { result-unbalanced } { splitting~or~extracting~submatches } - { \flag_height:n { regex_end } } - { \flag_height:n { regex_begin } } + { \flag_height:n { @@_end } } + { \flag_height:n { @@_begin } } } \use:x { \group_end: - \tl_set:Nn \exp_not:N #1 { \l_regex_internal_a_tl } + \tl_set:Nn \exp_not:N #1 { \l_@@_internal_a_tl } } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux, EXP]{\regex_extract_seq_aux:n, \regex_extract_seq_aux:ww} +% \begin{macro}[aux, EXP]{\@@_extract_seq_aux:n, \@@_extract_seq_aux:ww} % The \texttt{:n} auxiliary builds one item of the sequence of % submatches. First compute the brace balance of the submatch, then % extract the submatch from the query, adding the appropriate braces % and raising a flag if the submatch is not balanced. % \begin{macrocode} -\cs_new:Npn \regex_extract_seq_aux:n #1 +\cs_new:Npn \@@_extract_seq_aux:n #1 { - \seq_item:n + \__seq_item:n { - \exp_after:wN \regex_extract_seq_aux:ww - \int_value:w \regex_submatch_balance:n {#1} ; #1; + \exp_after:wN \@@_extract_seq_aux:ww + \__int_value:w \@@_submatch_balance:n {#1} ; #1; } } -\cs_new:Npn \regex_extract_seq_aux:ww #1; #2; +\cs_new:Npn \@@_extract_seq_aux:ww #1; #2; { - \if_num:w #1 < \c_zero - \flag_raise:n { regex_end } + \if_int_compare:w #1 < \c_zero + \flag_raise:n { @@_end } \prg_replicate:nn {-#1} { \exp_not:n { { \if_false: } \fi: } } \fi: - \regex_query_submatch:n {#2} - \if_num:w #1 > \c_zero - \flag_raise:n { regex_begin } + \@@_query_submatch:n {#2} + \if_int_compare:w #1 > \c_zero + \flag_raise:n { @@_begin } \prg_replicate:nn {#1} { \exp_not:n { \if_false: { \fi: } } } \fi: } @@ -5475,11 +5492,11 @@ % \end{macro} % % \begin{macro}[aux] -% {\regex_extract:, \regex_extract_aux_b:wn, \regex_extract_aux_e:wn} +% {\@@_extract:, \@@_extract_b:wn, \@@_extract_e:wn} % Our task here is to extract from the property list -% \cs{l_regex_success_submatches_prop} the list of end-points of +% \cs{l_@@_success_submatches_prop} the list of end-points of % submatches, and store them in \tn{skip} registers, from -% \cs{l_regex_zeroth_submatch_int} upwards. We begin by emptying those +% \cs{l_@@_zeroth_submatch_int} upwards. We begin by emptying those % \tn{skip} registers. Then for each \meta{key}--\meta{value} pair in % the property list update the appropriate \tn{skip} component. This % is somewhat a hack: the \meta{key} is a non-negative integer @@ -5487,35 +5504,35 @@ % end, store the information about the position at which the match % attempt started, as a shrink component. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_extract: +\cs_new_protected_nopar:Npn \@@_extract: { - \if_meaning:w \c_true_bool \g_regex_success_bool - \int_set_eq:NN \l_regex_zeroth_submatch_int \l_regex_submatch_int - \prg_replicate:nn \l_regex_capturing_group_int + \if_meaning:w \c_true_bool \g_@@_success_bool + \int_set_eq:NN \l_@@_zeroth_submatch_int \l_@@_submatch_int + \prg_replicate:nn \l_@@_capturing_group_int { - \tex_skip:D \l_regex_submatch_int \c_zero sp \scan_stop: - \int_incr:N \l_regex_submatch_int + \tex_skip:D \l_@@_submatch_int \c_zero sp \scan_stop: + \int_incr:N \l_@@_submatch_int } - \prop_map_inline:Nn \l_regex_success_submatches_prop + \prop_map_inline:Nn \l_@@_success_submatches_prop { - \if_num:w ##1 \c_minus_one - \exp_after:wN \regex_extract_aux_e:wn \int_value:w + \if_int_compare:w ##1 \c_minus_one + \exp_after:wN \@@_extract_e:wn \__int_value:w \else: - \exp_after:wN \regex_extract_aux_b:wn \int_value:w + \exp_after:wN \@@_extract_b:wn \__int_value:w \fi: - \int_eval:w \l_regex_zeroth_submatch_int + ##1 {##2} + \__int_eval:w \l_@@_zeroth_submatch_int + ##1 {##2} } - \tex_skip:D \l_regex_zeroth_submatch_int - = \tex_the:D \tex_skip:D \l_regex_zeroth_submatch_int - minus \l_regex_start_pos_int sp \scan_stop: + \tex_skip:D \l_@@_zeroth_submatch_int + = \tex_the:D \tex_skip:D \l_@@_zeroth_submatch_int + minus \l_@@_start_pos_int sp \scan_stop: \fi: } -\cs_new_protected:Npn \regex_extract_aux_b:wn #1 < #2 +\cs_new_protected:Npn \@@_extract_b:wn #1 < #2 { \tex_skip:D #1 = #2 sp plus \etex_gluestretch:D \tex_skip:D #1 \scan_stop: } -\cs_new_protected:Npn \regex_extract_aux_e:wn #1 > #2 +\cs_new_protected:Npn \@@_extract_e:wn #1 > #2 { \tex_skip:D #1 = 1 \tex_skip:D #1 plus #2 sp \scan_stop: @@ -5525,7 +5542,7 @@ % % \subsubsection{Replacement} % -% \begin{macro}[aux]{\regex_replace_once_aux:nnN} +% \begin{macro}[aux]{\@@_replace_once:nnN} % Build the \textsc{nfa} and the replacement functions, then find a % single match. If the match failed, simply exit the % group. Otherwise, we do the replacement. Extract submatches. Compute @@ -5539,106 +5556,106 @@ % \texttt{x}-expansion, and checks that braces are balanced in the % final result. % \begin{macrocode} -\cs_new_protected:Npn \regex_replace_once_aux:nnN #1#2#3 +\cs_new_protected:Npn \@@_replace_once:nnN #1#2#3 { \group_begin: - \regex_single_match: + \@@_single_match: #1 - \regex_replacement:n {#2} - \exp_args:No \regex_match:n { #3 } - \if_meaning:w \c_false_bool \g_regex_success_bool + \@@_replacement:n {#2} + \exp_args:No \@@_match:n { #3 } + \if_meaning:w \c_false_bool \g_@@_success_bool \group_end: \else: - \regex_extract: - \int_set:Nn \l_regex_balance_int + \@@_extract: + \int_set:Nn \l_@@_balance_int { - \regex_replacement_balance_one_match:n - { \l_regex_zeroth_submatch_int } + \@@_replacement_balance_one_match:n + { \l_@@_zeroth_submatch_int } } - \tl_set:Nx \l_regex_internal_a_tl + \tl_set:Nx \l_@@_internal_a_tl { - \regex_replacement_do_one_match:n { \l_regex_zeroth_submatch_int } - \regex_query_range:nn - { \etex_gluestretch:D \tex_skip:D \l_regex_zeroth_submatch_int } - { \l_regex_max_pos_int } + \@@_replacement_do_one_match:n { \l_@@_zeroth_submatch_int } + \@@_query_range:nn + { \etex_gluestretch:D \tex_skip:D \l_@@_zeroth_submatch_int } + { \l_@@_max_pos_int } } - \regex_group_end_replace:N #3 + \@@_group_end_replace:N #3 \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_replace_all_aux:nnN} +% \begin{macro}[aux]{\@@_replace_all:nnN} % Match multiple times, and for every match, extract submatches and % additionally store the position at which the match attempt started % (as the shrink component of a \tn{skip} register). The \tn{skip} -% registers from \cs{l_regex_max_state_int} to -% \cs{l_regex_submatch_int} hold information about submatches of every +% registers from \cs{l_@@_max_state_int} to +% \cs{l_@@_submatch_int} hold information about submatches of every % match in order; each match corresponds to -% \cs{l_regex_capturing_group_int} consecutive \tn{skip} registers. +% \cs{l_@@_capturing_group_int} consecutive \tn{skip} registers. % Compute the brace balance corresponding to doing all the % replacements: this is the sum of brace balances for replacing each % match. Join together the replacement texts for each match (including % the part of the query before the match), and the end of the query. % \begin{macrocode} -\cs_new_protected:Npn \regex_replace_all_aux:nnN #1#2#3 +\cs_new_protected:Npn \@@_replace_all:nnN #1#2#3 { \group_begin: - \regex_multi_match:n { \regex_extract: } + \@@_multi_match:n { \@@_extract: } #1 - \regex_replacement:n {#2} - \exp_args:No \regex_match:n {#3} - \int_set:Nn \l_regex_balance_int + \@@_replacement:n {#2} + \exp_args:No \@@_match:n {#3} + \int_set:Nn \l_@@_balance_int { 0 - \prg_stepwise_function:nnnN - { \c_two * \l_regex_max_state_int } - \l_regex_capturing_group_int - { \l_regex_submatch_int - \c_one } - \regex_replacement_balance_one_match:n + \int_step_function:nnnN + { \c_two * \l_@@_max_state_int } + \l_@@_capturing_group_int + { \l_@@_submatch_int - \c_one } + \@@_replacement_balance_one_match:n } - \tl_set:Nx \l_regex_internal_a_tl + \tl_set:Nx \l_@@_internal_a_tl { - \prg_stepwise_function:nnnN - { \c_two * \l_regex_max_state_int } - \l_regex_capturing_group_int - { \l_regex_submatch_int - \c_one } - \regex_replacement_do_one_match:n - \regex_query_range:nn - \l_regex_start_pos_int \l_regex_max_pos_int + \int_step_function:nnnN + { \c_two * \l_@@_max_state_int } + \l_@@_capturing_group_int + { \l_@@_submatch_int - \c_one } + \@@_replacement_do_one_match:n + \@@_query_range:nn + \l_@@_start_pos_int \l_@@_max_pos_int } - \regex_group_end_replace:N #3 + \@@_group_end_replace:N #3 } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux]{\regex_group_end_replace:N} +% \begin{macro}[aux]{\@@_group_end_replace:N} % If the brace balance is not $0$, raise an error. Then set the user's % variable |#1| to the \texttt{x}-expansion of -% \cs{l_regex_internal_a_tl}, adding the appropriate braces to produce +% \cs{l_@@_internal_a_tl}, adding the appropriate braces to produce % a balanced result. And end the group. % \begin{macrocode} -\cs_new_protected_nopar:Npn \regex_group_end_replace:N #1 +\cs_new_protected_nopar:Npn \@@_group_end_replace:N #1 { - \if_num:w \l_regex_balance_int = \c_zero + \if_int_compare:w \l_@@_balance_int = \c_zero \else: - \msg_kernel_error:nnxxx { regex } { result-unbalanced } + \__msg_kernel_error:nnxxx { regex } { result-unbalanced } { replacing } - { \int_max:nn { - \l_regex_balance_int } { \c_zero } } - { \int_max:nn { \l_regex_balance_int } { \c_zero } } + { \int_max:nn { - \l_@@_balance_int } { \c_zero } } + { \int_max:nn { \l_@@_balance_int } { \c_zero } } \fi: \use:x { \group_end: \tl_set:Nn \exp_not:N #1 { - \if_int_compare:w \l_regex_balance_int < \c_zero - \prg_replicate:nn { - \l_regex_balance_int } + \if_int_compare:w \l_@@_balance_int < \c_zero + \prg_replicate:nn { - \l_@@_balance_int } { { \if_false: } \fi: } \fi: - \l_regex_internal_a_tl - \if_int_compare:w \l_regex_balance_int > \c_zero - \prg_replicate:nn { \l_regex_balance_int } + \l_@@_internal_a_tl + \if_int_compare:w \l_@@_balance_int > \c_zero + \prg_replicate:nn { \l_@@_balance_int } { \if_false: { \fi: } } \fi: } @@ -5653,20 +5670,20 @@ % % Messages for the preparsing phase. % \begin{macrocode} -\msg_kernel_new:nnnn { regex } { trailing-backslash } +\__msg_kernel_new:nnnn { regex } { trailing-backslash } { Trailing~escape~character~\iow_char:N\\. } { A~regular~expression~or~its~replacement~text~ends~with~ the~escape~character~\iow_char:N\\.~It~will~be~ignored. } -\msg_kernel_new:nnnn { regex } { x-missing-rbrace } +\__msg_kernel_new:nnnn { regex } { x-missing-rbrace } { Missing~closing~brace~in~\iow_char:N\\x~hexadecimal~sequence. } { You~wrote~something~like~ `\iow_char:N\\x\{\int_to_hexadecimal:n{#1}'.~ The~closing~brace~is~missing. } -\msg_kernel_new:nnnn { regex } { x-overflow } +\__msg_kernel_new:nnnn { regex } { x-overflow } { Character~code~'#1'~too~large~in~\iow_char:N\\x~hexadecimal~sequence. } { You~wrote~something~like~ @@ -5677,7 +5694,7 @@ % % Invalid quantifier. % \begin{macrocode} -\msg_kernel_new:nnnn { regex } { invalid-quantifier } +\__msg_kernel_new:nnnn { regex } { invalid-quantifier } { Braced~quantifier~'#1'~may~not~be~followed~by~'#2'. } { The~character~'#2'~is~invalid~in~the~braced~quantifier~'#1'.~ @@ -5689,13 +5706,13 @@ % Messages for missing or extra closing brackets and parentheses, with % some fancy singular/plural handling for the case of parentheses. % \begin{macrocode} -\msg_kernel_new:nnnn { regex } { missing-rbrack } +\__msg_kernel_new:nnnn { regex } { missing-rbrack } { Missing~right~bracket~inserted~in~regular~expression. } { LaTeX~was~given~a~regular~expression~where~a~character~class~ was~started~with~'[',~but~the~matching~']'~is~missing. } -\msg_kernel_new:nnnn { regex } { missing-rparen } +\__msg_kernel_new:nnnn { regex } { missing-rparen } { Missing~right~parenthes\int_compare:nTF{#1=1}{i}{e}s~ inserted~in~regular~expression. @@ -5705,7 +5722,7 @@ more~left~parenthes\int_compare:nTF{#1=1}{i}{e}s~than~right~ parenthes\int_compare:nTF{#1=1}{i}{e}s. } -\msg_kernel_new:nnnn { regex } { extra-rparen } +\__msg_kernel_new:nnnn { regex } { extra-rparen } { Extra~right~parenthesis~ignored~in~regular~expression. } { LaTeX~came~across~a~closing~parenthesis~when~no~submatch~group~ @@ -5715,25 +5732,25 @@ % % Some escaped alphanumerics are not allowed everywhere. % \begin{macrocode} -\msg_kernel_new:nnnn { regex } { bad-escape } +\__msg_kernel_new:nnnn { regex } { bad-escape } { Invalid~escape~\c_backslash_str #1~ - \regex_if_in_cs:TF { within~a~control~sequence. } + \@@_if_in_cs:TF { within~a~control~sequence. } { - \regex_if_in_class:TF + \@@_if_in_class:TF { in~a~character~class. } { following~a~category~test. } } } { The~escape~sequence~\iow_char:N\\#1~may~not~appear~ - \regex_if_in_cs:TF + \@@_if_in_cs:TF { within~a~control~sequence~test~introduced~by~ \iow_char:N\\c\iow_char:N\{. } { - \regex_if_in_class:TF + \@@_if_in_class:TF { within~a~character~class~ } { following~a~category~test~such~as~\iow_char:N\\cL~ } because~it~does~not~match~exactly~one~character. @@ -5743,14 +5760,14 @@ % % Range errors. % \begin{macrocode} -\msg_kernel_new:nnnn { regex } { range-missing-end } +\__msg_kernel_new:nnnn { regex } { range-missing-end } { Invalid~end-point~for~range~'#1-#2'~in~character~class. } { The~end-point~'#2'~of~the~range~'#1-#2'~may~not~serve~as~an~ end-point~for~a~range:~alphanumeric~characters~should~not~be~ escaped,~and~non-alphanumeric~characters~should~be~escaped. } -\msg_kernel_new:nnnn { regex } { range-backwards } +\__msg_kernel_new:nnnn { regex } { range-backwards } { Range~[#1-#2]~out~of~order~in~character~class. } { In~ranges~of~characters~[x-y]~appearing~in~character~classes,~ @@ -5762,27 +5779,27 @@ % % Errors related to |\c| and |\u|. % \begin{macrocode} -\msg_kernel_new:nnnn { regex } { c-bad-mode } +\__msg_kernel_new:nnnn { regex } { c-bad-mode } { Invalid~nested~\iow_char:N\\c~escape~in~regular~expression. } { The~\iow_char:N\\c~escape~cannot~be~used~within~ a~control~sequence~test~'\iow_char:N\\c{...}'.~ To~combine~several~category~tests,~use~'\iow_char:N\\c[...]'. } -\msg_kernel_new:nnnn { regex } { c-missing-rbrace } +\__msg_kernel_new:nnnn { regex } { c-missing-rbrace } { Missing~right~brace~inserted~for~\iow_char:N\\c~escape. } { LaTeX~was~given~a~regular~expression~where~a~ '\iow_char:N\\c\iow_char:N\{...'~construction~was~not~ended~ with~a~closing~brace~'\iow_char:N\}'. } -\msg_kernel_new:nnnn { regex } { c-missing-rbrack } +\__msg_kernel_new:nnnn { regex } { c-missing-rbrack } { Missing~right~bracket~inserted~for~\iow_char:N\\c~escape. } { A~construction~'\iow_char:N\\c[...'~appears~in~a~ regular~expression,~but~the~closing~']'~is~not~present. } -\msg_kernel_new:nnnn { regex } { c-missing-category } +\__msg_kernel_new:nnnn { regex } { c-missing-category } { Invalid~character~'#1'~following~\iow_char:N\\c~escape. } { In~regular~expressions,~the~\iow_char:N\\c~escape~sequence~ @@ -5790,17 +5807,17 @@ capital~letter~representing~a~character~category,~namely~ one~of~ABCDELMOPSTU. } -\msg_kernel_new:nnnn { regex } { u-missing-lbrace } +\__msg_kernel_new:nnnn { regex } { u-missing-lbrace } { Missing~left~brace~following~\iow_char:N\\u~escape. } { The~\iow_char:N\\u~escape~sequence~must~be~followed~by~ a~brace~group~with~the~name~of~the~variable~to~use. } -\msg_kernel_new:nnnn { regex } { u-missing-rbrace } +\__msg_kernel_new:nnnn { regex } { u-missing-rbrace } { Missing~right~brace~inserted~for~\iow_char:N\\u~escape. } { LaTeX~ - \tl_if_empty:xTF {#2} + \str_if_eq_x:nnTF { } {#2} { reached~the~end~of~the~string~ } { encountered~an~escaped~alphanumeric~character '\iow_char:N\\#2'~ } when~parsing~the~argument~of~an~'\iow_char:N\\u\iow_char:N\{...\}'~escape. @@ -5809,14 +5826,14 @@ % % Errors when encountering the \textsc{posix} syntax |[:...:]|. % \begin{macrocode} -\msg_kernel_new:nnnn { regex } { posix-unsupported } +\__msg_kernel_new:nnnn { regex } { posix-unsupported } { POSIX~collating~element~'[#1 ~ #1]'~not~supported. } { The~[.foo.]~and~[=bar=]~syntaxes~have~a~special~meaning~in~POSIX~ regular~expressions.~This~is~not~supported~by~LaTeX.~Maybe~you~ forgot~to~escape~a~left~bracket~in~a~character~class? } -\msg_kernel_new:nnnn { regex } { posix-unknown } +\__msg_kernel_new:nnnn { regex } { posix-unknown } { POSIX~class~[:#1:]~unknown. } { [:#1:]~is~not~among~the~known~POSIX~classes~ @@ -5825,7 +5842,7 @@ [:print:],~[:punct:],~[:space:],~[:upper:],~ [:word:],~and~[:xdigit:]. } -\msg_kernel_new:nnnn { regex } { posix-missing-close } +\__msg_kernel_new:nnnn { regex } { posix-missing-close } { Missing~closing~':]'~for~POSIX~class. } { The~POSIX~syntax~'#1'~must~be~followed~by~':]',~not~'#2'. } % \end{macrocode} @@ -5834,7 +5851,7 @@ % with an unbalanced token list, which we must re-balance by adding % begin-group or end-group character tokens. % \begin{macrocode} -\msg_kernel_new:nnnn { regex } { result-unbalanced } +\__msg_kernel_new:nnnn { regex } { result-unbalanced } { Missing~brace~inserted~when~#1. } { LaTeX~was~asked~to~do~some~regular~expression~operation,~ @@ -5846,7 +5863,7 @@ % % Error message for unknown options. % \begin{macrocode} -\msg_kernel_new:nnnn { regex } { unknown-option } +\__msg_kernel_new:nnnn { regex } { unknown-option } { Unknown~option~`#1'~for~regular~expressions. } { The~only~available~option~is~'case-insensitive',~toggled~by~ @@ -5856,28 +5873,28 @@ % % Errors in the replacement text. % \begin{macrocode} -\msg_kernel_new:nnnn { regex } { replacement-c } +\__msg_kernel_new:nnnn { regex } { replacement-c } { Misused~\iow_char:N\\c~command~in~a~replacement~text. } { In~a~replacement~text,~the~\iow_char:N\\c~escape~sequence~ can~be~followed~by~one~of~the~letters~ABCDELMOPSTU~ or~a~brace~group,~not~by~'#1'. } -\msg_kernel_new:nnnn { regex } { replacement-u } +\__msg_kernel_new:nnnn { regex } { replacement-u } { Misused~\iow_char:N\\u~command~in~a~replacement~text. } { In~a~replacement~text,~the~\iow_char:N\\u~escape~sequence~ must~be~~followed~by~a~brace~group~holding~the~name~of~the~ variable~to~use. } -\msg_kernel_new:nnnn { regex } { replacement-g } +\__msg_kernel_new:nnnn { regex } { replacement-g } { Missing~brace~for~the~\iow_char:N\\g~construction~in~a~replacement~text. } { In~the~replacement~text~for~a~regular~expression~search,~ submatches~are~represented~either~as~\iow_char:N \\g{dd..d},~ or~\\d,~where~`d'~are~single~digits.~Here,~a~brace~is~missing. } -\msg_kernel_new:nnnn { regex } { replacement-catcode-end } +\__msg_kernel_new:nnnn { regex } { replacement-catcode-end } { Missing~character~for~the~\iow_char:N\\c~ construction~in~a~replacement~text. @@ -5888,7 +5905,7 @@ the~character~category.~Then,~a~character~must~follow.~LaTeX~ reached~the~end~of~the~replacement~when~looking~for~that. } -\msg_kernel_new:nnnn { regex } { replacement-null-space } +\__msg_kernel_new:nnnn { regex } { replacement-null-space } { TeX~cannot~build~a~space~token~with~character~code~0. } { You~asked~for~a~character~token~with~category~'space',~ @@ -5897,26 +5914,26 @@ This~specific~case~is~impossible~and~will~be~replaced~ by~a~normal~space. } -\msg_kernel_new:nnnn { regex } { replacement-missing-rbrace } +\__msg_kernel_new:nnnn { regex } { replacement-missing-rbrace } { Missing~right~brace~inserted~in~replacement~text. } { - There~were~\int_use:N \l_regex_replacement_csnames_int \ + There~were~\int_use:N \l_@@_replacement_csnames_int \ missing~right~braces. } % \end{macrocode} % -% \begin{macro}[aux]{\regex_msg_repeated:nnN} +% \begin{macro}[aux]{\@@_msg_repeated:nnN} % This is not technically a message, but seems related enough to go % there. The arguments are: |#1| is the minimum number of repetitions; % |#2| is the number of allowed extra repetitions ($-1$ for infinite % number), and |#3| tells us aboud lazyness. % \begin{macrocode} -\cs_new:Npn \regex_msg_repeated:nnN #1#2#3 +\cs_new:Npn \@@_msg_repeated:nnN #1#2#3 { - \str_if_eq:xxF { #1 #2 } { 1 0 } + \str_if_eq_x:nnF { #1 #2 } { 1 0 } { , ~ repeated ~ - \prg_case_int:nnn {#2} + \int_case:nnn {#2} { { -1 } { #1~or~more~times,~\bool_if:NTF #3 { lazy } { greedy } } { 0 } { #1~times } @@ -5935,18 +5952,18 @@ % The tracing code is still very experimental, and is meant to be used % with the \pkg{l3trace} package, currently in \texttt{l3trial}. % -% \begin{macro}[int]{\regex_trace_states:n} +% \begin{macro}[int]{\@@_trace_states:n} % This function lists the contents of all states of the \textsc{nfa}, -% stored in \tn{toks} from $0$ to \cs{l_regex_max_state_int} +% stored in \tn{toks} from $0$ to \cs{l_@@_max_state_int} % (excluded). % \begin{macrocode} %<*trace> -\cs_new_protected:Npn \regex_trace_states:n #1 +\cs_new_protected:Npn \@@_trace_states:n #1 { - \prg_stepwise_inline:nnnn - \l_regex_min_state_int + \int_step_inline:nnnn + \l_@@_min_state_int \c_one - { \l_regex_max_state_int - 1 } + { \l_@@_max_state_int - 1 } { \trace:nnx { regex } { #1 } { \iow_char:N \\toks ##1 = { \tex_the:D \tex_toks:D ##1 } } @@ -5957,11 +5974,12 @@ % \end{macro} % % \begin{macrocode} -% +% % \end{macrocode} % % \end{implementation} % +% \PrintIndex % \endinput %^^A NOT IMPLEMENTED %^^A \p{xx} a character with the xx property diff --git a/Master/texmf-dist/source/latex/l3experimental/l3str/l3str.dtx b/Master/texmf-dist/source/latex/l3experimental/l3str/l3str.dtx index ef2bc9b5460..51c42856c3c 100644 --- a/Master/texmf-dist/source/latex/l3experimental/l3str/l3str.dtx +++ b/Master/texmf-dist/source/latex/l3experimental/l3str/l3str.dtx @@ -35,8 +35,8 @@ % %<*driver|package> \RequirePackage{expl3} -\GetIdInfo$Id: l3str.dtx 3460 2012-02-26 12:16:41Z will $ - {L3 Experimental Strings} +\GetIdInfo$Id: l3str.dtx 3940 2012-07-09 00:41:44Z bruno $ + {L3 Experimental strings} % %<*driver> \documentclass[full]{l3doc} @@ -96,6 +96,11 @@ % % \section{Building strings} % +% \begin{variable}{\c_max_char_int} +% The maximum valid character code, $255$ for pdf\TeX{}, and $1114111$ +% for \XeTeX{} and \LuaTeX{}. +% \end{variable} +% % \begin{variable} % { % \c_backslash_str, @@ -192,15 +197,16 @@ % \section{Accessing the contents of a string} % % \begin{function}[EXP] -% {\str_length:N, \str_length:n, \str_length_ignore_spaces:n} +% {\str_count:N, \str_count:n, \str_count_ignore_spaces:n} % \begin{syntax} -% \cs{str_length:n} \Arg{token list} +% \cs{str_count:n} \Arg{token list} % \end{syntax} -% Leaves the length of the string representation of \meta{token list} +% Leaves the number of tokens in the string representation of +% \meta{token list} % in the input stream as an integer denotation. The functions differ -% in their treatment of spaces. In the case of \cs{str_length:N} and -% \cs{str_length:n}, all characters including spaces are counted. The -% \cs{str_length_ignore_spaces:n} leaves the number of non-space +% in their treatment of spaces. In the case of \cs{str_count:N} and +% \cs{str_count:n}, all characters including spaces are counted. The +% \cs{str_count_ignore_spaces:n} leaves the number of non-space % characters in the input stream. % \end{function} % @@ -256,7 +262,7 @@ % account. The \cs{str_item_ignore_spaces:nn} function skips spaces % in its argument. If the \meta{integer expression} is negative, % characters are counted from the end of the \meta{string}. Hence, -% $-1$ is the right-most character, \emph{etc.}, while $0$ is the +% $-1$ is the right-most character, \emph{etc.}, while $1$ is the % first (left-most) character. % \end{function} % @@ -267,28 +273,27 @@ % \end{syntax} % Converts the \meta{token list} to a \meta{string}, and leaves in the % input stream the characters from the \meta{start index} inclusive to -% the \meta{end index} exclusive. Note that the length of the +% the \meta{end index} exclusive. Note that the token count of the % substring is equal to the difference between the two \meta{indices}. % If either of \meta{start index} or \meta{end index} is negative, -% then it is incremented by the length of the list. If either of +% then it is incremented by the token count of the list. If either of % \meta{start index} or \meta{end index} is empty, it is replaced by % the corresponding end-point of the string. Both \meta{start index} -% and \meta{end index} count from $0$ for the first (left most) +% and \meta{end index} count from $1$ for the first (left most) % character. For instance, % \begin{verbatim} -% \iow_term:x { \str_substr:nnn { abcdef } { 1 } { 4 } } +% \iow_term:x { \str_substr:nnn { abcdef } { 2 } { 5 } } % \iow_term:x { \str_substr:nnn { abcdef } { -4 } { } } % \end{verbatim} -% will print \texttt{bcd} and \texttt{cdef}. +% will print \texttt{bcd} and \texttt{cdef} to the terminal. % \end{function} % % \section{String conditionals} % -% \begin{function}[EXP, pTF] +% \begin{function}[EXP,pTF] % { -% \str_if_eq:NN, % \str_if_eq:nn, \str_if_eq:Vn, \str_if_eq:on, \str_if_eq:no, -% \str_if_eq:nV, \str_if_eq:VV, \str_if_eq:xx +% \str_if_eq:nV, \str_if_eq:VV % } % \begin{syntax} % \cs{str_if_eq_p:nn} \Arg{tl_1} \Arg{tl_2} @@ -298,11 +303,63 @@ % basis, and is \texttt{true} if the two lists contain the same % characters in the same order. Thus for example % \begin{verbatim} -% \str_if_eq_p:xx { abc } { \tl_to_str:n { abc } } +% \str_if_eq_p:no { abc } { \tl_to_str:n { abc } } +% \end{verbatim} +% is logically \texttt{true}. +% \end{function} +% +% \begin{function}[EXP,pTF, added = 2012-06-05]{\str_if_eq_x:nn} +% \begin{syntax} +% \cs{str_if_eq_x_p:nn} \Arg{tl_1} \Arg{tl_2} +% \cs{str_if_eq_x:nnTF} \Arg{tl_1} \Arg{tl_2} \Arg{true code} \Arg{false code} +% \end{syntax} +% Compares the full expansion of two \meta{token lists} on a character by +% character basis, and is \texttt{true} if the two lists contain the same +% characters in the same order. Thus for example +% \begin{verbatim} +% \str_if_eq_x_p:nn { abc } { \tl_to_str:n { abc } } % \end{verbatim} -% is logically \texttt{true}. All versions of these functions are -% fully expandable (including those involving an \texttt{x}-type -% expansion). +% is logically \texttt{true}. +% \end{function} +% +% \begin{function}[added = 2012-06-03, EXP]{\str_case:nnn, \str_case:onn} +% \begin{syntax} +% \cs{str_case:nnn} \Arg{test string} \\ +% ~~|{| \\ +% ~~~~\Arg{string case1} \Arg{code case1} \\ +% ~~~~\Arg{string case2} \Arg{code case2} \\ +% ~~~~\ldots \\ +% ~~~~\Arg{string case$\sb{n}$} \Arg{code case$\sb{n}$} \\ +% ~~|}| \\ +% ~~\Arg{else case} +% \end{syntax} +% This function compares the \meta{test string} in turn with each +% of the \meta{string cases}. If the two are equal (as described for +% \cs{str_if_eq:nnTF} then the +% associated \meta{code} is left in the input stream. If none of +% the tests are \texttt{true} then the \texttt{else code} will be +% left in the input stream. +% \end{function} +% +% \begin{function}[added = 2012-06-05, EXP]{\str_case_x:nnn} +% \begin{syntax} +% \cs{str_case_x:nnn} \Arg{test string} \\ +% ~~|{| \\ +% ~~~~\Arg{string case1} \Arg{code case1} \\ +% ~~~~\Arg{string case2} \Arg{code case2} \\ +% ~~~~\ldots \\ +% ~~~~\Arg{string case$\sb{n}$} \Arg{code case$\sb{n}$} \\ +% ~~|}| \\ +% ~~\Arg{else case} +% \end{syntax} +% This function compares the full expansion of the \meta{test string} +% in turn with the full expansion of the \meta{string cases}. If the two +% full expansions are equal (as described for \cs{str_if_eq:nnTF} then the +% associated \meta{code} is left in the input stream. If none of the tests +% are \texttt{true} then the \texttt{else code} will be left in the input +% stream. The \meta{test string} is expanded in each comparison, and must +% always yield the same result: for example, random numbers must +% not be used within this string. % \end{function} % % \section{Viewing strings} @@ -314,6 +371,22 @@ % Displays the content of the \meta{str~var} on the terminal. % \end{function} % +% \section{Scratch strings} +% +% \begin{variable}{\l_tmpa_str, \l_tmpb_str} +% Scratch strings for local assignment. These are never used by +% the kernel code, and so are safe for use with any \LaTeX3-defined +% function. However, they may be overwritten by other non-kernel +% code and so should only be used for short-term storage. +% \end{variable} +% +% \begin{variable}{\g_tmpa_str, \g_tmpb_str} +% Scratch strings for global assignment. These are never used by +% the kernel code, and so are safe for use with any \LaTeX3-defined +% function. However, they may be overwritten by other non-kernel +% code and so should only be used for short-term storage. +% \end{variable} +% % \section{Encoding functions} % % Traditionally, string encodings only specify how strings of characters @@ -448,87 +521,39 @@ % % \section{Internal string functions} % -% \begin{variable}{\c_max_char_int} -% The maximum valid character code, $255$ for pdf\TeX{}, and $1114111$ -% for the two other engines. -% \end{variable} -% -% \begin{function}{\str_aux_hexadecimal_use:NTF} +% \begin{function}{\__str_gset_other:Nn} % \begin{syntax} -% \cs{str_aux_hexadecimal_use:NTF} \meta{token} \Arg{true code} \Arg{false code} -% \end{syntax} -% If the \meta{token} is a hexdecimal digit (upper case or lower -% case), its upper-case version is left in the input stream, -% \emph{followed} by the \meta{true code}. Otherwise, the \meta{false -% code} is left in the input stream. -% \begin{texnote} -% This function will fail if the escape character is a hexadecimal -% digit, or has a character code in the range $[0,5]$. We are thus -% careful to set the escape character to a known value before using -% it. -% \end{texnote} -% \end{function} -% -% \begin{function}{\str_aux_octal_use:NTF} -% \begin{syntax} -% \cs{str_aux_octal_use:NTF} \meta{token} \Arg{true code} \Arg{false code} -% \end{syntax} -% If the \meta{token} is an octal digit, it is left in the input -% stream, \emph{followed} by the \meta{true code}. Otherwise, the -% \meta{false code} is left in the input stream. -% \begin{texnote} -% This function will fail if the escape character is an octal -% digit. We are thus careful to set the escape character to a known -% value before using it. -% \end{texnote} -% \end{function} -% -% \begin{function}[EXP]{\tl_to_other_str:n} -% \begin{syntax} -% \cs{tl_to_other_str:n} \Arg{token list} -% \end{syntax} -% Converts the \meta{token list} to a \meta{other string}, where -% spaces have category code \enquote{other}. -% \begin{texnote} -% This function can be \texttt{f}-expanded without fear of losing a -% leading space, since spaces do not have category code $10$ in its -% result. This function takes a time quadratic in the length of the -% string; \cs{str_gset_other:Nn} is faster but not expandable. -% \end{texnote} -% \end{function} -% -% \begin{function}{\str_gset_other:Nn} -% \begin{syntax} -% \cs{str_gset_other:Nn} \meta{tl~var} \Arg{token list} +% \cs{__str_gset_other:Nn} \meta{tl~var} \Arg{token list} % \end{syntax} % Converts the \meta{token list} to an \meta{other string}, where % spaces have category code \enquote{other}, and assigns the result to % the \meta{tl~var}, globally. % \end{function} % -% \begin{function}[EXP] -% { -% \str_if_contains_char:NNT, -% \str_if_contains_char:NNTF, -% \str_if_contains_char:nNTF -% } +% \begin{function}{\__str_hexadecimal_use:NTF} % \begin{syntax} -% \cs{str_if_contains_char:nNTF} \Arg{token list} \meta{char} +% \cs{__str_hexadecimal_use:NTF} \meta{token} \Arg{true code} \Arg{false code} % \end{syntax} -% Converts the \meta{token list} to a \meta{string} and tests whether -% the \meta{char} is present in the \meta{string}. Spaces are -% ignored. +% If the \meta{token} is a hexadecimal digit (upper case or lower +% case), its upper-case version is left in the input stream, +% \emph{followed} by the \meta{true code}. Otherwise, the \meta{false +% code} is left in the input stream. +% \begin{texnote} +% This function fails on some inputs if the escape character is a +% hexadecimal digit. We are thus careful to set the escape +% character to a known (safe) value before using it. +% \end{texnote} % \end{function} % -% \begin{function}{\str_declare_eight_bit_encoding:nnn} +% \begin{function}[EXP]{\__str_output_byte:n} % \begin{syntax} -% \cs{str_declare_eight_bit_encoding:nnn} \Arg{name} \Arg{mapping} \Arg{missing} +% \cs{__str_output_byte:n} \Arg{intexpr} % \end{syntax} -% This declares the encoding \meta{name} to map bytes to Unicode -% characters according to the \meta{mapping}, and map those bytes -% which are not mentionned in the \meta{mapping} either to the -% replacement character (if they appear in \meta{missing}), or to -% themselves. +% Expands to a character token with category other and character code +% equal to the value of \meta{intexpr}. The value of \meta{intexpr} +% must be in the range $[-1, 255]$, and any value outside this range +% results in undefined behaviour. The special value $-1$ is used to +% produce an empty result. % \end{function} % % \section{Possibilities, and things to do} @@ -541,7 +566,7 @@ % \item Add documentation about each encoding and escaping method, and % add examples. % \item The \texttt{hex} unescaping should raise an error for -% odd-length strings. +% odd-token count strings. % \item Decide what bytes should be escaped in the \texttt{url} % escaping. Perhaps |!'()*-./0123456789_| are safe, and all other % characters should be escaped? @@ -555,6 +580,7 @@ % % Other string tasks. % \begin{itemize} +% \item Expandable \cs{str_if_in:nnTF}? % \item \cs{str_if_head_eq:nN} % \item \cs{str_if_numeric/decimal/integer:n}, perhaps in \pkg{l3fp}? % \item Should \cs{str_item:Nn} be \cs{str_char:Nn}? @@ -565,6 +591,7 @@ % either one integer expression, two integer expressions separated % by |:|, or three integer expressions separated by |:|, \emph{cf.} % Python's extended slice syntax? +% \item Analog of \texttt{printf}? % \end{itemize} % % \end{documentation} @@ -578,6 +605,10 @@ % \end{macrocode} % % \begin{macrocode} +%<@@=str> +% \end{macrocode} +% +% \begin{macrocode} \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} \RequirePackage{l3tl-analysis,l3tl-build,l3flag} @@ -587,13 +618,13 @@ % \pkg{l3kernel}. % \begin{itemize} % \item \cs{str_if_eq:nn}[pTF] and variants, -% \item \cs{str_if_eq_return:on}, \cs{str_if_eq_return:xx} +% \item \cs{str_if_eq_x_return:on}, \cs{str_if_eq_x_return:nn} % \item \cs{tl_to_str:n}, \cs{tl_to_str:N}, \cs{tl_to_str:c}, % \item \cs{token_to_str:N}, \cs{cs_to_str:N} -% \item \cs{str_head:n}, \cs{str_head_aux:w}, (copied here) -% \item \cs{str_tail:n}, \cs{str_tail_aux:w}, (copied here) -% \item \cs{str_length_skip_spaces} (deprecated) -% \item \cs{str_length_loop:NNNNNNNNN} (unchanged) +% \item \cs{str_head:n}, \cs{__str_head:w}, (copied here) +% \item \cs{str_tail:n}, \cs{__str_tail:w}, (copied here) +% \item \cs{str_count_ignore_spaces} (unchanged) +% \item \cs{str_count_loop:NNNNNNNNN} (unchanged) % \end{itemize} % % \subsection{Helpers} @@ -657,27 +688,27 @@ % % \subsubsection{Variables and constants} % -% \begin{macro}{\str_tmp:w} -% \begin{variable}{\l_str_internal_int} -% \begin{variable}{\l_str_internal_tl} +% \begin{macro}{\@@_tmp:w} +% \begin{variable}{\l_@@_internal_int} +% \begin{variable}{\l_@@_internal_tl} % Internal scratch space for some functions. % \begin{macrocode} -\cs_new_protected_nopar:Npn \str_tmp:w { } -\tl_new:N \l_str_internal_tl -\int_new:N \l_str_internal_int +\cs_new_protected_nopar:Npn \@@_tmp:w { } +\tl_new:N \l_@@_internal_tl +\int_new:N \l_@@_internal_int % \end{macrocode} % \end{variable} % \end{variable} % \end{macro} % -% \begin{variable}{\g_str_result_tl} -% The \cs{g_str_result_tl} variable is used to hold the result of +% \begin{variable}{\g_@@_result_tl} +% The \cs{g_@@_result_tl} variable is used to hold the result of % various internal string operations (mostly conversions) which are % typically performed in a group. The variable is global so that it % remains defined outside the group, to be assigned to a user-provided % variable. % \begin{macrocode} -\tl_new:N \g_str_result_tl +\tl_new:N \g_@@_result_tl % \end{macrocode} % \end{variable} % @@ -709,11 +740,11 @@ % \end{macrocode} % \end{variable} % -% \begin{variable}{\c_str_replacement_char_int} +% \begin{variable}{\c_@@_replacement_char_int} % When converting, invalid bytes are replaced by the Unicode % replacement character \hexnum{FFFD}. % \begin{macrocode} -\int_const:Nn \c_str_replacement_char_int { "FFFD } +\int_const:Nn \c_@@_replacement_char_int { "FFFD } % \end{macrocode} % \end{variable} % @@ -726,7 +757,7 @@ % \c_tilde_str, % \c_percent_str % } -% For all of those strings, \cs{cs_to_str:N} produces characters with +% For all of those strings, use \cs{cs_to_str:N} to get characters with % the correct category code. % \begin{macrocode} \tl_const:Nx \c_backslash_str { \cs_to_str:N \\ } @@ -738,36 +769,36 @@ % \end{macrocode} % \end{variable} % -% \begin{variable}{\g_str_file_prop} +% \begin{variable}{\g_@@_alias_prop} % To avoid needing one file per encoding/escaping alias, we keep track % of those in a property list. % \begin{macrocode} -\prop_new:N \g_str_file_prop -\prop_gput:Nnn \g_str_file_prop { latin1 } { iso88591 } -\prop_gput:Nnn \g_str_file_prop { latin2 } { iso88592 } -\prop_gput:Nnn \g_str_file_prop { latin3 } { iso88593 } -\prop_gput:Nnn \g_str_file_prop { latin4 } { iso88594 } -\prop_gput:Nnn \g_str_file_prop { latin5 } { iso88599 } -\prop_gput:Nnn \g_str_file_prop { latin6 } { iso885910 } -\prop_gput:Nnn \g_str_file_prop { latin7 } { iso885913 } -\prop_gput:Nnn \g_str_file_prop { latin8 } { iso885914 } -\prop_gput:Nnn \g_str_file_prop { latin9 } { iso885915 } -\prop_gput:Nnn \g_str_file_prop { latin10 } { iso885916 } -\prop_gput:Nnn \g_str_file_prop { utf16le } { utf16 } -\prop_gput:Nnn \g_str_file_prop { utf16be } { utf16 } -\prop_gput:Nnn \g_str_file_prop { utf32le } { utf32 } -\prop_gput:Nnn \g_str_file_prop { utf32be } { utf32 } -\prop_gput:Nnn \g_str_file_prop { hexadecimal } { hex } +\prop_new:N \g_@@_alias_prop +\prop_gput:Nnn \g_@@_alias_prop { latin1 } { iso88591 } +\prop_gput:Nnn \g_@@_alias_prop { latin2 } { iso88592 } +\prop_gput:Nnn \g_@@_alias_prop { latin3 } { iso88593 } +\prop_gput:Nnn \g_@@_alias_prop { latin4 } { iso88594 } +\prop_gput:Nnn \g_@@_alias_prop { latin5 } { iso88599 } +\prop_gput:Nnn \g_@@_alias_prop { latin6 } { iso885910 } +\prop_gput:Nnn \g_@@_alias_prop { latin7 } { iso885913 } +\prop_gput:Nnn \g_@@_alias_prop { latin8 } { iso885914 } +\prop_gput:Nnn \g_@@_alias_prop { latin9 } { iso885915 } +\prop_gput:Nnn \g_@@_alias_prop { latin10 } { iso885916 } +\prop_gput:Nnn \g_@@_alias_prop { utf16le } { utf16 } +\prop_gput:Nnn \g_@@_alias_prop { utf16be } { utf16 } +\prop_gput:Nnn \g_@@_alias_prop { utf32le } { utf32 } +\prop_gput:Nnn \g_@@_alias_prop { utf32be } { utf32 } +\prop_gput:Nnn \g_@@_alias_prop { hexadecimal } { hex } % \end{macrocode} % \end{variable} % -% \begin{variable}{\g_str_error_bool} +% \begin{variable}{\g_@@_error_bool} % In conversion functions with a built-in conditional, errors are not % reported directly to the user, but the information is collected in % this boolean, used at the end to decide on which branch of the % conditional to take. % \begin{macrocode} -\bool_new:N \g_str_error_bool +\bool_new:N \g_@@_error_bool % \end{macrocode} % \end{variable} % @@ -783,12 +814,20 @@ % % \subsubsection{Escaping spaces} % -% \begin{macro}[EXP]{\tl_to_other_str:n} -% \begin{macro}[EXP,aux]{\tl_to_other_str_loop:w, \tl_to_other_str_end:w} -% Replaces all spaces by \enquote{other} spaces, after converting the -% token list to a string via \cs{tl_to_str:n}. This function is -% \texttt{f}-expandable, storing the part of the string with escaped -% spaces between the \cs{q_mark} and \cs{q_stop} markers. +% \begin{macro}[EXP]{\@@_to_other:n} +% \begin{macro}[EXP, aux]{\@@_to_other_loop:w, \@@_to_other_end:w} +% \begin{syntax} +% \cs{@@_to_other:n} \Arg{token list} +% \end{syntax} +% Converts the \meta{token list} to a \meta{other string}, where +% spaces have category code \enquote{other}. First apply +% \cs{tl_to_str:n}, then replace all spaces by \enquote{other} spaces, +% $8$ at a time, storing the converted part of the string between the +% \cs{q_mark} and \cs{q_stop} markers. This function can be +% \texttt{f}-expanded without fear of losing a leading space, since +% spaces do not have category code $10$ in its result. This function +% takes a time quadratic in the token count of the string; +% \cs{@@_gset_other:Nn} is faster but not expandable. % \begin{macrocode} \group_begin: \char_set_lccode:nn { `\* } { `\ } @@ -796,31 +835,31 @@ \tl_to_lowercase:n { \group_end: - \cs_new:Npn \tl_to_other_str:n #1 + \cs_new:Npn \@@_to_other:n #1 { - \exp_after:wN \tl_to_other_str_loop:w \tl_to_str:n {#1} ~ % + \exp_after:wN \@@_to_other_loop:w \tl_to_str:n {#1} ~ % A ~ A ~ A ~ A ~ A ~ A ~ A ~ A ~ \q_mark \q_stop } - \cs_new:Npn \tl_to_other_str_loop:w + \cs_new:Npn \@@_to_other_loop:w #1 ~ #2 ~ #3 ~ #4 ~ #5 ~ #6 ~ #7 ~ #8 ~ #9 \q_stop { \if_meaning:w A #8 - \tl_to_other_str_end:w + \@@_to_other_end:w \fi: - \tl_to_other_str_loop:w + \@@_to_other_loop:w #9 #1 * #2 * #3 * #4 * #5 * #6 * #7 * #8 * \q_stop } - \cs_new:Npn \tl_to_other_str_end:w \fi: #1 \q_mark #2 * A #3 \q_stop + \cs_new:Npn \@@_to_other_end:w \fi: #1 \q_mark #2 * A #3 \q_stop { \fi: #2 } } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\str_gset_other:Nn} -% \begin{macro}[aux,EXP]{\str_gset_other_loop:w} -% \begin{macro}[aux,EXP]{\str_gset_other_end:w} -% This function could be done by using \cs{tl_to_other_str:n} within +% \begin{macro}[int]{\@@_gset_other:Nn} +% \begin{macro}[aux,EXP]{\@@_gset_other_loop:w} +% \begin{macro}[aux,EXP]{\@@_gset_other_end:w} +% This function could be done by using \cs{@@_to_other:n} within % an \texttt{x}-expansion, but that would take a time quadratic in the % size of the string. Instead, we can \enquote{leave the result behind % us} in the input stream, to be captured into the expanding @@ -832,24 +871,24 @@ \tl_to_lowercase:n { \group_end: - \cs_new_protected:Npn \str_gset_other:Nn #1#2 + \cs_new_protected:Npn \@@_gset_other:Nn #1#2 { \tl_gset:Nx #1 { - \exp_after:wN \str_gset_other_loop:w \tl_to_str:n {#2} ~ % + \exp_after:wN \@@_gset_other_loop:w \tl_to_str:n {#2} ~ % A ~ A ~ A ~ A ~ A ~ A ~ A ~ A ~ A ~ \q_stop } } - \cs_new:Npn \str_gset_other_loop:w + \cs_new:Npn \@@_gset_other_loop:w #1 ~ #2 ~ #3 ~ #4 ~ #5 ~ #6 ~ #7 ~ #8 ~ #9 ~ { \if_meaning:w A #9 - \str_gset_other_end:w + \@@_gset_other_end:w \fi: #1 * #2 * #3 * #4 * #5 * #6 * #7 * #8 * #9 - \str_gset_other_loop:w * + \@@_gset_other_loop:w * } - \cs_new:Npn \str_gset_other_end:w \fi: #1 * A #2 \q_stop + \cs_new:Npn \@@_gset_other_end:w \fi: #1 * A #2 \q_stop { \fi: #1 } } % \end{macrocode} @@ -861,7 +900,7 @@ % % \begin{macro}[EXP]{\str_count_spaces:N} % \begin{macro}[EXP]{\str_count_spaces:n} -% \begin{macro}[EXP,aux]{\str_count_spaces_loop:wwwwwwwww} +% \begin{macro}[EXP,aux]{\@@_count_spaces_loop:wwwwwwwww} % To speed up this function, we grab $9$ spaces in each step. The % loop stops when the last argument is one of the trailing % |X|\meta{number}, and that \meta{number} is added to the sum of $9$ @@ -873,31 +912,31 @@ { \int_eval:n { - \exp_after:wN \str_count_spaces_loop:wwwwwwwww + \exp_after:wN \@@_count_spaces_loop:wwwwwwwww \tl_to_str:n {#1} ~ X 7 ~ X 6 ~ X 5 ~ X 4 ~ X 3 ~ X 2 ~ X 1 ~ X 0 ~ X -1 ~ \q_stop } } -\cs_new:Npn \str_count_spaces_loop:wwwwwwwww #1~#2~#3~#4~#5~#6~#7~#8~#9~ +\cs_new:Npn \@@_count_spaces_loop:wwwwwwwww #1~#2~#3~#4~#5~#6~#7~#8~#9~ { \if_meaning:w X #9 \exp_after:wN \use_none_delimit_by_q_stop:w \fi: - \c_nine + \str_count_spaces_loop:wwwwwwwww + \c_nine + \@@_count_spaces_loop:wwwwwwwww } % \end{macrocode} % \end{macro} % \end{macro} % \end{macro} % -% \begin{macro}[EXP]{\str_length:N} -% \begin{macro}[EXP]{\str_length:n} -% \begin{macro}[EXP]{\str_length_unsafe:n} -% \begin{macro}[EXP]{\str_length_ignore_spaces:n} -% \begin{macro}[EXP,aux]{\str_length_aux:n, \str_length_loop:NNNNNNNNN} -% To measure the length of a string we could first escape all spaces -% using \cs{tl_to_other_str:N}, then measure the length of this token +% \begin{macro}[EXP]{\str_count:N} +% \begin{macro}[EXP]{\str_count:n} +% \begin{macro}[EXP]{\@@_count_unsafe:n} +% \begin{macro}[EXP]{\str_count_ignore_spaces:n} +% \begin{macro}[EXP,aux]{\@@_count:n, \@@_count_loop:NNNNNNNNN} +% To measure the token count of a string we could first escape all spaces +% using \cs{@@_to_other:o}, then measure the count of this token % list. However, this would be quadratic in the length of the string, % and we can do better. Namely, add the number of spaces (counted % using the functions defined above) to the length ignoring spaces. @@ -907,26 +946,26 @@ % \texttt{_unsafe} variant expects a token list consisting entirely of % category code $12$ characters. % \begin{macrocode} -\cs_new_nopar:Npn \str_length:N { \exp_args:No \str_length:n } -\cs_new:Npn \str_length:n #1 +\cs_new_nopar:Npn \str_count:N { \exp_args:No \str_count:n } +\cs_new:Npn \str_count:n #1 { - \str_length_aux:n + \@@_count:n { \str_count_spaces:n {#1} - + \exp_after:wN \str_length_loop:NNNNNNNNN \tl_to_str:n {#1} + + \exp_after:wN \@@_count_loop:NNNNNNNNN \tl_to_str:n {#1} } } -\cs_new:Npn \str_length_unsafe:n #1 +\cs_new:Npn \@@_count_unsafe:n #1 { - \str_length_aux:n - { \str_length_loop:NNNNNNNNN #1 } + \@@_count:n + { \@@_count_loop:NNNNNNNNN #1 } } -\cs_new:Npn \str_length_ignore_spaces:n #1 +\cs_new:Npn \str_count_ignore_spaces:n #1 { - \str_length_aux:n - { \exp_after:wN \str_length_loop:NNNNNNNNN \tl_to_str:n {#1} } + \@@_count:n + { \exp_after:wN \@@_count_loop:NNNNNNNNN \tl_to_str:n {#1} } } -\cs_new:Npn \str_length_aux:n #1 +\cs_new:Npn \@@_count:n #1 { \int_eval:n { @@ -937,12 +976,12 @@ \q_stop } } -\cs_set:Npn \str_length_loop:NNNNNNNNN #1#2#3#4#5#6#7#8#9 +\cs_set:Npn \@@_count_loop:NNNNNNNNN #1#2#3#4#5#6#7#8#9 { \if_meaning:w X #9 \exp_after:wN \use_none_delimit_by_q_stop:w \fi: - \c_nine + \str_length_loop:NNNNNNNNN + \c_nine + \@@_count_loop:NNNNNNNNN } % \end{macrocode} % \end{macro} @@ -954,16 +993,16 @@ % \begin{macro}[EXP]{\str_head:N} % \begin{macro}[EXP]{\str_head:n} % \begin{macro}[EXP]{\str_head_ignore_spaces:n} -% \begin{macro}[EXP,aux]{\str_head_aux:w} +% \begin{macro}[EXP,aux]{\@@_head:w} % The \texttt{_ignore_spaces} variant is almost identical to % \cs{tl_head:n}. As usual, \cs{str_head:N} expands its argument and % hands it to \cs{str_head:n}. To circumvent the fact that \TeX{} % skips spaces when grabbing undelimited macro parameters, -% \cs{str_head_aux:w} takes an argument delimited by a space. If |#1| +% \cs{@@_head:w} takes an argument delimited by a space. If |#1| % starts with a non-space character, \cs{use_i_delimit_by_q_stop:nw} % leaves that in the input stream. On the other hand, if |#1| starts -% with a space, the \cs{str_head_aux:w} takes an empty argument, and -% the single (braced) space in the definition of \cs{str_head_aux:w} +% with a space, the \cs{@@_head:w} takes an empty argument, and +% the single (braced) space in the definition of \cs{@@_head:w} % makes its way to the output. Finally, for an empty argument, the % (braced) empty brace group in the definition of \cs{str_head:n} % gives an empty result after passing through @@ -972,11 +1011,11 @@ \cs_new_nopar:Npn \str_head:N { \exp_args:No \str_head:n } \cs_set:Npn \str_head:n #1 { - \exp_after:wN \str_head_aux:w + \exp_after:wN \@@_head:w \tl_to_str:n {#1} { { } } ~ \q_stop } -\cs_set:Npn \str_head_aux:w #1 ~ % +\cs_set:Npn \@@_head:w #1 ~ % { \use_i_delimit_by_q_stop:nw #1 { ~ } } \cs_new:Npn \str_head_ignore_spaces:n #1 { @@ -992,8 +1031,8 @@ % \begin{macro}[EXP]{\str_tail:N} % \begin{macro}[EXP]{\str_tail:n} % \begin{macro}[EXP]{\str_tail_ignore_spaces:n} -% \begin{macro}[EXP,aux]{\str_tail_aux:w} -% \begin{macro}[EXP,aux]{\str_tail_aux_ii:w} +% \begin{macro}[EXP,aux]{\@@_tail:w} +% \begin{macro}[EXP,aux]{\@@_tail_ii:w} % As when fetching the head of a string, the \texttt{_ignore_spaces} % variant is similar to \cs{tl_tail:n}. The more commonly used % \cs{str_tail:n} function is a little bit more convoluted: hitting @@ -1007,17 +1046,17 @@ \cs_new_nopar:Npn \str_tail:N { \exp_args:No \str_tail:n } \cs_set:Npn \str_tail:n #1 { - \exp_after:wN \str_tail_aux:w + \exp_after:wN \@@_tail:w \reverse_if:N \if_charcode:w \scan_stop: \tl_to_str:n {#1} X X \q_stop } -\cs_set:Npn \str_tail_aux:w #1 X #2 \q_stop { \fi: #1 } +\cs_set:Npn \@@_tail:w #1 X #2 \q_stop { \fi: #1 } \cs_new:Npn \str_tail_ignore_spaces:n #1 { - \exp_after:wN \str_tail_aux_ii:w + \exp_after:wN \@@_tail_ii:w \tl_to_str:n {#1} X X \q_stop } -\cs_new:Npn \str_tail_aux_ii:w #1 #2 X #3 \q_stop { #2 } +\cs_new:Npn \@@_tail_ii:w #1 #2 X #3 \q_stop { #2 } % \end{macrocode} % \end{macro} % \end{macro} @@ -1025,88 +1064,88 @@ % \end{macro} % \end{macro} % -% \begin{macro}[EXP, int]{\str_skip_c_zero:w} -% \begin{macro}[EXP, aux]{\str_skip_aux:wNNNNNNNN} -% \begin{macro}[EXP, aux]{\str_skip_end:w, \str_skip_end_ii:NNNNNNNN} +% \begin{macro}[EXP, int]{\@@_skip_c_zero:w} +% \begin{macro}[EXP, aux]{\@@_skip_loop:wNNNNNNNN} +% \begin{macro}[EXP, aux]{\@@_skip_end:w, \@@_skip_end_ii:NNNNNNNN} % Removes |max(#1,0)| characters from the input stream, and then % leaves \cs{c_zero}. This should be expanded using % \cs{tex_romannumeral:D}. We remove characters $8$ at a time until % there are at most $8$ to remove. Then we do a dirty trick: the % \cs{if_case:w} construction leaves between $0$ and $8$ times the % \cs{or:} control sequence, and those \cs{or:} become arguments of -% \cs{str_skip_end_ii:NNNNNNNN}. If the number of characters to remove +% \cs{@@_skip_end_ii:NNNNNNNN}. If the number of characters to remove % is $6$, say, then there are two \cs{or:} left, and the $8$ arguments -% of \cs{str_skip_end_ii:NNNNNNNN} are the two \cs{or:}, and $6$ +% of \cs{@@_skip_end_ii:NNNNNNNN} are the two \cs{or:}, and $6$ % characters from the input stream, exactly what we wanted to % remove. Then close the \cs{if_case:w} conditional with \cs{fi:}, and % stop the initial expansion with \cs{c_zero} (see places where -% \cs{str_skip_c_zero:w} is called). +% \cs{@@_skip_c_zero:w} is called). % \begin{macrocode} -\cs_new:Npn \str_skip_c_zero:w #1; +\cs_new:Npn \@@_skip_c_zero:w #1; { - \if_num:w \int_eval:w #1 > \c_eight - \exp_after:wN \str_skip_aux:wNNNNNNNN + \if_int_compare:w \__int_eval:w #1 > \c_eight + \exp_after:wN \@@_skip_loop:wNNNNNNNN \else: - \exp_after:wN \str_skip_end:w - \int_use:N \int_eval:w + \exp_after:wN \@@_skip_end:w + \int_use:N \__int_eval:w \fi: #1 ; } -\cs_new:Npn \str_skip_aux:wNNNNNNNN #1; #2#3#4#5#6#7#8#9 - { \exp_after:wN \str_skip_c_zero:w \int_use:N \int_eval:w #1 - \c_eight ; } -\cs_new:Npn \str_skip_end:w #1 ; +\cs_new:Npn \@@_skip_loop:wNNNNNNNN #1; #2#3#4#5#6#7#8#9 + { \exp_after:wN \@@_skip_c_zero:w \int_use:N \__int_eval:w #1 - \c_eight ; } +\cs_new:Npn \@@_skip_end:w #1 ; { - \exp_after:wN \str_skip_end_ii:NNNNNNNN - \if_case:w \if_num:w #1 > \c_zero #1 \else: 0 \fi: \exp_stop_f: + \exp_after:wN \@@_skip_end_ii:NNNNNNNN + \if_case:w \if_int_compare:w #1 > \c_zero #1 \else: 0 \fi: \exp_stop_f: \or: \or: \or: \or: \or: \or: \or: \or: } -\cs_new:Npn \str_skip_end_ii:NNNNNNNN #1#2#3#4#5#6#7#8 { \fi: \c_zero } +\cs_new:Npn \@@_skip_end_ii:NNNNNNNN #1#2#3#4#5#6#7#8 { \fi: \c_zero } % \end{macrocode} % \end{macro} % \end{macro} % \end{macro} % -% \begin{macro}[EXP,int]{\str_collect_delimit_by_q_stop:w} -% \begin{macro}[EXP,aux]{\str_collect_aux:wn, \str_collect_aux:wnNNNNNNN} +% \begin{macro}[EXP,int]{\@@_collect_delimit_by_q_stop:w} +% \begin{macro}[EXP,aux]{\@@_collect_loop:wn, \@@_collect_loop_ii:wnNNNNNNN} % \begin{macro}[EXP,aux] -% {\str_collect_end:wn, \str_collect_end_ii:nnnnnnnnw} Collects +% {\@@_collect_end:wn, \@@_collect_end_ii:nnnnnnnnw} Collects % |max(#1,0)| characters, and removes everything else until -% \cs{q_stop}. This is somewhat similar to \cs{str_skip_c_zero:w}, but +% \cs{q_stop}. This is somewhat similar to \cs{@@_skip_c_zero:w}, but % this time we can only grab $7$ characters at a time. At the end, we % use an \cs{if_case:w} trick again, so that the $8$ first arguments -% of \cs{str_collect_end_ii:nnnnnnnnw} are some \cs{or:}, followed by -% an \cs{if:}, followed by |#1| characters from the input +% of \cs{@@_collect_end_ii:nnnnnnnnw} are some \cs{or:}, followed by +% an \cs{fi:}, followed by |#1| characters from the input % stream. Simply leaving this in the input stream will close the % conditional properly and the \cs{or:} disappear. % \begin{macrocode} -\cs_new:Npn \str_collect_delimit_by_q_stop:w #1; +\cs_new:Npn \@@_collect_delimit_by_q_stop:w #1; { - \exp_after:wN \str_collect_aux:wn - \int_use:N \int_eval:w #1 ; + \exp_after:wN \@@_collect_loop:wn + \int_use:N \__int_eval:w #1 ; { } } -\cs_new:Npn \str_collect_aux:wn #1 ; +\cs_new:Npn \@@_collect_loop:wn #1 ; { - \if_num:w #1 > \c_seven - \exp_after:wN \str_collect_aux:wnNNNNNNN + \if_int_compare:w #1 > \c_seven + \exp_after:wN \@@_collect_loop_ii:wnNNNNNNN \else: - \exp_after:wN \str_collect_end:wn + \exp_after:wN \@@_collect_end:wn \fi: #1 ; } -\cs_new:Npn \str_collect_aux:wnNNNNNNN #1; #2 #3#4#5#6#7#8#9 +\cs_new:Npn \@@_collect_loop_ii:wnNNNNNNN #1; #2 #3#4#5#6#7#8#9 { - \exp_after:wN \str_collect_aux:wn - \int_use:N \int_eval:w #1 - \c_seven ; + \exp_after:wN \@@_collect_loop:wn + \int_use:N \__int_eval:w #1 - \c_seven ; { #2 #3#4#5#6#7#8#9 } } -\cs_new:Npn \str_collect_end:wn #1 ; +\cs_new:Npn \@@_collect_end:wn #1 ; { - \exp_after:wN \str_collect_end_ii:nnnnnnnnw - \if_case:w \if_num:w #1 > \c_zero #1 \else: 0 \fi: \exp_stop_f: + \exp_after:wN \@@_collect_end_ii:nnnnnnnnw + \if_case:w \if_int_compare:w #1 > \c_zero #1 \else: 0 \fi: \exp_stop_f: \or: \or: \or: \or: \or: \or: \fi: } -\cs_new:Npn \str_collect_end_ii:nnnnnnnnw #1#2#3#4#5#6#7#8 #9 \q_stop +\cs_new:Npn \@@_collect_end_ii:nnnnnnnnw #1#2#3#4#5#6#7#8 #9 \q_stop { #1#2#3#4#5#6#7#8 } % \end{macrocode} % \end{macro} @@ -1116,33 +1155,36 @@ % \begin{macro}[EXP]{\str_item:Nn} % \begin{macro}[EXP]{\str_item:nn} % \begin{macro}[EXP]{\str_item_ignore_spaces:nn} -% \begin{macro}[EXP]{\str_item_unsafe:nn} -% \begin{macro}[EXP,aux]{\str_item_aux:ww} +% \begin{macro}[EXP]{\@@_item_unsafe:nn} +% \begin{macro}[EXP,aux]{\@@_item:ww} % This is mostly shuffling arguments around to avoid measuring the % length of the string more than once, and make sure that the -% parameters given to \cs{str_skip_c_zero:w} are necessarily within +% parameters given to \cs{@@_skip_c_zero:w} are necessarily within % the bounds of the length of the string. The \texttt{_ignore_spaces} % function cheats a little bit in that it doesn't hand to -% \cs{str_item_unsafe:nn} an \enquote{other string}. This is alright, +% \cs{@@_item_unsafe:nn} an \enquote{other string}. This is alright, % as everything else is done with undelimited arguments. % \begin{macrocode} \cs_new_nopar:Npn \str_item:Nn { \exp_args:No \str_item:nn } -\cs_new:Npn \str_item:nn #1 +\cs_new:Npn \str_item:nn #1#2 { - \exp_last_unbraced:Nf \token_to_str:N - \exp_args:Nf \str_item_unsafe:nn { \tl_to_other_str:n {#1} } + \exp_args:Nf \tl_to_str:n + { + \exp_args:Nf \@@_item_unsafe:nn + { \@@_to_other:n {#1} } {#2} + } } \cs_new:Npn \str_item_ignore_spaces:nn #1 - { \exp_args:No \str_item_unsafe:nn { \tl_to_str:n {#1} } } -\cs_new:Npn \str_item_unsafe:nn #1#2 + { \exp_args:No \@@_item_unsafe:nn { \tl_to_str:n {#1} } } +\cs_new:Npn \@@_item_unsafe:nn #1#2 { - \exp_after:wN \str_item_aux:ww - \int_use:N \int_eval:w #2 \exp_after:wN ; - \int_value:w \str_length_unsafe:n {#1} ; - #1 + \exp_after:wN \@@_item:ww + \int_use:N \__int_eval:w #2 \exp_after:wN ; + \__int_value:w \@@_count_unsafe:n {#1} ; + { } #1 \q_stop } -\cs_new:Npn \str_item_aux:ww #1; #2; +\cs_new:Npn \@@_item:ww #1; #2; { \int_compare:nNnTF {#1} < \c_zero { @@ -1150,16 +1192,16 @@ { \use_none_delimit_by_q_stop:w } { \exp_after:wN \use_i_delimit_by_q_stop:nw - \tex_romannumeral:D \str_skip_c_zero:w #1 + #2 ; + \tex_romannumeral:D \@@_skip_c_zero:w #1 + #2 + \c_one ; } } { - \int_compare:nNnTF {#1} < {#2} + \int_compare:nNnTF {#1} > {#2} + { \use_none_delimit_by_q_stop:w } { \exp_after:wN \use_i_delimit_by_q_stop:nw - \tex_romannumeral:D \str_skip_c_zero:w #1 ; + \tex_romannumeral:D \@@_skip_c_zero:w #1 ; } - { \use_none_delimit_by_q_stop:w } } } % \end{macrocode} @@ -1174,21 +1216,14 @@ % \begin{macro}[EXP]{\str_substr_ignore_spaces:nnn} % \begin{macro}[EXP,aux] % { -% \str_substr_unsafe:nnn, -% \str_substr_aux:nN, -% \str_substr_aux:www, -% \str_substr_aux:nnw, -% \str_substr_normalize_range:nn +% \@@_substr_unsafe:nnn, +% \@@_substr_i:nN, +% \@@_substr_ii:www, +% \@@_substr_iii:nnw, +% \@@_substr_normalize_range:nn % } % Sanitize the string. Then evaluate the arguments, replacing them by -% \cs{c_zero} or \cs{c_max_int} if they are empty. This is done by -% using the construction -% \begin{verbatim} -% \int_use:N \int_eval:w #2 \c_zero \c_zero ; -% \end{verbatim} -% which expands to the value of |#2|, followed by |\c_zero \c_zero ;| -% if |#2| is an expression, and expands to |0\c_zero ;| otherwise. -% The same is done to the end-point of the range. Then limit the +% \cs{c_zero} or \cs{c_max_int} if they are empty. Then limit the % range to be at most the length of the string (this avoids needing to % check for the end of the string when grabbing characters). % Afterwards, skip characters, then keep some more, and finally drop @@ -1199,51 +1234,51 @@ { \exp_args:Nf \tl_to_str:n { - \exp_args:Nf \str_substr_unsafe:nnn - { \tl_to_other_str:n {#1} } {#2} {#3} + \exp_args:Nf \@@_substr_unsafe:nnn + { \@@_to_other:n {#1} } {#2} {#3} } } \cs_new:Npn \str_substr_ignore_spaces:nnn #1 - { \exp_args:No \str_substr_unsafe:nnn { \tl_to_str:n {#1} } } -\cs_new:Npn \str_substr_unsafe:nnn #1#2#3 + { \exp_args:No \@@_substr_unsafe:nnn { \tl_to_str:n {#1} } } +\cs_new:Npn \@@_substr_unsafe:nnn #1#2#3 { - \exp_after:wN \str_substr_aux:www - \int_value:w \str_length_unsafe:n {#1} \exp_after:wN ; - \int_use:N \int_eval:w #2 + \c_zero \exp_after:wN ; - \int_use:N \int_eval:w - \exp_args:Nf \str_substr_aux:nN {#3} \c_max_int ; - #1 + \exp_after:wN \@@_substr_ii:www + \__int_value:w \@@_count_unsafe:n {#1} \exp_after:wN ; + \int_use:N \__int_eval:w #2 + \c_zero \exp_after:wN ; + \int_use:N \__int_eval:w + \exp_args:Nf \@@_substr_i:nN {#3} \c_max_int ; + { } #1 \q_stop } -\cs_new:Npn \str_substr_aux:nN #1 #2 +\cs_new:Npn \@@_substr_i:nN #1 #2 { \tl_if_empty:nTF {#1} {#2} {#1} } -\cs_new:Npn \str_substr_aux:www #1; #2; #3; +\cs_new:Npn \@@_substr_ii:www #1; #2; #3; { - \exp_args:Nf \str_substr_aux:nnw - { \str_substr_normalize_range:nn {#2} {#1} } - { \str_substr_normalize_range:nn {#3} {#1} } + \exp_args:Nf \@@_substr_iii:nnw + { \@@_substr_normalize_range:nn {#2} {#1} } + { \@@_substr_normalize_range:nn {#3} {#1} } } -\cs_new:Npn \str_substr_aux:nnw #1#2 +\cs_new:Npn \@@_substr_iii:nnw #1#2 { - \exp_after:wN \str_collect_delimit_by_q_stop:w - \int_use:N \int_eval:w #2 - #1 \exp_after:wN ; - \tex_romannumeral:D \str_skip_c_zero:w #1 ; + \exp_after:wN \@@_collect_delimit_by_q_stop:w + \int_use:N \__int_eval:w #2 + \c_one - #1 \exp_after:wN ; + \tex_romannumeral:D \@@_skip_c_zero:w #1 ; } -\cs_new:Npn \str_substr_normalize_range:nn #1#2 +\cs_new:Npn \@@_substr_normalize_range:nn #1#2 { \int_eval:n { - \if_num:w #1 < \c_zero - \if_num:w #1 < - #2 \exp_stop_f: + \if_int_compare:w #1 < \c_zero + \if_int_compare:w #1 < - #2 \exp_stop_f: \c_zero \else: - #1 + #2 + #1 + #2 + \c_one \fi: \else: - \if_num:w #1 < #2 \exp_stop_f: - #1 - \else: + \if_int_compare:w #1 > #2 \exp_stop_f: #2 + \else: + #1 \fi: \fi: } @@ -1257,9 +1292,8 @@ % \subsection{String conditionals} % % \begin{macro}[EXP,pTF]{\str_if_eq:NN} -% \begin{macro}[EXP,pTF]{\str_if_eq:nn,\str_if_eq:xx} -% The \texttt{nn} and \texttt{xx} variants are already defined in -% \pkg{l3basics}. Note that \cs{str_if_eq:NN} is different from +% \begin{macro}[EXP,pTF]{\str_if_eq:nn,\str_if_eq_x:nn} +% Note that \cs{str_if_eq:NN} is different from % \cs{tl_if_eq:NN} because it needs to ignore category codes. % \begin{macrocode} \prg_new_conditional:Npnn \str_if_eq:NN #1#2 { p , TF , T , F } @@ -1271,52 +1305,73 @@ % \end{macro} % \end{macro} % -% \begin{macro}[EXP]{\str_if_contains_char:NNT, \str_if_contains_char:NNTF} -% \begin{macro}[EXP]{\str_if_contains_char:nNTF} -% \begin{macro}[EXP,aux]{\str_if_contains_char_aux:NN} -% \begin{macro}[EXP,aux]{\str_if_contains_char_true:} +% \begin{macro}{\str_case:nnn, \str_case:onn, \str_case_x:nnn} +% Defined in \pkg{l3basics} at present. +% \end{macro} +% +% \begin{macro}[EXP]{\@@_if_contains_char:NNT, \@@_if_contains_char:NNTF} +% \begin{macro}[EXP]{\@@_if_contains_char:nNTF} +% \begin{macro}[EXP,aux]{\@@_if_contains_char_aux:NN} +% \begin{macro}[EXP,aux]{\@@_if_contains_char_true:} +% \begin{syntax} +% \cs{@@_if_contains_char:nNTF} \Arg{token list} \meta{char} +% \end{syntax} +% Converts the \meta{token list} to a \meta{string} and tests whether +% the \meta{char} is present in the \meta{string}. Spaces are +% ignored. % Loop over the characters of the string, comparing character codes. % The loop is broken if character codes match. Otherwise we return % \enquote{false}. % \begin{macrocode} -\prg_new_conditional:Npnn \str_if_contains_char:NN #1#2 { T , TF } +\prg_new_conditional:Npnn \@@_if_contains_char:NN #1#2 { T , TF } { - \exp_after:wN \str_if_contains_char_aux:NN \exp_after:wN #2 - \tl_to_str:n \exp_after:wN {#1} { \prg_map_break:n { ? \fi: } } - \prg_break_point:n { } + \exp_after:wN \@@_if_contains_char_aux:NN \exp_after:wN #2 + \etex_detokenize:D \exp_after:wN {#1} { \__prg_break:n { ? \fi: } } + \__prg_break_point: \prg_return_false: } -\prg_new_conditional:Npnn \str_if_contains_char:nN #1#2 { TF } +\prg_new_conditional:Npnn \@@_if_contains_char:nN #1#2 { TF } { - \exp_after:wN \str_if_contains_char_aux:NN \exp_after:wN #2 - \tl_to_str:n {#1} { \prg_map_break:n { ? \fi: } } - \prg_break_point:n { } + \exp_after:wN \@@_if_contains_char_aux:NN \exp_after:wN #2 + \tl_to_str:n {#1} { \__prg_break:n { ? \fi: } } + \__prg_break_point: \prg_return_false: } -\cs_new:Npn \str_if_contains_char_aux:NN #1#2 +\cs_new:Npn \@@_if_contains_char_aux:NN #1#2 { \if_charcode:w #1 #2 - \exp_after:wN \str_if_contains_char_true: + \exp_after:wN \@@_if_contains_char_true: \fi: - \str_if_contains_char_aux:NN #1 + \@@_if_contains_char_aux:NN #1 } -\cs_new_nopar:Npn \str_if_contains_char_true: - { \prg_map_break:n { \prg_return_true: \use_none:n } } +\cs_new_nopar:Npn \@@_if_contains_char_true: + { \__prg_break:n { \prg_return_true: \use_none:n } } % \end{macrocode} % \end{macro} % \end{macro} % \end{macro} % \end{macro} % -% \begin{macro}[aux, rEXP]{\str_aux_octal_use:NTF} +% \begin{macro}[aux, rEXP]{\@@_octal_use:NTF} +% \begin{syntax} +% \cs{@@_octal_use:NTF} \meta{token} \Arg{true code} \Arg{false code} +% \end{syntax} +% If the \meta{token} is an octal digit, it is left in the input +% stream, \emph{followed} by the \meta{true code}. Otherwise, the +% \meta{false code} is left in the input stream. +% \begin{texnote} +% This function will fail if the escape character is an octal +% digit. We are thus careful to set the escape character to a known +% value before using it. +% \end{texnote} % \TeX{} dutifully detects octal digits for us: if |#1| is an octal % digit, then the right-hand side of the comparison is |'1#1|, greater % than $1$. Otherwise, the right-hand side stops as |'1|, and the % conditional takes the \texttt{false} branch. % \begin{macrocode} -\prg_new_conditional:Npnn \str_aux_octal_use:N #1 { TF } +\prg_new_conditional:Npnn \@@_octal_use:N #1 { TF } { - \if_num:w \c_one < '1 \token_to_str:N #1 \exp_stop_f: + \if_int_compare:w \c_one < '1 \token_to_str:N #1 \exp_stop_f: #1 \prg_return_true: \else: \prg_return_false: @@ -1325,19 +1380,19 @@ % \end{macrocode} % \end{macro} % -% \begin{macro}[aux, rEXP]{\str_aux_hexadecimal_use:NTF} +% \begin{macro}[aux, rEXP]{\@@_hexadecimal_use:NTF} % \TeX{} detects uppercase hexadecimal digits for us (see -% \cs{str_aux_octal_use:NTF}), but not the lowercase letters, which we +% \cs{@@_octal_use:NTF}), but not the lowercase letters, which we % need to detect and replace by their uppercase counterpart. % \begin{macrocode} -\prg_new_conditional:Npnn \str_aux_hexadecimal_use:N #1 { TF } +\prg_new_conditional:Npnn \@@_hexadecimal_use:N #1 { TF } { - \if_num:w \c_two < "1 \token_to_str:N #1 \exp_stop_f: + \if_int_compare:w \c_two < "1 \token_to_str:N #1 \exp_stop_f: #1 \prg_return_true: \else: - \if_case:w \int_eval:w + \if_case:w \__int_eval:w \exp_after:wN ` \token_to_str:N #1 - `a - \int_eval_end: + \__int_eval_end: A \or: B \or: C @@ -1369,8 +1424,8 @@ % % \subsubsection{Producing one byte or character} % -% \begin{variable}{\c_str_byte_0_tl, \c_str_byte_1_tl, \c_str_byte_255_tl} -% \begin{variable}{\c_str_byte_-1_tl} +% \begin{variable}{\c_@@_byte_0_tl, \c_@@_byte_1_tl, \c_@@_byte_255_tl} +% \begin{variable}{\c_@@_byte_-1_tl} % For each integer $N$ in the range $[0,255]$, we create a constant % token list which holds three character tokens with category code % other: the character with character code $N$, followed by the @@ -1380,59 +1435,59 @@ % \begin{macrocode} \group_begin: \char_set_catcode_other:n { \c_zero } - \tl_set:Nx \l_str_internal_tl { \tl_to_str:n { 0123456789ABCDEF } } - \exp_args:No \tl_map_inline:nn { \l_str_internal_tl " } + \tl_set:Nx \l_@@_internal_tl { \tl_to_str:n { 0123456789ABCDEF } } + \exp_args:No \tl_map_inline:nn { \l_@@_internal_tl " } { \char_set_lccode:nn {`#1} { \c_zero } } - \tl_map_inline:Nn \l_str_internal_tl + \tl_map_inline:Nn \l_@@_internal_tl { - \tl_map_inline:Nn \l_str_internal_tl + \tl_map_inline:Nn \l_@@_internal_tl { \char_set_lccode:nn { \c_zero } {"#1##1} \tl_to_lowercase:n { \tl_const:cx - { c_str_byte_ \int_eval:n {"#1##1} _tl } + { c_@@_byte_ \int_eval:n {"#1##1} _tl } { ^^@ #1 ##1 } } } } \group_end: -\tl_const:cn { c_str_byte_-1_tl } { { } \use_none:n { } } +\tl_const:cn { c_@@_byte_-1_tl } { { } \use_none:n { } } % \end{macrocode} % \end{variable} % \end{variable} % -% \begin{macro}[int, EXP]{\str_output_byte:n} -% \begin{macro}[int, EXP]{\str_output_byte:w} -% \begin{macro}[int, EXP]{\str_output_hexadecimal:n} -% \begin{macro}[int, EXP]{\str_output_hexadecimal:w} -% \begin{macro}[int, EXP]{\str_output_end:} +% \begin{macro}[int, EXP]{\@@_output_byte:n} +% \begin{macro}[int, EXP]{\@@_output_byte:w} +% \begin{macro}[int, EXP]{\@@_output_hexadecimal:n} +% \begin{macro}[int, EXP]{\@@_output_hexadecimal:w} +% \begin{macro}[int, EXP]{\@@_output_end:} % Those functions must be used carefully: feeding them a value outside % the range $[-1,255]$ will attempt to use the undefined token list -% variable \cs{c_str_byte_\meta{number}_tl}. Assuming that the +% variable \cs{c_@@_byte_\meta{number}_tl}. Assuming that the % argument is in the right range, we expand the corresponding token % list, and pick either the byte (first token) or the hexadecimal % representations (second and third tokens). The value $-1$ produces % an empty result in both cases. % \begin{macrocode} -\cs_new:Npn \str_output_byte:n #1 - { \str_output_byte:w #1 \str_output_end: } -\cs_new_nopar:Npn \str_output_byte:w +\cs_new:Npn \@@_output_byte:n #1 + { \@@_output_byte:w #1 \@@_output_end: } +\cs_new_nopar:Npn \@@_output_byte:w { \exp_after:wN \exp_after:wN \exp_after:wN \use_i:nnn - \cs:w c_str_byte_ \int_use:N \int_eval:w + \cs:w c_@@_byte_ \int_use:N \__int_eval:w } -\cs_new:Npn \str_output_hexadecimal:n #1 - { \str_output_hexadecimal:w #1 \str_output_end: } -\cs_new_nopar:Npn \str_output_hexadecimal:w +\cs_new:Npn \@@_output_hexadecimal:n #1 + { \@@_output_hexadecimal:w #1 \@@_output_end: } +\cs_new_nopar:Npn \@@_output_hexadecimal:w { \exp_after:wN \exp_after:wN \exp_after:wN \use_none:n - \cs:w c_str_byte_ \int_use:N \int_eval:w + \cs:w c_@@_byte_ \int_use:N \__int_eval:w } -\cs_new_nopar:Npn \str_output_end: - { \int_eval_end: _tl \cs_end: } +\cs_new_nopar:Npn \@@_output_end: + { \__int_eval_end: _tl \cs_end: } % \end{macrocode} % \end{macro} % \end{macro} @@ -1440,27 +1495,27 @@ % \end{macro} % \end{macro} % -% \begin{macro}[int, rEXP]{\str_output_byte_pair_be:n} -% \begin{macro}[int, rEXP]{\str_output_byte_pair_le:n} -% \begin{macro}[aux, rEXP]{\str_output_byte_pair_aux:nnN} +% \begin{macro}[int, rEXP]{\@@_output_byte_pair_be:n} +% \begin{macro}[int, rEXP]{\@@_output_byte_pair_le:n} +% \begin{macro}[aux, rEXP]{\@@_output_byte_pair:nnN} % Convert a number in the range $[0,65535]$ to a pair of bytes, either % big-endian or little-endian. % \begin{macrocode} -\cs_new:Npn \str_output_byte_pair_be:n #1 +\cs_new:Npn \@@_output_byte_pair_be:n #1 { - \exp_args:Nf \str_output_byte_pair_aux:nnN + \exp_args:Nf \@@_output_byte_pair:nnN { \int_div_truncate:nn { #1 } { "100 } } {#1} \use:nn } -\cs_new:Npn \str_output_byte_pair_le:n #1 +\cs_new:Npn \@@_output_byte_pair_le:n #1 { - \exp_args:Nf \str_output_byte_pair_aux:nnN + \exp_args:Nf \@@_output_byte_pair:nnN { \int_div_truncate:nn { #1 } { "100 } } {#1} \use_ii_i:nn } -\cs_new:Npn \str_output_byte_pair_aux:nnN #1#2#3 +\cs_new:Npn \@@_output_byte_pair:nnN #1#2#3 { #3 - { \str_output_byte:n { #1 } } - { \str_output_byte:n { #2 - #1 * "100 } } + { \@@_output_byte:n { #1 } } + { \@@_output_byte:n { #2 - #1 * "100 } } } % \end{macrocode} % \end{macro} @@ -1469,52 +1524,52 @@ % % \subsubsection{Mapping functions for conversions} % -% \begin{macro}{\str_convert_gmap:N} -% \begin{macro}[aux, rEXP]{\str_convert_gmap_loop:NN} +% \begin{macro}{\@@_convert_gmap:N} +% \begin{macro}[aux, rEXP]{\@@_convert_gmap_loop:NN} % This maps the function |#1| over all characters in -% \cs{g_str_result_tl}, which should be a byte string in most cases, +% \cs{g_@@_result_tl}, which should be a byte string in most cases, % sometimes a native string. % \begin{macrocode} -\cs_new_protected:Npn \str_convert_gmap:N #1 +\cs_new_protected:Npn \@@_convert_gmap:N #1 { - \tl_gset:Nx \g_str_result_tl + \tl_gset:Nx \g_@@_result_tl { - \exp_after:wN \str_convert_gmap_loop:NN + \exp_after:wN \@@_convert_gmap_loop:NN \exp_after:wN #1 - \g_str_result_tl { ? \prg_map_break: } - \prg_break_point:n { } + \g_@@_result_tl { ? \__prg_break: } + \__prg_break_point: } } -\cs_new:Npn \str_convert_gmap_loop:NN #1#2 +\cs_new:Npn \@@_convert_gmap_loop:NN #1#2 { \use_none:n #2 #1#2 - \str_convert_gmap_loop:NN #1 + \@@_convert_gmap_loop:NN #1 } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}{\str_convert_gmap_internal:N} -% \begin{macro}[aux, rEXP]{\str_convert_gmap_internal_loop:Nw} +% \begin{macro}{\@@_convert_gmap_internal:N} +% \begin{macro}[aux, rEXP]{\@@_convert_gmap_internal_loop:Nw} % This maps the function |#1| over all character codes in -% \cs{g_str_result_tl}, which must be in the internal representation. +% \cs{g_@@_result_tl}, which must be in the internal representation. % \begin{macrocode} -\cs_new_protected:Npn \str_convert_gmap_internal:N #1 +\cs_new_protected:Npn \@@_convert_gmap_internal:N #1 { - \tl_gset:Nx \g_str_result_tl + \tl_gset:Nx \g_@@_result_tl { - \exp_after:wN \str_convert_gmap_internal_loop:Nww + \exp_after:wN \@@_convert_gmap_internal_loop:Nww \exp_after:wN #1 - \g_str_result_tl \s_tl \q_stop \prg_map_break: \s_tl - \prg_break_point:n { } + \g_@@_result_tl \s__tl \q_stop \__prg_break: \s__tl + \__prg_break_point: } } -\cs_new:Npn \str_convert_gmap_internal_loop:Nww #1 #2 \s_tl #3 \s_tl +\cs_new:Npn \@@_convert_gmap_internal_loop:Nww #1 #2 \s__tl #3 \s__tl { \use_none_delimit_by_q_stop:w #3 \q_stop #1 {#3} - \str_convert_gmap_internal_loop:Nww #1 + \@@_convert_gmap_internal_loop:Nww #1 } % \end{macrocode} % \end{macro} @@ -1522,38 +1577,38 @@ % % \subsubsection{Error-reporting during conversion} % -% \begin{macro}[int]{\str_if_flag_error:nnx} -% \begin{macro}[aux]{\str_if_flag_no_error:nnx} +% \begin{macro}[int]{\@@_if_flag_error:nnx} +% \begin{macro}[aux]{\@@_if_flag_no_error:nnx} % When converting using the function \cs{str_set_convert:Nnnn}, errors % should be reported to the user after each step in the % conversion. Errors are signalled by raising some flag (typically -% \texttt{str_error}), so here we test that flag: if it is raised, +% \texttt{@@_error}), so here we test that flag: if it is raised, % give the user an error, otherwise remove the arguments. On the other % hand, in the conditional functions \cs{str_set_convert:NnnnTF}, % errors should be suppressed. This is done by changing -% \cs{str_if_flag_error:nnx} into \cs{str_if_flag_no_error:nnx} +% \cs{@@_if_flag_error:nnx} into \cs{@@_if_flag_no_error:nnx} % locally. % \begin{macrocode} -\cs_new_protected:Npn \str_if_flag_error:nnx #1 +\cs_new_protected:Npn \@@_if_flag_error:nnx #1 { \flag_if_raised:nTF {#1} - { \msg_kernel_error:nnx { str } } + { \__msg_kernel_error:nnx { str } } { \use_none:nn } } -\cs_new_protected:Npn \str_if_flag_no_error:nnx #1#2#3 - { \flag_if_raised:nT {#1} { \bool_gset_true:N \g_str_error_bool } } +\cs_new_protected:Npn \@@_if_flag_no_error:nnx #1#2#3 + { \flag_if_raised:nT {#1} { \bool_gset_true:N \g_@@_error_bool } } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\str_if_flag_times:nT} +% \begin{macro}[int]{\@@_if_flag_times:nT} % At the end of each conversion step, we raise all relevant errors as % one error message, built on the fly. The height of each flag % indicates how many times a given error was encountered. This % function prints |#2| followed by the number of occurrences of an % error if it occurred, nothing otherwise. % \begin{macrocode} -\cs_new_protected:Npn \str_if_flag_times:nT #1#2 +\cs_new_protected:Npn \@@_if_flag_times:nT #1#2 { \flag_if_raised:nT {#1} { #2~(x \flag_height:n {#1} ) } } % \end{macrocode} % \end{macro} @@ -1575,7 +1630,7 @@ % to a list of Unicode characters in an internal representation, % with items of the form % \begin{quote} -% \meta{bytes} \cs{s_tl} \meta{Unicode code point} \cs{s_tl} +% \meta{bytes} \cs{s__tl} \meta{Unicode code point} \cs{s__tl} % \end{quote} % where we have collected the \meta{bytes} which combined to form % this particular Unicode character, and the \meta{Unicode code @@ -1593,63 +1648,63 @@ % % \begin{macro}{\str_set_convert:Nnnn, \str_gset_convert:Nnnn} % \begin{macro}[TF]{\str_set_convert:Nnnn, \str_gset_convert:Nnnn} -% \begin{macro}[aux]{\str_convert_aux_i:nNNnnn} -% The input string is stored in \cs{g_str_result_tl}, then we: +% \begin{macro}[aux]{\@@_convert_i:nNNnnn} +% The input string is stored in \cs{g_@@_result_tl}, then we: % unescape and decode; encode and escape; exit the group and store the % result in the user's variable. The various conversion functions all -% act on \cs{g_str_result_tl}. Errors are silenced for the conditional -% functions by redefining \cs{str_if_flag_error:nnx} locally. +% act on \cs{g_@@_result_tl}. Errors are silenced for the conditional +% functions by redefining \cs{@@_if_flag_error:nnx} locally. % \begin{macrocode} \cs_new_protected_nopar:Npn \str_set_convert:Nnnn - { \str_convert_aux_i:nNNnnn { } \tl_set_eq:NN } + { \@@_convert_i:nNNnnn { } \tl_set_eq:NN } \cs_new_protected_nopar:Npn \str_gset_convert:Nnnn - { \str_convert_aux_i:nNNnnn { } \tl_gset_eq:NN } + { \@@_convert_i:nNNnnn { } \tl_gset_eq:NN } \prg_new_protected_conditional:Npnn \str_set_convert:Nnnn #1#2#3#4 { T , F , TF } { - \bool_gset_false:N \g_str_error_bool - \str_convert_aux_i:nNNnnn - { \cs_set_eq:NN \str_if_flag_error:nnx \str_if_flag_no_error:nnx } + \bool_gset_false:N \g_@@_error_bool + \@@_convert_i:nNNnnn + { \cs_set_eq:NN \@@_if_flag_error:nnx \@@_if_flag_no_error:nnx } \tl_set_eq:NN #1 {#2} {#3} {#4} - \bool_if:NTF \g_str_error_bool \prg_return_false: \prg_return_true: + \bool_if:NTF \g_@@_error_bool \prg_return_false: \prg_return_true: } \prg_new_protected_conditional:Npnn \str_gset_convert:Nnnn #1#2#3#4 { T , F , TF } { - \bool_gset_false:N \g_str_error_bool - \str_convert_aux_i:nNNnnn - { \cs_set_eq:NN \str_if_flag_error:nnx \str_if_flag_no_error:nnx } + \bool_gset_false:N \g_@@_error_bool + \@@_convert_i:nNNnnn + { \cs_set_eq:NN \@@_if_flag_error:nnx \@@_if_flag_no_error:nnx } \tl_gset_eq:NN #1 {#2} {#3} {#4} - \bool_if:NTF \g_str_error_bool \prg_return_false: \prg_return_true: + \bool_if:NTF \g_@@_error_bool \prg_return_false: \prg_return_true: } -\cs_new_protected:Npn \str_convert_aux_i:nNNnnn #1#2#3#4#5#6 +\cs_new_protected:Npn \@@_convert_i:nNNnnn #1#2#3#4#5#6 { \group_begin: #1 - \str_gset_other:Nn \g_str_result_tl {#4} - \exp_after:wN \str_convert_aux_ii:wwwnn + \@@_gset_other:Nn \g_@@_result_tl {#4} + \exp_after:wN \@@_convert_ii:wwwnn \tl_to_str:n {#5} /// \q_stop { decode } { unescape } \prg_do_nothing: - \str_convert_decode_: - \exp_after:wN \str_convert_aux_ii:wwwnn + \@@_convert_decode_: + \exp_after:wN \@@_convert_ii:wwwnn \tl_to_str:n {#6} /// \q_stop { encode } { escape } \use_ii_i:nn - \str_convert_encode_: + \@@_convert_encode_: \group_end: - #2 #3 \g_str_result_tl + #2 #3 \g_@@_result_tl } % \end{macrocode} % \end{macro} % \end{macro} % \end{macro} % -% \begin{macro}[aux]{\str_convert_aux_ii:wwwnn} -% \begin{macro}[aux]{\str_convert_aux_iii:NNnNN} -% The task of \cs{str_convert_aux_ii:wwwnn} is to split +% \begin{macro}[aux]{\@@_convert_ii:wwwnn} +% \begin{macro}[aux]{\@@_convert_iii:NNnNN} +% The task of \cs{@@_convert_ii:wwwnn} is to split % \meta{encoding}/\meta{escaping} pairs into their components, |#1| -% and |#2|. Calls to \cs{str_convert_aux_iv:nnn} ensure that the +% and |#2|. Calls to \cs{@@_convert_iv:nnn} ensure that the % corresponding conversion functions are defined. The third auxiliary % does the main work. % \begin{itemize} @@ -1669,19 +1724,19 @@ % (native), then the escaping should be ignored, with an error if any % was given, and only the encoding, |#1|, should be performed. % \begin{macrocode} -\cs_new_protected:Npn \str_convert_aux_ii:wwwnn +\cs_new_protected:Npn \@@_convert_ii:wwwnn #1 / #2 // #3 \q_stop #4#5 { - \str_convert_aux_iv:nnn {enc} {#4} {#1} - \str_convert_aux_iv:nnn {esc} {#5} {#2} - \exp_args:Ncc \str_convert_aux_iii:NNnNN - { str_convert_#4_#1: } { str_convert_#5_#2: } {#2} + \@@_convert_iv:nnn {enc} {#4} {#1} + \@@_convert_iv:nnn {esc} {#5} {#2} + \exp_args:Ncc \@@_convert_iii:NNnNN + { @@_convert_#4_#1: } { @@_convert_#5_#2: } {#2} } -\cs_new_protected:Npn \str_convert_aux_iii:NNnNN #1#2#3#4#5 +\cs_new_protected:Npn \@@_convert_iii:NNnNN #1#2#3#4#5 { \if_meaning:w #1 #5 \tl_if_empty:nF {#3} - { \msg_kernel_error:nnx { str } { native-escaping } {#3} } + { \__msg_kernel_error:nnx { str } { native-escaping } {#3} } #1 \else: #4 #2 #1 @@ -1691,15 +1746,15 @@ % \end{macro} % \end{macro} % -% \begin{macro}[aux]{\str_convert_aux_iv:nnn} -% \begin{macro}[aux]{\str_convert_aux_v:nnnn} -% The arguments of \cs{str_convert_aux_iv:nnn} are: \texttt{enc} or +% \begin{macro}[aux]{\@@_convert_iv:nnn} +% \begin{macro}[aux]{\@@_convert_v:nnnn} +% The arguments of \cs{@@_convert_iv:nnn} are: \texttt{enc} or % \texttt{esc}, used to build filenames, the type of the conversion % (unescape, decode, encode, escape), and the encoding or escaping % name. If the function is already defined, no need to do anything. % Otherwise, filter out all non-alphanumerics in the name, and % lowercase it. Feed that, and the same three arguments, to -% \cs{str_convert_aux_v:nnnn}. The task is then to make sure that the +% \cs{@@_convert_v:nnnn}. The task is then to make sure that the % conversion function |#3_#1| corresponding to the type |#3| and % filtered name |#1| is defined, then set our initial conversion % function |#3_#4| equal to that. @@ -1707,10 +1762,10 @@ % How do we get the |#3_#1| conversion to be defined if it isn't? % Two main cases. % -% First, if |#1| is a key in \cs{g_str_file_prop}, then the value -% \cs{l_str_internal_tl} tells us what file to load. Loading is +% First, if |#1| is a key in \cs{g_@@_alias_prop}, then the value +% \cs{l_@@_internal_tl} tells us what file to load. Loading is % skipped if the file was already read, \emph{i.e.}, if the conversion -% command based on \cs{l_str_internal_tl} already exists. Otherwise, +% command based on \cs{l_@@_internal_tl} already exists. Otherwise, % try to load the file; if that fails, there is an error, use the % default empty name instead. % @@ -1720,96 +1775,96 @@ % third-party \texttt{.def} files). If the file is not found, there is % an error, use the default empty name instead. % -% In all cases, the conversion based on \cs{l_str_internal_tl} is +% In all cases, the conversion based on \cs{l_@@_internal_tl} is % defined, so we can set the |#3_#1| function equal to that. In some % cases (\emph{e.g.}, \texttt{utf16be}), the |#3_#1| function is % actually defined within the file we just loaded, and it is different -% from the \cs{l_str_internal_tl}-based function: we mustn't clobber +% from the \cs{l_@@_internal_tl}-based function: we mustn't clobber % that different definition. % \begin{macrocode} -\cs_new_protected:Npn \str_convert_aux_iv:nnn #1#2#3 +\cs_new_protected:Npn \@@_convert_iv:nnn #1#2#3 { - \cs_if_exist:cF { str_convert_#2_#3: } + \cs_if_exist:cF { @@_convert_#2_#3: } { - \exp_args:Nx \str_convert_aux_v:nnnn - { \str_convert_lowercase_alphanum:n {#3} } + \exp_args:Nx \@@_convert_v:nnnn + { \@@_convert_lowercase_alphanum:n {#3} } {#1} {#2} {#3} } } -\cs_new_protected:Npn \str_convert_aux_v:nnnn #1#2#3#4 +\cs_new_protected:Npn \@@_convert_v:nnnn #1#2#3#4 { - \cs_if_exist:cF { str_convert_#3_#1: } + \cs_if_exist:cF { @@_convert_#3_#1: } { - \prop_get:NnNF \g_str_file_prop {#1} \l_str_internal_tl - { \tl_set:Nn \l_str_internal_tl {#1} } - \cs_if_exist:cF { str_convert_#3_ \l_str_internal_tl : } + \prop_get:NnNF \g_@@_alias_prop {#1} \l_@@_internal_tl + { \tl_set:Nn \l_@@_internal_tl {#1} } + \cs_if_exist:cF { @@_convert_#3_ \l_@@_internal_tl : } { - \file_if_exist:nTF { l3str-#2- \l_str_internal_tl .def } + \file_if_exist:nTF { l3str-#2- \l_@@_internal_tl .def } { \group_begin: - \str_load_catcodes: - \file_input:n { l3str-#2- \l_str_internal_tl .def } + \@@_load_catcodes: + \file_input:n { l3str-#2- \l_@@_internal_tl .def } \group_end: } { - \tl_clear:N \l_str_internal_tl - \msg_kernel_error:nnxx { str } { unknown-#2 } {#4} {#1} + \tl_clear:N \l_@@_internal_tl + \__msg_kernel_error:nnxx { str } { unknown-#2 } {#4} {#1} } } - \cs_if_exist:cF { str_convert_#3_#1: } + \cs_if_exist:cF { @@_convert_#3_#1: } { - \cs_gset_eq:cc { str_convert_#3_#1: } - { str_convert_#3_ \l_str_internal_tl : } + \cs_gset_eq:cc { @@_convert_#3_#1: } + { @@_convert_#3_ \l_@@_internal_tl : } } } - \cs_gset_eq:cc { str_convert_#3_#4: } { str_convert_#3_#1: } + \cs_gset_eq:cc { @@_convert_#3_#4: } { @@_convert_#3_#1: } } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int, rEXP]{\str_convert_lowercase_alphanum:n} -% \begin{macro}[aux, rEXP]{\str_convert_lowercase_alphanum_loop:N} +% \begin{macro}[int, rEXP]{\@@_convert_lowercase_alphanum:n} +% \begin{macro}[aux, rEXP]{\@@_convert_lowercase_alphanum_loop:N} % This function keeps only letters and digits, with upper case letters % converted to lower case. % \begin{macrocode} -\cs_new:Npn \str_convert_lowercase_alphanum:n #1 +\cs_new:Npn \@@_convert_lowercase_alphanum:n #1 { - \exp_after:wN \str_convert_lowercase_alphanum_loop:N - \tl_to_str:n {#1} { ? \prg_map_break: } - \prg_break_point:n { } + \exp_after:wN \@@_convert_lowercase_alphanum_loop:N + \tl_to_str:n {#1} { ? \__prg_break: } + \__prg_break_point: } -\cs_new:Npn \str_convert_lowercase_alphanum_loop:N #1 +\cs_new:Npn \@@_convert_lowercase_alphanum_loop:N #1 { \use_none:n #1 - \if_num:w `#1 < \c_ninety_one - \if_num:w `#1 < \c_sixty_five - \if_num:w \c_one < 1#1 \exp_stop_f: + \if_int_compare:w `#1 < \c_ninety_one + \if_int_compare:w `#1 < \c_sixty_five + \if_int_compare:w \c_one < 1#1 \exp_stop_f: #1 \fi: \else: - \str_output_byte:n { `#1 + \c_thirty_two } + \@@_output_byte:n { `#1 + \c_thirty_two } \fi: \else: - \if_num:w `#1 < \c_one_hundred_twenty_three - \if_num:w `#1 < \c_ninety_seven + \if_int_compare:w `#1 < \c_one_hundred_twenty_three + \if_int_compare:w `#1 < \c_ninety_seven \else: #1 \fi: \fi: \fi: - \str_convert_lowercase_alphanum_loop:N + \@@_convert_lowercase_alphanum_loop:N } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\str_load_catcodes:} +% \begin{macro}[int]{\@@_load_catcodes:} % Since encoding files may be loaded at arbitrary places in a \TeX{} % document, including within verbatim mode, we set the catcodes of all % characters appearing in any encoding definition file. % \begin{macrocode} -\cs_new_protected:Npn \str_load_catcodes: +\cs_new_protected:Npn \@@_load_catcodes: { \char_set_catcode_escape:N \\ \char_set_catcode_group_begin:N \{ @@ -1837,84 +1892,84 @@ % needed. By default, on input any non-byte is filtered out, while the % output simply consists in letting bytes through. % -% \begin{macro}[int, rEXP]{\str_filter_bytes:n} -% \begin{macro}[aux, rEXP]{\str_filter_bytes_aux:N} +% \begin{macro}[int, rEXP]{\@@_filter_bytes:n} +% \begin{macro}[aux, rEXP]{\@@_filter_bytes_aux:N} % In the case of pdf\TeX{}, every character is a byte. For % Unicode-aware engines, test the character code; non-bytes cause us % to raise the flag \texttt{str_byte}. Spaces have already been given % the correct category code when this function is called. % \begin{macrocode} \pdftex_if_engine:TF - { \cs_new_eq:NN \str_filter_bytes:n \use:n } + { \cs_new_eq:NN \@@_filter_bytes:n \use:n } { - \cs_new:Npn \str_filter_bytes:n #1 + \cs_new:Npn \@@_filter_bytes:n #1 { - \str_filter_bytes_aux:N #1 - { ? \prg_map_break: } - \prg_break_point:n { } + \@@_filter_bytes_aux:N #1 + { ? \__prg_break: } + \__prg_break_point: } - \cs_new:Npn \str_filter_bytes_aux:N #1 + \cs_new:Npn \@@_filter_bytes_aux:N #1 { \use_none:n #1 - \if_num:w `#1 < 256 \exp_stop_f: + \if_int_compare:w `#1 < 256 \exp_stop_f: #1 \else: \flag_raise:n { str_byte } \fi: - \str_filter_bytes_aux:N + \@@_filter_bytes_aux:N } } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\str_convert_unescape_:} -% \begin{macro}[int]{\str_convert_unescape_bytes:} +% \begin{macro}[int]{\@@_convert_unescape_:} +% \begin{macro}[int]{\@@_convert_unescape_bytes:} % The simplest unescaping method removes non-bytes from -% \cs{g_str_result_tl}. +% \cs{g_@@_result_tl}. % \begin{macrocode} \pdftex_if_engine:TF - { \cs_new_protected_nopar:Npn \str_convert_unescape_: { } } + { \cs_new_protected_nopar:Npn \@@_convert_unescape_: { } } { - \cs_new_protected_nopar:Npn \str_convert_unescape_: + \cs_new_protected_nopar:Npn \@@_convert_unescape_: { \flag_clear:n { str_byte } - \tl_gset:Nx \g_str_result_tl - { \exp_args:No \str_filter_bytes:n \g_str_result_tl } - \str_if_flag_error:nnx { str_byte } { non-byte } { bytes } + \tl_gset:Nx \g_@@_result_tl + { \exp_args:No \@@_filter_bytes:n \g_@@_result_tl } + \@@_if_flag_error:nnx { str_byte } { non-byte } { bytes } } } -\cs_new_eq:NN \str_convert_unescape_bytes: \str_convert_unescape_: +\cs_new_eq:NN \@@_convert_unescape_bytes: \@@_convert_unescape_: % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\str_convert_escape_:} -% \begin{macro}[int]{\str_convert_escape_bytes:} +% \begin{macro}[int]{\@@_convert_escape_:} +% \begin{macro}[int]{\@@_convert_escape_bytes:} % The simplest form of escape leaves the bytes from the previous step % of the conversion unchanged. % \begin{macrocode} -\cs_new_protected_nopar:Npn \str_convert_escape_: { } -\cs_new_eq:NN \str_convert_escape_bytes: \str_convert_escape_: +\cs_new_protected_nopar:Npn \@@_convert_escape_: { } +\cs_new_eq:NN \@@_convert_escape_bytes: \@@_convert_escape_: % \end{macrocode} % \end{macro} % \end{macro} % % \subsubsection{Native strings} % -% \begin{macro}[int]{\str_convert_decode_:} -% \begin{macro}[aux, rEXP]{\str_decode_native_char:N} +% \begin{macro}[int]{\@@_convert_decode_:} +% \begin{macro}[aux, rEXP]{\@@_decode_native_char:N} % Convert each character to its character code, one at a time. % \begin{macrocode} -\cs_new_protected_nopar:Npn \str_convert_decode_: - { \str_convert_gmap:N \str_decode_native_char:N } -\cs_new:Npn \str_decode_native_char:N #1 - { #1 \s_tl \int_value:w `#1 \s_tl } +\cs_new_protected_nopar:Npn \@@_convert_decode_: + { \@@_convert_gmap:N \@@_decode_native_char:N } +\cs_new:Npn \@@_decode_native_char:N #1 + { #1 \s__tl \__int_value:w `#1 \s__tl } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\str_convert_encode_:} +% \begin{macro}[int]{\@@_convert_encode_:} % The conversion from an internal string to native character tokens is % very different in pdf\TeX{} and in other engines. For Unicode-aware % engines, we need the definitions to be read when the null byte has @@ -1924,7 +1979,7 @@ \char_set_catcode_other:n { 0 } \pdftex_if_engine:TF % \end{macrocode} -% \begin{macro}[aux, EXP]{\str_encode_native_char:n} +% \begin{macro}[aux, EXP]{\@@_encode_native_char:n} % Since pdf\TeX{} only supports 8-bit characters, and we have a table % of all bytes, the conversion can be done in linear time within an % \texttt{x}-expanding assignment. Look out for character codes larger @@ -1932,23 +1987,23 @@ % which then triggers a pdf\TeX{}-specific error. % \begin{macrocode} { - \cs_new_protected_nopar:Npn \str_convert_encode_: + \cs_new_protected_nopar:Npn \@@_convert_encode_: { \flag_clear:n { str_error } - \str_convert_gmap_internal:N \str_encode_native_char:n - \str_if_flag_error:nnx { str_error } + \@@_convert_gmap_internal:N \@@_encode_native_char:n + \@@_if_flag_error:nnx { str_error } { pdfTeX-native-overflow } { } } - \cs_new:Npn \str_encode_native_char:n #1 + \cs_new:Npn \@@_encode_native_char:n #1 { - \if_num:w #1 < \c_two_hundred_fifty_six - \str_output_byte:n {#1} + \if_int_compare:w #1 < \c_two_hundred_fifty_six + \@@_output_byte:n {#1} \else: \flag_raise:n { str_error } ? \fi: } - \msg_kernel_new:nnnn { str } { pdfTeX-native-overflow } + \__msg_kernel_new:nnnn { str } { pdfTeX-native-overflow } { Character~code~too~large~for~pdfTeX. } { The~pdfTeX~engine~only~supports~8-bit~characters:~ @@ -1958,9 +2013,9 @@ } % \end{macrocode} % \end{macro} -% \begin{macro}[aux]{\str_encode_native_loop:w} -% \begin{macro}[aux]{\str_encode_native_flush:} -% \begin{macro}[aux, rEXP]{\str_encode_native_filter:N} +% \begin{macro}[aux]{\@@_encode_native_loop:w} +% \begin{macro}[aux]{\@@_encode_native_flush:} +% \begin{macro}[aux, rEXP]{\@@_encode_native_filter:N} % In Unicode-aware engines, since building particular characters % cannot be done expandably in \TeX{}, we cannot hope to get a % linear-time function. However, we get quite close using the @@ -1968,28 +2023,28 @@ % linear time. Use the standard lowercase trick to produce an % arbitrary character from the null character, and add that character % to the end of the token list being built. At the end of the loop, -% put the token list together with \cs{tl_build_end:}. Note that we +% put the token list together with \cs{__tl_build_end:}. Note that we % use an \texttt{x}-expanding assignment because it is slightly % faster. Unicode-aware engines will never incur an overflow because % the internal string is guaranteed to only contain code points in % $[0,\hexnum{10FFFF}]$. % \begin{macrocode} { - \cs_new_protected_nopar:Npn \str_convert_encode_: + \cs_new_protected_nopar:Npn \@@_convert_encode_: { - \int_zero:N \l_tl_build_offset_int - \tl_gset_build_x:Nw \g_str_result_tl - \exp_after:wN \str_encode_native_loop:w - \g_str_result_tl \s_tl { \q_stop \prg_map_break: } \s_tl - \prg_break_point:n { } - \tl_build_end: + \int_zero:N \l__tl_build_offset_int + \__tl_gbuild_x:Nw \g_@@_result_tl + \exp_after:wN \@@_encode_native_loop:w + \g_@@_result_tl \s__tl { \q_stop \__prg_break: } \s__tl + \__prg_break_point: + \__tl_build_end: } - \cs_new_protected:Npn \str_encode_native_loop:w #1 \s_tl #2 \s_tl + \cs_new_protected:Npn \@@_encode_native_loop:w #1 \s__tl #2 \s__tl { \use_none_delimit_by_q_stop:w #2 \q_stop - \tex_lccode:D \l_str_internal_int \int_eval:w #2 \int_eval_end: - \tl_to_lowercase:n { \tl_build_one:n { ^^@ } } - \str_encode_native_loop:w + \tex_lccode:D \l_@@_internal_int \__int_eval:w #2 \__int_eval_end: + \tl_to_lowercase:n { \__tl_build_one:n { ^^@ } } + \@@_encode_native_loop:w } } % \end{macrocode} @@ -2020,9 +2075,18 @@ % All the 8-bit encodings which \pkg{l3str} supports rely on the same % internal functions. % -% \begin{macro}[int]{\str_declare_eight_bit_encoding:nnn} +% \begin{macro}[int]{\@@_declare_eight_bit_encoding:nnn} +% \begin{syntax} +% \cs{@@_declare_eight_bit_encoding:nnn} \Arg{name} \Arg{mapping} \Arg{missing} +% \end{syntax} +% This declares the encoding \meta{name} to map bytes to Unicode +% characters according to the \meta{mapping}, and map those bytes +% which are not mentionned in the \meta{mapping} either to the +% replacement character (if they appear in \meta{missing}), or to +% themselves. +% % All the 8-bit encoding definition file start with -% \cs{str_declare_eight_bit_encoding:nnn} \Arg{encoding name} +% \cs{@@_declare_eight_bit_encoding:nnn} \Arg{encoding name} % \Arg{mapping} \Arg{missing bytes}. The \meta{mapping} argument is a % token list of pairs \Arg{byte} \Arg{Unicode} expressed in uppercase % hexadecimal notation. The \meta{missing} argument is a token list @@ -2030,70 +2094,70 @@ % \meta{mapping} nor the \meta{missing} lists maps to the same code % point in Unicode. % \begin{macrocode} -\cs_new_protected:Npn \str_declare_eight_bit_encoding:nnn #1#2#3 +\cs_new_protected:Npn \@@_declare_eight_bit_encoding:nnn #1#2#3 { - \tl_set:Nn \l_str_internal_tl {#1} - \cs_new_protected_nopar:cpn { str_convert_decode_#1: } - { \str_convert_decode_eight_bit:n {#1} } - \cs_new_protected_nopar:cpn { str_convert_encode_#1: } - { \str_convert_encode_eight_bit:n {#1} } - \tl_const:cn { c_str_encoding_#1_tl } {#2} - \tl_const:cn { c_str_encoding_#1_missing_tl } {#3} + \tl_set:Nn \l_@@_internal_tl {#1} + \cs_new_protected_nopar:cpn { @@_convert_decode_#1: } + { \@@_convert_decode_eight_bit:n {#1} } + \cs_new_protected_nopar:cpn { @@_convert_encode_#1: } + { \@@_convert_encode_eight_bit:n {#1} } + \tl_const:cn { c_@@_encoding_#1_tl } {#2} + \tl_const:cn { c_@@_encoding_#1_missing_tl } {#3} } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\str_convert_decode_eight_bit:n} -% \begin{macro}[aux]{\str_decode_eight_bit_load:nn} -% \begin{macro}[aux]{\str_decode_eight_bit_load_missing:n} -% \begin{macro}[aux, EXP]{\str_decode_eight_bit_char:N} +% \begin{macro}[int]{\@@_convert_decode_eight_bit:n} +% \begin{macro}[aux]{\@@_decode_eight_bit_load:nn} +% \begin{macro}[aux]{\@@_decode_eight_bit_load_missing:n} +% \begin{macro}[aux, EXP]{\@@_decode_eight_bit_char:N} %^^A todo: document % \begin{macrocode} -\cs_new_protected:Npn \str_convert_decode_eight_bit:n #1 +\cs_new_protected:Npn \@@_convert_decode_eight_bit:n #1 { \group_begin: - \int_zero:N \l_str_internal_int - \exp_last_unbraced:Nx \str_decode_eight_bit_load:nn - { \tl_use:c { c_str_encoding_#1_tl } } - { \q_stop \prg_map_break: } { } - \prg_break_point:n { } - \exp_last_unbraced:Nx \str_decode_eight_bit_load_missing:n - { \tl_use:c { c_str_encoding_#1_missing_tl } } - { \q_stop \prg_map_break: } - \prg_break_point:n { } + \int_zero:N \l_@@_internal_int + \exp_last_unbraced:Nx \@@_decode_eight_bit_load:nn + { \tl_use:c { c_@@_encoding_#1_tl } } + { \q_stop \__prg_break: } { } + \__prg_break_point: + \exp_last_unbraced:Nx \@@_decode_eight_bit_load_missing:n + { \tl_use:c { c_@@_encoding_#1_missing_tl } } + { \q_stop \__prg_break: } + \__prg_break_point: \flag_clear:n { str_error } - \str_convert_gmap:N \str_decode_eight_bit_char:N - \str_if_flag_error:nnx { str_error } { decode-8-bit } {#1} + \@@_convert_gmap:N \@@_decode_eight_bit_char:N + \@@_if_flag_error:nnx { str_error } { decode-8-bit } {#1} \group_end: } -\cs_new_protected:Npn \str_decode_eight_bit_load:nn #1#2 +\cs_new_protected:Npn \@@_decode_eight_bit_load:nn #1#2 { \use_none_delimit_by_q_stop:w #1 \q_stop - \tex_dimen:D "#1 = \l_str_internal_int sp \scan_stop: - \tex_skip:D \l_str_internal_int = "#1 sp \scan_stop: - \tex_toks:D \l_str_internal_int \exp_after:wN { \int_value:w "#2 } - \tex_advance:D \l_str_internal_int \c_one - \str_decode_eight_bit_load:nn + \tex_dimen:D "#1 = \l_@@_internal_int sp \scan_stop: + \tex_skip:D \l_@@_internal_int = "#1 sp \scan_stop: + \tex_toks:D \l_@@_internal_int \exp_after:wN { \__int_value:w "#2 } + \tex_advance:D \l_@@_internal_int \c_one + \@@_decode_eight_bit_load:nn } -\cs_new_protected:Npn \str_decode_eight_bit_load_missing:n #1 +\cs_new_protected:Npn \@@_decode_eight_bit_load_missing:n #1 { \use_none_delimit_by_q_stop:w #1 \q_stop - \tex_dimen:D "#1 = \l_str_internal_int sp \scan_stop: - \tex_skip:D \l_str_internal_int = "#1 sp \scan_stop: - \tex_toks:D \l_str_internal_int \exp_after:wN - { \int_use:N \c_str_replacement_char_int } - \tex_advance:D \l_str_internal_int \c_one - \str_decode_eight_bit_load_missing:n + \tex_dimen:D "#1 = \l_@@_internal_int sp \scan_stop: + \tex_skip:D \l_@@_internal_int = "#1 sp \scan_stop: + \tex_toks:D \l_@@_internal_int \exp_after:wN + { \int_use:N \c_@@_replacement_char_int } + \tex_advance:D \l_@@_internal_int \c_one + \@@_decode_eight_bit_load_missing:n } -\cs_new:Npn \str_decode_eight_bit_char:N #1 +\cs_new:Npn \@@_decode_eight_bit_char:N #1 { - #1 \s_tl - \if_num:w \tex_dimen:D `#1 < \l_str_internal_int - \if_num:w \tex_skip:D \tex_dimen:D `#1 = `#1 \exp_stop_f: + #1 \s__tl + \if_int_compare:w \tex_dimen:D `#1 < \l_@@_internal_int + \if_int_compare:w \tex_skip:D \tex_dimen:D `#1 = `#1 \exp_stop_f: \tex_the:D \tex_toks:D \tex_dimen:D \fi: \fi: - \int_value:w `#1 \s_tl + \__int_value:w `#1 \s__tl } % \end{macrocode} % \end{macro} @@ -2101,53 +2165,53 @@ % \end{macro} % \end{macro} % -% \begin{macro}[int]{\str_convert_encode_eight_bit:n} -% \begin{macro}[aux]{\str_encode_eight_bit_load:nn} -% \begin{macro}[aux, rEXP]{\str_encode_eight_bit_char:n} -% \begin{macro}[aux, rEXP]{\str_encode_eight_bit_char_aux:n} +% \begin{macro}[int]{\@@_convert_encode_eight_bit:n} +% \begin{macro}[aux]{\@@_encode_eight_bit_load:nn} +% \begin{macro}[aux, rEXP]{\@@_encode_eight_bit_char:n} +% \begin{macro}[aux, rEXP]{\@@_encode_eight_bit_char_aux:n} %^^A todo: document % \begin{macrocode} -\cs_new_protected:Npn \str_convert_encode_eight_bit:n #1 +\cs_new_protected:Npn \@@_convert_encode_eight_bit:n #1 { \group_begin: - \int_zero:N \l_str_internal_int - \exp_last_unbraced:Nx \str_encode_eight_bit_load:nn - { \tl_use:c { c_str_encoding_#1_tl } } - { \q_stop \prg_map_break: } { } - \prg_break_point:n { } + \int_zero:N \l_@@_internal_int + \exp_last_unbraced:Nx \@@_encode_eight_bit_load:nn + { \tl_use:c { c_@@_encoding_#1_tl } } + { \q_stop \__prg_break: } { } + \__prg_break_point: \flag_clear:n { str_error } - \str_convert_gmap_internal:N \str_encode_eight_bit_char:n - \str_if_flag_error:nnx { str_error } { encode-8-bit } {#1} + \@@_convert_gmap_internal:N \@@_encode_eight_bit_char:n + \@@_if_flag_error:nnx { str_error } { encode-8-bit } {#1} \group_end: } -\cs_new_protected:Npn \str_encode_eight_bit_load:nn #1#2 +\cs_new_protected:Npn \@@_encode_eight_bit_load:nn #1#2 { \use_none_delimit_by_q_stop:w #1 \q_stop - \tex_dimen:D "#2 = \l_str_internal_int sp \scan_stop: - \tex_skip:D \l_str_internal_int = "#2 sp \scan_stop: - \exp_args:NNf \tex_toks:D \l_str_internal_int - { \str_output_byte:n { "#1 } } - \tex_advance:D \l_str_internal_int \c_one - \str_encode_eight_bit_load:nn + \tex_dimen:D "#2 = \l_@@_internal_int sp \scan_stop: + \tex_skip:D \l_@@_internal_int = "#2 sp \scan_stop: + \exp_args:NNf \tex_toks:D \l_@@_internal_int + { \@@_output_byte:n { "#1 } } + \tex_advance:D \l_@@_internal_int \c_one + \@@_encode_eight_bit_load:nn } -\cs_new:Npn \str_encode_eight_bit_char:n #1 +\cs_new:Npn \@@_encode_eight_bit_char:n #1 { - \if_num:w #1 > \c_max_register_int + \if_int_compare:w #1 > \c_max_register_int \flag_raise:n { str_error } \else: - \if_num:w \tex_dimen:D #1 < \l_str_internal_int - \if_num:w \tex_skip:D \tex_dimen:D #1 = #1 \exp_stop_f: + \if_int_compare:w \tex_dimen:D #1 < \l_@@_internal_int + \if_int_compare:w \tex_skip:D \tex_dimen:D #1 = #1 \exp_stop_f: \tex_the:D \tex_toks:D \tex_dimen:D #1 \exp_stop_f: \exp_after:wN \exp_after:wN \exp_after:wN \use_none:nn \fi: \fi: - \str_encode_eight_bit_char_aux:n {#1} + \@@_encode_eight_bit_char_aux:n {#1} \fi: } -\cs_new:Npn \str_encode_eight_bit_char_aux:n #1 +\cs_new:Npn \@@_encode_eight_bit_char_aux:n #1 { - \if_num:w #1 < \c_two_hundred_fifty_six - \str_output_byte:n {#1} + \if_int_compare:w #1 < \c_two_hundred_fifty_six + \@@_output_byte:n {#1} \else: \flag_raise:n { str_error } \fi: @@ -2163,18 +2227,18 @@ % General messages, and messages for the encodings and escapings loaded % by default (\enquote{native}, and \enquote{bytes}). % \begin{macrocode} -\msg_kernel_new:nnn { str } { unknown-esc } +\__msg_kernel_new:nnn { str } { unknown-esc } { Escaping~scheme~'#1'~(filtered:~'#2')~unknown. } -\msg_kernel_new:nnn { str } { unknown-enc } +\__msg_kernel_new:nnn { str } { unknown-enc } { Encoding~scheme~'#1'~(filtered:~'#2')~unknown. } -\msg_kernel_new:nnnn { str } { native-escaping } +\__msg_kernel_new:nnnn { str } { native-escaping } { The~'native'~encoding~scheme~does~not~support~any~escaping. } { Since~native~strings~do~not~consist~in~bytes,~ none~of~the~escaping~methods~make~sense.~ The~specified~escaping,~'#1',~will be ignored. } -\msg_kernel_new:nnn { str } { file-not-found } +\__msg_kernel_new:nnn { str } { file-not-found } { File~'l3str-#1.def'~not~found. } % \end{macrocode} % @@ -2186,7 +2250,7 @@ % \begin{macrocode} \pdftex_if_engine:F { - \msg_kernel_new:nnnn { str } { non-byte } + \__msg_kernel_new:nnnn { str } { non-byte } { String~invalid~in~escaping~'#1':~it~may~only~contain~bytes. } { Some~characters~in~the~string~you~asked~to~convert~are~not~ @@ -2204,13 +2268,13 @@ % % Those messages are used when converting to and from 8-bit encodings. % \begin{macrocode} -\msg_kernel_new:nnnn { str } { decode-8-bit } +\__msg_kernel_new:nnnn { str } { decode-8-bit } { Invalid~string~in~encoding~'#1'. } { LaTeX~came~across~a~byte~which~is~not~defined~to~represent~ any~character~in~the~encoding~'#1'. } -\msg_kernel_new:nnnn { str } { encode-8-bit } +\__msg_kernel_new:nnnn { str } { encode-8-bit } { Unicode~string~cannot~be~converted~to~encoding~'#1'. } { The~encoding~'#1'~only~contains~a~subset~of~all~Unicode~characters.~ @@ -2220,14 +2284,14 @@ % \end{macrocode} % % \subsection{Deprecated string functions} -% -% \begin{macro}{\str_length_skip_spaces:N, \str_length_skip_spaces:n} -% The naming scheme is a little bit more consistent with -% \enquote{ignore_spaces} instead of \enquote{skip_spaces}. +% +% Deprecated 2012-05-13 for removal by 2012-08-31. +% +% \begin{macro}{\str_length:N, \str_length:n, \str_length_ignore_spaces:n} % \begin{macrocode} -\cs_set:Npn \str_length_skip_spaces:N - { \exp_args:No \str_length_skip_spaces:n } -\cs_set_eq:NN \str_length_skip_spaces:n \str_length_ignore_spaces:n +\cs_new_eq:NN \str_length:N \str_count:N +\cs_new_eq:NN \str_length:n \str_count:n +\cs_new_eq:NN \str_length_ignore_spaces:n \str_count_ignore_spaces:n % \end{macrocode} % \end{macro} % @@ -2253,9 +2317,9 @@ % % \subsubsection{Unescape methods} % -% \begin{macro}[int]{\str_convert_unescape_hex:} -% \begin{macro}[aux, rEXP]{\str_unescape_hex_i:N} -% \begin{macro}[aux, rEXP]{\str_unescape_hex_ii:N} +% \begin{macro}[int]{\@@_convert_unescape_hex:} +% \begin{macro}[aux, rEXP]{\@@_unescape_hex_i:N} +% \begin{macro}[aux, rEXP]{\@@_unescape_hex_ii:N} % Take chars two by two, and interpret each pair as the hexadecimal % code for a byte. Anything else than hexadecimal digits is ignored, % raising the flag. A string which contains an odd number of @@ -2263,46 +2327,47 @@ % appending a |0| in all cases, and dropping it if it is alone. % \begin{macrocode} %<*hex> -\cs_new_protected_nopar:Npn \str_convert_unescape_hex: +\cs_new_protected_nopar:Npn \@@_convert_unescape_hex: { \group_begin: \flag_clear:n { str_error } \int_set:Nn \tex_escapechar:D { 92 } - \tl_gset:Nx \g_str_result_tl + \tl_gset:Nx \g_@@_result_tl { - \str_output_byte:w " - \exp_last_unbraced:Nf \str_unescape_hex_i:N - { \tl_to_str:N \g_str_result_tl } - 0 { ? 0 - \c_one \prg_map_break: } - \prg_break_point:n { \str_output_end: } + \@@_output_byte:w " + \exp_last_unbraced:Nf \@@_unescape_hex_i:N + { \tl_to_str:N \g_@@_result_tl } + 0 { ? 0 - \c_one \__prg_break: } + \__prg_break_point: + \@@_output_end: } - \str_if_flag_error:nnx { str_error } { unescape-hex } { } + \@@_if_flag_error:nnx { str_error } { unescape-hex } { } \group_end: } -\cs_new:Npn \str_unescape_hex_i:N #1 +\cs_new:Npn \@@_unescape_hex_i:N #1 { \use_none:n #1 - \str_aux_hexadecimal_use:NTF #1 - { \str_unescape_hex_ii:N } + \@@_hexadecimal_use:NTF #1 + { \@@_unescape_hex_ii:N } { \flag_raise:n { str_error } - \str_unescape_hex_i:N + \@@_unescape_hex_i:N } } -\cs_new:Npn \str_unescape_hex_ii:N #1 +\cs_new:Npn \@@_unescape_hex_ii:N #1 { \use_none:n #1 - \str_aux_hexadecimal_use:NTF #1 + \@@_hexadecimal_use:NTF #1 { - \str_output_end: - \str_output_byte:w " \str_unescape_hex_i:N + \@@_output_end: + \@@_output_byte:w " \@@_unescape_hex_i:N } { \flag_raise:n { str_error } - \str_unescape_hex_ii:N + \@@_unescape_hex_ii:N } } -\msg_kernel_new:nnnn { str } { unescape-hex } +\__msg_kernel_new:nnnn { str } { unescape-hex } { String~invalid~in~escaping~'hex':~only~hexadecimal~digits~allowed. } { Some~characters~in~the~string~you~asked~to~convert~are~not~ @@ -2314,58 +2379,58 @@ % \end{macro} % \end{macro} % -% \begin{macro}[int]{\str_convert_unescape_name:} -% \begin{macro}[aux, rEXP]{\str_unescape_name_loop:wNN} -% \begin{macro}[int]{\str_convert_unescape_url:} -% \begin{macro}[aux, rEXP]{\str_unescape_url_loop:wNN} -% The \cs{str_convert_unescape_name:} function replaces each +% \begin{macro}[int]{\@@_convert_unescape_name:} +% \begin{macro}[aux, rEXP]{\@@_unescape_name_loop:wNN} +% \begin{macro}[int]{\@@_convert_unescape_url:} +% \begin{macro}[aux, rEXP]{\@@_unescape_url_loop:wNN} +% The \cs{@@_convert_unescape_name:} function replaces each % occurrence of |#| followed by two hexadecimal digits in -% \cs{g_str_result_tl} by the corresponding byte. The \texttt{url} +% \cs{g_@@_result_tl} by the corresponding byte. The \texttt{url} % function is identical, with escape character |%| instead of |#|. -% Thus we define the two together. The arguments of \cs{str_tmp:w} are +% Thus we define the two together. The arguments of \cs{@@_tmp:w} are % the character code of |#| or |%| in hexadecimal, the name of the % main function to define, and the name of the auxiliary which % performs the loop. % % The looping auxiliary |#3| finds the next escape character, reads % the following two characters, and tests them. The test -% \cs{str_aux_hexadecimal_use:NTF} leaves the upper-case digit in the +% \cs{@@_hexadecimal_use:NTF} leaves the upper-case digit in the % input stream, hence we surround the test with -% \cs{str_output_byte:w}~|"| and \cs{str_output_end:}. If both +% \cs{@@_output_byte:w}~|"| and \cs{@@_output_end:}. If both % characters are hexadecimal digits, they should be removed before % looping: this is done by \cs{use_i:nnn}. If one of the characters % is not a hexadecimal digit, then feed |"#1| to -% \cs{str_output_byte:w} to produce the escape character, raise the +% \cs{@@_output_byte:w} to produce the escape character, raise the % flag, and call the looping function followed by the two characters % (remove \cs{use_i:nnn}). % \begin{macrocode} %<*name|url> -\cs_set_protected:Npn \str_tmp:w #1#2#3 +\cs_set_protected:Npn \@@_tmp:w #1#2#3 { - \cs_new_protected:cpn { str_convert_unescape_#2: } + \cs_new_protected:cpn { @@_convert_unescape_#2: } { \group_begin: \flag_clear:n { str_byte } \flag_clear:n { str_error } \int_set:Nn \tex_escapechar:D { 92 } - \tl_gset:Nx \g_str_result_tl + \tl_gset:Nx \g_@@_result_tl { - \exp_after:wN #3 \g_str_result_tl - #1 ? { ? \prg_map_break: } - \prg_break_point:n { } + \exp_after:wN #3 \g_@@_result_tl + #1 ? { ? \__prg_break: } + \__prg_break_point: } - \str_if_flag_error:nnx { str_byte } { non-byte } { #2 } - \str_if_flag_error:nnx { str_error } { unescape-#2 } { } + \@@_if_flag_error:nnx { str_byte } { non-byte } { #2 } + \@@_if_flag_error:nnx { str_error } { unescape-#2 } { } \group_end: } \cs_new:Npn #3 ##1#1##2##3 { - \str_filter_bytes:n {##1} + \@@_filter_bytes:n {##1} \use_none:n ##3 - \str_output_byte:w " - \str_aux_hexadecimal_use:NTF ##2 + \@@_output_byte:w " + \@@_hexadecimal_use:NTF ##2 { - \str_aux_hexadecimal_use:NTF ##3 + \@@_hexadecimal_use:NTF ##3 { } { \flag_raise:n { str_error } @@ -2376,10 +2441,10 @@ \flag_raise:n { str_error } 0 + `#1 \use_i:nn } - \str_output_end: + \@@_output_end: \use_i:nnn #3 ##2##3 } - \msg_kernel_new:nnnn { str } { unescape-#2 } + \__msg_kernel_new:nnnn { str } { unescape-#2 } { String~invalid~in~escaping~'#2'. } { LaTeX~came~across~the~escape~character~'#1'~not~followed~by~ @@ -2387,20 +2452,24 @@ } } % -%\exp_after:wN \str_tmp:w \c_hash_str { name } -% \str_unescape_name_loop:wNN -%\exp_after:wN \str_tmp:w \c_percent_str { url } -% \str_unescape_url_loop:wNN +%<*name> +\exp_after:wN \@@_tmp:w \c_hash_str { name } + \@@_unescape_name_loop:wNN +% +%<*url> +\exp_after:wN \@@_tmp:w \c_percent_str { url } + \@@_unescape_url_loop:wNN +% % \end{macrocode} % \end{macro} % \end{macro} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\str_convert_unescape_string:} -% \begin{macro}[aux, rEXP]{\str_unescape_string_newlines:wN} -% \begin{macro}[aux, rEXP]{\str_unescape_string_loop:wNNN} -% \begin{macro}[aux, rEXP]{\str_unescape_string_repeat:NNNNNN} +% \begin{macro}[int]{\@@_convert_unescape_string:} +% \begin{macro}[aux, rEXP]{\@@_unescape_string_newlines:wN} +% \begin{macro}[aux, rEXP]{\@@_unescape_string_loop:wNNN} +% \begin{macro}[aux, rEXP]{\@@_unescape_string_repeat:NNNNNN} % The \texttt{string} escaping is somewhat similar to the % \texttt{name} and \texttt{url} escapings, with escape character |\|. % The first step is to convert all three line endings, |^^J|, |^^M|, @@ -2432,51 +2501,51 @@ \char_set_catcode_other:N \^^M \tl_to_lowercase:n { - \cs_new_protected_nopar:Npn \str_convert_unescape_string: + \cs_new_protected_nopar:Npn \@@_convert_unescape_string: { \group_begin: \flag_clear:n { str_byte } \flag_clear:n { str_error } \int_set:Nn \tex_escapechar:D { 92 } - \tl_gset:Nx \g_str_result_tl + \tl_gset:Nx \g_@@_result_tl { - \exp_after:wN \str_unescape_string_newlines:wN - \g_str_result_tl \prg_map_break: ^^M ? - \prg_break_point:n { } + \exp_after:wN \@@_unescape_string_newlines:wN + \g_@@_result_tl \__prg_break: ^^M ? + \__prg_break_point: } - \tl_gset:Nx \g_str_result_tl + \tl_gset:Nx \g_@@_result_tl { - \exp_after:wN \str_unescape_string_loop:wNNN - \g_str_result_tl * ?? { ? \prg_map_break: } - \prg_break_point:n { } + \exp_after:wN \@@_unescape_string_loop:wNNN + \g_@@_result_tl * ?? { ? \__prg_break: } + \__prg_break_point: } - \str_if_flag_error:nnx { str_byte } { non-byte } { string } - \str_if_flag_error:nnx { str_error } { unescape-string } { } + \@@_if_flag_error:nnx { str_byte } { non-byte } { string } + \@@_if_flag_error:nnx { str_error } { unescape-string } { } \group_end: } - \cs_new:Npn \str_unescape_string_loop:wNNN #1 *#2#3#4 + \cs_new:Npn \@@_unescape_string_loop:wNNN #1 *#2#3#4 } { - \str_filter_bytes:n {#1} + \@@_filter_bytes:n {#1} \use_none:n #4 - \str_output_byte:w ' - \str_aux_octal_use:NTF #2 + \@@_output_byte:w ' + \@@_octal_use:NTF #2 { - \str_aux_octal_use:NTF #3 + \@@_octal_use:NTF #3 { - \str_aux_octal_use:NTF #4 + \@@_octal_use:NTF #4 { \if_int_compare:w #2 > \c_three - 256 \fi: - \str_unescape_string_repeat:NNNNNN + \@@_unescape_string_repeat:NNNNNN } - { \str_unescape_string_repeat:NNNNNN ? } + { \@@_unescape_string_repeat:NNNNNN ? } } - { \str_unescape_string_repeat:NNNNNN ?? } + { \@@_unescape_string_repeat:NNNNNN ?? } } { - \prg_case_str:xxn {#2} + \str_case_x:nnn {#2} { { \c_backslash_str } { 134 } { ( } { 50 } @@ -2493,18 +2562,18 @@ 0 - \c_one \use_i:nn } } - \str_output_end: - \use_i:nn \str_unescape_string_loop:wNNN #2#3#4 + \@@_output_end: + \use_i:nn \@@_unescape_string_loop:wNNN #2#3#4 } - \cs_new:Npn \str_unescape_string_repeat:NNNNNN #1#2#3#4#5#6 - { \str_output_end: \str_unescape_string_loop:wNNN } - \cs_new:Npn \str_unescape_string_newlines:wN #1 ^^M #2 + \cs_new:Npn \@@_unescape_string_repeat:NNNNNN #1#2#3#4#5#6 + { \@@_output_end: \@@_unescape_string_loop:wNNN } + \cs_new:Npn \@@_unescape_string_newlines:wN #1 ^^M #2 { #1 \if_charcode:w ^^J #2 \else: ^^J \fi: - \str_unescape_string_newlines:wN #2 + \@@_unescape_string_newlines:wN #2 } - \msg_kernel_new:nnnn { str } { unescape-string } + \__msg_kernel_new:nnnn { str } { unescape-string } { String~invalid~in~escaping~'string'. } { LaTeX~came~across~an~escape~character~'\c_backslash_str'~ @@ -2525,52 +2594,52 @@ % Currently, none of the escape methods can lead to errors, assuming % that their input is made out of bytes. % -% \begin{macro}[int]{\str_convert_escape_hex:} -% \begin{macro}[aux, rEXP]{\str_escape_hex_char:N} +% \begin{macro}[int]{\@@_convert_escape_hex:} +% \begin{macro}[aux, rEXP]{\@@_escape_hex_char:N} % Loop and convert each byte to hexadecimal. % \begin{macrocode} %<*hex> -\cs_new_protected_nopar:Npn \str_convert_escape_hex: - { \str_convert_gmap:N \str_escape_hex_char:N } -\cs_new:Npn \str_escape_hex_char:N #1 - { \str_output_hexadecimal:n { `#1 } } +\cs_new_protected_nopar:Npn \@@_convert_escape_hex: + { \@@_convert_gmap:N \@@_escape_hex_char:N } +\cs_new:Npn \@@_escape_hex_char:N #1 + { \@@_output_hexadecimal:n { `#1 } } % % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\str_convert_escape_name:} -% \begin{macro}[aux, rEXP]{\str_escape_name_char:N} -% \begin{macro}[aux, rEXP]{\str_if_escape_name:NTF} -% \begin{variable}{\c_str_escape_name_str} -% \begin{variable}{\c_str_escape_name_not_str} +% \begin{macro}[int]{\@@_convert_escape_name:} +% \begin{macro}[aux, rEXP]{\@@_escape_name_char:N} +% \begin{macro}[aux, rEXP]{\@@_if_escape_name:NTF} +% \begin{variable}{\c_@@_escape_name_str} +% \begin{variable}{\c_@@_escape_name_not_str} % For each byte, test whether it should be output as is, or be % \enquote{hash-encoded}. Roughly, bytes outside the range % $[\hexnum{2A},\hexnum{7E}]$ are hash-encoded. We keep two lists of -% exceptions: characters in \cs{c_str_escape_name_not_str} are not -% hash-encoded, and characters in the \cs{c_str_escape_name_str} are +% exceptions: characters in \cs{c_@@_escape_name_not_str} are not +% hash-encoded, and characters in the \cs{c_@@_escape_name_str} are % encoded. % \begin{macrocode} %<*name> -\str_const:Nn \c_str_escape_name_not_str { ! " $ & ' } %$ -\str_const:Nn \c_str_escape_name_str { {}/<>[] } -\cs_new_protected_nopar:Npn \str_convert_escape_name: - { \str_convert_gmap:N \str_escape_name_char:N } -\cs_new:Npn \str_escape_name_char:N #1 +\str_const:Nn \c_@@_escape_name_not_str { ! " $ & ' } %$ +\str_const:Nn \c_@@_escape_name_str { {}/<>[] } +\cs_new_protected_nopar:Npn \@@_convert_escape_name: + { \@@_convert_gmap:N \@@_escape_name_char:N } +\cs_new:Npn \@@_escape_name_char:N #1 { - \str_if_escape_name:NTF #1 {#1} - { \c_hash_str \str_output_hexadecimal:n {`#1} } + \@@_if_escape_name:NTF #1 {#1} + { \c_hash_str \@@_output_hexadecimal:n {`#1} } } -\prg_new_conditional:Npnn \str_if_escape_name:N #1 { TF } +\prg_new_conditional:Npnn \@@_if_escape_name:N #1 { TF } { - \if_num:w `#1 < "2A \exp_stop_f: - \str_if_contains_char:NNTF \c_str_escape_name_not_str #1 + \if_int_compare:w `#1 < "2A \exp_stop_f: + \@@_if_contains_char:NNTF \c_@@_escape_name_not_str #1 \prg_return_true: \prg_return_false: \else: - \if_num:w `#1 > "7E \exp_stop_f: + \if_int_compare:w `#1 > "7E \exp_stop_f: \prg_return_false: \else: - \str_if_contains_char:NNTF \c_str_escape_name_str #1 + \@@_if_contains_char:NNTF \c_@@_escape_name_str #1 \prg_return_false: \prg_return_true: \fi: \fi: @@ -2583,25 +2652,25 @@ % \end{macro} % \end{macro} % -% \begin{macro}[int]{\str_convert_escape_string:} -% \begin{macro}[aux, rEXP]{\str_escape_string_char:N} -% \begin{macro}[aux, rEXP]{\str_if_escape_string:NTF} -% \begin{variable}{\c_str_escape_string_str} +% \begin{macro}[int]{\@@_convert_escape_string:} +% \begin{macro}[aux, rEXP]{\@@_escape_string_char:N} +% \begin{macro}[aux, rEXP]{\@@_if_escape_string:NTF} +% \begin{variable}{\c_@@_escape_string_str} % Any character below (and including) space, and any character above % (and including) \texttt{del}, are converted to octal. One backslash % is added before each parenthesis and backslash. % \begin{macrocode} %<*string> -\str_const:Nx \c_str_escape_string_str +\str_const:Nx \c_@@_escape_string_str { \c_backslash_str ( ) } -\cs_new_protected_nopar:Npn \str_convert_escape_string: - { \str_convert_gmap:N \str_escape_string_char:N } -\cs_new:Npn \str_escape_string_char:N #1 +\cs_new_protected_nopar:Npn \@@_convert_escape_string: + { \@@_convert_gmap:N \@@_escape_string_char:N } +\cs_new:Npn \@@_escape_string_char:N #1 { - \str_if_escape_string:NTF #1 + \@@_if_escape_string:NTF #1 { - \str_if_contains_char:NNT - \c_str_escape_string_str #1 + \@@_if_contains_char:NNT + \c_@@_escape_string_str #1 { \c_backslash_str } #1 } @@ -2612,12 +2681,12 @@ \int_mod:nn {`#1} \c_eight } } -\prg_new_conditional:Npnn \str_if_escape_string:N #1 { TF } +\prg_new_conditional:Npnn \@@_if_escape_string:N #1 { TF } { - \if_num:w `#1 < "21 \exp_stop_f: + \if_int_compare:w `#1 < "21 \exp_stop_f: \prg_return_false: \else: - \if_num:w `#1 > "7E \exp_stop_f: + \if_int_compare:w `#1 > "7E \exp_stop_f: \prg_return_false: \else: \prg_return_true: @@ -2631,30 +2700,30 @@ % \end{macro} % \end{macro} % -% \begin{macro}[int]{\str_convert_escape_url:} -% \begin{macro}[aux, rEXP]{\str_escape_url_char:N} -% \begin{macro}[aux, rEXP]{\str_if_escape_url:NTF} -% This function is similar to \cs{str_convert_escape_name:}, escaping +% \begin{macro}[int]{\@@_convert_escape_url:} +% \begin{macro}[aux, rEXP]{\@@_escape_url_char:N} +% \begin{macro}[aux, rEXP]{\@@_if_escape_url:NTF} +% This function is similar to \cs{@@_convert_escape_name:}, escaping % different characters. % \begin{macrocode} %<*url> -\cs_new_protected_nopar:Npn \str_convert_escape_url: - { \str_convert_gmap:N \str_escape_url_char:N } -\cs_new:Npn \str_escape_url_char:N #1 +\cs_new_protected_nopar:Npn \@@_convert_escape_url: + { \@@_convert_gmap:N \@@_escape_url_char:N } +\cs_new:Npn \@@_escape_url_char:N #1 { - \str_if_escape_url:NTF #1 {#1} - { \c_percent_str \str_output_hexadecimal:n { `#1 } } + \@@_if_escape_url:NTF #1 {#1} + { \c_percent_str \@@_output_hexadecimal:n { `#1 } } } -\prg_new_conditional:Npnn \str_if_escape_url:N #1 { TF } +\prg_new_conditional:Npnn \@@_if_escape_url:N #1 { TF } { - \if_num:w `#1 < "41 \exp_stop_f: - \str_if_contains_char:nNTF { "-.<> } #1 + \if_int_compare:w `#1 < "41 \exp_stop_f: + \@@_if_contains_char:nNTF { "-.<> } #1 \prg_return_true: \prg_return_false: \else: - \if_num:w `#1 > "7E \exp_stop_f: + \if_int_compare:w `#1 > "7E \exp_stop_f: \prg_return_false: \else: - \str_if_contains_char:nNTF { [ ] } #1 + \@@_if_contains_char:nNTF { [ ] } #1 \prg_return_false: \prg_return_true: \fi: \fi: @@ -2683,9 +2752,9 @@ %<*utf8> % \end{macrocode} % -% \begin{macro}[int]{\str_convert_encode_utf8:} -% \begin{macro}[aux, rEXP]{\str_encode_utf_viii_char:n} -% \begin{macro}[aux, rEXP]{\str_encode_utf_viii_loop:wwnnw} +% \begin{macro}[int]{\@@_convert_encode_utf8:} +% \begin{macro}[aux, rEXP]{\@@_encode_utf_viii_char:n} +% \begin{macro}[aux, rEXP]{\@@_encode_utf_viii_loop:wwnnw} % Loop through the internal string, and convert each character to its % \textsc{utf-8} representation. The representation is built from the % right-most (least significant) byte to the left-most (most @@ -2706,7 +2775,7 @@ % repeat, producing a last continuation byte, and offset the quotient % by $240$ for the leading byte. % -% How is that implemented? \cs{str_encode_utf_vii_loop:wwnnw} takes +% How is that implemented? \cs{@@_encode_utf_vii_loop:wwnnw} takes % successive quotients as its first argument, the quotient from the % previous step as its second argument (except in step~$1$), the bound % for quotients that trigger one more step or not, and finally the @@ -2722,27 +2791,27 @@ % construction |\c_minus_one + \c_zero *| removes the spurious initial % continuation byte (better methods welcome). % \begin{macrocode} -\cs_new_protected_nopar:cpn { str_convert_encode_utf8: } - { \str_convert_gmap_internal:N \str_encode_utf_viii_char:n } -\cs_new:Npn \str_encode_utf_viii_char:n #1 +\cs_new_protected_nopar:cpn { @@_convert_encode_utf8: } + { \@@_convert_gmap_internal:N \@@_encode_utf_viii_char:n } +\cs_new:Npn \@@_encode_utf_viii_char:n #1 { - \str_encode_utf_viii_loop:wwnnw #1 ; \c_minus_one + \c_zero * ; + \@@_encode_utf_viii_loop:wwnnw #1 ; \c_minus_one + \c_zero * ; { 128 } { \c_zero } { 32 } { 192 } { 16 } { 224 } { 8 } { 240 } \q_stop } -\cs_new:Npn \str_encode_utf_viii_loop:wwnnw #1; #2; #3#4 #5 \q_stop +\cs_new:Npn \@@_encode_utf_viii_loop:wwnnw #1; #2; #3#4 #5 \q_stop { - \if_num:w #1 < #3 \exp_stop_f: - \str_output_byte:n { #1 + #4 } + \if_int_compare:w #1 < #3 \exp_stop_f: + \@@_output_byte:n { #1 + #4 } \exp_after:wN \use_none_delimit_by_q_stop:w \fi: - \exp_after:wN \str_encode_utf_viii_loop:wwnnw - \int_value:w \int_div_truncate:nn {#1} {64} ; #1 ; + \exp_after:wN \@@_encode_utf_viii_loop:wwnnw + \__int_value:w \int_div_truncate:nn {#1} {64} ; #1 ; #5 \q_stop - \str_output_byte:n { #2 - 64 * ( #1 - \c_two ) } + \@@_output_byte:n { #2 - 64 * ( #1 - \c_two ) } } % \end{macrocode} % \end{macro} @@ -2751,10 +2820,10 @@ % % \begin{variable} % { -% \l_str_missing_flag , -% \l_str_extra_flag , -% \l_str_overlong_flag , -% \l_str_overflow_flag , +% \l_@@_missing_flag , +% \l_@@_extra_flag , +% \l_@@_overlong_flag , +% \l_@@_overflow_flag , % } % When decoding a string that is purportedly in the \textsc{utf-8} % encoding, four different errors can occur, signalled by a specific @@ -2783,13 +2852,13 @@ \flag_clear_new:n { str_extra } \flag_clear_new:n { str_overlong } \flag_clear_new:n { str_overflow } -\msg_kernel_new:nnnn { str } { utf8-decode } +\__msg_kernel_new:nnnn { str } { utf8-decode } { Invalid~UTF-8~string: \exp_last_unbraced:Nf \use_none:n - \str_if_flag_times:nT { str_missing } { ,~missing~continuation~byte } - \str_if_flag_times:nT { str_extra } { ,~extra~continuation~byte } - \str_if_flag_times:nT { str_overlong } { ,~overlong~form } - \str_if_flag_times:nT { str_overflow } { ,~code~point~too~large } + \@@_if_flag_times:nT { str_missing } { ,~missing~continuation~byte } + \@@_if_flag_times:nT { str_extra } { ,~extra~continuation~byte } + \@@_if_flag_times:nT { str_overlong } { ,~overlong~form } + \@@_if_flag_times:nT { str_overflow } { ,~code~point~too~large } . } { @@ -2830,15 +2899,15 @@ % \end{macrocode} % \end{variable} % -% \begin{macro}[int]{\str_convert_decode_utf8:} +% \begin{macro}[int]{\@@_convert_decode_utf8:} % \begin{macro}[aux, rEXP] % { -% \str_decode_utf_viii_start:N, -% \str_decode_utf_viii_continuation:wwN, -% \str_decode_utf_viii_aux:wNnnwN +% \@@_decode_utf_viii_start:N, +% \@@_decode_utf_viii_continuation:wwN, +% \@@_decode_utf_viii_aux:wNnnwN % } % \begin{macro}[aux, rEXP] -% {\str_decode_utf_viii_overflow:w, \str_decode_utf_viii_end:} +% {\@@_decode_utf_viii_overflow:w, \@@_decode_utf_viii_end:} % Decoding is significantly harder than encoding. As before, lower % some flags, which are tested at the end (in bulk, to trigger at most % one \LaTeX3 error, as explained above). We expect successive @@ -2889,105 +2958,107 @@ % need to check whether we reached the end of the string. In a correct % \textsc{utf-8} string, this happens automatically when the % \texttt{_start} auxiliary leaves its first argument in the input -% stream: the end-marker begins with \cs{prg_map_break:}, which ends +% stream: the end-marker begins with \cs{__prg_break:}, which ends % the loop. On the other hand, if the end is reached when looking for % a continuation byte, the \cs{use_none:n} |#3| construction removes % the first token from the end-marker, and leaves the \texttt{_end} % auxiliary, which raises the appropriate error flag before ending the % mapping. % \begin{macrocode} -\cs_new_protected_nopar:cpn { str_convert_decode_utf8: } +\cs_new_protected_nopar:cpn { @@_convert_decode_utf8: } { \flag_clear:n { str_error } \flag_clear:n { str_missing } \flag_clear:n { str_extra } \flag_clear:n { str_overlong } \flag_clear:n { str_overflow } - \tl_gset:Nx \g_str_result_tl + \tl_gset:Nx \g_@@_result_tl { - \exp_after:wN \str_decode_utf_viii_start:N \g_str_result_tl - { \prg_map_break: \str_decode_utf_viii_end: } - \prg_break_point:n { } + \exp_after:wN \@@_decode_utf_viii_start:N \g_@@_result_tl + { \__prg_break: \@@_decode_utf_viii_end: } + \__prg_break_point: } - \str_if_flag_error:nnx { str_error } { utf8-decode } { } + \@@_if_flag_error:nnx { str_error } { utf8-decode } { } } -\cs_new:Npn \str_decode_utf_viii_start:N #1 +\cs_new:Npn \@@_decode_utf_viii_start:N #1 { #1 - \if_num:w `#1 < "C0 \exp_stop_f: - \s_tl - \if_num:w `#1 < "80 \exp_stop_f: - \int_value:w `#1 + \if_int_compare:w `#1 < "C0 \exp_stop_f: + \s__tl + \if_int_compare:w `#1 < "80 \exp_stop_f: + \__int_value:w `#1 \else: \flag_raise:n { str_extra } \flag_raise:n { str_error } - \int_use:N \c_str_replacement_char_int + \int_use:N \c_@@_replacement_char_int \fi: \else: - \exp_after:wN \str_decode_utf_viii_continuation:wwN - \int_use:N \int_eval:w `#1 - "C0 \exp_after:wN \int_eval_end: + \exp_after:wN \@@_decode_utf_viii_continuation:wwN + \int_use:N \__int_eval:w `#1 - "C0 \exp_after:wN \__int_eval_end: \fi: - \s_tl + \s__tl \use_none_delimit_by_q_stop:w {"80} {"800} {"10000} {"110000} \q_stop - \str_decode_utf_viii_start:N + \@@_decode_utf_viii_start:N } -\cs_new:Npn \str_decode_utf_viii_continuation:wwN - #1 \s_tl #2 \str_decode_utf_viii_start:N #3 +\cs_new:Npn \@@_decode_utf_viii_continuation:wwN + #1 \s__tl #2 \@@_decode_utf_viii_start:N #3 { \use_none:n #3 - \if_num:w `#3 < \if_num:w `#3 < "80 \exp_stop_f: - \fi: "C0 \exp_stop_f: + \if_int_compare:w `#3 < + \if_int_compare:w `#3 < "80 \exp_stop_f: - \fi: + "C0 \exp_stop_f: #3 - \exp_after:wN \str_decode_utf_viii_aux:wNnnwN - \int_use:N \int_eval:w + \exp_after:wN \@@_decode_utf_viii_aux:wNnnwN + \int_use:N \__int_eval:w #1 * "40 + `#3 - "80 - \exp_after:wN \int_eval_end: + \exp_after:wN \__int_eval_end: \else: - \s_tl + \s__tl \flag_raise:n { str_missing } \flag_raise:n { str_error } - \int_use:N \c_str_replacement_char_int + \int_use:N \c_@@_replacement_char_int \fi: - \s_tl + \s__tl #2 - \str_decode_utf_viii_start:N #3 + \@@_decode_utf_viii_start:N #3 } -\cs_new:Npn \str_decode_utf_viii_aux:wNnnwN - #1 \s_tl #2#3#4 #5 \str_decode_utf_viii_start:N #6 +\cs_new:Npn \@@_decode_utf_viii_aux:wNnnwN + #1 \s__tl #2#3#4 #5 \@@_decode_utf_viii_start:N #6 { - \if_num:w #1 < #4 \exp_stop_f: - \s_tl - \if_num:w #1 < #3 \exp_stop_f: + \if_int_compare:w #1 < #4 \exp_stop_f: + \s__tl + \if_int_compare:w #1 < #3 \exp_stop_f: \flag_raise:n { str_overlong } \flag_raise:n { str_error } - \int_use:N \c_str_replacement_char_int + \int_use:N \c_@@_replacement_char_int \else: #1 \fi: \else: \if_meaning:w \q_stop #5 - \str_decode_utf_viii_overflow:w #1 + \@@_decode_utf_viii_overflow:w #1 \fi: - \exp_after:wN \str_decode_utf_viii_continuation:wwN - \int_use:N \int_eval:w #1 - #4 \exp_after:wN \int_eval_end: + \exp_after:wN \@@_decode_utf_viii_continuation:wwN + \int_use:N \__int_eval:w #1 - #4 \exp_after:wN \__int_eval_end: \fi: - \s_tl + \s__tl #2 {#4} #5 - \str_decode_utf_viii_start:N + \@@_decode_utf_viii_start:N } -\cs_new:Npn \str_decode_utf_viii_overflow:w #1 \fi: #2 \fi: +\cs_new:Npn \@@_decode_utf_viii_overflow:w #1 \fi: #2 \fi: { \fi: \fi: \flag_raise:n { str_overflow } \flag_raise:n { str_error } - \int_use:N \c_str_replacement_char_int + \int_use:N \c_@@_replacement_char_int } -\cs_new_nopar:Npn \str_decode_utf_viii_end: +\cs_new_nopar:Npn \@@_decode_utf_viii_end: { - \s_tl + \s__tl \flag_raise:n { str_missing } \flag_raise:n { str_error } - \int_use:N \c_str_replacement_char_int \s_tl - \prg_map_break: + \int_use:N \c_@@_replacement_char_int \s__tl + \__prg_break: } % \end{macrocode} % \end{macro} @@ -3011,14 +3082,14 @@ % % \begin{macro}[int] % { -% \str_convert_encode_utf16: , -% \str_convert_encode_utf16be: , -% \str_convert_encode_utf16le: , +% \@@_convert_encode_utf16: , +% \@@_convert_encode_utf16be: , +% \@@_convert_encode_utf16le: , % } % \begin{macro}[aux, rEXP] % { -% \str_encode_utf_xvi_aux:N , -% \str_encode_utf_xvi_char:n , +% \@@_encode_utf_xvi_aux:N , +% \@@_encode_utf_xvi_char:n , % } % When the endianness is not specified, it is big-endian by default, % and we add a byte-order mark. Convert characters one by one in a @@ -3033,45 +3104,45 @@ % surrogates, each two bytes. The magic \hexnum{D7C0} is % $\hexnum{D800}-\hexnum{10000}/\hexnum{400}$. % \end{itemize} -% For the duration of this operation, \cs{str_tmp:w} is defined as a +% For the duration of this operation, \cs{@@_tmp:w} is defined as a % function to convert a number in the range $[0, \hexnum{FFFF}]$ to a % pair of bytes (either big endian or little endian), by feeding the % quotient of the division of |#1| by \hexnum{100}, followed by |#1| -% to \cs{str_encode_utf_xvi_be:nn} or its \texttt{le} analog: those +% to \cs{@@_encode_utf_xvi_be:nn} or its \texttt{le} analog: those % compute the remainder, and output two bytes for the quotient and % remainder. % \begin{macrocode} - \cs_new_protected_nopar:cpn { str_convert_encode_utf16: } + \cs_new_protected_nopar:cpn { @@_convert_encode_utf16: } { - \str_encode_utf_xvi_aux:N \str_output_byte_pair_be:n - \tl_gput_left:Nx \g_str_result_tl { ^^fe ^^ff } + \@@_encode_utf_xvi_aux:N \@@_output_byte_pair_be:n + \tl_gput_left:Nx \g_@@_result_tl { ^^fe ^^ff } } - \cs_new_protected_nopar:cpn { str_convert_encode_utf16be: } - { \str_encode_utf_xvi_aux:N \str_output_byte_pair_be:n } - \cs_new_protected_nopar:cpn { str_convert_encode_utf16le: } - { \str_encode_utf_xvi_aux:N \str_output_byte_pair_le:n } - \cs_new_protected:Npn \str_encode_utf_xvi_aux:N #1 + \cs_new_protected_nopar:cpn { @@_convert_encode_utf16be: } + { \@@_encode_utf_xvi_aux:N \@@_output_byte_pair_be:n } + \cs_new_protected_nopar:cpn { @@_convert_encode_utf16le: } + { \@@_encode_utf_xvi_aux:N \@@_output_byte_pair_le:n } + \cs_new_protected:Npn \@@_encode_utf_xvi_aux:N #1 { \flag_clear:n { str_error } - \cs_set_eq:NN \str_tmp:w #1 - \str_convert_gmap_internal:N \str_encode_utf_xvi_char:n - \str_if_flag_error:nnx { str_error } { utf16-encode } { } + \cs_set_eq:NN \@@_tmp:w #1 + \@@_convert_gmap_internal:N \@@_encode_utf_xvi_char:n + \@@_if_flag_error:nnx { str_error } { utf16-encode } { } } - \cs_new:Npn \str_encode_utf_xvi_char:n #1 + \cs_new:Npn \@@_encode_utf_xvi_char:n #1 { \if_int_compare:w #1 < "D800 \exp_stop_f: - \str_tmp:w {#1} + \@@_tmp:w {#1} \else: \if_int_compare:w #1 < "10000 \exp_stop_f: \if_int_compare:w #1 < "E000 \exp_stop_f: \flag_raise:n { str_error } - \str_tmp:w { \c_str_replacement_char_int } + \@@_tmp:w { \c_@@_replacement_char_int } \else: - \str_tmp:w {#1} + \@@_tmp:w {#1} \fi: \else: - \exp_args:Nf \str_tmp:w { \int_div_truncate:nn {#1} {"400} + "D7C0 } - \exp_args:Nf \str_tmp:w { \int_mod:nn {#1} {"400} + "DC00 } + \exp_args:Nf \@@_tmp:w { \int_div_truncate:nn {#1} {"400} + "D7C0 } + \exp_args:Nf \@@_tmp:w { \int_mod:nn {#1} {"400} + "DC00 } \fi: \fi: } @@ -3081,14 +3152,14 @@ % % \begin{variable} % { -% \l_str_missing_flag , -% \l_str_extra_flag , -% \l_str_end_flag , +% \l_@@_missing_flag , +% \l_@@_extra_flag , +% \l_@@_end_flag , % } % When encoding a Unicode string to \textsc{utf-16}, only one error % can occur: code points in the range $[\hexnum{D800}, % \hexnum{DFFF}]$, corresponding to surrogates, cannot be encoded. We -% use the all-purpose flag \texttt{str_error} to signal that error. +% use the all-purpose flag \texttt{@@_error} to signal that error. % % When decoding a Unicode string which is purportedly in % \textsc{utf-16}, three errors can occur: a missing trail surrogate, @@ -3098,19 +3169,19 @@ \flag_clear_new:n { str_missing } \flag_clear_new:n { str_extra } \flag_clear_new:n { str_end } - \msg_kernel_new:nnnn { str } { utf16-encode } + \__msg_kernel_new:nnnn { str } { utf16-encode } { Unicode~string~cannot~be~expressed~in~UTF-16:~surrogate. } { Surrogate~code~points~(in~the~range~[U+D800,~U+DFFF])~ can~be~expressed~in~the~UTF-8~and~UTF-32~encodings,~ but~not~in~the~UTF-16~encoding. } - \msg_kernel_new:nnnn { str } { utf16-decode } + \__msg_kernel_new:nnnn { str } { utf16-decode } { Invalid~UTF-16~string: \exp_last_unbraced:Nf \use_none:n - \str_if_flag_times:nT { str_missing } { ,~missing~trail~surrogate } - \str_if_flag_times:nT { str_extra } { ,~extra~trail~surrogate } - \str_if_flag_times:nT { str_end } { ,~odd~number~of~bytes } + \@@_if_flag_times:nT { str_missing } { ,~missing~trail~surrogate } + \@@_if_flag_times:nT { str_extra } { ,~extra~trail~surrogate } + \@@_if_flag_times:nT { str_end } { ,~odd~number~of~bytes } . } { @@ -3148,11 +3219,11 @@ % % \begin{macro}[int] % { -% \str_convert_decode_utf16: , -% \str_convert_decode_utf16be: , -% \str_convert_decode_utf16le: , +% \@@_convert_decode_utf16: , +% \@@_convert_decode_utf16be: , +% \@@_convert_decode_utf16le: , % } -% \begin{macro}[aux]{\str_decode_utf_xvi_bom:NN, \str_decode_utf_xvi:Nw} +% \begin{macro}[aux]{\@@_decode_utf_xvi_bom:NN, \@@_decode_utf_xvi:Nw} % As for \textsc{utf-8}, decoding \textsc{utf-16} is harder than % encoding it. If the endianness is unknown, check the first two % bytes: if those are \hexnum{FE} and \hexnum{FF} in either order, @@ -3160,48 +3231,48 @@ % big-endianness. The three endianness cases are based on a common % auxiliary whose first argument is $1$ for big-endian and $2$ for % little-endian, and whose second argument, delimited by the scan mark -% \cs{s_stop}, is expanded once (the string may be long; passing -% \cs{g_str_result_tl} as an argument before expansion is cheaper). +% \cs{s__stop}, is expanded once (the string may be long; passing +% \cs{g_@@_result_tl} as an argument before expansion is cheaper). % -% The \cs{strr_decode_utf_xvi:Nw} function defines \cs{str_tmp:w} to +% The \cs{@@_decode_utf_xvi:Nw} function defines \cs{@@_tmp:w} to % take two arguments and return the character code of the first one if % the string is big-endian, and the second one if the string is % little-endian, then loops over the string using -% \cs{str_decode_utf_xvi_pair:NN} described below. +% \cs{@@_decode_utf_xvi_pair:NN} described below. % \begin{macrocode} - \cs_new_protected_nopar:cpn { str_convert_decode_utf16be: } - { \str_decode_utf_xvi:Nw 1 \g_str_result_tl \s_stop } - \cs_new_protected_nopar:cpn { str_convert_decode_utf16le: } - { \str_decode_utf_xvi:Nw 2 \g_str_result_tl \s_stop } - \cs_new_protected_nopar:cpn { str_convert_decode_utf16: } + \cs_new_protected_nopar:cpn { @@_convert_decode_utf16be: } + { \@@_decode_utf_xvi:Nw 1 \g_@@_result_tl \s__stop } + \cs_new_protected_nopar:cpn { @@_convert_decode_utf16le: } + { \@@_decode_utf_xvi:Nw 2 \g_@@_result_tl \s__stop } + \cs_new_protected_nopar:cpn { @@_convert_decode_utf16: } { - \exp_after:wN \str_decode_utf_xvi_bom:NN - \g_str_result_tl \s_stop \s_stop \s_stop + \exp_after:wN \@@_decode_utf_xvi_bom:NN + \g_@@_result_tl \s__stop \s__stop \s__stop } - \cs_new_protected:Npn \str_decode_utf_xvi_bom:NN #1#2 + \cs_new_protected:Npn \@@_decode_utf_xvi_bom:NN #1#2 { - \str_if_eq:xxTF { #1#2 } { ^^ff ^^fe } - { \str_decode_utf_xvi:Nw 2 } + \str_if_eq_x:nnTF { #1#2 } { ^^ff ^^fe } + { \@@_decode_utf_xvi:Nw 2 } { - \str_if_eq:xxTF { #1#2 } { ^^fe ^^ff } - { \str_decode_utf_xvi:Nw 1 } - { \str_decode_utf_xvi:Nw 1 #1#2 } + \str_if_eq_x:nnTF { #1#2 } { ^^fe ^^ff } + { \@@_decode_utf_xvi:Nw 1 } + { \@@_decode_utf_xvi:Nw 1 #1#2 } } } - \cs_new_protected:Npn \str_decode_utf_xvi:Nw #1#2 \s_stop + \cs_new_protected:Npn \@@_decode_utf_xvi:Nw #1#2 \s__stop { \flag_clear:n { str_error } \flag_clear:n { str_missing } \flag_clear:n { str_extra } \flag_clear:n { str_end } - \cs_set:Npn \str_tmp:w ##1 ##2 { ` ## #1 } - \tl_gset:Nx \g_str_result_tl + \cs_set:Npn \@@_tmp:w ##1 ##2 { ` ## #1 } + \tl_gset:Nx \g_@@_result_tl { - \exp_after:wN \str_decode_utf_xvi_pair:NN + \exp_after:wN \@@_decode_utf_xvi_pair:NN #2 \q_nil \q_nil - \prg_break_point:n { } + \__prg_break_point: } - \str_if_flag_error:nnx { str_error } { utf16-decode } { } + \@@_if_flag_error:nnx { str_error } { utf16-decode } { } } % \end{macrocode} % \end{macro} @@ -3209,16 +3280,16 @@ % % \begin{macro}[aux, rEXP] % { -% \str_decode_utf_xvi_pair:NN , -% \str_decode_utf_xvi_quad:NNwNN , -% \str_decode_utf_xvi_pair_end:Nw , +% \@@_decode_utf_xvi_pair:NN , +% \@@_decode_utf_xvi_quad:NNwNN , +% \@@_decode_utf_xvi_pair_end:Nw , % } % \begin{macro}[aux, rEXP] % { -% \str_decode_utf_xvi_error:nNN , -% \str_decode_utf_xvi_extra:NNw , +% \@@_decode_utf_xvi_error:nNN , +% \@@_decode_utf_xvi_extra:NNw , % } -% Bytes are read two at a time. At this stage, |\str_tmp:w #1#2| +% Bytes are read two at a time. At this stage, |\@@_tmp:w #1#2| % expands to the character code of the most significant byte, and we % distinguish cases depending on which range it lies in: % \begin{itemize} @@ -3239,9 +3310,9 @@ % range $[\hexnum{DC}, \hexnum{DF}]$ (trail surrogate). The test is % similar to the test used for continuation bytes in the % \textsc{utf-8} decoding functions. In the case where |#4#5| is -% indeed a trail surrogate, leave |#1#2#4#5| \cs{s_tl} -% \meta{code~point} \cs{s_tl}, and remove the pair |#4#5| before -% looping with \cs{str_decode_utf_xvi_pair:NN}. Otherwise, of course, +% indeed a trail surrogate, leave |#1#2#4#5| \cs{s__tl} +% \meta{code~point} \cs{s__tl}, and remove the pair |#4#5| before +% looping with \cs{@@_decode_utf_xvi_pair:NN}. Otherwise, of course, % complain about the missing surrogate. % % The magic number \hexnum{D7F7} is such that @@ -3253,63 +3324,63 @@ % string had an even length. Also, if the end is reached when % expecting a trail surrogate, we treat that as a missing surrogate. % \begin{macrocode} - \cs_new:Npn \str_decode_utf_xvi_pair:NN #1#2 + \cs_new:Npn \@@_decode_utf_xvi_pair:NN #1#2 { \if_meaning:w \q_nil #2 - \str_decode_utf_xvi_pair_end:Nw #1 + \@@_decode_utf_xvi_pair_end:Nw #1 \fi: \if_case:w - \int_eval:w ( \str_tmp:w #1#2 - "D6 ) / \c_four \int_eval_end: - \or: \exp_after:wN \str_decode_utf_xvi_quad:NNwNN - \or: \exp_after:wN \str_decode_utf_xvi_extra:NNw + \__int_eval:w ( \@@_tmp:w #1#2 - "D6 ) / \c_four \__int_eval_end: + \or: \exp_after:wN \@@_decode_utf_xvi_quad:NNwNN + \or: \exp_after:wN \@@_decode_utf_xvi_extra:NNw \fi: - #1#2 \s_tl - \int_eval:n { "100 * \str_tmp:w #1#2 + \str_tmp:w #2#1 } \s_tl - \str_decode_utf_xvi_pair:NN + #1#2 \s__tl + \int_eval:n { "100 * \@@_tmp:w #1#2 + \@@_tmp:w #2#1 } \s__tl + \@@_decode_utf_xvi_pair:NN } - \cs_new:Npn \str_decode_utf_xvi_quad:NNwNN - #1#2 #3 \str_decode_utf_xvi_pair:NN #4#5 + \cs_new:Npn \@@_decode_utf_xvi_quad:NNwNN + #1#2 #3 \@@_decode_utf_xvi_pair:NN #4#5 { \if_meaning:w \q_nil #5 - \str_decode_utf_xvi_error:nNN { missing } #1#2 - \str_decode_utf_xvi_pair_end:Nw #4 + \@@_decode_utf_xvi_error:nNN { missing } #1#2 + \@@_decode_utf_xvi_pair_end:Nw #4 \fi: - \if_num:w - \if_num:w \str_tmp:w #4#5 < "DC \exp_stop_f: + \if_int_compare:w + \if_int_compare:w \@@_tmp:w #4#5 < "DC \exp_stop_f: \c_zero = \c_one \else: - \str_tmp:w #4#5 < "E0 \exp_stop_f: + \@@_tmp:w #4#5 < "E0 \exp_stop_f: \fi: - #1 #2 #4 #5 \s_tl + #1 #2 #4 #5 \s__tl \int_eval:n { - ( "100 * \str_tmp:w #1#2 + \str_tmp:w #2#1 - "D7F7 ) * "400 - + "100 * \str_tmp:w #4#5 + \str_tmp:w #5#4 + ( "100 * \@@_tmp:w #1#2 + \@@_tmp:w #2#1 - "D7F7 ) * "400 + + "100 * \@@_tmp:w #4#5 + \@@_tmp:w #5#4 } - \s_tl + \s__tl \exp_after:wN \use_i:nnn \else: - \str_decode_utf_xvi_error:nNN { missing } #1#2 + \@@_decode_utf_xvi_error:nNN { missing } #1#2 \fi: - \str_decode_utf_xvi_pair:NN #4#5 + \@@_decode_utf_xvi_pair:NN #4#5 } - \cs_new:Npn \str_decode_utf_xvi_pair_end:Nw #1 \fi: + \cs_new:Npn \@@_decode_utf_xvi_pair_end:Nw #1 \fi: { \fi: \if_meaning:w \q_nil #1 \else: - \str_decode_utf_xvi_error:nNN { end } #1 \prg_do_nothing: + \@@_decode_utf_xvi_error:nNN { end } #1 \prg_do_nothing: \fi: - \prg_map_break: + \__prg_break: } - \cs_new:Npn \str_decode_utf_xvi_extra:NNw #1#2 \s_tl #3 \s_tl - { \str_decode_utf_xvi_error:nNN { extra } #1#2 } - \cs_new:Npn \str_decode_utf_xvi_error:nNN #1#2#3 + \cs_new:Npn \@@_decode_utf_xvi_extra:NNw #1#2 \s__tl #3 \s__tl + { \@@_decode_utf_xvi_error:nNN { extra } #1#2 } + \cs_new:Npn \@@_decode_utf_xvi_error:nNN #1#2#3 { \flag_raise:n { str_error } \flag_raise:n { str_#1 } - #2 #3 \s_tl - \int_use:N \c_str_replacement_char_int \s_tl + #2 #3 \s__tl + \int_use:N \c_@@_replacement_char_int \s__tl } % \end{macrocode} % \end{macro} @@ -3336,51 +3407,51 @@ % % \begin{macro}[int] % { -% \str_convert_encode_utf32: , -% \str_convert_encode_utf32be: , -% \str_convert_encode_utf32le: , +% \@@_convert_encode_utf32: , +% \@@_convert_encode_utf32be: , +% \@@_convert_encode_utf32le: , % } % \begin{macro}[aux, rEXP] % { -% \str_encode_utf_xxxii_be:n , -% \str_encode_utf_xxxii_be_aux:nn , -% \str_encode_utf_xxxii_le:n , -% \str_encode_utf_xxxii_le_aux:nn , +% \@@_encode_utf_xxxii_be:n , +% \@@_encode_utf_xxxii_be_aux:nn , +% \@@_encode_utf_xxxii_le:n , +% \@@_encode_utf_xxxii_le_aux:nn , % } -% Convert each integer in the comma-list \cs{g_str_result_tl} to a +% Convert each integer in the comma-list \cs{g_@@_result_tl} to a % sequence of four bytes. The functions for big-endian and % little-endian encodings are very similar, but the -% \cs{str_output_byte:n} instructions are reversed. +% \cs{@@_output_byte:n} instructions are reversed. % \begin{macrocode} - \cs_new_protected_nopar:cpn { str_convert_encode_utf32: } + \cs_new_protected_nopar:cpn { @@_convert_encode_utf32: } { - \str_convert_gmap_internal:N \str_encode_utf_xxxii_be:n - \tl_gput_left:Nx \g_str_result_tl { ^^00 ^^00 ^^fe ^^ff } + \@@_convert_gmap_internal:N \@@_encode_utf_xxxii_be:n + \tl_gput_left:Nx \g_@@_result_tl { ^^00 ^^00 ^^fe ^^ff } } - \cs_new_protected_nopar:cpn { str_convert_encode_utf32be: } - { \str_convert_gmap_internal:N \str_encode_utf_xxxii_be:n } - \cs_new_protected_nopar:cpn { str_convert_encode_utf32le: } - { \str_convert_gmap_internal:N \str_encode_utf_xxxii_le:n } - \cs_new:Npn \str_encode_utf_xxxii_be:n #1 + \cs_new_protected_nopar:cpn { @@_convert_encode_utf32be: } + { \@@_convert_gmap_internal:N \@@_encode_utf_xxxii_be:n } + \cs_new_protected_nopar:cpn { @@_convert_encode_utf32le: } + { \@@_convert_gmap_internal:N \@@_encode_utf_xxxii_le:n } + \cs_new:Npn \@@_encode_utf_xxxii_be:n #1 { - \exp_args:Nf \str_encode_utf_xxxii_be_aux:nn + \exp_args:Nf \@@_encode_utf_xxxii_be_aux:nn { \int_div_truncate:nn {#1} { "100 } } {#1} } - \cs_new:Npn \str_encode_utf_xxxii_be_aux:nn #1#2 + \cs_new:Npn \@@_encode_utf_xxxii_be_aux:nn #1#2 { ^^00 - \str_output_byte_pair_be:n {#1} - \str_output_byte:n { #2 - #1 * "100 } + \@@_output_byte_pair_be:n {#1} + \@@_output_byte:n { #2 - #1 * "100 } } - \cs_new:Npn \str_encode_utf_xxxii_le:n #1 + \cs_new:Npn \@@_encode_utf_xxxii_le:n #1 { - \exp_args:Nf \str_encode_utf_xxxii_le_aux:nn + \exp_args:Nf \@@_encode_utf_xxxii_le_aux:nn { \int_div_truncate:nn {#1} { "100 } } {#1} } - \cs_new:Npn \str_encode_utf_xxxii_le_aux:nn #1#2 + \cs_new:Npn \@@_encode_utf_xxxii_le_aux:nn #1#2 { - \str_output_byte:n { #2 - #1 * "100 } - \str_output_byte_pair_le:n {#1} + \@@_output_byte:n { #2 - #1 * "100 } + \@@_output_byte_pair_le:n {#1} ^^00 } % \end{macrocode} @@ -3396,11 +3467,11 @@ % \begin{macrocode} \flag_clear_new:n { str_overflow } \flag_clear_new:n { str_end } - \msg_kernel_new:nnnn { str } { utf32-decode } + \__msg_kernel_new:nnnn { str } { utf32-decode } { Invalid~UTF-32~string: \exp_last_unbraced:Nf \use_none:n - \str_if_flag_times:nT { str_overflow } { ,~code~point~too~large } - \str_if_flag_times:nT { str_end } { ,~truncated~string } + \@@_if_flag_times:nT { str_overflow } { ,~code~point~too~large } + \@@_if_flag_times:nT { str_end } { ,~truncated~string } . } { @@ -3425,103 +3496,103 @@ % % \begin{macro}[int] % { -% \str_convert_decode_utf32: , -% \str_convert_decode_utf32be: , -% \str_convert_decode_utf32le: , +% \@@_convert_decode_utf32: , +% \@@_convert_decode_utf32be: , +% \@@_convert_decode_utf32le: , % } % \begin{macro}[aux] -% {\str_decode_utf_xxxii_bom:NNNN, \str_decode_utf_xxxii:Nw} +% {\@@_decode_utf_xxxii_bom:NNNN, \@@_decode_utf_xxxii:Nw} % \begin{macro}[aux, rEXP] -% {\str_decode_utf_xxxii_loop:NNNN, \str_decode_utf_xxxii_end:w} +% {\@@_decode_utf_xxxii_loop:NNNN, \@@_decode_utf_xxxii_end:w} % % The structure is similar to \textsc{utf-16} decoding functions. If % the endianness is not given, test the first $4$ bytes of the string -% (possibly \cs{s_stop} if the string is too short) for the presence +% (possibly \cs{s__stop} if the string is too short) for the presence % of a byte-order mark. If there is a byte-order mark, use that % endianness, and remove the $4$ bytes, otherwise default to % big-endian, and leave the $4$ bytes in place. The -% \cs{str_decode_utf_xxxii:Nw} auxiliary recieves $1$ or $2$ as its +% \cs{@@_decode_utf_xxxii:Nw} auxiliary recieves $1$ or $2$ as its % first argument indicating endianness, and the string to convert as -% its second argument (expanded or not). It sets \cs{str_tmp:w} to +% its second argument (expanded or not). It sets \cs{@@_tmp:w} to % expand to the character code of either of its two arguments % depending on endianness, then triggers the \texttt{_loop} auxiliary -% inside an \texttt{x}-expanding assignment to \cs{g_str_result_tl}. +% inside an \texttt{x}-expanding assignment to \cs{g_@@_result_tl}. % % The \texttt{_loop} auxiliary first checks for the end-of-string -% marker \cs{s_stop}, calling the \texttt{_end} auxiliary if -% appropriate. Otherwise, leave the \meta{4~bytes} \cs{s_tl} behind, +% marker \cs{s__stop}, calling the \texttt{_end} auxiliary if +% appropriate. Otherwise, leave the \meta{4~bytes} \cs{s__tl} behind, % then check that the code point is not overflowing: the leading byte % must be $0$, and the following byte at most $16$. % % In the ending code, we check that there remains no byte: there -% should be nothing left until the first \cs{s_stop}. Break the map. +% should be nothing left until the first \cs{s__stop}. Break the map. % \begin{macrocode} - \cs_new_protected_nopar:cpn { str_convert_decode_utf32be: } - { \str_decode_utf_xxxii:Nw 1 \g_str_result_tl \s_stop } - \cs_new_protected_nopar:cpn { str_convert_decode_utf32le: } - { \str_decode_utf_xxxii:Nw 2 \g_str_result_tl \s_stop } - \cs_new_protected_nopar:cpn { str_convert_decode_utf32: } + \cs_new_protected_nopar:cpn { @@_convert_decode_utf32be: } + { \@@_decode_utf_xxxii:Nw 1 \g_@@_result_tl \s__stop } + \cs_new_protected_nopar:cpn { @@_convert_decode_utf32le: } + { \@@_decode_utf_xxxii:Nw 2 \g_@@_result_tl \s__stop } + \cs_new_protected_nopar:cpn { @@_convert_decode_utf32: } { - \exp_after:wN \str_decode_utf_xxxii_bom:NNNN \g_str_result_tl - \s_stop \s_stop \s_stop \s_stop \s_stop + \exp_after:wN \@@_decode_utf_xxxii_bom:NNNN \g_@@_result_tl + \s__stop \s__stop \s__stop \s__stop \s__stop } - \cs_new_protected:Npn \str_decode_utf_xxxii_bom:NNNN #1#2#3#4 + \cs_new_protected:Npn \@@_decode_utf_xxxii_bom:NNNN #1#2#3#4 { - \str_if_eq:xxTF { #1#2#3#4 } { ^^ff ^^fe ^^00 ^^00 } - { \str_decode_utf_xxxii:Nw 2 } + \str_if_eq_x:nnTF { #1#2#3#4 } { ^^ff ^^fe ^^00 ^^00 } + { \@@_decode_utf_xxxii:Nw 2 } { - \str_if_eq:xxTF { #1#2#3#4 } { ^^00 ^^00 ^^fe ^^ff } - { \str_decode_utf_xxxii:Nw 1 } - { \str_decode_utf_xxxii:Nw 1 #1#2#3#4 } + \str_if_eq_x:nnTF { #1#2#3#4 } { ^^00 ^^00 ^^fe ^^ff } + { \@@_decode_utf_xxxii:Nw 1 } + { \@@_decode_utf_xxxii:Nw 1 #1#2#3#4 } } } - \cs_new_protected:Npn \str_decode_utf_xxxii:Nw #1#2 \s_stop + \cs_new_protected:Npn \@@_decode_utf_xxxii:Nw #1#2 \s__stop { \flag_clear:n { str_overflow } \flag_clear:n { str_end } \flag_clear:n { str_error } - \cs_set:Npn \str_tmp:w ##1 ##2 { ` ## #1 } - \tl_gset:Nx \g_str_result_tl + \cs_set:Npn \@@_tmp:w ##1 ##2 { ` ## #1 } + \tl_gset:Nx \g_@@_result_tl { - \exp_after:wN \str_decode_utf_xxxii_loop:NNNN - #2 \s_stop \s_stop \s_stop \s_stop - \prg_break_point:n { } + \exp_after:wN \@@_decode_utf_xxxii_loop:NNNN + #2 \s__stop \s__stop \s__stop \s__stop + \__prg_break_point: } - \str_if_flag_error:nnx { str_error } { utf32-decode } { } + \@@_if_flag_error:nnx { str_error } { utf32-decode } { } } - \cs_new:Npn \str_decode_utf_xxxii_loop:NNNN #1#2#3#4 + \cs_new:Npn \@@_decode_utf_xxxii_loop:NNNN #1#2#3#4 { - \if_meaning:w \s_stop #4 - \exp_after:wN \str_decode_utf_xxxii_end:w + \if_meaning:w \s__stop #4 + \exp_after:wN \@@_decode_utf_xxxii_end:w \fi: - #1#2#3#4 \s_tl - \if_num:w \str_tmp:w #1#4 > \c_zero + #1#2#3#4 \s__tl + \if_int_compare:w \@@_tmp:w #1#4 > \c_zero \flag_raise:n { str_overflow } \flag_raise:n { str_error } - \int_use:N \c_str_replacement_char_int + \int_use:N \c_@@_replacement_char_int \else: - \if_num:w \str_tmp:w #2#3 > \c_sixteen + \if_int_compare:w \@@_tmp:w #2#3 > \c_sixteen \flag_raise:n { str_overflow } \flag_raise:n { str_error } - \int_use:N \c_str_replacement_char_int + \int_use:N \c_@@_replacement_char_int \else: \int_eval:n - { \str_tmp:w #2#3*"10000 + \str_tmp:w #3#2*"100 + \str_tmp:w #4#1 } + { \@@_tmp:w #2#3*"10000 + \@@_tmp:w #3#2*"100 + \@@_tmp:w #4#1 } \fi: \fi: - \s_tl - \str_decode_utf_xxxii_loop:NNNN + \s__tl + \@@_decode_utf_xxxii_loop:NNNN } - \cs_new:Npn \str_decode_utf_xxxii_end:w #1 \s_stop + \cs_new:Npn \@@_decode_utf_xxxii_end:w #1 \s__stop { \tl_if_empty:nF {#1} { \flag_raise:n { str_end } \flag_raise:n { str_error } - #1 \s_tl - \int_use:N \c_str_replacement_char_int \s_tl + #1 \s__tl + \int_use:N \c_@@_replacement_char_int \s__tl } - \prg_map_break: + \__prg_break: } % \end{macrocode} % \end{macro} @@ -3541,7 +3612,7 @@ % family, we keep track only of differences, and of unassigned bytes. % \begin{macrocode} %<*iso88591> -\str_declare_eight_bit_encoding:nnn { iso88591 } +\@@_declare_eight_bit_encoding:nnn { iso88591 } { } { @@ -3551,7 +3622,7 @@ % % \begin{macrocode} %<*iso88592> -\str_declare_eight_bit_encoding:nnn { iso88592 } +\@@_declare_eight_bit_encoding:nnn { iso88592 } { { A1 } { 0104 } { A2 } { 02D8 } @@ -3618,7 +3689,7 @@ % % \begin{macrocode} %<*iso88593> -\str_declare_eight_bit_encoding:nnn { iso88593 } +\@@_declare_eight_bit_encoding:nnn { iso88593 } { { A1 } { 0126 } { A2 } { 02D8 } @@ -3663,7 +3734,7 @@ % % \begin{macrocode} %<*iso88594> -\str_declare_eight_bit_encoding:nnn { iso88594 } +\@@_declare_eight_bit_encoding:nnn { iso88594 } { { A1 } { 0104 } { A2 } { 0138 } @@ -3723,7 +3794,7 @@ % % \begin{macrocode} %<*iso88595> -\str_declare_eight_bit_encoding:nnn { iso88595 } +\@@_declare_eight_bit_encoding:nnn { iso88595 } { { A1 } { 0401 } { A2 } { 0402 } @@ -3827,7 +3898,7 @@ % % \begin{macrocode} %<*iso88596> -\str_declare_eight_bit_encoding:nnn { iso88596 } +\@@_declare_eight_bit_encoding:nnn { iso88596 } { { AC } { 060C } { BB } { 061B } @@ -3917,7 +3988,7 @@ % % \begin{macrocode} %<*iso88597> -\str_declare_eight_bit_encoding:nnn { iso88597 } +\@@_declare_eight_bit_encoding:nnn { iso88597 } { { A1 } { 2018 } { A2 } { 2019 } @@ -4006,7 +4077,7 @@ % % \begin{macrocode} %<*iso88598> -\str_declare_eight_bit_encoding:nnn { iso88598 } +\@@_declare_eight_bit_encoding:nnn { iso88598 } { { AA } { 00D7 } { BA } { 00F7 } @@ -4083,7 +4154,7 @@ % % \begin{macrocode} %<*iso88599> -\str_declare_eight_bit_encoding:nnn { iso88599 } +\@@_declare_eight_bit_encoding:nnn { iso88599 } { { D0 } { 011E } { DD } { 0130 } @@ -4099,7 +4170,7 @@ % % \begin{macrocode} %<*iso885910> -\str_declare_eight_bit_encoding:nnn { iso885910 } +\@@_declare_eight_bit_encoding:nnn { iso885910 } { { A1 } { 0104 } { A2 } { 0112 } @@ -4155,7 +4226,7 @@ % % \begin{macrocode} %<*iso885911> -\str_declare_eight_bit_encoding:nnn { iso885911 } +\@@_declare_eight_bit_encoding:nnn { iso885911 } { { A1 } { 0E01 } { A2 } { 0E02 } @@ -4256,7 +4327,7 @@ % % \begin{macrocode} %<*iso885913> -\str_declare_eight_bit_encoding:nnn { iso885913 } +\@@_declare_eight_bit_encoding:nnn { iso885913 } { { A1 } { 201D } { A5 } { 201E } @@ -4322,7 +4393,7 @@ % % \begin{macrocode} %<*iso885914> -\str_declare_eight_bit_encoding:nnn { iso885914 } +\@@_declare_eight_bit_encoding:nnn { iso885914 } { { A1 } { 1E02 } { A2 } { 1E03 } @@ -4363,7 +4434,7 @@ % % \begin{macrocode} %<*iso885915> -\str_declare_eight_bit_encoding:nnn { iso885915 } +\@@_declare_eight_bit_encoding:nnn { iso885915 } { { A4 } { 20AC } { A6 } { 0160 } @@ -4381,7 +4452,7 @@ % % \begin{macrocode} %<*iso885916> -\str_declare_eight_bit_encoding:nnn { iso885916 } +\@@_declare_eight_bit_encoding:nnn { iso885916 } { { A1 } { 0104 } { A2 } { 0105 } diff --git a/Master/texmf-dist/source/latex/l3experimental/l3str/l3str.ins b/Master/texmf-dist/source/latex/l3experimental/l3str/l3str.ins index 7dc0c00357d..d8429f908fa 100644 --- a/Master/texmf-dist/source/latex/l3experimental/l3str/l3str.ins +++ b/Master/texmf-dist/source/latex/l3experimental/l3str/l3str.ins @@ -16,7 +16,7 @@ The released version of this bundle is available from CTAN. \fi -\input docstrip.tex +\input l3docstrip.tex \askforoverwritefalse \preamble diff --git a/Master/texmf-dist/source/latex/l3experimental/l3str/l3tl-analysis.dtx b/Master/texmf-dist/source/latex/l3experimental/l3str/l3tl-analysis.dtx index 89ce9c22f3d..ef4d26bfe75 100644 --- a/Master/texmf-dist/source/latex/l3experimental/l3str/l3tl-analysis.dtx +++ b/Master/texmf-dist/source/latex/l3experimental/l3str/l3tl-analysis.dtx @@ -36,7 +36,7 @@ %<*driver|package> \RequirePackage{expl3} \GetIdInfo$Id: l3tl-analysis.dtx 3039 2011-12-08 09:22:35Z bruno $ - {L3 Experimental Analysis of Token lists} + {L3 Experimental token lists analysis} % %<*driver> \documentclass[full]{l3doc} @@ -88,14 +88,14 @@ % % \subsection{Internal functions} % -% \begin{variable}{\s_tl} +% \begin{variable}{\s__tl} % The format used to store token lists internally uses the scan mark -% \cs{s_tl} as a delimiter. +% \cs{s__tl} as a delimiter. % \end{variable} % -% \begin{function}{\tl_analysis_map_inline:nn} +% \begin{function}{\__tl_analysis_map_inline:nn} % \begin{syntax} -% \cs{tl_analysis_map_inline:nn} \Arg{token list} \Arg{inline function} +% \cs{__tl_analysis_map_inline:nn} \Arg{token list} \Arg{inline function} % \end{syntax} % Applies the \meta{inline function} to each individual \meta{token} % in the \meta{token list}. The \meta{inline function} receives three @@ -116,8 +116,8 @@ % \end{function} % % For optimizations in \pkg{l3regex} (when matching control sequences), -% it may be useful to provide a \cs{tl_analysis_from_str_map_inline:nn} -% function, perhaps named \cs{str_analysis_map_inline:nn}. +% it may be useful to provide a \cs{__tl_analysis_from_str_map_inline:nn} +% function, perhaps named \cs{__str_analysis_map_inline:nn}. % % \subsection{Internal format} % @@ -138,7 +138,7 @@ % regexes are catcode-agnostic. The internal format thus takes the form % of a succession of items of the form % \begin{quote} -% \meta{tokens} \cs{s_tl} \meta{catcode} \meta{char code} \cs{s_tl} +% \meta{tokens} \cs{s__tl} \meta{catcode} \meta{char code} \cs{s__tl} % \end{quote} % The \meta{tokens} \texttt{o}- \emph{and} \texttt{x}-expand to the % original token in the token list or to the cluster of tokens @@ -149,7 +149,7 @@ % % Using delimited arguments lets us build the \meta{tokens} % progressively when doing an encoding conversion in \pkg{l3str}. On the -% other hand, the delimiter \cs{s_tl} may not appear unbraced in +% other hand, the delimiter \cs{s__tl} may not appear unbraced in % \meta{tokens}. This is not a problem because we are careful to wrap % control sequences in braces (as an argument to \cs{exp_not:n}) when % converting from a general token list to the internal format. @@ -159,15 +159,15 @@ % it is the following. % \begin{itemize} % \item A control sequence |\cs| becomes |\exp_not:n { \cs }| -% \cs{s_tl} $0$ $-1$ \cs{s_tl}. +% \cs{s__tl} $0$ $-1$ \cs{s__tl}. % \item A begin-group character |{| becomes \cs{exp_after:wN} |{| -% \cs{if_false:} |}| \cs{fi:} \cs{s_tl} $1$ \meta{char code} -% \cs{s_tl}. +% \cs{if_false:} |}| \cs{fi:} \cs{s__tl} $1$ \meta{char code} +% \cs{s__tl}. % \item An end-group character |}| becomes \cs{if_false:} |{| \cs{fi:} -% |}| \cs{s_tl} $2$ \meta{char code} \cs{s_tl}. +% |}| \cs{s__tl} $2$ \meta{char code} \cs{s__tl}. % \item A character with any other category code becomes -% \cs{exp_not:n} \Arg{character} \cs{s_tl} \meta{hex catcode} -% \meta{char code} \cs{s_tl}. +% \cs{exp_not:n} \Arg{character} \cs{s__tl} \meta{hex catcode} +% \meta{char code} \cs{s__tl}. % \end{itemize} % % ^^A todo: ask LuaTeX list for an \ifx\undefined @@ -184,6 +184,10 @@ % \end{macrocode} % % \begin{macrocode} +%<@@=tl_analysis> +% \end{macrocode} +% +% \begin{macrocode} \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} \RequirePackage{l3str} @@ -191,127 +195,127 @@ % % \subsection{Variables and helper functions} % -% \begin{variable}{\s_tl} -% The scan mark \cs{s_tl} is used as a delimiter in the internal +% \begin{variable}{\s__tl} +% The scan mark \cs{s__tl} is used as a delimiter in the internal % format. This is more practical than using a quark, because we would % then need to control expansion much more carefully: compare -% \cs{int_value:w} |`#1| \cs{s_tl} with \cs{int_value:w} |`#1| +% \cs{__int_value:w} |`#1| \cs{s__tl} with \cs{__int_value:w} |`#1| % \cs{exp_stop_f:} \cs{exp_not:N} \cs{q_mark} to extract a character % code followed by the delimiter in an \texttt{x}-expansion. % \begin{macrocode} -\scan_new:N \s_tl +\__scan_new:N \s__tl % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_tl_analysis_internal_tl} +% \begin{variable}{\l_@@_internal_tl} % This token list variable is used to hand the argument of % \cs{tl_show_analysis:n} to \cs{tl_show_analysis:N}. % \begin{macrocode} -\tl_new:N \l_tl_analysis_internal_tl +\tl_new:N \l_@@_internal_tl % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_tl_analysis_token} -% \begin{variable}{\l_tl_analysis_char_token} +% \begin{variable}{\l_@@_token} +% \begin{variable}{\l_@@_char_token} % The tokens in the token list are probed with the \TeX{} primitive -% \tn{futurelet}. We use \cs{l_tl_analysis_token} in that +% \tn{futurelet}. We use \cs{l_@@_token} in that % construction. In some cases, we convert the following token to a % string before probing it: then the token variable used is -% \cs{l_tl_analysis_char_token}. +% \cs{l_@@_char_token}. % \begin{macrocode} -\cs_new_eq:NN \l_tl_analysis_token ? -\cs_new_eq:NN \l_tl_analysis_char_token ? +\cs_new_eq:NN \l_@@_token ? +\cs_new_eq:NN \l_@@_char_token ? % \end{macrocode} % \end{variable} % \end{variable} % -% \begin{variable}{\l_tl_analysis_normal_int} +% \begin{variable}{\l_@@_normal_int} % The number of normal (\texttt{N}-type argument) tokens since the % last special token. % \begin{macrocode} -\int_new:N \l_tl_analysis_normal_int +\int_new:N \l_@@_normal_int % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_tl_analysis_index_int} +% \begin{variable}{\l_@@_index_int} % During the first pass, this is the index in the array being built. % During the second pass, it is equal to the maximum index in the % array from the first pass. % \begin{macrocode} -\int_new:N \l_tl_analysis_index_int +\int_new:N \l_@@_index_int % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_tl_analysis_nesting_int} +% \begin{variable}{\l_@@_nesting_int} % Nesting depth of explicit begin-group and end-group characters % during the first pass. This lets us detect the end of the token list % without a reserved end-marker. % \begin{macrocode} -\int_new:N \l_tl_analysis_nesting_int +\int_new:N \l_@@_nesting_int % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_tl_analysis_type_int} +% \begin{variable}{\l_@@_type_int} % When encountering special characters, we record their \enquote{type} % in this integer. % \begin{macrocode} -\int_new:N \l_tl_analysis_type_int +\int_new:N \l_@@_type_int % \end{macrocode} % \end{variable} % -% \begin{variable}{\g_tl_analysis_result_tl} +% \begin{variable}{\g_@@_result_tl} % The result of the conversion is stored in this token list, with a % succession of items of the form % \begin{quote} -% \meta{tokens} \cs{s_tl} \meta{catcode} \meta{char code} \cs{s_tl} +% \meta{tokens} \cs{s__tl} \meta{catcode} \meta{char code} \cs{s__tl} % \end{quote} % \begin{macrocode} -\tl_new:N \g_tl_analysis_result_tl +\tl_new:N \g_@@_result_tl % \end{macrocode} % \end{variable} % -% \begin{macro}[int, EXP]{\tl_analysis_extract_charcode:} -% \begin{macro}[aux, EXP]{\tl_analysis_extract_charcode_aux:w} +% \begin{macro}[int, EXP]{\@@_extract_charcode:} +% \begin{macro}[aux, EXP]{\@@_extract_charcode_aux:w} % Extracting the character code from the meaning of -% \cs{l_tl_analysis_token}. This has no error checking, and should +% \cs{l_@@_token}. This has no error checking, and should % only be assumed to work for begin-group and end-group character % tokens. It produces a number in the form |`|\meta{char}. % \begin{macrocode} -\cs_new_nopar:Npn \tl_analysis_extract_charcode: +\cs_new_nopar:Npn \@@_extract_charcode: { - \exp_after:wN \tl_analysis_extract_charcode_aux:w - \token_to_meaning:N \l_tl_analysis_token + \exp_after:wN \@@_extract_charcode_aux:w + \token_to_meaning:N \l_@@_token } -\cs_new:Npn \tl_analysis_extract_charcode_aux:w #1 ~ #2 ~ { ` } +\cs_new:Npn \@@_extract_charcode_aux:w #1 ~ #2 ~ { ` } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int, EXP]{\tl_analysis_cs_space_count:NN} -% \begin{macro}[aux, EXP]{\tl_analysis_cs_space_count:w} -% \begin{macro}[aux, EXP]{\tl_analysis_cs_space_count_end:w} +% \begin{macro}[int, EXP]{\@@_cs_space_count:NN} +% \begin{macro}[aux, EXP]{\@@_cs_space_count:w} +% \begin{macro}[aux, EXP]{\@@_cs_space_count_end:w} % Counts the number of spaces in the string representation of its % second argument, as well as the number of characters following the % last space in that representation, and feeds the two numbers as % semicolon-delimited arguments to the first argument. When this % function is used, the escape character is printable and non-space. % \begin{macrocode} -\cs_new:Npn \tl_analysis_cs_space_count:NN #1 #2 +\cs_new:Npn \@@_cs_space_count:NN #1 #2 { \exp_after:wN #1 - \int_value:w \int_eval:w \c_zero - \exp_after:wN \tl_analysis_cs_space_count:w + \__int_value:w \__int_eval:w \c_zero + \exp_after:wN \@@_cs_space_count:w \token_to_str:N #2 - \fi: \tl_analysis_cs_space_count_end:w ; ~ ! + \fi: \@@_cs_space_count_end:w ; ~ ! } -\cs_new:Npn \tl_analysis_cs_space_count:w #1 ~ +\cs_new:Npn \@@_cs_space_count:w #1 ~ { \if_false: #1 #1 \fi: + \c_one - \tl_analysis_cs_space_count:w + \@@_cs_space_count:w } -\cs_new:Npn \tl_analysis_cs_space_count_end:w ; #1 \fi: #2 ! - { \exp_after:wN ; \int_value:w \str_length_ignore_spaces:n {#1} ; } +\cs_new:Npn \@@_cs_space_count_end:w ; #1 \fi: #2 ! + { \exp_after:wN ; \__int_value:w \str_count_ignore_spaces:n {#1} ; } % \end{macrocode} % \end{macro} % \end{macro} @@ -321,10 +325,10 @@ % % Our goal is to produce a token list of the form roughly % \begin{quote} -% \meta{token 1} \cs{s_tl} \meta{catcode 1} \meta{char code 1} \cs{s_tl} \\ -% \meta{token 2} \cs{s_tl} \meta{catcode 2} \meta{char code 2} \cs{s_tl} \\ +% \meta{token 1} \cs{s__tl} \meta{catcode 1} \meta{char code 1} \cs{s__tl} \\ +% \meta{token 2} \cs{s__tl} \meta{catcode 2} \meta{char code 2} \cs{s__tl} \\ % \ldots{} -% \meta{token N} \cs{s_tl} \meta{catcode N} \meta{char code N} \cs{s_tl} +% \meta{token N} \cs{s__tl} \meta{catcode N} \meta{char code N} \cs{s__tl} % \end{quote} % Most but not all tokens can be grabbed as an undelimited % (\texttt{N}-type) argument by \TeX{}. The plan is to have a two pass @@ -337,7 +341,7 @@ % result one item at a time. % % To ease the difficult first pass, we first do some setup with -% \cs{tl_analysis_setup:n}. Active characters set equal to non-active +% \cs{@@_setup:n}. Active characters set equal to non-active % characters cause trouble, so we disable all active characters by % setting them equal to \texttt{undefined} locally. We also set there % the escape character to be printable (backslash, but this later @@ -373,19 +377,19 @@ % % The second pass is a simple exercise in expandable loops. % -% \begin{macro}[int]{\tl_analysis:n} +% \begin{macro}[int]{\@@:n} % Everything is done within a group, and all definitions will be % local. We use \cs{group_align_safe_begin/end:} to avoid problems in -% case \cs{tl_analysis:n} is used within an alignment and its argument +% case \cs{@@:n} is used within an alignment and its argument % contains alignment tab tokens. % \begin{macrocode} -\cs_new_protected:Npn \tl_analysis:n #1 +\cs_new_protected:Npn \@@:n #1 { \group_begin: \group_align_safe_begin: - \tl_analysis_setup:n {#1} - \tl_analysis_i:n {#1} - \tl_analysis_ii:n {#1} + \@@_setup:n {#1} + \@@_i:n {#1} + \@@_ii:n {#1} \group_align_safe_end: \group_end: } @@ -394,8 +398,8 @@ % % \subsection{Setup} % -% \begin{macro}[int]{\tl_analysis_setup:n} -% \begin{macro}[aux]{\tl_analysis_disable_loop:N} +% \begin{macro}[int]{\@@_setup:n} +% \begin{macro}[aux]{\@@_disable_loop:N} % Active characters can cause problems later on in the processing, % so the first step is to disable them, by setting them to % \texttt{undefined}. Since Unicode contains too many characters @@ -407,20 +411,20 @@ % and we end the loop by feeding an odd non-\texttt{N}-type % argument to the looping macro. % \begin{macrocode} -\cs_new_protected:Npn \tl_analysis_setup:n #1 +\cs_new_protected:Npn \@@_setup:n #1 { \int_set_eq:NN \tex_escapechar:D \c_minus_one - \exp_after:wN \tl_analysis_disable_loop:N - \tl_to_str:n {#1} { ~ } { ? ~ \prg_map_break: } - \prg_break_point:n { } + \exp_after:wN \@@_disable_loop:N + \tl_to_str:n {#1} { ~ } { ? ~ \__prg_break: } + \__prg_break_point: } \group_begin: \char_set_catcode_active:N \^^@ - \cs_new_protected:Npn \tl_analysis_disable_loop:N #1 + \cs_new_protected:Npn \@@_disable_loop:N #1 { \tex_lccode:D \c_zero `#1 ~ \tl_to_lowercase:n { \tex_let:D ^^@ } \c_undefined:D - \tl_analysis_disable_loop:N + \@@_disable_loop:N } \group_end: % \end{macrocode} @@ -463,36 +467,36 @@ % \cs{str_tail:n} \Arg{token} is non-empty, because the % escape character is printable. % -% \begin{macro}[int]{\tl_analysis_i:n} +% \begin{macro}[int]{\@@_i:n} % We read tokens one by one using \tn{futurelet}. % While performing the loop, we keep track of the number of % true begin-group characters minus the number of -% true end-group characters in \cs{l_tl_analysis_nesting_int}. +% true end-group characters in \cs{l_@@_nesting_int}. % This reaches $-1$ when we read the closing brace. % \begin{macrocode} -\cs_new_protected:Npn \tl_analysis_i:n #1 +\cs_new_protected:Npn \@@_i:n #1 { \int_set:Nn \tex_escapechar:D { 92 } - \int_zero:N \l_tl_analysis_normal_int - \int_zero:N \l_tl_analysis_index_int - \int_zero:N \l_tl_analysis_nesting_int - \if_false: { \fi: \tl_analysis_i_loop:w #1 } - \int_decr:N \l_tl_analysis_index_int + \int_zero:N \l_@@_normal_int + \int_zero:N \l_@@_index_int + \int_zero:N \l_@@_nesting_int + \if_false: { \fi: \@@_i_loop:w #1 } + \int_decr:N \l_@@_index_int } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\tl_analysis_i_loop:w} +% \begin{macro}[int]{\@@_i_loop:w} % Read one character and check its type. % \begin{macrocode} -\cs_new_protected_nopar:Npn \tl_analysis_i_loop:w - { \tex_futurelet:D \l_tl_analysis_token \tl_analysis_i_type:w } +\cs_new_protected_nopar:Npn \@@_i_loop:w + { \tex_futurelet:D \l_@@_token \@@_i_type:w } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\tl_analysis_i_type:w} -% At this point, \cs{l_tl_analysis_token} holds the meaning -% of the following token. We store in \cs{l_tl_analysis_type_int} +% \begin{macro}[int]{\@@_i_type:w} +% At this point, \cs{l_@@_token} holds the meaning +% of the following token. We store in \cs{l_@@_type_int} % the meaning of the token ahead: % \begin{itemize} % \item 0 space token; @@ -504,45 +508,45 @@ % character changes the nesting level ($2$ is used only here, % and is irrelevant later). Then call the auxiliary for each case. % Note that nesting conditionals here is safe because we only skip -% over \cs{l_tl_analysis_token} if it matches with one of the +% over \cs{l_@@_token} if it matches with one of the % character tokens (hence is not a primitive conditional). % \begin{macrocode} -\cs_new_protected_nopar:Npn \tl_analysis_i_type:w +\cs_new_protected_nopar:Npn \@@_i_type:w { - \l_tl_analysis_type_int = - \if_meaning:w \l_tl_analysis_token \c_space_token + \l_@@_type_int = + \if_meaning:w \l_@@_token \c_space_token \c_zero \else: - \if_catcode:w \exp_not:N \l_tl_analysis_token \c_group_begin_token + \if_catcode:w \exp_not:N \l_@@_token \c_group_begin_token \c_one \else: - \if_catcode:w \exp_not:N \l_tl_analysis_token \c_group_end_token + \if_catcode:w \exp_not:N \l_@@_token \c_group_end_token \c_minus_one \else: \c_two \fi: \fi: \fi: - \if_case:w \l_tl_analysis_type_int - \exp_after:wN \tl_analysis_i_space:w - \or: \exp_after:wN \tl_analysis_i_bgroup:w - \or: \exp_after:wN \tl_analysis_i_safe:N - \else: \exp_after:wN \tl_analysis_i_egroup:w + \if_case:w \l_@@_type_int + \exp_after:wN \@@_i_space:w + \or: \exp_after:wN \@@_i_bgroup:w + \or: \exp_after:wN \@@_i_safe:N + \else: \exp_after:wN \@@_i_egroup:w \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\tl_analysis_i_space:w} -% \begin{macro}[aux]{\tl_analysis_i_space_test:w} +% \begin{macro}[int]{\@@_i_space:w} +% \begin{macro}[aux]{\@@_i_space_test:w} % In this branch, the following token's meaning is a blank space. % Apply \tn{string} to that token: if it is a control sequence % the result starts with the escape character; otherwise it is % a true blank space, whose string representation is also a blank space. -% We test for that in \cs{tl_analysis_i_space_test:w}, -% after grabbing as \cs{l_tl_analysis_char_token} the first character +% We test for that in \cs{@@_i_space_test:w}, +% after grabbing as \cs{l_@@_char_token} the first character % of the string representation. -% Also, since \cs{tl_analysis_i_store:} expects the special token to be +% Also, since \cs{@@_i_store:} expects the special token to be % stored in the relevant \tn{toks} register, we do that. The extra % \cs{exp_not:n} is unnecessary of course, but it makes the treatment % of all tokens more homogeneous. @@ -555,30 +559,30 @@ % so that the second pass does not need to test the meaning of tokens, % only strings. % \begin{macrocode} -\cs_new_protected_nopar:Npn \tl_analysis_i_space:w +\cs_new_protected_nopar:Npn \@@_i_space:w { - \tex_afterassignment:D \tl_analysis_i_space_test:w + \tex_afterassignment:D \@@_i_space_test:w \exp_after:wN \cs_set_eq:NN - \exp_after:wN \l_tl_analysis_char_token + \exp_after:wN \l_@@_char_token \token_to_str:N } -\cs_new_protected_nopar:Npn \tl_analysis_i_space_test:w +\cs_new_protected_nopar:Npn \@@_i_space_test:w { - \if_meaning:w \l_tl_analysis_char_token \c_space_token - \tex_toks:D \l_tl_analysis_index_int { \exp_not:n { ~ } } - \tl_analysis_i_store: + \if_meaning:w \l_@@_char_token \c_space_token + \tex_toks:D \l_@@_index_int { \exp_not:n { ~ } } + \@@_i_store: \else: - \int_incr:N \l_tl_analysis_normal_int + \int_incr:N \l_@@_normal_int \fi: - \tl_analysis_i_loop:w + \@@_i_loop:w } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\tl_analysis_i_bgroup:w, \tl_analysis_i_egroup:w} -% \begin{macro}[aux]{\tl_analysis_i_group:nw} -% \begin{macro}[aux]{\tl_analysis_i_group_test:w} +% \begin{macro}[int]{\@@_i_bgroup:w, \@@_i_egroup:w} +% \begin{macro}[aux]{\@@_i_group:nw} +% \begin{macro}[aux]{\@@_i_group_test:w} % The token might be either a true character token with % catcode $1$ or $2$, or it could be a control sequence. % The only tricky case is if the character code happens @@ -588,50 +592,50 @@ % and of a control sequence set equal to it start differently. % Then probe what the first character of that string % representation is: this is the place where we need -% \cs{l_tl_analysis_char_token} to be a separate control -% sequence from \cs{l_tl_analysis_token}, to compare them. +% \cs{l_@@_char_token} to be a separate control +% sequence from \cs{l_@@_token}, to compare them. % \begin{macrocode} \group_begin: \char_set_catcode_group_begin:N \^^@ \char_set_catcode_group_end:N \^^E - \cs_new_protected_nopar:Npn \tl_analysis_i_bgroup:w - { \tl_analysis_i_group:nw { \exp_after:wN ^^@ \if_false: ^^E \fi: } } + \cs_new_protected_nopar:Npn \@@_i_bgroup:w + { \@@_i_group:nw { \exp_after:wN ^^@ \if_false: ^^E \fi: } } \char_set_catcode_group_begin:N \^^B \char_set_catcode_group_end:N \^^@ - \cs_new_protected_nopar:Npn \tl_analysis_i_egroup:w - { \tl_analysis_i_group:nw { \if_false: ^^B \fi: ^^@ } } + \cs_new_protected_nopar:Npn \@@_i_egroup:w + { \@@_i_group:nw { \if_false: ^^B \fi: ^^@ } } \group_end: -\cs_new_protected:Npn \tl_analysis_i_group:nw #1 +\cs_new_protected:Npn \@@_i_group:nw #1 { - \tex_lccode:D \c_zero = \tl_analysis_extract_charcode: \scan_stop: - \tl_to_lowercase:n { \tex_toks:D \l_tl_analysis_index_int {#1} } - \if_num:w \tex_lccode:D \c_zero = \tex_escapechar:D + \tex_lccode:D \c_zero = \@@_extract_charcode: \scan_stop: + \tl_to_lowercase:n { \tex_toks:D \l_@@_index_int {#1} } + \if_int_compare:w \tex_lccode:D \c_zero = \tex_escapechar:D \int_set:Nn \tex_escapechar:D { 139 - \tex_escapechar:D } \fi: - \tex_afterassignment:D \tl_analysis_i_group_test:w + \tex_afterassignment:D \@@_i_group_test:w \exp_after:wN \cs_set_eq:NN - \exp_after:wN \l_tl_analysis_char_token + \exp_after:wN \l_@@_char_token \token_to_str:N } -\cs_new_protected_nopar:Npn \tl_analysis_i_group_test:w +\cs_new_protected_nopar:Npn \@@_i_group_test:w { - \if_charcode:w \l_tl_analysis_token \l_tl_analysis_char_token - \tl_analysis_i_store: + \if_charcode:w \l_@@_token \l_@@_char_token + \@@_i_store: \else: - \int_incr:N \l_tl_analysis_normal_int + \int_incr:N \l_@@_normal_int \fi: - \tl_analysis_i_loop:w + \@@_i_loop:w } % \end{macrocode} % \end{macro} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\tl_analysis_i_store:} +% \begin{macro}[int]{\@@_i_store:} % This function is called each time we meet a special token; -% at this point, the \tn{toks} register \cs{l_tl_analysis_index_int} +% at this point, the \tn{toks} register \cs{l_@@_index_int} % holds a token list which expands to the given special token. -% Also, the value of \cs{l_tl_analysis_type_int} indicates which case +% Also, the value of \cs{l_@@_type_int} indicates which case % we are in: % \begin{itemize} % \item -1 end-group character; @@ -651,31 +655,31 @@ % \item 2 space begin-group character. % \end{itemize} % This has the property that non-space characters correspond to odd -% values of \cs{l_tl_analysis_type_int}. +% values of \cs{l_@@_type_int}. % The number of normal tokens, and the type of special token, % are packed into a \tn{skip} register. % Finally, we check whether we reached the last closing brace, in which % case we stop by disabling the looping function (locally). % \begin{macrocode} -\cs_new_protected_nopar:Npn \tl_analysis_i_store: +\cs_new_protected_nopar:Npn \@@_i_store: { - \tex_advance:D \l_tl_analysis_nesting_int \l_tl_analysis_type_int - \if_num:w \tex_lccode:D \c_zero = \c_thirty_two - \tex_multiply:D \l_tl_analysis_type_int \c_two + \tex_advance:D \l_@@_nesting_int \l_@@_type_int + \if_int_compare:w \tex_lccode:D \c_zero = \c_thirty_two + \tex_multiply:D \l_@@_type_int \c_two \fi: - \tex_skip:D \l_tl_analysis_index_int - = \l_tl_analysis_normal_int sp plus \l_tl_analysis_type_int sp \scan_stop: - \int_incr:N \l_tl_analysis_index_int - \int_zero:N \l_tl_analysis_normal_int - \if_num:w \l_tl_analysis_nesting_int = \c_minus_one - \cs_set_eq:NN \tl_analysis_i_loop:w \scan_stop: + \tex_skip:D \l_@@_index_int + = \l_@@_normal_int sp plus \l_@@_type_int sp \scan_stop: + \int_incr:N \l_@@_index_int + \int_zero:N \l_@@_normal_int + \if_int_compare:w \l_@@_nesting_int = \c_minus_one + \cs_set_eq:NN \@@_i_loop:w \scan_stop: \fi: } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\tl_analysis_i_safe:N} -% \begin{macro}[aux]{\tl_analysis_i_cs:ww} +% \begin{macro}[int]{\@@_i_safe:N} +% \begin{macro}[aux]{\@@_i_cs:ww} % This should be the simplest case: since the upcoming token is safe, % we can simply grab it in a second pass. However, other branches of % the code must pass their tokens through \tn{string}, hence we do it @@ -689,32 +693,32 @@ % if the control sequence name contains no space, count that token % as a number of normal tokens equal to its string length. If the % control sequence contains spaces, they should be registered as -% special characters by increasing \cs{l_tl_analysis_index_int} +% special characters by increasing \cs{l_@@_index_int} % (no need to carefully count character between each space), and % all characters after the last space should be counted in the % following sequence of \enquote{normal} tokens. % \begin{macrocode} -\cs_new_protected:Npn \tl_analysis_i_safe:N #1 +\cs_new_protected:Npn \@@_i_safe:N #1 { \if_charcode:w \scan_stop: \exp_after:wN \use_none:n \token_to_str:N #1 \prg_do_nothing: \scan_stop: - \int_incr:N \l_tl_analysis_normal_int + \int_incr:N \l_@@_normal_int \else: - \tl_analysis_cs_space_count:NN \tl_analysis_i_cs:ww #1 + \@@_cs_space_count:NN \@@_i_cs:ww #1 \fi: - \tl_analysis_i_loop:w + \@@_i_loop:w } -\cs_new_protected:Npn \tl_analysis_i_cs:ww #1; #2; +\cs_new_protected:Npn \@@_i_cs:ww #1; #2; { - \if_num:w #1 > \c_zero - \tex_skip:D \l_tl_analysis_index_int - = \int_eval:w \l_tl_analysis_normal_int + \c_one sp \scan_stop: - \tex_advance:D \l_tl_analysis_index_int #1 \exp_stop_f: - \l_tl_analysis_normal_int #2 \exp_stop_f: + \if_int_compare:w #1 > \c_zero + \tex_skip:D \l_@@_index_int + = \__int_eval:w \l_@@_normal_int + \c_one sp \scan_stop: + \tex_advance:D \l_@@_index_int #1 \exp_stop_f: + \l_@@_normal_int #2 \exp_stop_f: \else: - \tex_advance:D \l_tl_analysis_normal_int #2 \exp_stop_f: + \tex_advance:D \l_@@_normal_int #2 \exp_stop_f: \fi: } % \end{macrocode} @@ -727,60 +731,60 @@ % All the necessary information is stored in \tn{skip} % and \tn{toks} registers. % -% \begin{macro}[int]{\tl_analysis_ii:n} -% \begin{macro}[int, EXP]{\tl_analysis_ii_loop:w} +% \begin{macro}[int]{\@@_ii:n} +% \begin{macro}[int, EXP]{\@@_ii_loop:w} % Start the loop with the index $0$. No need for an end-marker: % the loop will stop by itself when the last index is read. % We will repeatedly oscillate between reading long stretches % of normal tokens, and reading special tokens. % \begin{macrocode} -\cs_new_protected:Npn \tl_analysis_ii:n #1 +\cs_new_protected:Npn \@@_ii:n #1 { - \tl_gset:Nx \g_tl_analysis_result_tl + \tl_gset:Nx \g_@@_result_tl { - \tl_analysis_ii_loop:w 0; #1 - \prg_break_point:n { } + \@@_ii_loop:w 0; #1 + \__prg_break_point: } } -\cs_new:Npn \tl_analysis_ii_loop:w #1; +\cs_new:Npn \@@_ii_loop:w #1; { - \exp_after:wN \tl_analysis_ii_normals:ww - \int_value:w \tex_skip:D #1 ; #1 ; + \exp_after:wN \@@_ii_normals:ww + \__int_value:w \tex_skip:D #1 ; #1 ; } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int, EXP]{\tl_analysis_ii_normals:ww} -% \begin{macro}[aux, EXP]{\tl_analysis_ii_normal:wwN} +% \begin{macro}[int, EXP]{\@@_ii_normals:ww} +% \begin{macro}[aux, EXP]{\@@_ii_normal:wwN} % The first argument is the number of normal tokens which remain % to be read, and the second argument is the index in the array % produced in the first step. % A character's string representation is always one character long, % while a control sequence is always longer (we have set the escape % character to a printable value). In both cases, we leave -% \cs{exp_not:n} \Arg{token} \cs{s_tl} in the input stream +% \cs{exp_not:n} \Arg{token} \cs{s__tl} in the input stream % (after \texttt{x}-expansion). Here, \cs{exp_not:n} is used -% rather than \cs{exp_not:N} because |#3| could be \cs{s_tl}, +% rather than \cs{exp_not:N} because |#3| could be \cs{s__tl}, % hence must be hidden behind braces in the result. % \begin{macrocode} -\cs_new:Npn \tl_analysis_ii_normals:ww #1; +\cs_new:Npn \@@_ii_normals:ww #1; { - \if_num:w #1 = \c_zero - \tl_analysis_ii_special:w + \if_int_compare:w #1 = \c_zero + \@@_ii_special:w \fi: - \tl_analysis_ii_normal:wwN #1; + \@@_ii_normal:wwN #1; } -\cs_new:Npn \tl_analysis_ii_normal:wwN #1; #2; #3 +\cs_new:Npn \@@_ii_normal:wwN #1; #2; #3 { - \exp_not:n { \exp_not:n { #3 } } \s_tl + \exp_not:n { \exp_not:n { #3 } } \s__tl \if_charcode:w \scan_stop: \exp_after:wN \use_none:n \token_to_str:N #3 \prg_do_nothing: \scan_stop: - \exp_after:wN \tl_analysis_ii_char:Nww + \exp_after:wN \@@_ii_char:Nww \else: - \exp_after:wN \tl_analysis_ii_cs:Nww + \exp_after:wN \@@_ii_cs:Nww \fi: #3 #1; #2; } @@ -788,10 +792,10 @@ % \end{macro} % \end{macro} % -% \begin{macro}[int, EXP]{\tl_analysis_ii_char:Nww} +% \begin{macro}[int, EXP]{\@@_ii_char:Nww} % If the normal token we grab is a character, leave -% \meta{catcode} \meta{charcode} followed by \cs{s_tl} -% in the input stream, and call \cs{tl_analysis_ii_normals:ww} +% \meta{catcode} \meta{charcode} followed by \cs{s__tl} +% in the input stream, and call \cs{@@_ii_normals:ww} % with its first argument decremented. % \begin{macrocode} \group_begin: @@ -801,7 +805,7 @@ \char_set_uccode:nn { `? } { `D } \tl_to_uppercase:n { - \cs_new:Npn \tl_analysis_ii_char:Nww #1 + \cs_new:Npn \@@_ii_char:Nww #1 { \if_meaning:w #1 \c_undefined:D ? \else: \if_catcode:w #1 \c_catcode_other_token C \else: @@ -813,47 +817,47 @@ \if_catcode:w #1 \c_space_token A \else: 6 \fi: \fi: \fi: \fi: \fi: \fi: \fi: \fi: - \int_value:w `#1 \s_tl - \exp_after:wN \tl_analysis_ii_normals:ww - \int_use:N \int_eval:w \c_minus_one + + \__int_value:w `#1 \s__tl + \exp_after:wN \@@_ii_normals:ww + \int_use:N \__int_eval:w \c_minus_one + } } \group_end: % \end{macrocode} % \end{macro} % -% \begin{macro}[int, EXP]{\tl_analysis_ii_cs:Nww} -% \begin{macro}[aux, EXP]{\tl_analysis_ii_cs_test:ww} +% \begin{macro}[int, EXP]{\@@_ii_cs:Nww} +% \begin{macro}[aux, EXP]{\@@_ii_cs_test:ww} % If the token we grab is a control sequence, leave % |0 -1| (as category code and character code) in the input stream, -% followed by \cs{s_tl}, -% and call \cs{tl_analysis_ii_normals:ww} with updated arguments. +% followed by \cs{s__tl}, +% and call \cs{@@_ii_normals:ww} with updated arguments. % \begin{macrocode} -\cs_new:Npn \tl_analysis_ii_cs:Nww #1 +\cs_new:Npn \@@_ii_cs:Nww #1 { - 0 -1 \s_tl - \tl_analysis_cs_space_count:NN \tl_analysis_ii_cs_test:ww #1 + 0 -1 \s__tl + \@@_cs_space_count:NN \@@_ii_cs_test:ww #1 } -\cs_new:Npn \tl_analysis_ii_cs_test:ww #1 ; #2 ; #3 ; #4 ; +\cs_new:Npn \@@_ii_cs_test:ww #1 ; #2 ; #3 ; #4 ; { - \exp_after:wN \tl_analysis_ii_normals:ww - \int_use:N \int_eval:w - \if_num:w #1 = \c_zero + \exp_after:wN \@@_ii_normals:ww + \int_use:N \__int_eval:w + \if_int_compare:w #1 = \c_zero #3 \else: - \tex_skip:D \int_eval:w #4 + #1 \int_eval_end: + \tex_skip:D \__int_eval:w #4 + #1 \__int_eval_end: \fi: - #2 \exp_after:wN ; - \int_use:N \int_eval:w #4 + #1 ; + \int_use:N \__int_eval:w #4 + #1 ; } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[int, EXP]{\tl_analysis_ii_special:w} -% \begin{macro}[aux, EXP]{\tl_analysis_ii_special_char:wN} -% \begin{macro}[aux, EXP]{\tl_analysis_ii_special_space:w} +% \begin{macro}[int, EXP]{\@@_ii_special:w} +% \begin{macro}[aux, EXP]{\@@_ii_special_char:wN} +% \begin{macro}[aux, EXP]{\@@_ii_special_space:w} % Here, |#1| is the current index in the array built in the first pass. % Check now whether we reached the end (we shouldn't keep the trailing % end-group character that marked the end of the token list in the @@ -861,18 +865,18 @@ % Unpack the \tn{toks} register: when \texttt{x}-expanding again, % we will get the special token. % Then leave the category code in the input stream, followed by -% the character code, and call \cs{tl_analysis_ii_loop:w} with the next index. +% the character code, and call \cs{@@_ii_loop:w} with the next index. % \begin{macrocode} \group_begin: \char_set_catcode_other:N A - \cs_new:Npn \tl_analysis_ii_special:w - \fi: \tl_analysis_ii_normal:wwN 0 ; #1 ; + \cs_new:Npn \@@_ii_special:w + \fi: \@@_ii_normal:wwN 0 ; #1 ; { \fi: - \if_num:w #1 = \l_tl_analysis_index_int - \exp_after:wN \prg_map_break: + \if_int_compare:w #1 = \l_@@_index_int + \exp_after:wN \__prg_break: \fi: - \tex_the:D \tex_toks:D #1 \s_tl + \tex_the:D \tex_toks:D #1 \s__tl \if_case:w \etex_gluestretch:D \tex_skip:D #1 \exp_stop_f: A \or: 1 @@ -880,23 +884,23 @@ \else: 2 \fi: \if_int_odd:w \etex_gluestretch:D \tex_skip:D #1 \exp_stop_f: - \exp_after:wN \tl_analysis_ii_special_char:wN \int_use:N + \exp_after:wN \@@_ii_special_char:wN \int_use:N \else: - \exp_after:wN \tl_analysis_ii_special_space:w \int_use:N + \exp_after:wN \@@_ii_special_space:w \int_use:N \fi: - \int_eval:w \c_one + #1 \exp_after:wN ; + \__int_eval:w \c_one + #1 \exp_after:wN ; \token_to_str:N } \group_end: -\cs_new:Npn \tl_analysis_ii_special_char:wN #1 ; #2 +\cs_new:Npn \@@_ii_special_char:wN #1 ; #2 { - \int_value:w `#2 \s_tl - \tl_analysis_ii_loop:w #1 ; + \__int_value:w `#2 \s__tl + \@@_ii_loop:w #1 ; } -\cs_new:Npn \tl_analysis_ii_special_space:w #1 ; ~ +\cs_new:Npn \@@_ii_special_space:w #1 ; ~ { - 32 \s_tl - \tl_analysis_ii_loop:w #1 ; + 32 \s__tl + \@@_ii_loop:w #1 ; } % \end{macrocode} % \end{macro} @@ -905,38 +909,38 @@ % % \subsection{Mapping through the analysis} % -% \begin{macro}[int]{\tl_analysis_map_inline:nn} -% \begin{macro}[aux]{\tl_analysis_map_inline_aux:Nn} +% \begin{macro}[int]{\@@_map_inline:nn} +% \begin{macro}[aux]{\@@_map_inline_aux:Nn} % First obtain the analysis of the token list into -% \cs{g_tl_analysis_result_tl}. To allow nested mappings, increase the -% nesting depth \cs{g_prg_map_int} (shared between all modules), then +% \cs{g_@@_result_tl}. To allow nested mappings, increase the +% nesting depth \cs{g__prg_map_int} (shared between all modules), then % define the looping macro, which has a name specific to that nesting % depth. That looping grabs the \meta{tokens}, \meta{catcode} and % \meta{char code}; it checks for the end of the loop with % \cs{use_none:n} |##2|, normally empty, but which becomes -% \cs{prg_map_break:} at the end; it then performs the user's code +% \cs{tl_map_break:} at the end; it then performs the user's code % |#2|, and loops by calling itself. When the loop ends, remember to % decrease the nesting depth. % \begin{macrocode} -\cs_new_protected:Npn \tl_analysis_map_inline:nn #1 +\cs_new_protected:Npn \@@_map_inline:nn #1 { - \tl_analysis:n {#1} - \int_gincr:N \g_prg_map_int - \exp_args:Nc \tl_analysis_map_inline_aux:Nn - { tl_analysis_map_inline_ \int_use:N \g_prg_map_int :wNw } + \@@:n {#1} + \int_gincr:N \g__prg_map_int + \exp_args:Nc \@@_map_inline_aux:Nn + { @@_map_inline_ \int_use:N \g__prg_map_int :wNw } } -\cs_new_protected:Npn \tl_analysis_map_inline_aux:Nn #1#2 +\cs_new_protected:Npn \@@_map_inline_aux:Nn #1#2 { - \cs_gset_protected:Npn #1 ##1 \s_tl ##2 ##3 \s_tl + \cs_gset_protected:Npn #1 ##1 \s__tl ##2 ##3 \s__tl { \use_none:n ##2 #2 #1 } \exp_after:wN #1 - \g_tl_analysis_result_tl - \s_tl { ? \prg_map_break: } \s_tl - \prg_break_point:n { \int_gdecr:N \g_prg_map_int } + \g_@@_result_tl + \s__tl { ? \tl_map_break: } \s__tl + \__prg_break_point:Nn \tl_map_break: { \int_gdecr:N \g__prg_map_int } } % \end{macrocode} % \end{macro} @@ -946,29 +950,29 @@ % % \begin{macro}{\tl_show_analysis:N} % \begin{macro}{\tl_show_analysis:n} -% Add to \cs{tl_analysis:n} a third pass to display tokens to the terminal. +% Add to \cs{@@:n} a third pass to display tokens to the terminal. % \begin{macrocode} \cs_new_protected:Npn \tl_show_analysis:N #1 { - \exp_args:No \tl_analysis:n {#1} - \msg_aux_show:Nnx #1 + \exp_args:No \@@:n {#1} + \__msg_show_variable:Nnn #1 { tl-analysis } { - \exp_after:wN \tl_show_analysis_loop:wNw \g_tl_analysis_result_tl - \s_tl { ? \prg_map_break: } \s_tl - \prg_break_point:n { } + \exp_after:wN \@@_show_loop:wNw \g_@@_result_tl + \s__tl { ? \__prg_break: } \s__tl + \__prg_break_point: } } \cs_new_protected:Npn \tl_show_analysis:n #1 { - \tl_set:Nn \l_tl_analysis_internal_tl {#1} - \tl_show_analysis:N \l_tl_analysis_internal_tl + \tl_set:Nn \l_@@_internal_tl {#1} + \tl_show_analysis:N \l_@@_internal_tl } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}[aux, rEXP]{\tl_show_analysis_loop:wNw} +% \begin{macro}[aux, rEXP]{\@@_show_loop:wNw} % Here, |#1| \texttt{o}- and \texttt{x}-expands to the token; % |#2| is the category code (one uppercase hexadecimal digit), % $0$ for control sequences; @@ -977,34 +981,34 @@ % the meaning may overflow one line, and we want to truncate % it. Those cases are thus separated out. % \begin{macrocode} -\cs_new:Npn \tl_show_analysis_loop:wNw #1 \s_tl #2 #3 \s_tl +\cs_new:Npn \@@_show_loop:wNw #1 \s__tl #2 #3 \s__tl { \use_none:n #2 \iow_newline: > \c_space_tl \c_space_tl - \if_num:w "#2 = \c_zero - \exp_after:wN \tl_show_analysis_cs:n + \if_int_compare:w "#2 = \c_zero + \exp_after:wN \@@_show_cs:n \else: - \if_num:w "#2 = \c_thirteen + \if_int_compare:w "#2 = \c_thirteen \exp_after:wN \exp_after:wN - \exp_after:wN \tl_show_analysis_active:n + \exp_after:wN \@@_show_active:n \else: \exp_after:wN \exp_after:wN - \exp_after:wN \tl_show_analysis_normal:n + \exp_after:wN \@@_show_normal:n \fi: \fi: {#1} - \tl_show_analysis_loop:wNw + \@@_show_loop:wNw } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux, rEXP]{\tl_show_analysis_normal:n} +% \begin{macro}[aux, rEXP]{\@@_show_normal:n} % Non-active characters are a simple matter of printing % the character, and its meaning. Our test suite checks that % begin-group and end-group characters do not mess up % \TeX{}'s alignment status. % \begin{macrocode} -\cs_new:Npn \tl_show_analysis_normal:n #1 +\cs_new:Npn \@@_show_normal:n #1 { \exp_after:wN \token_to_str:N #1 ~ ( \exp_after:wN \token_to_meaning:N #1 ) @@ -1012,63 +1016,63 @@ % \end{macrocode} % \end{macro} % -% \begin{macro}[EXP]{\tl_show_analysis_value:N} +% \begin{macro}[EXP]{\@@_show_value:N} % This expands to the value of |#1| if it has any. % \begin{macrocode} -\cs_new:Npn \tl_show_analysis_value:N #1 +\cs_new:Npn \@@_show_value:N #1 { \token_if_expandable:NF #1 { - \token_if_chardef:NTF #1 \prg_map_break: { } - \token_if_mathchardef:NTF #1 \prg_map_break: { } - \token_if_dim_register:NTF #1 \prg_map_break: { } - \token_if_int_register:NTF #1 \prg_map_break: { } - \token_if_skip_register:NTF #1 \prg_map_break: { } - \token_if_toks_register:NTF #1 \prg_map_break: { } - \use_none:nnnn - \prg_break_point:n { } + \token_if_chardef:NTF #1 \__prg_break: { } + \token_if_mathchardef:NTF #1 \__prg_break: { } + \token_if_dim_register:NTF #1 \__prg_break: { } + \token_if_int_register:NTF #1 \__prg_break: { } + \token_if_skip_register:NTF #1 \__prg_break: { } + \token_if_toks_register:NTF #1 \__prg_break: { } + \use_none:nnn + \__prg_break_point: \use:n { = \tex_the:D #1 } } } % \end{macrocode} % \end{macro} % -% \begin{macro}[aux, rEXP]{\tl_show_analysis_cs:n} -% \begin{macro}[aux, rEXP]{\tl_show_analysis_active:n} -% \begin{macro}[aux, rEXP]{\tl_show_analysis_long:nn} -% \begin{macro}[aux, rEXP]{\tl_show_analysis_long_aux:nnnn} +% \begin{macro}[aux, rEXP]{\@@_show_cs:n} +% \begin{macro}[aux, rEXP]{\@@_show_active:n} +% \begin{macro}[aux, rEXP]{\@@_show_long:nn} +% \begin{macro}[aux, rEXP]{\@@_show_long_aux:nnnn} % Control sequences and active characters are printed in the same way, -% making sure not to go beyond the \cs{l_iow_line_length_int}. In case +% making sure not to go beyond the \cs{l_iow_line_count_int}. In case % of an overflow, we replace the last characters by -% \cs{c_tl_show_analysis_etc_str}. +% \cs{c_@@_show_etc_str}. % \begin{macrocode} -\cs_new:Npn \tl_show_analysis_cs:n #1 - { \exp_args:No \tl_show_analysis_long:nn {#1} { control~sequence= } } -\cs_new:Npn \tl_show_analysis_active:n #1 - { \exp_args:No \tl_show_analysis_long:nn {#1} { active~character= } } -\cs_new:Npn \tl_show_analysis_long:nn #1 +\cs_new:Npn \@@_show_cs:n #1 + { \exp_args:No \@@_show_long:nn {#1} { control~sequence= } } +\cs_new:Npn \@@_show_active:n #1 + { \exp_args:No \@@_show_long:nn {#1} { active~character= } } +\cs_new:Npn \@@_show_long:nn #1 { - \tl_show_analysis_long_aux:oofn + \@@_show_long_aux:oofn { \token_to_str:N #1 } { \token_to_meaning:N #1 } - { \tl_show_analysis_value:N #1 } + { \@@_show_value:N #1 } } -\cs_new:Npn \tl_show_analysis_long_aux:nnnn #1#2#3#4 +\cs_new:Npn \@@_show_long_aux:nnnn #1#2#3#4 { \int_compare:nNnTF - { \str_length:n { #1 ~ ( #4 #2 #3 ) } } - > { \l_iow_line_length_int - \c_three } + { \str_count:n { #1 ~ ( #4 #2 #3 ) } } + > { \l_iow_line_count_int - \c_three } { - \str_substr:nnn { #1 ~ ( #4 #2 #3 ) } \c_zero + \str_substr:nnn { #1 ~ ( #4 #2 #3 ) } \c_one { - \l_iow_line_length_int - \c_three - - \str_length:N \c_tl_show_analysis_etc_str + \l_iow_line_count_int - \c_three + - \str_count:N \c_@@_show_etc_str } - \c_tl_show_analysis_etc_str + \c_@@_show_etc_str } { #1 ~ ( #4 #2 #3 ) } } -\cs_generate_variant:Nn \tl_show_analysis_long_aux:nnnn { oof } +\cs_generate_variant:Nn \@@_show_long_aux:nnnn { oof } % \end{macrocode} % \end{macro} % \end{macro} @@ -1077,21 +1081,21 @@ % % \subsection{Messages} % -% \begin{variable}{\c_tl_show_analysis_etc_str} +% \begin{variable}{\c_@@_show_etc_str} % When a control sequence (or active character) % and its meaning are too long to fit in one line % of the terminal, the end is replaced by this token list. % \begin{macrocode} -\tl_const:Nx \c_tl_show_analysis_etc_str % ( +\tl_const:Nx \c_@@_show_etc_str % ( { \token_to_str:N \ETC.) } % \end{macrocode} % \end{variable} % % \begin{macrocode} -\msg_kernel_new:nnn { tl-analysis } { show } +\__msg_kernel_new:nnn { kernel } { show-tl-analysis } { The~token~list~ - \str_if_eq:nnF {#1} { \l_tl_analysis_internal_tl } + \str_if_eq:nnF {#1} { \l_@@_internal_tl } { \token_to_str:N #1 ~ } \tl_if_empty:NTF #1 { is~empty } diff --git a/Master/texmf-dist/source/latex/l3experimental/l3str/l3tl-build.dtx b/Master/texmf-dist/source/latex/l3experimental/l3str/l3tl-build.dtx index d3fa2d4e366..ad4d20f7aba 100644 --- a/Master/texmf-dist/source/latex/l3experimental/l3str/l3tl-build.dtx +++ b/Master/texmf-dist/source/latex/l3experimental/l3str/l3tl-build.dtx @@ -36,7 +36,7 @@ %<*driver|package> \RequirePackage{expl3} \GetIdInfo$Id: l3tl-build.dtx 3039 2011-12-08 09:22:35Z bruno $ - {L3 Experimental Build of Token lists} + {L3 Experimental token list construction} % %<*driver> \documentclass[full]{l3doc} @@ -71,7 +71,8 @@ % % \section{\pkg{l3tl-build} documentation} % -% This module provides no user function. +% This module provides no user function: it is meant for kernel use +% only. % % There are two main ways of building token lists from individual % tokens. Either in one go within an \texttt{x}-expanding assignment, or @@ -103,37 +104,37 @@ % % \begin{function} % { -% \tl_set_build:Nw, \tl_gset_build:Nw, -% \tl_set_build_x:Nw, \tl_gset_build_x:Nw +% \__tl_build:Nw, \__tl_gbuild:Nw, +% \__tl_build_x:Nw, \__tl_gbuild_x:Nw % } % \begin{syntax} -% \cs{tl_set_build:Nw} \meta{tl~var} \texttt{\ldots{}} -% \cs{tl_build_one:n} \Arg{tokens_1} \texttt{\ldots{}} -% \cs{tl_build_one:n} \Arg{tokens_2} \texttt{\ldots{}} +% \cs{__tl_build:Nw} \meta{tl~var} \texttt{\ldots{}} +% \cs{__tl_build_one:n} \Arg{tokens_1} \texttt{\ldots{}} +% \cs{__tl_build_one:n} \Arg{tokens_2} \texttt{\ldots{}} % \ldots{} -% \cs{tl_build_end:} +% \cs{__tl_build_end:} % \end{syntax} % Defines the \meta{tl~var} to contain the contens of \meta{tokens1} % followed by \meta{tokens2}, \emph{etc.} This is built in such a way % to be more efficient than repeatedly using \cs{tl_put_right:Nn}. The % code in \enquote{\texttt{\ldots{}}} does not need to be -% expandable. The commands \cs{tl_set_build:Nw} and \cs{tl_build_end:} +% expandable. The commands \cs{__tl_build:Nw} and \cs{__tl_build_end:} % start and end a group. The assignment to the \meta{tl~var} occurs % just after the end of that group, using \cs{tl_set:Nn}, % \cs{tl_gset:Nn}, \cs{tl_set:Nx}, or \cs{tl_gset:Nx}. % \end{function} % -% \begin{function}{\tl_build_one:n, \tl_build_one:o, \tl_build_one:x} +% \begin{function}{\__tl_build_one:n, \__tl_build_one:o, \__tl_build_one:x} % \begin{syntax} -% \cs{tl_build_one:n} \Arg{tokens} +% \cs{__tl_build_one:n} \Arg{tokens} % \end{syntax} % This function may only be used within the scope of a -% \cs{tl_set_build:Nw} function. It adds the \meta{tokens} on the +% \cs{__tl_build:Nw} function. It adds the \meta{tokens} on the % right of the current token list. % \end{function} % -% \begin{function}{\tl_build_end:} -% Ends the scope started by \cs{tl_set_build:Nw}, and performs the +% \begin{function}{\__tl_build_end:} +% Ends the scope started by \cs{__tl_build:Nw}, and performs the % relevant assignment. % \end{function} % @@ -148,36 +149,40 @@ % \end{macrocode} % % \begin{macrocode} +%<@@=tl_build> +% \end{macrocode} +% +% \begin{macrocode} \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} % \end{macrocode} % % \subsection{Variables and helper functions} % -% \begin{variable}{\l_tl_build_start_index_int, \l_tl_build_index_int} +% \begin{variable}{\l_@@_start_index_int, \l_@@_index_int} % Integers pointing to the starting index (currently always starts at % zero), and the current index. The corresponding \tn{toks} are % accessed directly by number. % \begin{macrocode} -\int_new:N \l_tl_build_start_index_int -\int_new:N \l_tl_build_index_int +\int_new:N \l_@@_start_index_int +\int_new:N \l_@@_index_int % \end{macrocode} % \end{variable} % -% \begin{variable}{\l_tl_build_result_tl} +% \begin{variable}{\l_@@_result_tl} % The resulting token list is normally built in one go by unpacking % all \tn{toks} in some range. In the rare cases where there are too -% many \cs{tl_build_one:n} commands, leading to the depletion of +% many \cs{@@_one:n} commands, leading to the depletion of % registers, the contents of the current set of \tn{toks} is unpacked -% into \cs{l_tl_build_result_tl}. This prevents overflow from +% into \cs{l_@@_result_tl}. This prevents overflow from % affecting the end-user (beyond an obvious performance hit). % \begin{macrocode} -\tl_new:N \l_tl_build_result_tl +\tl_new:N \l_@@_result_tl % \end{macrocode} % \end{variable} % -% \begin{macro}{\tl_build_unpack:} -% \begin{macro}[aux, EXP]{\tl_build_unpack_loop:w} +% \begin{macro}{\@@_unpack:} +% \begin{macro}[aux, EXP]{\@@_unpack_loop:w} % The various pieces of the token list are built in \tn{toks} from the % \texttt{start_index} (inclusive) to the (current) \texttt{index} % (excluded). Those \tn{toks} are unpacked and stored in order in the @@ -186,23 +191,23 @@ % expanding to |\the\toks#10...\the\toks#19|, but this should be kept % for much later. % \begin{macrocode} -\cs_new_protected_nopar:Npn \tl_build_unpack: +\cs_new_protected_nopar:Npn \@@_unpack: { - \tl_put_right:Nx \l_tl_build_result_tl + \tl_put_right:Nx \l_@@_result_tl { - \exp_after:wN \tl_build_unpack_loop:w - \int_use:N \l_tl_build_start_index_int ; - \prg_break_point:n { } + \exp_after:wN \@@_unpack_loop:w + \int_use:N \l_@@_start_index_int ; + \__prg_break_point: } } -\cs_new:Npn \tl_build_unpack_loop:w #1 ; +\cs_new:Npn \@@_unpack_loop:w #1 ; { - \if_num:w #1 = \l_tl_build_index_int - \exp_after:wN \prg_map_break: + \if_int_compare:w #1 = \l_@@_index_int + \exp_after:wN \__prg_break: \fi: \tex_the:D \tex_toks:D #1 \exp_stop_f: - \exp_after:wN \tl_build_unpack_loop:w - \int_use:N \int_eval:w #1 + \c_one ; + \exp_after:wN \@@_unpack_loop:w + \int_use:N \__int_eval:w #1 + \c_one ; } % \end{macrocode} % \end{macro} @@ -212,83 +217,83 @@ % % \begin{macro} % { -% \tl_set_build:Nw , \tl_set_build_x:Nw , -% \tl_gset_build:Nw , \tl_gset_build_x:Nw +% \@@:Nw , \@@_x:Nw , +% \__tl_gbuild:Nw , \__tl_gbuild_x:Nw % } -% \begin{macro}[aux]{\tl_set_build_aux:NNw} +% \begin{macro}[aux]{\@@_aux:NNw} % Similar to what is done for coffins: redefine some command, here -% \cs{tl_build_end_aux:n} to hold the relevant assignment (see -% \cs{tl_build_end:} for details). Then initialize the start index and +% \cs{@@_end_aux:n} to hold the relevant assignment (see +% \cs{@@_end:} for details). Then initialize the start index and % the current index at zero, and empty the \texttt{result} token list. % \begin{macrocode} -\cs_new_protected_nopar:Npn \tl_set_build:Nw - { \tl_set_build_aux:NNw \tl_set:Nn } -\cs_new_protected_nopar:Npn \tl_set_build_x:Nw - { \tl_set_build_aux:NNw \tl_set:Nx } -\cs_new_protected_nopar:Npn \tl_gset_build:Nw - { \tl_set_build_aux:NNw \tl_gset:Nn } -\cs_new_protected_nopar:Npn \tl_gset_build_x:Nw - { \tl_set_build_aux:NNw \tl_gset:Nx } -\cs_new_protected:Npn \tl_set_build_aux:NNw #1#2 +\cs_new_protected_nopar:Npn \@@:Nw + { \@@_aux:NNw \tl_set:Nn } +\cs_new_protected_nopar:Npn \@@_x:Nw + { \@@_aux:NNw \tl_set:Nx } +\cs_new_protected_nopar:Npn \__tl_gbuild:Nw + { \@@_aux:NNw \tl_gset:Nn } +\cs_new_protected_nopar:Npn \__tl_gbuild_x:Nw + { \@@_aux:NNw \tl_gset:Nx } +\cs_new_protected:Npn \@@_aux:NNw #1#2 { \group_begin: - \cs_set_nopar:Npn \tl_build_end_assignment:n + \cs_set_nopar:Npn \@@_end_assignment:n { \group_end: #1 #2 } - \int_zero:N \l_tl_build_start_index_int - \int_zero:N \l_tl_build_index_int - \tl_clear:N \l_tl_build_result_tl + \int_zero:N \l_@@_start_index_int + \int_zero:N \l_@@_index_int + \tl_clear:N \l_@@_result_tl } % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}{\tl_build_end:} -% \begin{macro}[aux]{\tl_build_end_assignment:n} +% \begin{macro}{\@@_end:} +% \begin{macro}[aux]{\@@_end_assignment:n} % When we are done building a token list, unpack all \tn{toks} into % the \texttt{result} token list, and expand this list before closing -% the group. The \cs{tl_build_end_assignment:n} function is defined by -% \cs{tl_set_build_aux:NNw} to end the group and hold the relevant +% the group. The \cs{@@_end_assignment:n} function is defined by +% \cs{@@_aux:NNw} to end the group and hold the relevant % assignment. Its value outside is irrelevant, but just in case, we % set it to a function which would clean up the contents of -% \cs{l_tl_build_result_tl}. +% \cs{l_@@_result_tl}. % \begin{macrocode} -\cs_new_protected_nopar:Npn \tl_build_end: +\cs_new_protected_nopar:Npn \@@_end: { - \tl_build_unpack: + \@@_unpack: \exp_args:No - \tl_build_end_assignment:n \l_tl_build_result_tl + \@@_end_assignment:n \l_@@_result_tl } -\cs_new_eq:NN \tl_build_end_assignment:n \use_none:n +\cs_new_eq:NN \@@_end_assignment:n \use_none:n % \end{macrocode} % \end{macro} % \end{macro} % -% \begin{macro}{\tl_build_one:n, \tl_build_one:o, \tl_build_one:x} +% \begin{macro}{\@@_one:n, \@@_one:o, \@@_one:x} % Store the tokens in a free \tn{toks}, then move the pointer to the % next one. If we overflow, unpack the current \tn{toks}, and reset % the current index, preparing to fill more \tn{toks}. This could be % optimized by avoiding to read |#1|, using \tn{afterassignment}. % \begin{macrocode} -\cs_new_protected:Npn \tl_build_one:n #1 +\cs_new_protected:Npn \@@_one:n #1 { - \tex_toks:D \l_tl_build_index_int {#1} - \tex_advance:D \l_tl_build_index_int \c_one - \if_num:w \l_tl_build_index_int > \c_max_register_int - \tl_build_unpack: - \l_tl_build_index_int \l_tl_build_start_index_int + \tex_toks:D \l_@@_index_int {#1} + \tex_advance:D \l_@@_index_int \c_one + \if_int_compare:w \l_@@_index_int > \c_max_register_int + \@@_unpack: + \l_@@_index_int \l_@@_start_index_int \fi: } -\cs_new_protected:Npn \tl_build_one:o #1 +\cs_new_protected:Npn \@@_one:o #1 { - \tex_toks:D \l_tl_build_index_int \exp_after:wN {#1} - \tex_advance:D \l_tl_build_index_int \c_one - \if_num:w \l_tl_build_index_int > \c_max_register_int - \tl_build_unpack: - \l_tl_build_index_int \l_tl_build_start_index_int + \tex_toks:D \l_@@_index_int \exp_after:wN {#1} + \tex_advance:D \l_@@_index_int \c_one + \if_int_compare:w \l_@@_index_int > \c_max_register_int + \@@_unpack: + \l_@@_index_int \l_@@_start_index_int \fi: } -\cs_new_protected:Npn \tl_build_one:x #1 - { \use:x { \tl_build_one:n {#1} } } +\cs_new_protected:Npn \@@_one:x #1 + { \use:x { \@@_one:n {#1} } } % \end{macrocode} % \end{macro} % diff --git a/Master/texmf-dist/source/latex/l3experimental/xcoffins/xcoffins.dtx b/Master/texmf-dist/source/latex/l3experimental/xcoffins/xcoffins.dtx index f8639cd6807..0cee83f61ab 100644 --- a/Master/texmf-dist/source/latex/l3experimental/xcoffins/xcoffins.dtx +++ b/Master/texmf-dist/source/latex/l3experimental/xcoffins/xcoffins.dtx @@ -36,7 +36,7 @@ % %<*driver|package> \RequirePackage{xparse} -\GetIdInfo$Id: xcoffins.dtx 3483 2012-03-03 18:56:29Z bruno $ +\GetIdInfo$Id: xcoffins.dtx 3876 2012-06-30 15:35:39Z joseph $ {L3 Experimental design-level coffins} % %<*driver> @@ -335,7 +335,7 @@ % % \begin{function}{\ResizeCoffin} % \begin{syntax} -% \cs{RotateCoffin} \meta{coffin} \Arg{width} \Arg{total-height} +% \cs{ResizeCoffin} \meta{coffin} \Arg{width} \Arg{total-height} % \end{syntax} % Resized the \meta{coffin} to \meta{width} and \meta{total-height}, % both of which should be given as dimension expressions. These may @@ -618,7 +618,7 @@ % % \begin{function}{\CoffinWidth} % \begin{syntax} -% \cs{CoffinHeight} \meta{coffin} +% \cs{CoffinWidth} \meta{coffin} % \end{syntax} % Calculates the width of the \meta{coffin} in a form % suitable for use in a \meta{dimension expression}, for example @@ -702,6 +702,10 @@ % \begin{macrocode} %<*package> % \end{macrocode} +% +% \begin{macrocode} +%<@@=coffin> +% \end{macrocode} % % \begin{macrocode} \ProvidesExplPackage @@ -711,13 +715,13 @@ % % \begin{variable} % { -% \l_coffin_A_hpole_tl , -% \l_coffin_A_vpole_tl , -% \l_coffin_B_hpole_tl , -% \l_coffin_B_vpole_tl , -% \l_coffin_bound_box_grow_bool , -% \l_coffin_hoffset_dim , -% \l_coffin_voffset_dim +% \l_@@_A_hpole_tl , +% \l_@@_A_vpole_tl , +% \l_@@_B_hpole_tl , +% \l_@@_B_vpole_tl , +% \l_@@_bound_box_grow_bool , +% \l_@@_hoffset_dim , +% \l_@@_voffset_dim % } % Key--value definitions for the alignment system. With the exception % of \texttt{grow-bounding-box}, all of these have to be given with a @@ -725,19 +729,19 @@ % \begin{macrocode} \keys_define:nn { coffin } { - coffin1-hpole .tl_set:N = \l_coffin_A_hpole_tl , + coffin1-hpole .tl_set:N = \l_@@_A_hpole_tl , coffin1-hpole .value_required: , - coffin1-vpole .tl_set:N = \l_coffin_A_vpole_tl , + coffin1-vpole .tl_set:N = \l_@@_A_vpole_tl , coffin1-vpole .value_required: , - coffin2-hpole .tl_set:N = \l_coffin_B_hpole_tl , + coffin2-hpole .tl_set:N = \l_@@_B_hpole_tl , coffin2-hpole .value_required: , - coffin2-vpole .tl_set:N = \l_coffin_B_vpole_tl , + coffin2-vpole .tl_set:N = \l_@@_B_vpole_tl , coffin2-vpole .value_required: , - grow-bounding-box .bool_set:N = \l_coffin_bound_box_grow_bool , + grow-bounding-box .bool_set:N = \l_@@_bound_box_grow_bool , grow-bounding-box .default:n = true , - hoffset .dim_set:N = \l_coffin_hoffset_dim , + hoffset .dim_set:N = \l_@@_hoffset_dim , hoffset .value_required: , - voffset .dim_set:N = \l_coffin_voffset_dim , + voffset .dim_set:N = \l_@@_voffset_dim , voffset .value_required: } \keys_set:nn { coffin } @@ -768,10 +772,10 @@ % \begin{macrocode} \NewDocumentCommand \BoxToCoffin { m } { - \prop_clear_new:c { l_coffin_corners_ \int_value:w #1 _prop } - \prop_clear_new:c { l_coffin_poles_ \int_value:w #1 _prop } - \coffin_update_corners:N #1 - \coffin_update_poles:N #1 + \prop_clear_new:c { l_@@_corners_ \__int_value:w #1 _prop } + \prop_clear_new:c { l_@@_poles_ \__int_value:w #1 _prop } + \@@_update_corners:N #1 + \@@_update_poles:N #1 } % \end{macrocode} % \end{macro} @@ -828,22 +832,22 @@ { \group_begin: \keys_set:nn { coffin } {#1} - \tl_set:Nx \l_coffin_tmp_tl + \tl_set:Nx \l_@@_tmp_tl { \group_end: - \bool_if:NTF \l_coffin_bound_box_grow_bool + \bool_if:NTF \l_@@_bound_box_grow_bool { \coffin_join:NnnNnnnn } { \coffin_attach:NnnNnnnn } \exp_not:N #3 - { \exp_not:o { \l_coffin_A_hpole_tl } } - { \exp_not:o { \l_coffin_A_vpole_tl } } + { \exp_not:o { \l_@@_A_hpole_tl } } + { \exp_not:o { \l_@@_A_vpole_tl } } \exp_not:N #5 - { \exp_not:o { \l_coffin_B_hpole_tl } } - { \exp_not:o { \l_coffin_B_vpole_tl } } - { \dim_use:N \l_coffin_hoffset_dim } - { \dim_use:N \l_coffin_voffset_dim } + { \exp_not:o { \l_@@_B_hpole_tl } } + { \exp_not:o { \l_@@_B_vpole_tl } } + { \dim_use:N \l_@@_hoffset_dim } + { \dim_use:N \l_@@_voffset_dim } } - \l_coffin_tmp_tl + \l_@@_tmp_tl } } % \end{macrocode} @@ -890,7 +894,7 @@ \NewDocumentCommand \CoffinHeight { m } { \coffin_ht:N #1 } \NewDocumentCommand \CoffinTotalHeight { m } - { \dim_eval:w \coffin_ht:N #1 + \coffin_dp:N #1 \dim_eval_end: } + { \__dim_eval:w \coffin_ht:N #1 + \coffin_dp:N #1 \__dim_eval_end: } \NewDocumentCommand \CoffinWidth { m } { \coffin_wd:N #1 } % \end{macrocode} diff --git a/Master/texmf-dist/source/latex/l3experimental/xcoffins/xcoffins.ins b/Master/texmf-dist/source/latex/l3experimental/xcoffins/xcoffins.ins index f57dc329b3c..f4289cf66fc 100644 --- a/Master/texmf-dist/source/latex/l3experimental/xcoffins/xcoffins.ins +++ b/Master/texmf-dist/source/latex/l3experimental/xcoffins/xcoffins.ins @@ -16,7 +16,7 @@ The released version of this bundle is available from CTAN. \fi -\input docstrip.tex +\input l3docstrip.tex \askforoverwritefalse \preamble diff --git a/Master/texmf-dist/source/latex/l3experimental/xgalley/l3galley.dtx b/Master/texmf-dist/source/latex/l3experimental/xgalley/l3galley.dtx index bb07281e01b..77b6456d15e 100644 --- a/Master/texmf-dist/source/latex/l3experimental/xgalley/l3galley.dtx +++ b/Master/texmf-dist/source/latex/l3experimental/xgalley/l3galley.dtx @@ -37,7 +37,7 @@ % %<*driver|package> \RequirePackage{xparse} -\GetIdInfo$Id: l3galley.dtx 3325 2012-02-06 22:07:00Z joseph $ +\GetIdInfo$Id: l3galley.dtx 3918 2012-07-05 00:32:37Z bruno $ {L3 Experimental galley code} % %<*driver> @@ -172,6 +172,12 @@ % \end{function} % % \subsection{Between paragraphs} +% +% \begin{variable}{\g_galley_previous_par_lines_int} +% The number of lines in the preceding conceptual paragraph. This may not +% correspond to the \TeX{} \tn{prevgraf} primitive value as one conceptual +% paragraph may contain several \TeX{} \tn{par} primitive tokens. +% \end{variable} % % \begin{variable}{\g_galley_restore_running_tl} % When galley settings need to be reset at the end of a paragraph, the @@ -298,7 +304,7 @@ % % \begin{variable}{\l_galley_par_indent_dim} % Fixed space added to the start of each paragraph except for those where -% \cs{l_galley_omit_next_indent_bool} is \texttt{true}. +% \cs{g_galley_omit_next_indent_bool} is \texttt{true}. % \end{variable} % % \begin{variable}{\l_galley_last_line_fit_int} @@ -351,6 +357,12 @@ % the interaction of a display block with the paragraphs before and after it % can be adjusted independent of the design of text. % \end{function} +% +% \subsection{Hyphenation} +% +% \begin{variable}{\l_galley_hyphen_left_int} +% THIS IS A HACK: SEE CODE! +% \end{variable} % % \subsection{Line breaking} % @@ -552,9 +564,9 @@ % % \section{Additional effects} % -% \begin{function}{\galley_end_par:n} +% \begin{function}{\@@_end_par:n} % \begin{syntax} -% \cs{galley_end_par:n} \Arg{tokens} +% \cs{@@_end_par:n} \Arg{tokens} % \end{syntax} % Adds the \meta{tokens} to the material collected for the last paragraph % before finalising the last paragraph in the usual way. This function should @@ -655,12 +667,15 @@ % \begin{macrocode} %<*initex|package> % \end{macrocode} +% +% \begin{macrocode} +%<@@=galley> +% \end{macrocode} % % \begin{macrocode} %<*package> \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} -\package_check_loaded_expl: % % \end{macrocode} % @@ -669,14 +684,6 @@ % Functions or settings which are needed by the galley but perhaps also % elsewhere. % -% \begin{macro}{\galley_leave_vmode:} -% The standard mode to leave vertical mode, starting a paragraph. -% \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_leave_vmode: - { \hbox_unpack:N \c_empty_box } -% \end{macrocode} -% \end{macro} -% % The default hyphenation character should be set and hyphenation should be % enabled. % \begin{macrocode} @@ -692,24 +699,24 @@ % % \begin{variable} % { -% \l_galley_parshape_left_indent_clist, -% \l_galley_parshape_right_indent_clist +% \l_@@_parshape_left_indent_clist, +% \l_@@_parshape_right_indent_clist % } % \begin{variable} % { -% \l_galley_parshape_multipar_bool, -% \l_galley_parshape_resume_std_bool, -% \l_galley_parshape_fixed_lines_bool +% \l_@@_parshape_multipar_bool, +% \l_@@_parshape_resume_std_bool, +% \l_@@_parshape_fixed_lines_bool % } % Setting up paragraph shape interacts with setting up the measure. The only % way to keep things flexible is to have a rather \enquote{rich} set of data % available. % \begin{macrocode} -\clist_new:N \l_galley_parshape_left_indent_clist -\clist_new:N \l_galley_parshape_right_indent_clist -\bool_new:N \l_galley_parshape_multipar_bool -\bool_new:N \l_galley_parshape_resume_std_bool -\bool_new:N \l_galley_parshape_fixed_lines_bool +\clist_new:N \l_@@_parshape_left_indent_clist +\clist_new:N \l_@@_parshape_right_indent_clist +\bool_new:N \l_@@_parshape_multipar_bool +\bool_new:N \l_@@_parshape_resume_std_bool +\bool_new:N \l_@@_parshape_fixed_lines_bool \int_new:N \l_galley_parshape_std_lines_int % \end{macrocode} % \end{variable} @@ -745,7 +752,7 @@ % \end{variable} % % \begin{variable} -% {\l_galley_interpar_penalty_skip, \l_galley_interpar_vspace_skip} +% {\l_galley_interpar_penalty_int, \l_galley_interpar_vspace_skip} % Items between paragraphs at the design level. % \begin{macrocode} \int_new:N \l_galley_interpar_penalty_int @@ -770,18 +777,18 @@ % globally. To allow different galley levels to exist, a local variant is % defined for each one to save the value when starting a new level. % -% \begin{variable}{\g_galley_begin_level_bool, \l_galley_begin_level_bool} +% \begin{variable}{\g_@@_begin_level_bool, \l_@@_begin_level_bool} % Indicates that the galley is at the very beginning of the level, and that % no material has yet been set. As a result, the global version is set % \texttt{true} to begin with. % \begin{macrocode} -\bool_new:N \g_galley_begin_level_bool -\bool_new:N \l_galley_begin_level_bool +\bool_new:N \g_@@_begin_level_bool +\bool_new:N \l_@@_begin_level_bool % \end{macrocode} % \end{variable} % % \begin{variable} -% {\g_galley_omit_next_indent_bool, \l_galley_omit_next_indent_bool} +% {\g_galley_omit_next_indent_bool, \l_@@_omit_next_indent_bool} % A global flag is needed for suppressing indentation of the next % paragraph. This does not need a \enquote{running} version since that % should be handled using the \texttt{justification} object: the two @@ -790,65 +797,65 @@ % following a heading. % \begin{macrocode} \bool_new:N \g_galley_omit_next_indent_bool -\bool_new:N \l_galley_omit_next_indent_bool +\bool_new:N \l_@@_omit_next_indent_bool % \end{macrocode} % \end{variable} % -% \begin{variable}{\g_galley_parshape_set_bool, \l_galley_parshape_set_bool} +% \begin{variable}{\g_@@_parshape_set_bool, \l_@@_parshape_set_bool} % This is not a setting for paragraph shape, but rather a tracker for the % galley system. % \begin{macrocode} -\bool_new:N \g_galley_parshape_set_bool -\bool_new:N \l_galley_parshape_set_bool +\bool_new:N \g_@@_parshape_set_bool +\bool_new:N \l_@@_parshape_set_bool % \end{macrocode} % \end{variable} % -% \begin{variable}{\g_galley_nobreak_next_bool, \l_galley_nobreak_next_bool} +% \begin{variable}{\g_galley_no_break_next_bool, \l_@@_no_break_next_bool} % Dealing with the no-break flag is pretty much the same as the case % for the indent: this applies on a single paragraph basis. % \begin{macrocode} -\bool_new:N \g_galley_nobreak_next_bool -\bool_new:N \l_galley_nobreak_next_bool +\bool_new:N \g_galley_no_break_next_bool +\bool_new:N \l_@@_no_break_next_bool % \end{macrocode} % \end{variable} % % \begin{variable} % { -% \g_galley_par_begin_hook_tl, \l_galley_par_begin_hook_tl, -% \g_galley_par_end_hook_tl, \l_galley_par_end_hook_tl, +% \g_galley_par_begin_hook_tl, \l_@@_galley_par_begin_hook_tl, +% \g_galley_par_end_hook_tl, \l_@@_galley_par_end_hook_tl, % } % Hooks for user-level code: these are not used by the galley itself. % \begin{macrocode} \tl_new:N \g_galley_par_begin_hook_tl -\tl_new:N \l_galley_par_begin_hook_tl +\tl_new:N \l_@@_galley_par_begin_hook_tl \tl_new:N \g_galley_par_end_hook_tl -\tl_new:N \l_galley_par_end_hook_tl +\tl_new:N \l_@@_galley_par_end_hook_tl % \end{macrocode} % \end{variable} % % \begin{variable} -% {\g_galley_par_after_hook_tl, \l_galley_par_after_hook_tl} +% {\g_galley_par_after_hook_tl, \l_@@_par_after_hook_tl} % This one is used by the galley: it happens \enquote{after} the current % paragraph, and is used for reset purposes. % \begin{macrocode} \tl_new:N \g_galley_par_after_hook_tl -\tl_new:N \l_galley_par_after_hook_tl +\tl_new:N \l_@@_par_after_hook_tl % \end{macrocode} % \end{variable} % % \begin{variable} -% {\g_galley_previous_par_lines_int, \l_galley_previous_par_lines_int} +% {\g_galley_previous_par_lines_int, \l_@@_previous_par_lines_int} % The number of lines in the previous typeset paragraph. This is reset at % the start of the paragraph and \emph{added to} when each \cs{tex_par:D} % primitive is used: \LaTeX{} uses the primitive in places that do not end % a (conceptual) paragraph. % \begin{macrocode} \int_new:N \g_galley_previous_par_lines_int -\int_new:N \l_galley_previous_par_lines_int +\int_new:N \l_@@_previous_par_lines_int % \end{macrocode} % \end{variable} % -% \begin{variable}{\g_galley_restore_running_tl, \l_galley_restore_running_tl} +% \begin{variable}{\g_galley_restore_running_tl, \l_@@_restore_running_tl} % When a parameter is altered from the \enquote{running} value to a % different \enquote{current} one, there needs to be a method to restore % the \enquote{running} value. This is done by adding the necessary @@ -857,14 +864,14 @@ % structure, and so there has to be a local save version. % \begin{macrocode} \tl_new:N \g_galley_restore_running_tl -\tl_new:N \l_galley_restore_running_tl +\tl_new:N \l_@@_restore_running_tl % \end{macrocode} % \end{variable} % % \begin{variable} % { -% \g_galley_whatsit_next_tl, \l_galley_whatsit_next_tl, -% \g_galley_whatsit_previous_tl, \l_galley_whatsit_previous_tl +% \g_galley_whatsit_next_tl, \l_@@_whatsit_next_tl, +% \g_galley_whatsit_previous_tl, \l_@@_whatsit_previous_tl % } % Whatsits only apply on a per-paragraph basis and so there is no need % to differentiate between current and running values. However, there @@ -873,30 +880,30 @@ % paragraph. % \begin{macrocode} \tl_new:N \g_galley_whatsit_next_tl -\tl_new:N \l_galley_whatsit_next_tl +\tl_new:N \l_@@_whatsit_next_tl \tl_new:N \g_galley_whatsit_previous_tl -\tl_new:N \l_galley_whatsit_previous_tl +\tl_new:N \l_@@_whatsit_previous_tl % \end{macrocode} % \end{variable} % % \begin{variable} -% {\g_galley_interpar_penalty_user_tl, \l_galley_interpar_penalty_user_tl} +% {\g_@@_interpar_penalty_user_tl, \l_@@_interpar_penalty_user_tl} % The user may want to over-ride the penalty for a break between % paragraphs, for example to prevent a break when the overall design % allows one. This is handled using an additional penalty. % \begin{macrocode} -\tl_new:N \g_galley_interpar_penalty_user_tl -\tl_new:N \l_galley_interpar_penalty_user_tl +\tl_new:N \g_@@_interpar_penalty_user_tl +\tl_new:N \l_@@_interpar_penalty_user_tl % \end{macrocode} % \end{variable} % % \begin{variable} -% {\g_galley_interpar_vspace_user_tl, \l_galley_interpar_vspace_user_tl} +% {\g_@@_interpar_vspace_user_tl, \l_@@_interpar_vspace_user_tl} % Arbitrary vertical space can be inserted by the user on a one-off % basis. This is used in place of any running space between paragraphs. % \begin{macrocode} -\tl_new:N \g_galley_interpar_vspace_user_tl -\tl_new:N \l_galley_interpar_vspace_user_tl +\tl_new:N \g_@@_interpar_vspace_user_tl +\tl_new:N \l_@@_interpar_vspace_user_tl % \end{macrocode} % \end{variable} % @@ -908,38 +915,38 @@ % achieved by storing the data structures in \emph{local} variables. The % later are only used to save and restore the global value, and so \TeX{} % grouping will manage the values correctly. This implies that each -% galley level must form a group: galley levels are tided to vertical +% galley level must form a group: galley levels are tied to vertical % boxes and so this is a reasonable requirements. % -% \begin{macro}{\galley_initialise_variables:} +% \begin{macro}{\@@_initialise_variables:} % At the start of a galley level, both the global and local variables % will need to be reset to standard values. For example, the measure % is set to the galley width and any paragraph shape is cleared. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_initialise_variables: +\cs_new_protected_nopar:Npn \@@_initialise_variables: { - \bool_gset_true:N \g_galley_begin_level_bool - \tl_gclear:N \g_galley_interpar_penalty_user_tl - \tl_gclear:N \g_galley_interpar_vspace_user_tl + \bool_gset_true:N \g_@@_begin_level_bool + \tl_gclear:N \g_@@_interpar_penalty_user_tl + \tl_gclear:N \g_@@_interpar_vspace_user_tl \bool_gset_true:N \g_galley_omit_next_indent_bool - \bool_gset_false:N \g_galley_nobreak_next_bool + \bool_gset_false:N \g_galley_no_break_next_bool \tl_gclear:N \g_galley_par_begin_hook_tl \tl_gclear:N \g_galley_par_end_hook_tl \tl_gclear:N \g_galley_par_after_hook_tl - \bool_gset_false:N \g_galley_parshape_set_bool + \bool_gset_false:N \g_@@_parshape_set_bool \int_gzero:N \g_galley_previous_par_lines_int \tl_gclear:N \g_galley_restore_running_tl \tl_gclear:N \g_galley_whatsit_previous_tl \tl_gclear:N \g_galley_whatsit_next_tl } -\galley_initialise_variables: +\@@_initialise_variables: % \end{macrocode} % \end{macro} % -% \begin{macro}{\galley_initialise_settings:} +% \begin{macro}{\@@_initialise_settings:} % This sets the local values of the various galley settings. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_initialise_settings: +\cs_new_protected_nopar:Npn \@@_initialise_settings: { \dim_set_eq:NN \l_galley_text_width_dim \l_galley_width_dim \dim_zero:N \l_galley_total_left_margin_dim @@ -948,84 +955,84 @@ % \end{macrocode} % \end{macro} % -% \begin{macro}{\galley_save_parameters:, \galley_restore_parameters:} +% \begin{macro}{\@@_save_parameters:, \@@_restore_parameters:} % Saving and restoring parameters is carried out by a series of copy % functions. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_save_parameters: +\cs_new_protected_nopar:Npn \@@_save_parameters: { - \bool_set_eq:NN \l_galley_begin_level_bool - \g_galley_begin_level_bool - \tl_set_eq:NN \l_galley_interpar_penalty_user_tl - \g_galley_interpar_penalty_user_tl - \tl_set_eq:NN \l_galley_interpar_vspace_user_tl - \g_galley_interpar_vspace_user_tl - \bool_set_eq:NN \l_galley_omit_next_indent_bool + \bool_set_eq:NN \l_@@_begin_level_bool + \g_@@_begin_level_bool + \tl_set_eq:NN \l_@@_interpar_penalty_user_tl + \g_@@_interpar_penalty_user_tl + \tl_set_eq:NN \l_@@_interpar_vspace_user_tl + \g_@@_interpar_vspace_user_tl + \bool_set_eq:NN \l_@@_omit_next_indent_bool \g_galley_omit_next_indent_bool - \bool_set_eq:NN \l_galley_nobreak_next_bool - \g_galley_nobreak_next_bool - \tl_set_eq:NN \l_galley_par_begin_hook_tl + \bool_set_eq:NN \l_@@_no_break_next_bool + \g_galley_no_break_next_bool + \tl_set_eq:NN \l_@@_galley_par_begin_hook_tl \g_galley_par_begin_hook_tl - \tl_set_eq:NN \l_galley_par_end_hook_tl + \tl_set_eq:NN \l_@@_galley_par_end_hook_tl \g_galley_par_end_hook_tl - \tl_set_eq:NN \l_galley_par_after_hook_tl + \tl_set_eq:NN \l_@@_par_after_hook_tl \g_galley_par_after_hook_tl - \bool_set_eq:NN \l_galley_parshape_set_bool - \g_galley_parshape_set_bool - \int_set_eq:NN \l_galley_previous_par_lines_int + \bool_set_eq:NN \l_@@_parshape_set_bool + \g_@@_parshape_set_bool + \int_set_eq:NN \l_@@_previous_par_lines_int \g_galley_previous_par_lines_int - \tl_set_eq:NN \l_galley_restore_running_tl + \tl_set_eq:NN \l_@@_restore_running_tl \g_galley_restore_running_tl - \tl_set_eq:NN \l_galley_whatsit_previous_tl + \tl_set_eq:NN \l_@@_whatsit_previous_tl \g_galley_whatsit_previous_tl - \tl_set_eq:NN \l_galley_whatsit_next_tl + \tl_set_eq:NN \l_@@_whatsit_next_tl \g_galley_whatsit_next_tl } -\cs_new_protected_nopar:Npn \galley_restore_parameters: +\cs_new_protected_nopar:Npn \@@_restore_parameters: { - \bool_gset_eq:NN \g_galley_begin_level_bool - \l_galley_begin_level_bool - \tl_gset_eq:NN \g_galley_interpar_penalty_user_tl - \l_galley_interpar_penalty_user_tl - \tl_gset_eq:NN \g_galley_interpar_vspace_user_tl - \l_galley_interpar_vspace_user_tl + \bool_gset_eq:NN \g_@@_begin_level_bool + \l_@@_begin_level_bool + \tl_gset_eq:NN \g_@@_interpar_penalty_user_tl + \l_@@_interpar_penalty_user_tl + \tl_gset_eq:NN \g_@@_interpar_vspace_user_tl + \l_@@_interpar_vspace_user_tl \bool_gset_eq:NN \g_galley_omit_next_indent_bool - \l_galley_omit_next_indent_bool - \bool_gset_eq:NN \g_galley_nobreak_next_bool - \l_galley_nobreak_next_bool + \l_@@_omit_next_indent_bool + \bool_gset_eq:NN \g_galley_no_break_next_bool + \l_@@_no_break_next_bool \tl_gset_eq:NN \g_galley_par_begin_hook_tl - \l_galley_par_begin_hook_tl + \l_@@_galley_par_begin_hook_tl \tl_gset_eq:NN \g_galley_par_end_hook_tl - \l_galley_par_end_hook_tl + \l_@@_galley_par_end_hook_tl \tl_gset_eq:NN \g_galley_par_after_hook_tl - \l_galley_par_after_hook_tl - \bool_gset_eq:NN \g_galley_parshape_set_bool - \l_galley_parshape_set_bool + \l_@@_par_after_hook_tl + \bool_gset_eq:NN \g_@@_parshape_set_bool + \l_@@_parshape_set_bool \int_gset_eq:NN \g_galley_previous_par_lines_int - \l_galley_previous_par_lines_int + \l_@@_previous_par_lines_int \tl_gset_eq:NN \g_galley_restore_running_tl - \l_galley_restore_running_tl + \l_@@_restore_running_tl \tl_gset_eq:NN \g_galley_whatsit_previous_tl - \l_galley_whatsit_previous_tl + \l_@@_whatsit_previous_tl \tl_gset_eq:NN \g_galley_whatsit_next_tl - \l_galley_whatsit_next_tl + \l_@@_whatsit_next_tl } % \end{macrocode} % \end{macro} % % \begin{macro}{\galley_level:} -% \begin{macro}[aux]{\galley_level_end:} +% \begin{macro}[aux]{\@@_level_end:} % Galley levels are created by saving all of the current global % settings, starting a group then initialising both the local and global % variables. % \begin{macrocode} \cs_new_protected_nopar:Npn \galley_level: { - \galley_save_parameters: + \@@_save_parameters: \group_begin: - \galley_initialise_variables: - \galley_initialise_settings: - \group_insert_after:N \galley_level_end: + \@@_initialise_variables: + \@@_initialise_settings: + \group_insert_after:N \@@_level_end: } % \end{macrocode} % At the end of the level, the global values are restored using the @@ -1035,10 +1042,10 @@ % must come in a fixed location. All of this relies on the the % \enquote{colour safe} group used inside a box. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_level_end: +\cs_new_protected_nopar:Npn \@@_level_end: { \par - \galley_restore_parameters: + \@@_restore_parameters: \group_end: } % \end{macrocode} @@ -1047,44 +1054,44 @@ % % \subsection{The \cs{par} token} % -% \begin{variable}{\s_par_omit} +% \begin{variable}{\s__par_omit} % Used to indicate that a paragraph should be omitted. % \begin{macrocode} -\scan_new:N \s_par_omit +\__scan_new:N \s__par_omit % \end{macrocode} % \end{variable} % -% \begin{macro}{\galley_std_par:} -% \begin{macro}[aux]{\galley_std_par_aux_i:, \galley_std_par_aux_ii:} -% \begin{macro}[aux]{\galley_std_par_aux:N} +% \begin{macro}{\@@_std_par:} +% \begin{macro}[aux]{\@@_std_par_aux_i:, \@@_std_par_aux_ii:} +% \begin{macro}[aux]{\@@_std_par_aux:N} % The idea here is to expand the next token in exactly the same way as \TeX{} % would do anyway. The \texttt{f}-type expansion will ignore any protection, % but will stop at a scan marker. Thus the code can test for an % \enquote{omit paragraph} marker. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_std_par: +\cs_new_protected_nopar:Npn \@@_std_par: { - \s_par_omit - \exp_after:wN \galley_std_par_aux_i: \tex_romannumeral:D - `0 + \s__par_omit + \exp_after:wN \@@_std_par_aux_i: \tex_romannumeral:D - `0 } -\cs_new_protected:Npn \galley_std_par_aux_i: +\cs_new_protected:Npn \@@_std_par_aux_i: { - \peek_meaning:NTF \s_par_omit - { \galley_std_par_aux:N } - { \galley_std_par_aux_ii: } + \peek_meaning:NTF \s__par_omit + { \@@_std_par_aux:N } + { \@@_std_par_aux_ii: } } -\cs_new_protected:Npn \galley_std_par_aux:N #1 +\cs_new_protected:Npn \@@_std_par_aux:N #1 { - \str_if_eq:xxF {#1} { \s_par_omit } + \str_if_eq_x:nnF {#1} { \s__par_omit } { - \galley_std_par_aux_ii: + \@@_std_par_aux_ii: #1 } } % \end{macrocode} % No marker, so really insert a paragraph. In vertical mode, % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_std_par_aux_ii: +\cs_new_protected_nopar:Npn \@@_std_par_aux_ii: { \mode_if_vertical:TF { \tex_par:D } @@ -1100,7 +1107,7 @@ % \begin{macrocode} { \g_galley_par_end_hook_tl - \galley_set_measure_and_parshape: + \@@_set_measure_and_parshape: \group_begin: \tex_par:D \group_end: @@ -1121,17 +1128,17 @@ % \end{macro} % \end{macro} % -% \begin{macro}{\galley_end_par:n} +% \begin{macro}{\@@_end_par:n} % Inserts tokens such that they are appended to the end of the last % paragraph, using the paragraph-omitting system. % \begin{macrocode} -\cs_new_protected:Npn \galley_end_par:n #1 +\cs_new_protected:Npn \@@_end_par:n #1 { - \s_par_omit - \bool_if:nF \g_galley_begin_level_bool + \s__par_omit + \bool_if:nF \g_@@_begin_level_bool { #1 - \galley_std_par: + \@@_std_par: } } % \end{macrocode} @@ -1141,7 +1148,7 @@ % The meaning of the token \cs{par} itself starts off as a standard % paragraph. % \begin{macrocode} -\cs_set_protected_nopar:Npn \par { \galley_std_par: } +\cs_set_protected_nopar:Npn \par { \@@_std_par: } % \end{macrocode} % \end{macro} % @@ -1151,7 +1158,7 @@ % so this will only be needed in package mode. % \begin{macrocode} %<*package> -\tl_set:Nn \@par { \galley_std_par: } +\tl_set:Nn \@par { \@@_std_par: } % % \end{macrocode} % \end{macro} @@ -1159,8 +1166,8 @@ % \subsection{Display levels} % % \begin{macro}{\galley_display_begin:, \galley_display_end:} -% \begin{macro}[aux]{\galley_display_penalty:N, \galley_display_vspace:N} -% \begin{macro}[aux]{\galley_display_par_setup:, \galley_display_par:} +% \begin{macro}[aux]{\@@_display_penalty:N, \@@_display_vspace:N} +% \begin{macro}[aux]{\@@_display_par_setup:, \@@_display_par:} % Display items within the galley are a bit like galley levels: they % may have different paragraph settings to the main part of the galley. % On the other hand, unlike independent galleys they should inherit the @@ -1170,15 +1177,15 @@ \cs_new_protected_nopar:Npn \galley_display_begin: { \group_begin: - \galley_save_parameters: + \@@_save_parameters: \mode_if_vertical:TF { - \galley_display_penalty:N \l_galley_display_begin_par_penalty_tl - \galley_display_vspace:N \l_galley_display_begin_par_vspace_tl + \@@_display_penalty:N \l_galley_display_begin_par_penalty_tl + \@@_display_vspace:N \l_galley_display_begin_par_vspace_tl } { - \galley_display_penalty:N \l_galley_display_begin_penalty_tl - \galley_display_vspace:N \l_galley_display_begin_vspace_tl + \@@_display_penalty:N \l_galley_display_begin_penalty_tl + \@@_display_vspace:N \l_galley_display_begin_vspace_tl } \par } @@ -1188,7 +1195,7 @@ % token list, before setting up the value to the special value needed % in this one case. % \begin{macrocode} -\cs_new_protected:Npn \galley_display_penalty:N #1 +\cs_new_protected:Npn \@@_display_penalty:N #1 { \tl_if_empty:NF #1 { @@ -1200,7 +1207,7 @@ \int_gset:Nn \g_galley_penalty_int {#1} } } -\cs_new_protected:Npn \galley_display_vspace:N #1 +\cs_new_protected:Npn \@@_display_vspace:N #1 { \tl_if_empty:NF #1 { @@ -1223,12 +1230,12 @@ \cs_new_protected_nopar:Npn \galley_display_end: { \par - \galley_restore_parameters: + \@@_restore_parameters: \group_end: %<*package> \group_insert_after:N \group_insert_after:N % - \group_insert_after:N \galley_display_par_setup: + \group_insert_after:N \@@_display_par_setup: } % \end{macrocode} % The method used here is to assume that the next piece of horizontal @@ -1237,10 +1244,10 @@ % \cs{par} token is then altered so that a check can be made to see if % this assumption was correct. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_display_par_setup: +\cs_new_protected_nopar:Npn \@@_display_par_setup: { \bool_gset_false:N \g_galley_omit_next_indent_bool - \cs_set_eq:NN \par \galley_display_par: + \cs_set_eq:NN \par \@@_display_par: } % \end{macrocode} % The \enquote{special} meaning of the paragraph token starts by putting @@ -1253,24 +1260,24 @@ % case where \TeX{} is in horizontal mode then a \cs{tex_par:D} primitive % is required in the same way as in \cs{galley_standard_par:}. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_display_par: +\cs_new_protected_nopar:Npn \@@_display_par: { - \cs_set_eq:NN \par \galley_std_par: + \cs_set_eq:NN \par \@@_std_par: \mode_if_vertical:TF { \par \bool_gset_false:N \g_galley_omit_next_indent_bool - \galley_display_penalty:N \l_galley_display_end_par_penalty_tl - \galley_display_vspace:N \l_galley_display_end_par_vspace_tl + \@@_display_penalty:N \l_galley_display_end_par_penalty_tl + \@@_display_vspace:N \l_galley_display_end_par_vspace_tl } { - \galley_set_measure_and_parshape: + \@@_set_measure_and_parshape: \group_begin: \tex_par:D \group_end: \int_gadd:Nn \g_galley_previous_par_lines_int \tex_prevgraf:D - \galley_display_penalty:N \l_galley_display_end_penalty_tl - \galley_display_vspace:N \l_galley_display_end_vspace_tl + \@@_display_penalty:N \l_galley_display_end_penalty_tl + \@@_display_vspace:N \l_galley_display_end_vspace_tl } } % \end{macrocode} @@ -1298,9 +1305,9 @@ %<*initex> \tex_everypar:D % TEMP { - \bool_if:NTF \g_galley_begin_level_bool - { \galley_start_paragraph_first: } - { \galley_start_paragraph_std: } + \bool_if:NTF \g_@@_begin_level_bool + { \@@_start_paragraph_first: } + { \@@_start_paragraph_std: } } % %<*package> @@ -1310,9 +1317,9 @@ { \tex_everypar:D { - \bool_if:NTF \g_galley_begin_level_bool - { \galley_start_paragraph_first: } - { \galley_start_paragraph_std: } + \bool_if:NTF \g_@@_begin_level_bool + { \@@_start_paragraph_first: } + { \@@_start_paragraph_std: } \tex_the:D \everypar } } @@ -1322,13 +1329,13 @@ % % \subsection{The galley mechanism} % -% \begin{variable}{\g_galley_last_box} +% \begin{variable}{\g_@@_last_box} % A temporary box to hold the box inserted by \TeX{} when a paragraph % is inserted with an indent. The galley actually inserts the space % (\emph{i.e.}~\cs{tex_parindent:D} is globally zero), but there is % still an empty box to test for. % \begin{macrocode} -\box_new:N \g_galley_last_box +\box_new:N \g_@@_last_box % \end{macrocode} % \end{variable} % @@ -1341,7 +1348,7 @@ % \end{verbatim} % and so anything applied here must be done globally. % -% \begin{macro}[int]{\galley_start_paragraph_std:} +% \begin{macro}[int]{\@@_start_paragraph_std:} % The routine at the start of a paragraph starts by removing any % (empty) indent box from the vertical list. As there may be vertical % mode items still to insert, a \cs{tex_par:D} primitive is used to @@ -1352,48 +1359,48 @@ % tidied up before sorting out any items which have been set on a % single-paragraph basis. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_start_paragraph_std: +\cs_new_protected_nopar:Npn \@@_start_paragraph_std: { \group_begin: - \box_gset_to_last:N \g_galley_last_box + \box_gset_to_last:N \g_@@_last_box \tex_par:D - \galley_insert_vertical_items: + \@@_insert_vertical_items: \tex_everypar:D { } \tex_noindent:D \group_end: \int_gzero:N \g_galley_previous_par_lines_int - \galley_insert_horizontal_items: - \galley_restore_running_parameters: + \@@_insert_horizontal_items: + \@@_restore_running_parameters: } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\galley_start_paragraph_first:} +% \begin{macro}[int]{\@@_start_paragraph_first:} % For the very first paragraph in a galley, the code needs to avoid % adding any unnecessary vertical items at the top as it will interfere with % vertical positioning in \cs{tex_vtop:D}. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_start_paragraph_first: +\cs_new_protected_nopar:Npn \@@_start_paragraph_first: { - \bool_gset_false:N \g_galley_begin_level_bool + \bool_gset_false:N \g_@@_begin_level_bool \mode_if_horizontal:TF { \group_begin: - \box_gset_to_last:N \g_galley_last_box + \box_gset_to_last:N \g_@@_last_box \tex_par:D - \galley_insert_vspace: + \@@_insert_vspace: \tex_everypar:D { } \tex_noindent:D \group_end: } - { \galley_insert_vspace: } - \galley_insert_horizontal_items: - \galley_restore_running_parameters: + { \@@_insert_vspace: } + \@@_insert_horizontal_items: + \@@_restore_running_parameters: } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\galley_insert_vertical_items, \galley_insert_vspace:} +% \begin{macro}[int]{\@@_insert_vertical_items, \@@_insert_vspace:} % The aim here is to insert the vertical items such that they attach to % the correct place. This function is used as part of the \cs{tex_everypar:D} % mechanism, meaning that the immediately-preceding item on the vertical @@ -1404,52 +1411,52 @@ % or the running penalty unless the no-break flag is set. Finally, % the inter-paragraph space is applied. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_insert_vertical_items: +\cs_new_protected_nopar:Npn \@@_insert_vertical_items: { \g_galley_whatsit_previous_tl \tl_gclear:N \g_galley_whatsit_previous_tl - \tl_if_empty:NTF \g_galley_interpar_penalty_user_tl + \tl_if_empty:NTF \g_@@_interpar_penalty_user_tl { - \bool_if:NTF \g_galley_nobreak_next_bool + \bool_if:NTF \g_galley_no_break_next_bool { \tex_penalty:D \c_ten_thousand } { \tex_penalty:D \l_galley_interpar_penalty_int } } { \tex_penalty:D - \int_eval:w \g_galley_interpar_penalty_user_tl \int_eval_end: - \tl_gclear:N \g_galley_interpar_penalty_user_tl + \__int_eval:w \g_@@_interpar_penalty_user_tl \__int_eval_end: + \tl_gclear:N \g_@@_interpar_penalty_user_tl } - \bool_gset_false:N \g_galley_nobreak_next_bool - \galley_insert_vspace: + \bool_gset_false:N \g_galley_no_break_next_bool + \@@_insert_vspace: } % \end{macrocode} % Inserting vertical space is set up as a separate function as it comes % up in a few places. The idea here is that any user-set space will % override the design value, and only one space is ever inserted. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_insert_vspace: +\cs_new_protected_nopar:Npn \@@_insert_vspace: { - \tl_if_empty:NTF \g_galley_interpar_vspace_user_tl + \tl_if_empty:NTF \g_@@_interpar_vspace_user_tl { \skip_vertical:N \l_galley_interpar_vspace_skip } { - \skip_vertical:n { \g_galley_interpar_vspace_user_tl } - \tl_gclear:N \g_galley_interpar_vspace_user_tl + \skip_vertical:n { \g_@@_interpar_vspace_user_tl } + \tl_gclear:N \g_@@_interpar_vspace_user_tl } } % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\galley_insert_horizontal_items:} +% \begin{macro}[int]{\@@_insert_horizontal_items:} % Horizontal mode objects start with the whatsits for the next paragraph. An % indent is then included if the removed box was not void. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_insert_horizontal_items: +\cs_new_protected_nopar:Npn \@@_insert_horizontal_items: { \g_galley_whatsit_next_tl \tl_gclear:N \g_galley_whatsit_next_tl \bool_if:NF \g_galley_omit_next_indent_bool { - \box_if_empty:NF \g_galley_last_box + \box_if_empty:NF \g_@@_last_box { \hbox_to_wd:nn \l_galley_par_indent_dim { } } } \skip_horizontal:N \l_galley_par_begin_skip @@ -1459,12 +1466,12 @@ % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\galley_restore_running_parameters:} +% \begin{macro}[int]{\@@_restore_running_parameters:} % Restoring the ongoing parameters just means using the token list % variable in which the appropriate assignments are stored. The % list can then be cleared. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_restore_running_parameters: +\cs_new_protected_nopar:Npn \@@_restore_running_parameters: { \g_galley_restore_running_tl \tl_gclear:N \g_galley_restore_running_tl @@ -1521,32 +1528,32 @@ % \begin{macrocode} \cs_new_protected:Npn \galley_parshape_fixed_lines:nnn #1#2#3 { - \bool_gset_true:N \g_galley_parshape_set_bool - \bool_set_true:N \l_galley_parshape_fixed_lines_bool + \bool_gset_true:N \g_@@_parshape_set_bool + \bool_set_true:N \l_@@_parshape_fixed_lines_bool \int_set:Nn \l_galley_parshape_std_lines_int {#1} - \clist_set:Nn \l_galley_parshape_left_indent_clist {#2} - \clist_set:Nn \l_galley_parshape_right_indent_clist {#3} - \bool_set_true:N \l_galley_parshape_resume_std_bool + \clist_set:Nn \l_@@_parshape_left_indent_clist {#2} + \clist_set:Nn \l_@@_parshape_right_indent_clist {#3} + \bool_set_true:N \l_@@_parshape_resume_std_bool } \cs_new_protected:Npn \galley_parshape_multi_par:nnnN #1#2#3#4 { - \bool_gset_true:N \g_galley_parshape_set_bool - \bool_set_true:N \l_galley_parshape_multipar_bool - \bool_set_false:N \l_galley_parshape_fixed_lines_bool + \bool_gset_true:N \g_@@_parshape_set_bool + \bool_set_true:N \l_@@_parshape_multipar_bool + \bool_set_false:N \l_@@_parshape_fixed_lines_bool \int_set:Nn \l_galley_parshape_std_lines_int {#1} - \clist_set:Nn \l_galley_parshape_left_indent_clist {#2} - \clist_set:Nn \l_galley_parshape_right_indent_clist {#3} - \bool_set_eq:NN \l_galley_parshape_resume_std_bool #4 + \clist_set:Nn \l_@@_parshape_left_indent_clist {#2} + \clist_set:Nn \l_@@_parshape_right_indent_clist {#3} + \bool_set_eq:NN \l_@@_parshape_resume_std_bool #4 } \cs_new_protected:Npn \galley_parshape_single_par:nnnN #1#2#3#4 { - \bool_gset_true:N \g_galley_parshape_set_bool - \bool_set_false:N \l_galley_parshape_multipar_bool - \bool_set_false:N \l_galley_parshape_fixed_lines_bool + \bool_gset_true:N \g_@@_parshape_set_bool + \bool_set_false:N \l_@@_parshape_multipar_bool + \bool_set_false:N \l_@@_parshape_fixed_lines_bool \int_set:Nn \l_galley_parshape_std_lines_int {#1} - \clist_set:Nn \l_galley_parshape_left_indent_clist {#2} - \clist_set:Nn \l_galley_parshape_right_indent_clist {#3} - \bool_set_eq:NN \l_galley_parshape_resume_std_bool #4 + \clist_set:Nn \l_@@_parshape_left_indent_clist {#2} + \clist_set:Nn \l_@@_parshape_right_indent_clist {#3} + \bool_set_eq:NN \l_@@_parshape_resume_std_bool #4 } \cs_generate_variant:Nn \galley_parshape_fixed_lines:nnn { nVV } \cs_generate_variant:Nn \galley_parshape_multi_par:nnnN { nVV } @@ -1554,25 +1561,25 @@ % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\galley_set_measure_and_parshape:} +% \begin{macro}[int]{\@@_set_measure_and_parshape:} % To set the paragraph shape for the current paragraph, there is a check to % see if the measure alone should be used. If not, then the shape may be % built by paragraph or based on the number of lines required. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_set_measure_and_parshape: +\cs_new_protected_nopar:Npn \@@_set_measure_and_parshape: { - \bool_if:NTF \g_galley_parshape_set_bool + \bool_if:NTF \g_@@_parshape_set_bool { - \bool_if:NTF \l_galley_parshape_fixed_lines_bool + \bool_if:NTF \l_@@_parshape_fixed_lines_bool { \int_compare:nNnTF \g_galley_previous_par_lines_int > \c_zero - { \galley_generate_parshape_lines: } - { \galley_generate_parshape: } + { \@@_generate_parshape_lines: } + { \@@_generate_parshape: } } { - \bool_gset_eq:NN \g_galley_parshape_set_bool - \l_galley_parshape_multipar_bool - \galley_generate_parshape: + \bool_gset_eq:NN \g_@@_parshape_set_bool + \l_@@_parshape_multipar_bool + \@@_generate_parshape: } } { @@ -1586,23 +1593,23 @@ % \end{macrocode} % \end{macro} % -% \begin{macro}[int]{\galley_generate_parshape:} -% \begin{macro}[aux]{\galley_set_parshape_map:nn, \galley_set_parshape_map:oo} -% \begin{macro}[aux]{\galley_set_parshape_map_aux:nw} +% \begin{macro}[int]{\@@_generate_parshape:} +% \begin{macro}[aux]{\@@_set_parshape_map:nn, \@@_set_parshape_map:oo} +% \begin{macro}[aux]{\@@_set_parshape_map_aux:nw} % For a shape to apply on a paragraph basis, the two user-supplied comma % lists are taken and converted into left-side offsets and line lengths. % This is all dependent on the current measure. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_generate_parshape: +\cs_new_protected_nopar:Npn \@@_generate_parshape: { \tex_global:D \tex_parshape:D - \int_eval:w + \__int_eval:w \l_galley_parshape_std_lines_int + \int_min:nn - { \clist_length:N \l_galley_parshape_left_indent_clist } - { \clist_length:N \l_galley_parshape_right_indent_clist } - \bool_if:NT \l_galley_parshape_resume_std_bool { + 1 } - \int_eval_end: + { \clist_count:N \l_@@_parshape_left_indent_clist } + { \clist_count:N \l_@@_parshape_right_indent_clist } + \bool_if:NT \l_@@_parshape_resume_std_bool { + 1 } + \__int_eval_end: \prg_replicate:nn \l_galley_parshape_std_lines_int { \dim_use:N \l_galley_total_left_margin_dim @@ -1610,10 +1617,10 @@ \dim_use:N \l_galley_text_width_dim \c_space_tl } - \galley_set_parshape_map:oo - \l_galley_parshape_left_indent_clist - \l_galley_parshape_right_indent_clist - \bool_if:NT \l_galley_parshape_resume_std_bool + \@@_set_parshape_map:oo + \l_@@_parshape_left_indent_clist + \l_@@_parshape_right_indent_clist + \bool_if:NT \l_@@_parshape_resume_std_bool { \c_space_tl \dim_use:N \l_galley_total_left_margin_dim @@ -1621,10 +1628,10 @@ \dim_use:N \l_galley_text_width_dim } } -\cs_new:Npn \galley_set_parshape_map:nn #1#2 - { \galley_set_parshape_map_aux:nw { } #1 , \q_mark #2 , \q_stop } -\cs_generate_variant:Nn \galley_set_parshape_map:nn { oo } -\cs_new:Npn \galley_set_parshape_map_aux:nw #1#2 , #3 \q_mark #4 , #5 \q_stop +\cs_new:Npn \@@_set_parshape_map:nn #1#2 + { \@@_set_parshape_map_aux:nw { } #1 , \q_mark #2 , \q_stop } +\cs_generate_variant:Nn \@@_set_parshape_map:nn { oo } +\cs_new:Npn \@@_set_parshape_map_aux:nw #1#2 , #3 \q_mark #4 , #5 \q_stop { \bool_if:nTF { \tl_if_empty_p:n {#3} || \tl_if_empty_p:n {#5} } { @@ -1634,7 +1641,7 @@ \dim_eval:n { \l_galley_text_width_dim - ( ( #2 ) + ( #4 ) ) } } { - \galley_set_parshape_map_aux:nw + \@@_set_parshape_map_aux:nw { #1 \dim_eval:n { \l_galley_total_left_margin_dim + ( #2 ) } @@ -1650,27 +1657,27 @@ % \end{macro} % \end{macro} % -% \begin{macro}[int]{\galley_generate_parshape_lines:} -% \begin{macro}[aux]{\galley_generate_parshape_lines_aux:n} +% \begin{macro}[int]{\@@_generate_parshape_lines:} +% \begin{macro}[aux]{\@@_generate_parshape_lines:n} % The idea here is to construct a paragraph shape based on the remaining % lines from the shape in the previous paragraph. If the previous paragraph % was sufficiently long, then life is \enquote{back to normal} and the % standard shape is set. If a special shape is needed, this is recovered % from the paragraph shape using the \eTeX{} primitives. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_generate_parshape_lines: +\cs_new_protected_nopar:Npn \@@_generate_parshape_lines: { \int_compare:nNnTF \tex_parshape:D > \g_galley_previous_par_lines_int { \tex_global:D \tex_parshape:D - \int_eval:w \tex_parshape:D - \g_galley_previous_par_lines_int - \int_eval_end: - \prg_stepwise_function:nnnN + \__int_eval:w \tex_parshape:D - \g_galley_previous_par_lines_int + \__int_eval_end: + \int_step_function:nnnN { \g_galley_previous_par_lines_int + \c_one } - \c_one \tex_parshape:D \galley_generate_parshape_lines_aux:n + \c_one \tex_parshape:D \@@_generate_parshape_lines:n } { - \bool_gset_false:N \g_galley_parshape_set_bool + \bool_gset_false:N \g_@@_parshape_set_bool \tex_global:D \tex_parshape:D \c_one \dim_use:N \l_galley_total_left_margin_dim @@ -1678,7 +1685,7 @@ \dim_use:N \l_galley_text_width_dim } } -\cs_new:Npn \galley_generate_parshape_lines_aux:n #1 +\cs_new:Npn \@@_generate_parshape_lines:n #1 { \etex_parshapeindent:D #1 ~ @@ -1697,9 +1704,9 @@ % deals with this nicely. % \begin{macrocode} \cs_new_protected:Npn \galley_set_user_penalty:n #1 - { \tl_gset:Nx \g_galley_interpar_penalty_user_tl { \int_eval:n {#1} } } + { \tl_gset:Nx \g_@@_interpar_penalty_user_tl { \int_eval:n {#1} } } \cs_new_protected:Npn \galley_set_user_vspace:n #1 - { \tl_gset:Nx \g_galley_interpar_vspace_user_tl { \skip_eval:n {#1} } } + { \tl_gset:Nx \g_@@_interpar_vspace_user_tl { \skip_eval:n {#1} } } % \end{macrocode} % \end{macro} % @@ -1837,7 +1844,7 @@ \cs_new_protected:Npn \galley_break_line:Nn #1#2 { \mode_if_vertical:TF - { \msg_kernel_error:nn { galley } { no-line-to-end } } + { \__msg_kernel_error:nn { galley } { no-line-to-end } } { \tex_unskip:D \bool_if:NF #1 @@ -1905,7 +1912,7 @@ \cs_new_protected:Npn \galley_set_display_widow_penalties:n #1 { \etex_displaywidowpenalties:D - \int_eval:w \clist_length:n {#1} + \c_one \int_eval_end: + \__int_eval:w \clist_count:n {#1} + \c_one \__int_eval_end: \clist_map_function:nN {#1} \galley_set_aux:n \c_zero } @@ -1913,7 +1920,7 @@ \cs_new_protected:Npn \galley_set_widow_penalties:n #1 { \etex_widowpenalties:D - \int_eval:w \clist_length:n {#1} + \c_one \int_eval_end: + \__int_eval:w \clist_count:n {#1} + \c_one \__int_eval_end: \clist_map_function:nN {#1} \galley_set_aux:n \c_zero } @@ -1940,13 +1947,13 @@ \cs_new_protected:Npn \galley_set_club_penalties:n #1 { \clist_set:Nn \l_galley_club_penalties_clist {#1} - \galley_calc_interline_penalties: + \@@_calc_interline_penalties: } \cs_generate_variant:Nn \galley_set_club_penalties:n { V , v } \cs_new_protected:Npn \galley_set_interline_penalties:n #1 { \clist_set:Nn \l_galley_line_penalties_clist {#1} - \galley_calc_interline_penalties: + \@@_calc_interline_penalties: } \cs_generate_variant:Nn \galley_set_interline_penalties:n { V , v } % \end{macrocode} @@ -1964,21 +1971,21 @@ \cs_new_protected:Npn \galley_set_display_club_penalties:n #1 { \etex_clubpenalties:D - \int_eval:w \clist_length:n {#1} + \c_one \int_eval_end: + \__int_eval:w \clist_count:n {#1} + \c_one \__int_eval_end: \clist_map_function:nN {#1} \galley_set_aux:n \c_zero - \galley_calc_interline_penalties: + \@@_calc_interline_penalties: } \cs_generate_variant:Nn \galley_set_display_club_penalties:n { V , v } % \end{macrocode} % \end{macro} % % \begin{macro}{\galley_set_interline_penalty:n} -% \begin{macro}[aux]{\galley_set_interline_penalty_aux:nn} +% \begin{macro}[aux]{\@@_set_interline_penalty:nn} % \begin{macro}[aux] % { -% \galley_set_interline_penalty_aux_i:n, -% \galley_set_interline_penalty_aux_ii:n +% \@@_set_interline_penalty_i:n, +% \@@_set_interline_penalty_ii:n % } % Dealing with the general interline penalty is handled in one shot. % The idea is that for lines with no special penalty, the old general @@ -1991,47 +1998,47 @@ { \int_compare:nNnTF { \etex_interlinepenalties:D \c_zero } = \c_zero { - \etex_interlinepenalties:D \c_one \int_eval:w #1 \int_eval_end: - \galley_calc_interline_penalties: + \etex_interlinepenalties:D \c_one \__int_eval:w #1 \__int_eval_end: + \@@_calc_interline_penalties: } { - \cs_set:Npn \galley_set_interline_penalty_aux_ii:n ##1 + \cs_set:Npn \@@_set_interline_penalty_ii:n ##1 { - \int_eval:w + \__int_eval:w \etex_interlinepenalties:D ##1 - \etex_interlinepenalties:D \etex_interlinepenalties:D \c_zero + #1 - \int_eval_end: + \__int_eval_end: } - \exp_args:Nf \galley_set_interline_penalty_aux:nn - { \clist_length:N \l_galley_line_penalties_clist } {#1} + \exp_args:Nf \@@_set_interline_penalty:nn + { \clist_count:N \l_galley_line_penalties_clist } {#1} } } -\cs_new_protected:Npn \galley_set_interline_penalty_aux:nn #1#2 +\cs_new_protected:Npn \@@_set_interline_penalty:nn #1#2 { \etex_interlinepenalties:D \etex_interlinepenalties:D \c_zero - \prg_stepwise_function:nnnN \c_one \c_one {#1} - \galley_set_interline_penalty_aux_i:n - \prg_stepwise_function:nnnN { #1 + \c_one } \c_one + \int_step_function:nnnN \c_one \c_one {#1} + \@@_set_interline_penalty_i:n + \int_step_function:nnnN { #1 + \c_one } \c_one { \etex_interlinepenalties:D \c_zero - \c_one } - \galley_set_interline_penalty_aux_ii:n - \int_eval:w #2 \int_eval_end: + \@@_set_interline_penalty_ii:n + \__int_eval:w #2 \__int_eval_end: } -\cs_new:Npn \galley_set_interline_penalty_aux_i:n #1 - { \etex_interlinepenalties:D \int_eval:w #1 \int_eval_end: } -\cs_new:Npn \galley_set_interline_penalty_aux_ii:n #1 { } +\cs_new:Npn \@@_set_interline_penalty_i:n #1 + { \etex_interlinepenalties:D \__int_eval:w #1 \__int_eval_end: } +\cs_new:Npn \@@_set_interline_penalty_ii:n #1 { } % \end{macrocode} % \end{macro} % \end{macro} % \end{macro} % -% \begin{macro}[int]{\galley_calc_interline_penalties:} -% \begin{macro}[aux]{\galley_calc_interline_penalties_aux:nn} +% \begin{macro}[int]{\@@_calc_interline_penalties:} +% \begin{macro}[aux]{\@@_calc_interline_penalties:nn} % \begin{macro}[aux] % { -% \galley_calc_interline_penalties_aux_i:n, -% \galley_calc_interline_penalties_aux_ii:n +% \@@_calc_interline_penalties_i:n, +% \@@_calc_interline_penalties_ii:n % } % The underlying interline penalty array has to deal with club penalties, % display club penalties and any special line penalties, and include @@ -2040,28 +2047,28 @@ % so an \texttt{f}-type expansion is used to make life a little less % complex. % \begin{macrocode} -\cs_new_protected_nopar:Npn \galley_calc_interline_penalties: +\cs_new_protected_nopar:Npn \@@_calc_interline_penalties: { - \exp_args:Nff \galley_calc_interline_penalties_aux:nn + \exp_args:Nff \@@_calc_interline_penalties:nn { \int_eval:n { \int_max:nn { - \clist_length:N \l_galley_club_penalties_clist + \clist_count:N \l_galley_club_penalties_clist + \c_one } { \int_max:nn { - \clist_length:N \l_galley_line_penalties_clist + \clist_count:N \l_galley_line_penalties_clist + \c_one } { \etex_clubpenalties:D \c_zero } } } } - { \clist_length:N \l_galley_line_penalties_clist } + { \clist_count:N \l_galley_line_penalties_clist } } % \end{macrocode} % The idea is now to calculate the correct penalties. Two auxiliary functions @@ -2069,32 +2076,30 @@ % normal lines. At the end of the process, the standard interline % penalty is always included. % \begin{macrocode} -\cs_new_protected:Npn \galley_calc_interline_penalties_aux:nn #1#2 +\cs_new_protected:Npn \@@_calc_interline_penalties:nn #1#2 { \etex_interlinepenalties:D #1 ~ - \prg_stepwise_function:nnnN \c_one \c_one {#2} - \galley_calc_interline_penalties_aux_i:n - \prg_stepwise_function:nnnN { #2 + \c_one } \c_one { #1 - \c_one } - \galley_calc_interline_penalties_aux_ii:n + \int_step_function:nnnN \c_one \c_one {#2} + \@@_calc_interline_penalties_i:n + \int_step_function:nnnN { #2 + \c_one } \c_one { #1 - \c_one } + \@@_calc_interline_penalties_ii:n \etex_interlinepenalties:D \etex_interlinepenalties:D \c_zero } -\cs_new:Npn \galley_calc_interline_penalties_aux_i:n #1 +\cs_new:Npn \@@_calc_interline_penalties_i:n #1 { - \int_eval:w - \clist_item:Nn \l_galley_line_penalties_clist { #1 - \c_one } - + 0 \clist_item:Nn \l_galley_club_penalties_clist - { #1 - \c_one } + \__int_eval:w + \clist_item:Nn \l_galley_line_penalties_clist {#1} + + 0 \clist_item:Nn \l_galley_club_penalties_clist {#1} - \etex_clubpenalties:D #1 ~ - \int_eval_end: + \__int_eval_end: } -\cs_new:Npn \galley_calc_interline_penalties_aux_ii:n #1 +\cs_new:Npn \@@_calc_interline_penalties_ii:n #1 { - \int_eval:w + \__int_eval:w \etex_interlinepenalties:D \etex_interlinepenalties:D \c_zero - + 0 \clist_item:Nn \l_galley_club_penalties_clist - { #1 - \c_one } + + 0 \clist_item:Nn \l_galley_club_penalties_clist {#1} - \etex_clubpenalties:D #1 ~ - \int_eval_end: + \__int_eval_end: } % \end{macrocode} % \end{macro} @@ -2111,9 +2116,9 @@ % } % \begin{macro}[aux] % { -% \galley_save_display_club_penalties_aux:n, -% \galley_save_display_widow_penalties_aux:n, -% \galley_save_widow_penalties_aux:n +% \@@_save_display_club_penalties:n, +% \@@_save_display_widow_penalties:n, +% \@@_save_widow_penalties:n % } % \begin{macro}{\galley_interline_penalty:} % Saving the array penalties varies in complexity depending on how they are @@ -2131,35 +2136,35 @@ { \clist_set:Nx #1 { - \prg_stepwise_function:nnnN \c_one \c_one + \int_step_function:nnnN \c_one \c_one { \etex_clubpenalties:D \c_zero - \c_one } \galley_save_display_club_penalties:_aux:n } } \cs_new:Npn \galley_save_display_club_penalties:_aux:n #1 - { \int_use:N \etex_clubpenaltes:D \int_eval:w #1 \int_eval_end: , } + { \int_use:N \etex_clubpenaltes:D \__int_eval:w #1 \__int_eval_end: , } \cs_new_protected:Npn \galley_save_display_widow_penalties:N #1 { \clist_set:Nx #1 { - \prg_stepwise_function:nnnN \c_one \c_one + \int_step_function:nnnN \c_one \c_one { \etex_displaywidowpenalties:D \c_zero - \c_one } \galley_save_display_widow_penalties:_aux:n } } \cs_new:Npn \galley_save_display_widow_penalties:_aux:n #1 - { \int_use:N \etex_displaywidowpenalties:D \int_eval:w #1 \int_eval_end: , } + { \int_use:N \etex_displaywidowpenalties:D \__int_eval:w #1 \__int_eval_end: , } \cs_new_protected:Npn \galley_save_widow_penalties:N #1 { \clist_set:Nx #1 { - \prg_stepwise_function:nnnN \c_one \c_one + \int_step_function:nnnN \c_one \c_one { \etex_widowpenalties:D \c_zero - \c_one } \galley_save_widow_penalties:_aux:n } } \cs_new:Npn \galley_save_widow_penalties:_aux:n #1 - { \int_use:N \etex_widowpenalties:D \int_eval:w #1 \int_eval_end: , } + { \int_use:N \etex_widowpenalties:D \__int_eval:w #1 \__int_eval_end: , } % \end{macrocode} % This one is not an array, but is stored in a primitive, so there is % a simple conversion. The general interline penalty is always the @@ -2175,7 +2180,7 @@ % \subsection{Messages} % % \begin{macrocode} -\msg_kernel_new:nnn { galley } { no-line-to-end } +\__msg_kernel_new:nnn { galley } { no-line-to-end } { There's~no~line~here~to~end. } % \end{macrocode} % @@ -2262,7 +2267,7 @@ % \begin{macro}{\@@par} % The primitive is moved as otherwise the clever skipping code will fail. % \begin{macrocode} -\cs_set_eq:NN \@@par \galley_std_par: +\cs_set_eq:NN \@@par \@@_std_par: % \end{macrocode} % \end{macro} % diff --git a/Master/texmf-dist/source/latex/l3experimental/xgalley/xgalley.dtx b/Master/texmf-dist/source/latex/l3experimental/xgalley/xgalley.dtx index bba0907bbd8..87b30d0ce25 100644 --- a/Master/texmf-dist/source/latex/l3experimental/xgalley/xgalley.dtx +++ b/Master/texmf-dist/source/latex/l3experimental/xgalley/xgalley.dtx @@ -37,7 +37,7 @@ % %<*driver|package> \RequirePackage{l3names} -\GetIdInfo$Id: xgalley.dtx 3253 2012-01-28 14:59:55Z joseph $ +\GetIdInfo$Id: xgalley.dtx 3879 2012-07-01 09:17:47Z joseph $ {L3 Experimental galley} % %<*driver> @@ -758,6 +758,10 @@ % \begin{macrocode} %<*package> % \end{macrocode} +% +% \begin{macrocode} +%<@@=galley> +% \end{macrocode} % % \begin{macrocode} \ProvidesExplPackage @@ -767,11 +771,11 @@ % % \subsection{Variables} % -% \begin{variable}{\l_galley_tmpa_clist, \l_galley_tmpb_clist} +% \begin{variable}{\l_@@_tmpa_clist, \l_@@_tmpb_clist} % Scratch space. % \begin{macrocode} -\clist_new:N \l_galley_tmpa_clist -\clist_new:N \l_galley_tmpb_clist +\clist_new:N \l_@@_tmpa_clist +\clist_new:N \l_@@_tmpb_clist % \end{macrocode} % \end{variable} % @@ -799,26 +803,26 @@ } % \end{macrocode} % -% \begin{variable}{\l_galley_left_margin_dim, \l_galley_right_margin_dim} -% In the \texttt{absolute} template, the two margin values are relative -% to the edges of the galley. This means that any existing offset or -% line-length adjustment are ignored. +% \begin{variable}{\l_@@_left_margin_dim, \l_@@_right_margin_dim} +% In the \texttt{absolute} template, the two margin values are relative +% to the edges of the galley. This means that any existing offset or +% line-length adjustment are ignored. % \begin{macrocode} %<*package> -\cs_new_eq:NN \l_galley_left_margin_dim \leftmargin +\cs_new_eq:NN \l_@@_left_margin_dim \leftmargin % %<*package> -\cs_new_eq:NN \l_galley_right_margin_dim \rightmargin +\cs_new_eq:NN \l_@@_right_margin_dim \rightmargin % \DeclareTemplateCode { measure } { absolute } { 0 } { - left-margin = \l_galley_left_margin_dim , - right-margin = \l_galley_right_margin_dim + left-margin = \l_@@_left_margin_dim , + right-margin = \l_@@_right_margin_dim } { \AssignTemplateKeys - \galley_margins_set_absolute:nn \l_galley_left_margin_dim - \l_galley_right_margin_dim + \galley_margins_set_absolute:nn \l_@@_left_margin_dim + \l_@@_right_margin_dim } % \end{macrocode} % On the other hand, the \texttt{relative} template works relative to @@ -826,13 +830,13 @@ % \begin{macrocode} \DeclareTemplateCode { measure } { relative } { 0 } { - left-margin = \l_galley_left_margin_dim , - right-margin = \l_galley_right_margin_dim + left-margin = \l_@@_left_margin_dim , + right-margin = \l_@@_right_margin_dim } { \AssignTemplateKeys - \galley_margins_set_relative:nn \l_galley_left_margin_dim - \l_galley_right_margin_dim + \galley_margins_set_relative:nn \l_@@_left_margin_dim + \l_@@_right_margin_dim } % \end{macrocode} % \end{variable} @@ -866,66 +870,66 @@ lines : integer = 2 } % \end{macrocode} -% \begin{variable}{\l_galley_parshape_indent_dim} -% \begin{variable}{\l_galley_parshape_on_left_bool} -% \begin{variable}{\l_galley_parshape_lines_int} -% Both of the templates are implemented as special cases of the more -% general function defined earlier. +% \begin{variable}{\l_@@_parshape_indent_dim} +% \begin{variable}{\l_@@_parshape_on_left_bool} +% \begin{variable}{\l_@@_parshape_lines_int} +% Both of the templates are implemented as special cases of the more +% general function defined earlier. % \begin{macrocode} \DeclareTemplateCode { parshape } { hang } { 0 } { - indent = \l_galley_parshape_indent_dim , - on-left-side = \l_galley_parshape_on_left_bool , - lines = \l_galley_parshape_lines_int + indent = \l_@@_parshape_indent_dim , + on-left-side = \l_@@_parshape_on_left_bool , + lines = \l_@@_parshape_lines_int } { \AssignTemplateKeys - \bool_if:NTF \l_galley_parshape_on_left_bool + \bool_if:NTF \l_@@_parshape_on_left_bool { \galley_parshape_single_par:nVVN - \l_galley_parshape_lines_int - \l_galley_parshape_indent_dim + \l_@@_parshape_lines_int + \l_@@_parshape_indent_dim \c_zero_dim \c_false_bool } { \galley_parshape_single_par:nVVN - \l_galley_parshape_lines_int + \l_@@_parshape_lines_int \c_zero_dim - \l_galley_parshape_indent_dim + \l_@@_parshape_indent_dim \c_false_bool } } \DeclareTemplateCode { parshape } { initial } { 0 } { - indent = \l_galley_parshape_indent_dim , - on-left-side = \l_galley_parshape_on_left_bool , - lines = \l_galley_parshape_lines_int + indent = \l_@@_parshape_indent_dim , + on-left-side = \l_@@_parshape_on_left_bool , + lines = \l_@@_parshape_lines_int } { \AssignTemplateKeys - \clist_clear:N \l_galley_tmpa_clist - \clist_clear:N \l_galley_tmpb_clist - \prg_replicate:nn { \l_galley_parshape_lines_int } + \clist_clear:N \l_@@_tmpa_clist + \clist_clear:N \l_@@_tmpb_clist + \prg_replicate:nn { \l_@@_parshape_lines_int } { - \clist_put_right:Nn \l_galley_tmpa_clist - { \l_galley_parshape_indent_dim } - \clist_put_right:Nn \l_galley_tmpb_clist + \clist_put_right:Nn \l_@@_tmpa_clist + { \l_@@_parshape_indent_dim } + \clist_put_right:Nn \l_@@_tmpb_clist { \c_zero_dim } } - \bool_if:NTF \l_galley_parshape_on_left_bool + \bool_if:NTF \l_@@_parshape_on_left_bool { \galley_parshape_single_par:nVVN \c_zero - \l_galley_tmpa_clist - \l_galley_tmpb_clist + \l_@@_tmpa_clist + \l_@@_tmpb_clist \c_true_bool } { \galley_parshape_single_par:nVVN \c_zero - \l_galley_tmpb_clist - \l_galley_tmpa_clist + \l_@@_tmpb_clist + \l_@@_tmpa_clist \c_true_bool } } @@ -1012,12 +1016,12 @@ \tex_defaulthyphenchar:D 45 \scan_stop: % \end{macrocode} % -% \begin{variable}{\l_galley_justification_other_tl} +% \begin{variable}{\l_@@_justification_other_tl} % Used for the reset system for justification: using this token list means % that there is no need to remove anything from % \cs{g_galley_restore_running_tl}. % \begin{macrocode} -\tl_new:N \l_galley_justification_other_tl +\tl_new:N \l_@@_justification_other_tl % \end{macrocode} % \end{variable} % @@ -1068,7 +1072,7 @@ } { \AssignTemplateKeys - \tl_clear:N \l_galley_justification_other_tl + \tl_clear:N \l_@@_justification_other_tl \galley_set_interword_spacing:N \l_galley_fixed_spacing_bool \bool_if:NTF \l_galley_par_stretch_last_bool { \int_set_eq:NN \l_galley_last_line_fit_int \c_one_thousand } @@ -1092,7 +1096,7 @@ stretch-last-line = \l_galley_par_stretch_last_bool } { - \tl_put_left:Nx \l_galley_justification_other_tl + \tl_put_left:Nx \l_@@_justification_other_tl { \skip_set:Nn \exp_not:N \l_galley_par_end_skip { \skip_use:N \l_galley_par_end_skip } @@ -1115,7 +1119,7 @@ { \skip_use:N \l_galley_line_right_skip } } \tl_gput_right:Nn \g_galley_restore_running_tl - { \l_galley_justification_other_tl } + { \l_@@_justification_other_tl } \AssignTemplateKeys \galley_set_interword_spacing:N \l_galley_fixed_spacing_bool \bool_if:NTF \l_galley_par_stretch_last_bool @@ -1191,7 +1195,7 @@ } % \end{macrocode} % -% \begin{macro}{\galley_justification_first:, \galley_justification_other:} +% \begin{macro}{\@@_justification_first:, \@@_justification_other:} % A second form of justification template is the case where the first % paragraph is different from all of the others. This is set up by % getting the justification to reset itself after the first paragraph. @@ -1205,16 +1209,16 @@ } \DeclareTemplateCode { justification } { compound } { 0 } { - first-paragraph = \galley_justification_first: , - other-paragraphs = \galley_justification_other: + first-paragraph = \@@_justification_first: , + other-paragraphs = \@@_justification_other: } { \AssignTemplateKeys - \galley_justification_first: - \tl_set:Nn \l_galley_justification_other_tl - { \galley_justification_other: } + \@@_justification_first: + \tl_set:Nn \l_@@_justification_other_tl + { \@@_justification_other: } \tl_gput_right:Nn \g_galley_restore_running_tl - { \l_galley_justification_other_tl } + { \l_@@_justification_other_tl } } % \end{macrocode} % \end{macro} @@ -1251,15 +1255,15 @@ \DeclareTemplateCode{ line-breaking } { std } { 0 } { badness = \l_galley_linebreak_badness_int , - binop-penalty = \l_galley_binop_penalty_int , + binop-penalty = \l_@@_binop_penalty_int , double-hyphen-demerits = \l_galley_double_hyphen_demerits_int , emergency-stretch = \l_galley_emergency_stretch_skip , final-hyphen-demerits = \l_galley_final_hyphen_demerits_int , fuzz = \l_galley_linebreak_fuzz_dim , - line-penalty = \l_galley_linebreak_penalty_int , + line-penalty = \l_@@_linebreak_penalty_int , mismatch-demerits = \l_galley_mismatch_demerits_int , pretolerance = \l_galley_linebreak_pretolerance_int , - relation-penalty = \l_galley_relation_penalty_int , + relation-penalty = \l_@@_relation_penalty_int , tolerance = \l_galley_linebreak_tolerance_int } { \AssignTemplateKeys } @@ -1278,11 +1282,11 @@ % % \begin{variable} % { -% \l_galley_club_penalty_int , -% \l_galley_display_club_penalty_int , -% \l_galley_display_widow_penalty_int , -% \l_galley_interline_penalty_int , -% \l_galley_widow_penalty_int +% \l_@@_club_penalty_int , +% \l_@@_display_club_penalty_int , +% \l_@@_display_widow_penalty_int , +% \l_@@_interline_penalty_int , +% \l_@@_widow_penalty_int % } % The second object here sets up how \TeX{} acts to break paragraphs at % page boundaries. As with the \texttt{line-breaking} object, there is not @@ -1307,23 +1311,23 @@ \DeclareTemplateCode { paragraph-breaking } { std } { 0 } { badness = \l_galley_parbreak_badness_int , - broken-penalty = \l_galley_broken_penalty_int , - club-penalty = \l_galley_club_penalty_int , - display-club-penalty = \l_galley_display_club_penalty_int , - display-widow-penalty = \l_galley_display_widow_penalty_int , + broken-penalty = \l_@@_broken_penalty_int , + club-penalty = \l_@@_club_penalty_int , + display-club-penalty = \l_@@_display_club_penalty_int , + display-widow-penalty = \l_@@_display_widow_penalty_int , fuzz = \l_galley_parbreak_fuzz_dim , - interline-penalty = \l_galley_interline_penalty_int , - post-display-penalty = \l_galley_post_display_penalty_int , - pre-display-penalty = \l_galley_pre_display_penalty_int , - widow-penalty = \l_galley_widow_penalty_int + interline-penalty = \l_@@_interline_penalty_int , + post-display-penalty = \l_@@_post_display_penalty_int , + pre-display-penalty = \l_@@_pre_display_penalty_int , + widow-penalty = \l_@@_widow_penalty_int } { \AssignTemplateKeys - \galley_set_club_penalties:V \l_galley_club_penalty_int - \galley_set_display_club_penalties:V \l_galley_display_club_penalty_int - \galley_set_display_widow_penalties:V \l_galley_display_widow_penalty_int - \galley_set_interline_penalty:n \l_galley_interline_penalty_int - \galley_set_widow_penalties:V \l_galley_widow_penalty_int + \galley_set_club_penalties:V \l_@@_club_penalty_int + \galley_set_display_club_penalties:V \l_@@_display_club_penalty_int + \galley_set_display_widow_penalties:V \l_@@_display_widow_penalty_int + \galley_set_interline_penalty:n \l_@@_interline_penalty_int + \galley_set_widow_penalties:V \l_@@_widow_penalty_int } % \end{macrocode} % \end{variable} @@ -1351,18 +1355,18 @@ % \end{macrocode} % \begin{variable} % { -% \l_galley_parbreak_badness_tl , -% \l_galley_broken_penalty_tl , -% \l_galley_club_penalties_tl , -% \l_galley_display_club_penalties_tl , -% \l_galley_display_widow_penalties_tl , -% \l_galley_parbreak_fuzz_tl , -% \l_galley_interline_penalty_tl , -% \l_galley_post_display_penalty_tl , -% \l_galley_pre_display_penalty_tl , -% \l_galley_widow_penalties_tl , -% \c_galley_parbreak_multi_seq , -% \c_galley_parbreak_single_seq +% \l_@@_parbreak_badness_tl , +% \l_@@_broken_penalty_tl , +% \l_@@_club_penalties_tl , +% \l_@@_display_club_penalties_tl , +% \l_@@_display_widow_penalties_tl , +% \l_@@_parbreak_fuzz_tl , +% \l_@@_interline_penalty_tl , +% \l_@@_post_display_penalty_tl , +% \l_@@_pre_display_penalty_tl , +% \l_@@_widow_penalties_tl , +% \c_@@_parbreak_multi_seq , +% \c_@@_parbreak_single_seq % } % There is also a version of this code which applies only to one % paragraph. This is done by storing the input in token list variables @@ -1383,16 +1387,16 @@ } \DeclareTemplateCode { paragraph-breaking } { single } { 0 } { - badness = \l_galley_parbreak_badness_tl , - broken-penalty = \l_galley_broken_penalty_tl , - club-penalty = \l_galley_club_penalties_tl , - display-club-penalty = \l_galley_display_club_penalties_tl , - display-widow-penalty = \l_galley_display_widow_penalties_tl , - fuzz = \l_galley_parbreak_fuzz_tl , - interline-penalty = \l_galley_interline_penalty_tl , - post-display-penalty = \l_galley_post_display_penalty_tl , - pre-display-penalty = \l_galley_pre_display_penalty_tl , - widow-penalty = \l_galley_widow_penalties_tl + badness = \l_@@_parbreak_badness_tl , + broken-penalty = \l_@@_broken_penalty_tl , + club-penalty = \l_@@_club_penalties_tl , + display-club-penalty = \l_@@_display_club_penalties_tl , + display-widow-penalty = \l_@@_display_widow_penalties_tl , + fuzz = \l_@@_parbreak_fuzz_tl , + interline-penalty = \l_@@_interline_penalty_tl , + post-display-penalty = \l_@@_post_display_penalty_tl , + pre-display-penalty = \l_@@_pre_display_penalty_tl , + widow-penalty = \l_@@_widow_penalties_tl } { \AssignTemplateKeys @@ -1400,30 +1404,30 @@ % The fuzz and interline penalties are handled explicitly as they have % particular requirements. % \begin{macrocode} - \tl_if_empty:NF \l_galley_interline_penalty_tl + \tl_if_empty:NF \l_@@_interline_penalty_tl { \tl_gput_right:Nx \g_galley_par_after_hook_tl { - \int_set:Nn \exp_not:N \l_galley_interline_penalty_int + \int_set:Nn \exp_not:N \l_@@_interline_penalty_int { \galley_interline_penalty: } } - \int_set:Nn \l_galley_interline_penalty_int - { \l_galley_interline_penalty_tl } + \int_set:Nn \l_@@_interline_penalty_int + { \l_@@_interline_penalty_tl } } - \tl_if_empty:NF \l_galley_parbreak_fuzz_tl + \tl_if_empty:NF \l_@@_parbreak_fuzz_tl { \tl_gput_right:Nx \g_galley_par_after_hook_tl { \dim_set:Nn \exp_not:N \l_galley_parbreak_fuzz_dim { \dim_use:N \l_galley_parbreak_fuzz_dim } } - \dim_set:Nn \l_galley_parbreak_fuzz_dim { \l_galley_parbreak_fuzz_tl } + \dim_set:Nn \l_galley_parbreak_fuzz_dim { \l_@@_parbreak_fuzz_tl } } % \end{macrocode} % For the single integer penalties, a simple check is needed to save the % value. % \begin{macrocode} - \seq_map_inline:Nn \c_galley_parbreak_single_seq + \seq_map_inline:Nn \c_@@_parbreak_single_seq { \tl_if_empty:cF { l_galley_ ##1 _tl } { @@ -1440,30 +1444,30 @@ % A bit more complex for the array penalties. Although the interface here % does not expose the arrays, it is necessary to correctly save them. % \begin{macrocode} - \seq_map_inline:Nn \c_galley_parbreak_multi_seq + \seq_map_inline:Nn \c_@@_parbreak_multi_seq { \tl_if_empty:cF { l_galley_ ##1 _tl } { - \use:c { galley_save_ ##1 :N } \l_galley_tmpa_clist + \use:c { galley_save_ ##1 :N } \l_@@_tmpa_clist \tl_gput_right:Nx \g_galley_par_after_hook_tl { \exp_not:c { galley_set_ ##1 :n } - { \exp_not:o \l_galley_tmpa_clist } + { \exp_not:o \l_@@_tmpa_clist } } \use:c { galley_set_ ##1 :v } { l_galley_ ##1 _tl } } } } -\seq_new:N \c_galley_parbreak_multi_seq -\seq_gput_right:Nn \c_galley_parbreak_multi_seq { club_penalties } -\seq_gput_right:Nn \c_galley_parbreak_multi_seq { display_club_penalties } -\seq_gput_right:Nn \c_galley_parbreak_multi_seq { display_widow_penalties } -\seq_gput_right:Nn \c_galley_parbreak_multi_seq { widow_penalties } -\seq_new:N \c_galley_parbreak_single_seq -\seq_gput_right:Nn \c_galley_parbreak_single_seq { parbreak_badness } -\seq_gput_right:Nn \c_galley_parbreak_single_seq { broken_penalty } -\seq_gput_right:Nn \c_galley_parbreak_single_seq { post_display_penalty } -\seq_gput_right:Nn \c_galley_parbreak_single_seq { pre_display_penalty } +\seq_new:N \c_@@_parbreak_multi_seq +\seq_gput_right:Nn \c_@@_parbreak_multi_seq { club_penalties } +\seq_gput_right:Nn \c_@@_parbreak_multi_seq { display_club_penalties } +\seq_gput_right:Nn \c_@@_parbreak_multi_seq { display_widow_penalties } +\seq_gput_right:Nn \c_@@_parbreak_multi_seq { widow_penalties } +\seq_new:N \c_@@_parbreak_single_seq +\seq_gput_right:Nn \c_@@_parbreak_single_seq { parbreak_badness } +\seq_gput_right:Nn \c_@@_parbreak_single_seq { broken_penalty } +\seq_gput_right:Nn \c_@@_parbreak_single_seq { post_display_penalty } +\seq_gput_right:Nn \c_@@_parbreak_single_seq { pre_display_penalty } % \end{macrocode} % \end{variable} % \begin{macrocode} diff --git a/Master/texmf-dist/source/latex/l3experimental/xgalley/xgalley.ins b/Master/texmf-dist/source/latex/l3experimental/xgalley/xgalley.ins index cadc1d2a98f..d29df2ffb2f 100644 --- a/Master/texmf-dist/source/latex/l3experimental/xgalley/xgalley.ins +++ b/Master/texmf-dist/source/latex/l3experimental/xgalley/xgalley.ins @@ -1,6 +1,6 @@ \iffalse meta-comment -File xgalley.ins Copyright (C) 2010,2011 The LaTeX3 Project +File xgalley.ins Copyright (C) 2010-2012 The LaTeX3 Project It may be distributed and/or modified under the conditions of the LaTeX Project Public License (LPPL), either version 1.3c of this @@ -16,7 +16,7 @@ The released version of this bundle is available from CTAN. \fi -\input docstrip.tex +\input l3docstrip.tex \askforoverwritefalse \preamble diff --git a/Master/texmf-dist/tex/latex/l3experimental/l3dt/l3dt.sty b/Master/texmf-dist/tex/latex/l3experimental/l3dt/l3dt.sty index 99bd83e7764..a672181b903 100644 --- a/Master/texmf-dist/tex/latex/l3experimental/l3dt/l3dt.sty +++ b/Master/texmf-dist/tex/latex/l3experimental/l3dt/l3dt.sty @@ -47,21 +47,20 @@ %% ----------------------------------------------------------------------- %% \RequirePackage{expl3} -\GetIdInfo$Id: l3dt.dtx 3459 2012-02-26 08:58:53Z joseph $ +\GetIdInfo$Id: l3dt.dtx 3877 2012-06-30 22:24:42Z joseph $ {L3 Experimental data tables} \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} -\package_check_loaded_expl: -\quark_new:N \q_dt -\quark_new:N \q_dt_row -\quark_new:N \q_dt_header +\quark_new:N \q__dt +\quark_new:N \q__dt_row +\quark_new:N \q__dt_header \tl_const:Nn \c_empty_dt { { 0 } - \q_dt + \q__dt \q_nil - \q_dt_header - \q_dt_row + \q__dt_header + \q__dt_row } \cs_new_protected:Npn \dt_new:N #1 { \cs_new_eq:NN #1 \c_empty_dt } \cs_new_protected:Npn \dt_clear:N #1 { \cs_set_eq:NN #1 \c_empty_dt } @@ -76,168 +75,166 @@ \dt_new:N \l_tmpb_dt \dt_new:N \g_tmpa_dt \dt_new:N \g_tmpb_dt -\cs_new_protected:Npn \dt_split_aux:nnnn #1#2#3#4 { #3 #2 } -\cs_new_protected:Npn \dt_split_aux:w { } -\cs_new:Npn \dt_split_header:NT #1#2 - { \exp_after:wN \dt_split_header_aux:wn #1 \q_stop {#2} } -\cs_new:Npn \dt_split_header_aux:wn #1 \q_nil \q_dt_header #2 \q_stop #3 - { #3 {#1} { \q_dt_header #2 } } -\cs_new_protected:Npn \dt_split_key:nnTF #1#2 - { \exp_args:No \dt_split_key_aux:nnTF { \tl_to_str:n {#2} } {#1} } -\cs_new_protected:Npn \dt_split_key_aux:nnTF #1#2 - { - \cs_set_protected:Npn \dt_split_aux:w - ##1 \q_dt #1 \q_dt ##2##3##4 \q_mark ##5 \q_stop - { \dt_split_aux:nnnn ##3 { { ##1 \q_dt } {##2} {##4} } } - \dt_split_aux:w #2 \q_mark - \q_dt #1 \q_dt { } { ? \use_ii:nn { } } \q_mark \q_stop - } -\cs_new_protected:Npn \dt_split_key_list:NnTF #1#2 - { \exp_args:NNo \dt_split_key_list_aux:NnTF #1 { \tl_to_str:n {#2} } } -\cs_new_protected:Npn \dt_split_key_list_aux:NnTF #1#2 - { - \cs_set_protected:Npn \dt_split_aux:w - ##1##2 \q_dt #2 \q_dt ##3##4 \q_dt_header ##5 \q_mark ##6 \q_stop +\cs_new_protected:Npn \__dt_split:nnnn #1#2#3#4 { #3 #2 } +\cs_new_protected:Npn \__dt_split:w { } +\cs_new:Npn \__dt_split_header:NT #1#2 + { \exp_after:wN \__dt_split_header:wn #1 \q_stop {#2} } +\cs_new:Npn \__dt_split_header:wn #1 \q_nil \q__dt_header #2 \q_stop #3 + { #3 {#1} { \q__dt_header #2 } } +\cs_new_protected:Npn \__dt_split_key:nnTF #1#2 + { \exp_args:No \__dt_split_key_aux:nnTF { \tl_to_str:n {#2} } {#1} } +\cs_new_protected:Npn \__dt_split_key_aux:nnTF #1#2 + { + \cs_set_protected:Npn \__dt_split:w + ##1 \q__dt #1 \q__dt ##2##3##4 \q_mark ##5 \q_stop + { \__dt_split:nnnn ##3 { { ##1 \q__dt } {##2} {##4} } } + \__dt_split:w #2 \q_mark + \q__dt #1 \q__dt { } { ? \use_ii:nn { } } \q_mark \q_stop + } +\cs_new_protected:Npn \__dt_split_key_list:NnTF #1#2 + { \exp_args:NNo \__dt_split_key_list_aux:NnTF #1 { \tl_to_str:n {#2} } } +\cs_new_protected:Npn \__dt_split_key_list_aux:NnTF #1#2 + { + \cs_set_protected:Npn \__dt_split:w + ##1##2 \q__dt #2 \q__dt ##3##4 \q__dt_header ##5 \q_mark ##6 \q_stop { - \dt_split_aux:nnnn ##3 - { { {##1} ##2 \q_dt } { ##3##4 \q_dt_header ##5 } } + \__dt_split:nnnn ##3 + { { {##1} ##2 \q__dt } { ##3##4 \q__dt_header ##5 } } } - \exp_after:wN \dt_split_aux:w #1 \q_mark - \q_dt #2 \q_dt { ? \use_ii:nn { } } \q_dt_header \q_mark \q_stop + \exp_after:wN \__dt_split:w #1 \q_mark + \q__dt #2 \q__dt { ? \use_ii:nn { } } \q__dt_header \q_mark \q_stop } -\cs_new_protected:Npn \dt_split_row:NnTF #1#2 - { \dt_split_row_aux:NfTF #1 { \int_eval:n {#2} } } -\cs_new_protected:Npn \dt_split_row_aux:NnTF #1#2 +\cs_new_protected:Npn \__dt_split_row:NnTF #1#2 + { \__dt_split_row_aux:NfTF #1 { \int_eval:n {#2} } } +\cs_new_protected:Npn \__dt_split_row_aux:NnTF #1#2 { - \cs_set_protected:Npn \dt_split_aux:w - ##1 \q_dt_row #2 \q_dt ##2##3 \q_dt_row ##4 \q_mark ##5 \q_stop + \cs_set_protected:Npn \__dt_split:w + ##1 \q__dt_row #2 \q__dt ##2##3 \q__dt_row ##4 \q_mark ##5 \q_stop { - \dt_split_aux:nnnn ##2 - { { ##1 \q_dt_row } { #2 \q_dt ##2##3 } {##4} } + \__dt_split:nnnn ##2 + { { ##1 \q__dt_row } { #2 \q__dt ##2##3 } {##4} } } - \exp_after:wN \dt_split_aux:w #1 \q_mark - \q_dt_row #2 \q_dt { ? \use_ii:nn { } } \q_dt_row \q_mark \q_stop + \exp_after:wN \__dt_split:w #1 \q_mark + \q__dt_row #2 \q__dt { ? \use_ii:nn { } } \q__dt_row \q_mark \q_stop } -\cs_generate_variant:Nn \dt_split_row_aux:NnTF { Nf } -\cs_new_protected_nopar:Npn \dt_add_key:Nn { \dt_add_key_aux:NNn \tl_set:Nx } -\cs_new_protected_nopar:Npn \dt_gadd_key:Nn { \dt_add_key_aux:NNn \tl_gset:Nx } -\cs_new_protected:Npn \dt_add_key_aux:NNn #1#2#3 +\cs_generate_variant:Nn \__dt_split_row_aux:NnTF { Nf } +\cs_new_protected_nopar:Npn \dt_add_key:Nn { \__dt_add_key:NNn \tl_set:Nx } +\cs_new_protected_nopar:Npn \dt_gadd_key:Nn { \__dt_add_key:NNn \tl_gset:Nx } +\cs_new_protected:Npn \__dt_add_key:NNn #1#2#3 { - \dt_split_key_list:NnTF #2 {#3} + \__dt_split_key_list:NnTF #2 {#3} { \use_none:nn } { - \dt_split_header:NT #2 - { \dt_add_key_aux:NNnnn #1 #2 {#3} } + \__dt_split_header:NT #2 + { \__dt_add_key:NNnnn #1 #2 {#3} } } } -\cs_new_protected:Npn \dt_add_key_aux:NNnnn #1#2#3#4#5 +\cs_new_protected:Npn \__dt_add_key:NNnnn #1#2#3#4#5 { #1 #2 { \exp_not:n {#4} \tl_to_str:n {#3} - \exp_not:n { \q_dt \q_nil #5 } + \exp_not:n { \q__dt \q_nil #5 } } } -\cs_new_protected_nopar:Npn \dt_add_row:N { \dt_add_row_aux:NN \tl_set:Nx } -\cs_new_protected_nopar:Npn \dt_gadd_row:N { \dt_add_row_aux:NN \tl_gset:Nx } -\cs_new_protected:Npn \dt_add_row_aux:NN #1#2 - { \exp_args:NNf \dt_add_row_aux:NnN #1 { \dt_rows:N #2 } #2 } -\cs_new_protected:Npn \dt_add_row_aux:NnN #1#2#3 +\cs_new_protected_nopar:Npn \dt_add_row:N { \__dt_add_row:NN \tl_set:Nx } +\cs_new_protected_nopar:Npn \dt_gadd_row:N { \__dt_add_row:NN \tl_gset:Nx } +\cs_new_protected:Npn \__dt_add_row:NN #1#2 + { \exp_args:NNf \__dt_add_row:NnN #1 { \dt_rows:N #2 } #2 } +\cs_new_protected:Npn \__dt_add_row:NnN #1#2#3 { #1 #3 { { \int_eval:n { #2 + \c_one } } - \exp_after:wN \dt_add_row_aux:nw #3 \q_stop + \exp_after:wN \__dt_add_row:nw #3 \q_stop #2 - \exp_not:n { \q_dt \q_nil \q_dt_row } + \exp_not:n { \q__dt \q_nil \q__dt_row } } } -\cs_new:Npn \dt_add_row_aux:nw #1#2 \q_stop { \exp_not:n {#2} } +\cs_new:Npn \__dt_add_row:nw #1#2 \q_stop { \exp_not:n {#2} } \cs_new_protected:Npn \dt_put:Nnn #1 { \dt_put:Nnnn #1 { \dt_rows:N #1 - \c_one } } \cs_new_protected:Npn \dt_gput:Nnn #1 { \dt_gput:Nnnn #1 { \dt_rows:N #1 - \c_one } } \cs_new_protected_nopar:Npn \dt_put:Nnnn - { \dt_put_aux:NNNnnn \dt_add_key:Nn \tl_set:Nx } + { \__dt_put:NNNnnn \dt_add_key:Nn \tl_set:Nx } \cs_new_protected_nopar:Npn \dt_gput:Nnnn - { \dt_put_aux:NNNnnn \dt_gadd_key:Nn \tl_gset:Nx } -\cs_new_protected:Npn \dt_put_aux:NNNnnn #1#2#3#4#5#6 + { \__dt_put:NNNnnn \dt_gadd_key:Nn \tl_gset:Nx } +\cs_new_protected:Npn \__dt_put:NNNnnn #1#2#3#4#5#6 { #1 #3 {#5} - \dt_split_row:NnTF #3 {#4} - { \dt_put_aux:NNnnnnn #2 #3 {#5} {#6} } + \__dt_split_row:NnTF #3 {#4} + { \__dt_put:NNnnnnn #2 #3 {#5} {#6} } { - \msg_kernel_error:nnxxx { dt } { unknown-row } + \__msg_kernel_error:nnxxx { dt } { unknown-row } { \token_to_str:N #3 } { \int_eval:n {#4} } { \dt_rows:N #3 } } } -\cs_new_protected:Npn \dt_put_aux:NNnnnnn #1#2#3#4#5#6#7 +\cs_new_protected:Npn \__dt_put:NNnnnnn #1#2#3#4#5#6#7 { - \dt_split_key:nnTF {#6} {#3} - { \dt_put_update:NNnnnnnnn #1 #2 {#3} {#4} {#5} {#7} } - { \dt_put_add_to_row:NNnnnnn #1 #2 {#3} {#4} {#5} {#6} {#7} } + \__dt_split_key:nnTF {#6} {#3} + { \__dt_put_update:NNnnnnnnn #1 #2 {#3} {#4} {#5} {#7} } + { \__dt_put_add_to_row:NNnnnnn #1 #2 {#3} {#4} {#5} {#6} {#7} } } -\cs_new_protected:Npn \dt_put_update:NNnnnnnnn #1#2#3#4#5#6#7#8#9 +\cs_new_protected:Npn \__dt_put_update:NNnnnnnnn #1#2#3#4#5#6#7#8#9 { #1 #2 { \exp_not:n { #5 #7 } \tl_to_str:n {#3} - \exp_not:n { \q_dt {#4} \q_dt #9 \q_dt_row #6 } + \exp_not:n { \q__dt {#4} \q__dt #9 \q__dt_row #6 } } } -\cs_new_protected:Npn \dt_put_add_to_row:NNnnnnn #1#2#3#4#5#6#7 +\cs_new_protected:Npn \__dt_put_add_to_row:NNnnnnn #1#2#3#4#5#6#7 { #1 #2 { \exp_not:n {#5} - \exp_not:o { \dt_put_add_row_aux:w #6 } + \exp_not:o { \__dt_put_add_to_row_aux:w #6 } \tl_to_str:n {#3} - \exp_not:n { \q_dt {#4} \q_dt \q_nil \q_dt_row #7 } + \exp_not:n { \q__dt {#4} \q__dt \q_nil \q__dt_row #7 } } } -\cs_new:Npn \dt_put_add_row_aux:w #1 \q_nil {#1} +\cs_new:Npn \__dt_put_add_to_row_aux:w #1 \q_nil {#1} \cs_new:Npn \dt_keys:N #1 - { \dt_split_header:NT #1 { \dt_keys_aux:nn } } -\cs_new:Npn \dt_keys_aux:nn #1#2 + { \__dt_split_header:NT #1 { \__dt_keys:nn } } +\cs_new:Npn \__dt_keys:nn #1#2 { \int_eval:n { 0 - \exp_after:wN \dt_keys_aux:wN \use_none:nn #1 \q_recursion_tail \q_dt - \prg_break_point:n { } + \exp_after:wN \__dt_keys:wN \use_none:nn #1 \q_recursion_tail \q__dt + \__prg_break_point:Nn \dt_map_break: { } } } -\cs_new:Npn \dt_keys_aux:wN #1 \q_dt +\cs_new:Npn \__dt_keys:wN #1 \q__dt { - \if_meaning:w \q_recursion_tail #1 - \exp_after:wN \prg_map_break: - \fi: + \__quark_if_recursion_tail_break:nN {#1} \dt_map_break: +1 - \dt_keys_aux:wN + \__dt_keys:wN } \cs_new:Npn \dt_rows:N #1 { \exp_after:wN \use_i_delimit_by_q_stop:nw #1 \q_stop } -\cs_new_protected:Npn \dt_del:Nn #1 { \dt_del:Nnn #1 +\cs_new_protected:Npn \dt_remove:Nn #1 { \dt_remove:Nnn #1 { \dt_rows:N #1 - \c_one } } -\cs_new_protected:Npn \dt_gdel:Nn #1 { \dt_gdel:Nnn #1 +\cs_new_protected:Npn \dt_gremove:Nn #1 { \dt_gremove:Nnn #1 { \dt_rows:N #1 - \c_one } } -\cs_new_protected_nopar:Npn \dt_del:Nnn { \dt_del_aux:NNnn \tl_set:Nn } -\cs_new_protected_nopar:Npn \dt_gdel:Nnn { \dt_del_aux:NNnn \tl_gset:Nn } -\cs_new_protected:Npn \dt_del_aux:NNnn #1#2#3#4 +\cs_new_protected_nopar:Npn \dt_remove:Nnn { \dt_remove_aux:NNnn \tl_set:Nn } +\cs_new_protected_nopar:Npn \dt_gremove:Nnn { \dt_remove_aux:NNnn \tl_gset:Nn } +\cs_new_protected:Npn \dt_remove_aux:NNnn #1#2#3#4 { - \dt_split_row:NnTF #2 {#3} - { \dt_del_aux:NNnnnn #1 #2 {#4} } + \__dt_split_row:NnTF #2 {#3} + { \dt_remove_aux:NNnnnn #1 #2 {#4} } { } } -\cs_new_protected:Npn \dt_del_aux:NNnnnn #1#2#3#4#5#6 +\cs_new_protected:Npn \dt_remove_aux:NNnnnn #1#2#3#4#5#6 { - \dt_split_key:nnTF {#5} {#3} - { \dt_del_aux:NNnnnnn #1 #2 {#4} {#6} } + \__dt_split_key:nnTF {#5} {#3} + { \dt_remove_aux:NNnnnnn #1 #2 {#4} {#6} } { } } -\cs_new_protected:Npn \dt_del_aux:NNnnnnn #1#2#3#4#5#6#7 +\cs_new_protected:Npn \dt_remove_aux:NNnnnnn #1#2#3#4#5#6#7 { #1 #2 { #3 #5 #7 #4 } } \cs_new_protected_nopar:Npn \dt_remove_key:Nn { \dt_remove_key_aux:NNn \tl_set:Nx } @@ -245,25 +242,23 @@ { \dt_remove_key_aux:NNn \tl_gset:Nx } \cs_new_protected:Npn \dt_remove_key_aux:NNn #1#2#3 { - \dt_split_key_list:NnTF #2 {#3} + \__dt_split_key_list:NnTF #2 {#3} { \exp_args:No \dt_remove_key_aux:nNNnn { \tl_to_str:n {#3} } #1 #2 } { } } \cs_new_protected:Npn \dt_remove_key_aux:nNNnn #1#2#3#4#5 { - \cs_set:Npn \dt_remove_key_aux:w ##1 \q_dt #1 \q_dt ##2 ##3 + \cs_set:Npn \dt_remove_key_aux:w ##1 \q__dt #1 \q__dt ##2 ##3 { \exp_not:n {##1} - \if_meaning:w \q_recursion_tail ##3 - \exp_after:wN \prg_map_break: - \fi: + \__quark_if_recursion_tail_break:nN {##3} \dt_map_break: \dt_remove_key_aux:w ##3 } #2 #3 { \exp_not:n {#4} - \dt_remove_key_aux:w #5 \q_dt #1 \q_dt { } \q_recursion_tail - \prg_break_point:n { } + \dt_remove_key_aux:w #5 \q__dt #1 \q__dt { } \q_recursion_tail + \__prg_break_point:Nn \dt_map_break: { } } } \cs_new:Npn \dt_remove_key_aux:w { } @@ -273,7 +268,7 @@ { \dt_remove_row_aux:NNn \tl_gset:Nx } \cs_new_protected:Npn \dt_remove_row_aux:NNn #1#2#3 { - \dt_split_row:NnTF #2 {#3} + \__dt_split_row:NnTF #2 {#3} { \dt_remove_row_aux:NNnnn #1 #2 } { } } @@ -283,46 +278,44 @@ { { \int_eval:n { \dt_rows:N #2 - \c_one } } \dt_remove_row_aux:nw #3 \q_stop - \dt_remove_row_loop:nw #5 \q_recursion_tail \q_dt_row - \prg_break_point:n { } + \dt_remove_row_loop:nw #5 \q_recursion_tail \q__dt_row + \__prg_break_point:Nn \dt_map_break: { } } } -\cs_new_eq:NN \dt_remove_row_aux:nw \dt_add_row_aux:nw -\cs_new:Npn \dt_remove_row_loop:nw #1#2 \q_dt_row +\cs_new_eq:NN \dt_remove_row_aux:nw \__dt_add_row:nw +\cs_new:Npn \dt_remove_row_loop:nw #1#2 \q__dt_row { - \if_meaning:w \q_recursion_tail #1 - \exp_after:wN \prg_map_break: - \fi: + \__quark_if_recursion_tail_break:nN {#1} \dt_map_break: \int_eval:n { #1 - \c_one } - \exp_not:n { #2 \q_dt_row } + \exp_not:n { #2 \q__dt_row } \dt_remove_row_loop:nw } \cs_new_protected:Npn \dt_get:NnnN #1#2#3#4 { - \dt_split_row:NnTF #1 {#2} + \__dt_split_row:NnTF #1 {#2} { \dt_get_aux:nNnnn {#3} #4 } { \tl_set:Nn #4 { \q_no_value } } } \cs_new_protected:Npn \dt_get_aux:nNnnn #1#2#3#4#5 { - \dt_split_key:nnTF {#4} {#1} + \__dt_split_key:nnTF {#4} {#1} { \dt_get_aux:Nnnn #2 } { \tl_set:Nn #2 { \q_no_value } } } \cs_new_protected:Npn \dt_get_aux:Nnnn #1#2#3#4 { \tl_set:Nn #1 {#3} } \prg_new_protected_conditional:Npnn \dt_get:NnnN #1#2#3#4 { T , F , TF } { - \dt_split_row:NnTF #1 {#2} - { \dt_get_aux_true:nNnnn {#3} #4 } + \__dt_split_row:NnTF #1 {#2} + { \__dt_get_true:nNnnn {#3} #4 } { \prg_return_false: } } -\cs_new_protected:Npn \dt_get_aux_true:nNnnn #1#2#3#4#5 +\cs_new_protected:Npn \__dt_get_true:nNnnn #1#2#3#4#5 { - \dt_split_key:nnTF {#4} {#1} - { \dt_get_aux_true:Nnnn #2 } + \__dt_split_key:nnTF {#4} {#1} + { \__dt_get_true:Nnnn #2 } { \prg_return_false: } } -\cs_new_protected:Npn \dt_get_aux_true:Nnnn #1#2#3#4 +\cs_new_protected:Npn \__dt_get_true:Nnnn #1#2#3#4 { \tl_set:Nn #1 {#3} \prg_return_true: @@ -340,36 +333,37 @@ { \int_gincr:N \g_dt_map_level_int \seq_gclear_new:c { g_dt_map_ \int_use:N \g_dt_map_level_int _seq } - \keyval_parse:NNn \use_none:n \dt_map_variables_key:nn {#2} - \dt_split_header:NT #1 { \dt_map_variables_aux:nnn {#3} } + \keyval_parse:NNn \use_none:n \__dt_map_variables_key:nn {#2} + \__dt_split_header:NT #1 { \__dt_map_variables:nnn {#3} } } -\cs_new_protected:Npn \dt_map_variables_key:nn #1#2 +\cs_new_protected:Npn \__dt_map_variables_key:nn #1#2 { \seq_gput_right:cn { g_dt_map_ \int_use:N \g_dt_map_level_int _seq } { {#1} #2 } } -\cs_new_protected:Npn \dt_map_variables_aux:nnn #1#2#3 - { \dt_map_variables_aux:nNNw {#1} #3 \q_stop } -\cs_new_protected:Npn \dt_map_variables_aux:nNNw - #1 \q_dt_header \q_dt_row #2 \q_stop +\cs_new_protected:Npn \__dt_map_variables:nnn #1#2#3 + { \__dt_map_variables:nNNw {#1} #3 \q_stop } +\cs_new_protected:Npn \__dt_map_variables:nNNw + #1 \q__dt_header \q__dt_row #2 \q_stop { \int_zero_new:c { l_dt_map_ \int_use:N \g_dt_map_level_int _row_int } - \dt_map_variables_aux:nnw {#1} #2 { } \q_recursion_tail \q_dt_row - \prg_break_point:n { \int_gdecr:N \g_dt_map_level_int } + \__dt_map_variables:nnw {#1} #2 { } \q_recursion_tail \q__dt_row + \__prg_break_point:Nn \dt_map_break: + { \int_gdecr:N \g_dt_map_level_int } } -\cs_new_protected:Npn \dt_map_variables_aux:nnw #1#2#3#4 \q_dt_row +\cs_new_protected:Npn \__dt_map_variables:nnw #1#2#3#4 \q__dt_row { - \if_meaning:w \q_recursion_tail #3 - \exp_after:wN \dt_map_break: - \fi: + \__quark_if_recursion_tail_break:nN {#3} \dt_map_break: \seq_map_inline:cn { g_dt_map_ \int_use:N \g_dt_map_level_int _seq } { \dt_get_aux:nNnnn ##1 { } {#3#4} { } } #1 \int_incr:c { l_dt_map_ \int_use:N \g_dt_map_level_int _row_int } - \dt_map_variables_aux:nnw {#1} + \__dt_map_variables:nnw {#1} } -\cs_new_eq:NN \dt_map_break: \prg_map_break: -\cs_new_eq:NN \dt_map_break:n \prg_map_break:n +\cs_new_nopar:Npn \dt_map_break: + { \__prg_map_break:Nn \dt_map_break: { } } +\cs_new_nopar:Npn \dt_map_break:n + { \__prg_map_break:Nn \dt_map_break: } \prg_new_conditional:Npnn \dt_if_empty:N #1 { T , F , TF , p } { \if_meaning:w #1 \c_empty_dt @@ -379,36 +373,36 @@ \fi: } \prg_new_conditional:Npnn \dt_if_in:Nn #1#2 { p , T , F , TF } - { \dt_split_header:NT #1 { \dt_if_in_aux:nnn {#2} } } -\cs_new:Npn \dt_if_in_aux:nnn #1#2#3 + { \__dt_split_header:NT #1 { \__dt_if_in:nnn {#2} } } +\cs_new:Npn \__dt_if_in:nnn #1#2#3 { - \exp_last_unbraced:Nno \dt_if_in_aux:nwN {#1} { \use_none:nn #2 } - \q_recursion_tail \q_dt - \prg_break_point:n { } + \exp_last_unbraced:Nno \__dt_if_in:nwN {#1} { \use_none:nn #2 } + \q_recursion_tail \q__dt + \__prg_break_point: } -\cs_new:Npn \dt_if_in_aux:nwN #1#2 \q_dt +\cs_new:Npn \__dt_if_in:nwN #1#2 \q__dt { \if_meaning:w \q_recursion_tail #2 - \exp_after:wN \prg_map_break:n + \exp_after:wN \__prg_break:n \else: \exp_after:wN \use_none:n \fi: { \prg_return_false: } \str_if_eq:nnTF {#1} {#2} - { \prg_map_break:n { \prg_return_true: } } - { \dt_if_in_aux:nwN {#1} } + { \__prg_break:n { \prg_return_true: } } + { \__dt_if_in:nwN {#1} } } \prg_new_conditional:Npnn \dt_if_in_row:Nnn #1#2#3 { p , T , F , TF } { - \exp_last_unbraced:Nno \dt_if_in_row_aux:nw {#2} #1 + \exp_last_unbraced:Nno \__dt_if_in_row:nw {#2} #1 \q_recursion_tail \q_nil - \prg_break_point:n { } + \__prg_break_point: { \tl_to_str:n {#3} } } -\cs_new:Npn \dt_if_in_row_aux:nw #1#2 \q_dt_row #3#4 \q_nil +\cs_new:Npn \__dt_if_in_row:nw #1#2 \q__dt_row #3#4 \q_nil { \if_meaning:w \q_recursion_tail #3 - \exp_after:wN \prg_map_break:n + \exp_after:wN \__prg_break:n \else: \exp_after:wN \use_none:n \fi: @@ -417,28 +411,28 @@ \prg_return_false: } \int_compare:nNnTF {#1} = {#3} - { \prg_map_break:n { \exp_args:Nno \dt_if_in_row_aux:nn {#4} } } - { \dt_if_in_row_aux:nw {#1} } + { \__prg_break:n { \exp_args:Nno \__dt_if_in_row:nn {#4} } } + { \__dt_if_in_row:nw {#1} } } -\cs_new:Npn \dt_if_in_row_aux:nn #1#2 +\cs_new:Npn \__dt_if_in_row:nn #1#2 { - \dt_if_in_row_aux:nwn {#2} #1 {#2} \q_dt { } \q_recursion_tail - \prg_break_point:n { } + \__dt_if_in_row:nwn {#2} #1 {#2} \q__dt { } \q_recursion_tail + \__prg_break_point: } -\cs_new:Npn \dt_if_in_row_aux:nwn #1 \q_dt #2 \q_dt #3 +\cs_new:Npn \__dt_if_in_row:nwn #1 \q__dt #2 \q__dt #3 { \str_if_eq:xxTF {#1} {#2} - { \dt_if_in_row_aux:N } - { \dt_if_in_row_aux:nwn {#1} } + { \__dt_if_in_row:N } + { \__dt_if_in_row:nwn {#1} } } -\cs_new:Npn \dt_if_in_row_aux:N #1 +\cs_new:Npn \__dt_if_in_row:N #1 { - \if_meaning:w \q_dt #1 + \if_meaning:w \q__dt #1 \prg_return_true: \else: \prg_return_false: \fi: - \prg_map_break: + \__prg_break: } \cs_new:Npn \dt_if_in_row_p:Nn #1 { \dt_if_in_row_p:Nnn #1 { \dt_rows:N #1 - \c_one } } @@ -448,7 +442,7 @@ { \dt_rows:N #1 - \c_one } } \cs_new:Npn \dt_if_in_row:NnTF #1 { \dt_if_in_row:NnnTF #1 { \dt_rows:N #1 - \c_one } } -\msg_kernel_new:nnnn { dt } { unknown-row } +\__msg_kernel_new:nnnn { dt } { unknown-row } { Data~table~#1~does~not~contain~a~row~'#2'. } { Data~table~#1~contains~#3~rows.~These~must~be~accessed~by~number:~row~ diff --git a/Master/texmf-dist/tex/latex/l3experimental/l3sort/l3sort.sty b/Master/texmf-dist/tex/latex/l3experimental/l3sort/l3sort.sty index b24bacd3794..b9adcd3eeeb 100644 --- a/Master/texmf-dist/tex/latex/l3experimental/l3sort/l3sort.sty +++ b/Master/texmf-dist/tex/latex/l3experimental/l3sort/l3sort.sty @@ -47,176 +47,202 @@ %% ----------------------------------------------------------------------- %% \RequirePackage{expl3} -\GetIdInfo$Id: l3sort.dtx 3467 2012-02-26 15:39:35Z bruno $ +\GetIdInfo$Id: l3sort.dtx 3953 2012-07-11 16:15:09Z bruno $ {L3 Experimental sorting functions} \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} -\package_check_loaded_expl: -\int_const:Nn \c_sort_max_length_int +\int_const:Nn \c__sort_max_length_int { \luatex_if_engine:TF { 49152 } { 24576 } } -\int_new:N \l_sort_length_int -\int_new:N \l_sort_block_int -\int_new:N \l_sort_begin_int -\int_new:N \l_sort_end_int -\int_new:N \l_sort_A_int -\int_new:N \l_sort_B_int -\int_new:N \l_sort_C_int -\cs_new_protected:Npn \sort_aux:NNnNn #1#2#3#4#5 +\int_new:N \l__sort_length_int +\int_new:N \l__sort_block_int +\int_new:N \l__sort_begin_int +\int_new:N \l__sort_end_int +\int_new:N \l__sort_A_int +\int_new:N \l__sort_B_int +\int_new:N \l__sort_C_int +\cs_new:Npn \tl_sort:nN #1#2 { \__sort_quick_i:nNn {#1} #2 { } } +\cs_new:Npn \__sort_quick_i:nNn #1#2 + { + \exp_after:wN \__sort_quick_ii:wn + \use_none:n #1 \__prg_break_point: + \__sort_quick_iii:nnNnn { } { } #2 + #1 + { \__prg_break_point: } + \__prg_break_point: + } +\cs_new:Npn \__sort_quick_ii:wn #1 \__prg_break_point: #2 {#2} +\cs_new:Npn \__sort_quick_iii:nnNnn #1#2#3#4#5 + { + \__prg_break: #5 + \__sort_quick_iv:nnNnwn {#1} {#2} + \__prg_break_point: + #3 {#4} {#5} + { \__sort_quick_iii:nnNnn {#1} { #2 {#5} } } + { \__sort_quick_iii:nnNnn { #1 {#5} } {#2} } + #3 {#4} + } +\cs_new:Npn \__sort_quick_iv:nnNnwn + #1#2 \__prg_break_point: #3#4 #5 \__prg_break_point: #6 + { + \__sort_quick_i:nNn {#2} #3 + { \__sort_quick_i:nNn {#1} #3 {#6} {#4} } + } +\cs_new_protected:Npn \__sort_main:NNNnNn #1#2#3#4#5#6 { \group_begin: - \l_sort_length_int \c_zero - #2 #4 + \l__sort_length_int \c_zero + #2 #5 { - \if_num:w \l_sort_length_int = \c_sort_max_length_int - \sort_too_long_error:Nw #4 + \if_int_compare:w \l__sort_length_int = \c__sort_max_length_int + \__sort_too_long_error:NNw #3 #5 \fi: - \tex_toks:D \l_sort_length_int {##1} - \tex_advance:D \l_sort_length_int \c_one + \tex_toks:D \l__sort_length_int {##1} + \tex_advance:D \l__sort_length_int \c_one } - \cs_set:Npn \sort_compare:nn ##1 ##2 { #5 } - \l_sort_block_int \c_one - \sort_level: + \cs_set:Npn \sort_compare:nn ##1 ##2 { #6 } + \l__sort_block_int \c_one + \__sort_level: \use:x { \group_end: - #1 \exp_not:N #4 {#3} + #1 \exp_not:N #5 {#4} } } \cs_new_protected_nopar:Npn \seq_sort:Nn { - \sort_aux:NNnNn \tl_set:Nn - \seq_map_inline:Nn - { \sort_toks:NNw \exp_not:N \seq_item:n 0 ; } + \__sort_main:NNNnNn \tl_set:Nn + \seq_map_inline:Nn \seq_map_break: + { \__sort_toks:NNw \exp_not:N \__seq_item:n 0 ; } } \cs_new_protected_nopar:Npn \seq_gsort:Nn { - \sort_aux:NNnNn \tl_gset:Nn - \seq_map_inline:Nn - { \sort_toks:NNw \exp_not:N \seq_item:n 0 ; } + \__sort_main:NNNnNn \tl_gset:Nn + \seq_map_inline:Nn \seq_map_break: + { \__sort_toks:NNw \exp_not:N \__seq_item:n 0 ; } } \cs_new_protected_nopar:Npn \tl_sort:Nn { - \sort_aux:NNnNn \tl_set:Nn - \tl_map_inline:Nn - { \sort_toks:NNw \prg_do_nothing: \prg_do_nothing: 0 ; } + \__sort_main:NNNnNn \tl_set:Nn + \tl_map_inline:Nn \tl_map_break: + { \__sort_toks:NNw \prg_do_nothing: \prg_do_nothing: 0 ; } } \cs_new_protected_nopar:Npn \tl_gsort:Nn { - \sort_aux:NNnNn \tl_gset:Nn - \tl_map_inline:Nn - { \sort_toks:NNw \prg_do_nothing: \prg_do_nothing: 0 ; } + \__sort_main:NNNnNn \tl_gset:Nn + \tl_map_inline:Nn \tl_map_break: + { \__sort_toks:NNw \prg_do_nothing: \prg_do_nothing: 0 ; } } \cs_new_protected_nopar:Npn \clist_sort:Nn - { \clist_sort_aux:NNn \tl_set:Nn } + { \__sort_sort:NNn \tl_set:Nn } \cs_new_protected_nopar:Npn \clist_gsort:Nn - { \clist_sort_aux:NNn \tl_gset:Nn } -\cs_new_protected:Npn \clist_sort_aux:NNn #1#2#3 + { \__sort_sort:NNn \tl_gset:Nn } +\cs_new_protected:Npn \__sort_sort:NNn #1#2#3 { \clist_if_empty:NF #2 { - \sort_aux:NNnNn #1 - \clist_map_inline:Nn + \__sort_main:NNNnNn #1 + \clist_map_inline:Nn \clist_map_break: { \exp_last_unbraced:Nf \use_none:n - { \sort_toks:NNw \exp_args:No \clist_wrap_item:n 0 ; } + { \__sort_toks:NNw \exp_args:No \__clist_wrap_item:n 0 ; } } #2 {#3} } } -\cs_new:Npn \sort_toks:NNw #1#2#3 ; +\cs_new:Npn \__sort_toks:NNw #1#2#3 ; { - \if_num:w #3 < \l_sort_length_int + \if_int_compare:w #3 < \l__sort_length_int #1 #2 { \tex_the:D \tex_toks:D #3 } - \exp_after:wN \sort_toks:NNw \exp_after:wN #1 \exp_after:wN #2 - \int_use:N \int_eval:w #3 + \c_one \exp_after:wN ; + \exp_after:wN \__sort_toks:NNw \exp_after:wN #1 \exp_after:wN #2 + \int_use:N \__int_eval:w #3 + \c_one \exp_after:wN ; \fi: } -\cs_new_protected_nopar:Npn \sort_level: +\cs_new_protected_nopar:Npn \__sort_level: { - \if_num:w \l_sort_block_int < \l_sort_length_int - \l_sort_end_int \c_zero - \sort_merge_blocks: - \tex_multiply:D \l_sort_block_int \c_two - \exp_after:wN \sort_level: + \if_int_compare:w \l__sort_block_int < \l__sort_length_int + \l__sort_end_int \c_zero + \__sort_merge_blocks: + \tex_multiply:D \l__sort_block_int \c_two + \exp_after:wN \__sort_level: \fi: } -\cs_new_protected_nopar:Npn \sort_merge_blocks: +\cs_new_protected_nopar:Npn \__sort_merge_blocks: { - \l_sort_begin_int \l_sort_end_int - \tex_advance:D \l_sort_end_int \l_sort_block_int - \if_num:w \int_eval:w \l_sort_end_int < \l_sort_length_int - \l_sort_A_int \l_sort_end_int - \tex_advance:D \l_sort_end_int \l_sort_block_int - \if_num:w \l_sort_end_int > \l_sort_length_int - \l_sort_end_int \l_sort_length_int + \l__sort_begin_int \l__sort_end_int + \tex_advance:D \l__sort_end_int \l__sort_block_int + \if_int_compare:w \__int_eval:w \l__sort_end_int < \l__sort_length_int + \l__sort_A_int \l__sort_end_int + \tex_advance:D \l__sort_end_int \l__sort_block_int + \if_int_compare:w \l__sort_end_int > \l__sort_length_int + \l__sort_end_int \l__sort_length_int \fi: - \l_sort_B_int \l_sort_A_int - \l_sort_C_int \l_sort_length_int + \l__sort_B_int \l__sort_A_int + \l__sort_C_int \l__sort_length_int \sort_copy_block: - \tex_advance:D \l_sort_A_int \c_minus_one - \tex_advance:D \l_sort_B_int \c_minus_one - \tex_advance:D \l_sort_C_int \c_minus_one - \sort_merge_blocks_aux: - \exp_after:wN \sort_merge_blocks: + \tex_advance:D \l__sort_A_int \c_minus_one + \tex_advance:D \l__sort_B_int \c_minus_one + \tex_advance:D \l__sort_C_int \c_minus_one + \__sort_merge_blocks_aux: + \exp_after:wN \__sort_merge_blocks: \fi: } \cs_new_protected_nopar:Npn \sort_copy_block: { - \tex_toks:D \l_sort_C_int \tex_toks:D \l_sort_B_int - \tex_advance:D \l_sort_C_int \c_one - \tex_advance:D \l_sort_B_int \c_one - \if_num:w \l_sort_B_int = \l_sort_end_int + \tex_toks:D \l__sort_C_int \tex_toks:D \l__sort_B_int + \tex_advance:D \l__sort_C_int \c_one + \tex_advance:D \l__sort_B_int \c_one + \if_int_compare:w \l__sort_B_int = \l__sort_end_int \use_i:nn \fi: \sort_copy_block: } -\cs_new_protected_nopar:Npn \sort_merge_blocks_aux: +\cs_new_protected_nopar:Npn \__sort_merge_blocks_aux: { \exp_after:wN \sort_compare:nn \exp_after:wN - { \tex_the:D \tex_toks:D \exp_after:wN \l_sort_A_int \exp_after:wN } - \exp_after:wN { \tex_the:D \tex_toks:D \l_sort_C_int } + { \tex_the:D \tex_toks:D \exp_after:wN \l__sort_A_int \exp_after:wN } + \exp_after:wN { \tex_the:D \tex_toks:D \l__sort_C_int } } \cs_new_protected_nopar:Npn \sort_ordered: { - \tex_toks:D \l_sort_B_int \tex_toks:D \l_sort_C_int - \tex_advance:D \l_sort_B_int \c_minus_one - \tex_advance:D \l_sort_C_int \c_minus_one - \if_num:w \l_sort_C_int < \l_sort_length_int + \tex_toks:D \l__sort_B_int \tex_toks:D \l__sort_C_int + \tex_advance:D \l__sort_B_int \c_minus_one + \tex_advance:D \l__sort_C_int \c_minus_one + \if_int_compare:w \l__sort_C_int < \l__sort_length_int \use_i:nn \fi: - \sort_merge_blocks_aux: + \__sort_merge_blocks_aux: } \cs_new_protected_nopar:Npn \sort_reversed: { - \tex_toks:D \l_sort_B_int \tex_toks:D \l_sort_A_int - \tex_advance:D \l_sort_B_int \c_minus_one - \tex_advance:D \l_sort_A_int \c_minus_one - \if_num:w \l_sort_A_int < \l_sort_begin_int - \sort_merge_blocks_end: \use_i:nn + \tex_toks:D \l__sort_B_int \tex_toks:D \l__sort_A_int + \tex_advance:D \l__sort_B_int \c_minus_one + \tex_advance:D \l__sort_A_int \c_minus_one + \if_int_compare:w \l__sort_A_int < \l__sort_begin_int + \__sort_merge_blocks_end: \use_i:nn \fi: - \sort_merge_blocks_aux: + \__sort_merge_blocks_aux: } -\cs_new_protected_nopar:Npn \sort_merge_blocks_end: +\cs_new_protected_nopar:Npn \__sort_merge_blocks_end: { - \tex_toks:D \l_sort_B_int \tex_toks:D \l_sort_C_int - \tex_advance:D \l_sort_B_int \c_minus_one - \tex_advance:D \l_sort_C_int \c_minus_one - \if_num:w \l_sort_B_int < \l_sort_begin_int + \tex_toks:D \l__sort_B_int \tex_toks:D \l__sort_C_int + \tex_advance:D \l__sort_B_int \c_minus_one + \tex_advance:D \l__sort_C_int \c_minus_one + \if_int_compare:w \l__sort_B_int < \l__sort_begin_int \use_i:nn \fi: - \sort_merge_blocks_end: + \__sort_merge_blocks_end: } -\cs_new_protected:Npn \sort_too_long_error:Nw #1 \fi: +\cs_new_protected:Npn \__sort_too_long_error:NNw #1#2 \fi: { \fi: - \msg_kernel_error:nnx { sort } { too-large } { \token_to_str:N #1 } - \prg_map_break: + \__msg_kernel_error:nnx { sort } { too-large } { \token_to_str:N #2 } + #1 } -\msg_kernel_new:nnnn { sort } { too-large } +\__msg_kernel_new:nnnn { sort } { too-large } { The~list~#1~is~too~long~to~be~sorted~by~TeX. } { - TeX~has~\int_use:N \c_max_register_int \ registers~available:~ - this~only~allows~to~sorts~with~up~to~\int_use:N \c_sort_max_length_int + TeX~has~\int_eval:n { \c_max_register_int + 1 }~registers~available:~ + this~only~allows~to~sorts~with~up~to~\int_use:N \c__sort_max_length_int \ items.~All~extra~items~will~be~ignored. } %% diff --git a/Master/texmf-dist/tex/latex/l3experimental/l3str/l3flag.sty b/Master/texmf-dist/tex/latex/l3experimental/l3str/l3flag.sty index b56100853ab..63970cdd45d 100644 --- a/Master/texmf-dist/tex/latex/l3experimental/l3str/l3flag.sty +++ b/Master/texmf-dist/tex/latex/l3experimental/l3str/l3flag.sty @@ -47,59 +47,59 @@ %% ----------------------------------------------------------------------- \RequirePackage{expl3} \GetIdInfo$Id: l3flag.dtx 3039 2011-12-08 09:22:35Z bruno $ - {L3 Experimental Flags} + {L3 Experimental flags} \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} \cs_new_protected:Npn \flag_new:n #1 { - \cs_new:cpn { flag_trap_#1:w } ##1 ; - { \exp_after:wN \use_none:n \cs:w l_#1_##1_flag \cs_end: } + \cs_new:cpn { __flag_trap_#1:w } ##1 ; + { \exp_after:wN \use_none:n \cs:w __flag_#1_##1: \cs_end: } } \cs_new_protected:Npn \flag_clear:n #1 - { \flag_clear_aux:ww 0 ; #1 \q_stop } -\cs_new_protected:Npn \flag_clear_aux:ww #1 ; #2 \q_stop + { \__flag_clear:ww 0 ; #1 \q_stop } +\cs_new_protected:Npn \__flag_clear:ww #1 ; #2 \q_stop { - \if_cs_exist:w l_#2_#1_flag \cs_end: + \if_cs_exist:w __flag_#2_#1: \cs_end: \else: \exp_after:wN \use_none_delimit_by_q_stop:w \fi: - \cs_set_eq:cN { l_#2_#1_flag } \c_undefined:D - \exp_after:wN \flag_clear_aux:ww - \int_use:N \int_eval:w \c_one + #1 ; + \cs_set_eq:cN { __flag_#2_#1: } \c_undefined:D + \exp_after:wN \__flag_clear:ww + \int_use:N \__int_eval:w \c_one + #1 ; #2 \q_stop } \cs_new_protected:Npn \flag_clear_new:n #1 { \flag_if_exist:nTF {#1} { \flag_clear:n } { \flag_new:n } {#1} } \cs_new_protected:Npn \flag_set_trap:nn #1#2 - { \cs_set:cpn { flag_trap_#1:w } ##1 ; {#2} } + { \cs_set:cpn { __flag_trap_#1:w } ##1 ; {#2} } \prg_new_conditional:Npnn \flag_if_exist:n #1 { p , T , F , TF } { - \cs_if_exist:cTF { flag_trap_#1:w } + \cs_if_exist:cTF { __flag_trap_#1:w } { \prg_return_true: } { \prg_return_false: } } \prg_new_conditional:Npnn \flag_if_raised:n #1 { p , T , F , TF } { - \if_cs_exist:w l_#1_0_flag \cs_end: + \if_cs_exist:w __flag_#1_0: \cs_end: \prg_return_true: \else: \prg_return_false: \fi: } -\cs_new:Npn \flag_height:n #1 { \flag_height_loop:ww 0; #1 \q_stop } -\cs_new:Npn \flag_height_loop:ww #1 ; #2 \q_stop +\cs_new:Npn \flag_height:n #1 { \__flag_height_loop:ww 0; #1 \q_stop } +\cs_new:Npn \__flag_height_loop:ww #1 ; #2 \q_stop { - \if_cs_exist:w l_#2_#1_flag \cs_end: - \exp_after:wN \flag_height_loop:ww \int_use:N \int_eval:w \c_one + + \if_cs_exist:w __flag_#2_#1: \cs_end: + \exp_after:wN \__flag_height_loop:ww \int_use:N \__int_eval:w \c_one + \else: - \exp_after:wN \flag_height_end:ww + \exp_after:wN \__flag_height_end:ww \fi: #1 ; #2 \q_stop } -\cs_new:Npn \flag_height_end:ww #1 ; #2 \q_stop { #1 } +\cs_new:Npn \__flag_height_end:ww #1 ; #2 \q_stop { #1 } \cs_new:Npn \flag_raise:n #1 { - \cs:w flag_trap_#1:w \exp_after:wN \cs_end: - \int_value:w \flag_height:n {#1} ; + \cs:w __flag_trap_#1:w \exp_after:wN \cs_end: + \__int_value:w \flag_height:n {#1} ; } %% %% diff --git a/Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex-trace.sty b/Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex-trace.sty index 0208088744a..7b31ebc4613 100644 --- a/Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex-trace.sty +++ b/Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex-trace.sty @@ -46,448 +46,448 @@ %% %% ----------------------------------------------------------------------- \RequirePackage{expl3} -\GetIdInfo$Id: l3regex.dtx 3488 2012-03-03 19:49:03Z bruno $ - {L3 Experimental Regular Expressions} +\GetIdInfo$Id: l3regex.dtx 3940 2012-07-09 00:41:44Z bruno $ + {L3 Experimental regular expressions} \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} \RequirePackage{l3tl-build, l3tl-analysis, l3flag, l3str} \cs_generate_variant:Nn \tl_to_str:n { V } -\cs_new:Npn \regex_tmp:w { } -\tl_new:N \l_regex_internal_a_tl -\tl_new:N \l_regex_internal_b_tl -\int_new:N \l_regex_internal_a_int -\int_new:N \l_regex_internal_b_int -\int_new:N \l_regex_internal_c_int -\bool_new:N \l_regex_internal_bool -\seq_new:N \l_regex_internal_seq -\tl_new:N \g_regex_internal_tl -\tl_const:Nn \c_regex_no_match_regex - { - \regex_branch:n - { \regex_class:NnnnN \c_true_bool { } { 1 } { 0 } \c_true_bool } - } -\int_new:N \l_regex_balance_int -\cs_new_protected:Npn \regex_break_true:w - #1 \regex_break_point:TF #2 #3 {#2} -\cs_new_protected:Npn \regex_break_point:TF #1 #2 { #2 } -\cs_new_protected:Npn \regex_item_reverse:n #1 +\cs_new:Npn \__regex_tmp:w { } +\tl_new:N \l__regex_internal_a_tl +\tl_new:N \l__regex_internal_b_tl +\int_new:N \l__regex_internal_a_int +\int_new:N \l__regex_internal_b_int +\int_new:N \l__regex_internal_c_int +\bool_new:N \l__regex_internal_bool +\seq_new:N \l__regex_internal_seq +\tl_new:N \g__regex_internal_tl +\tl_const:Nn \c__regex_no_match_regex + { + \__regex_branch:n + { \__regex_class:NnnnN \c_true_bool { } { 1 } { 0 } \c_true_bool } + } +\int_new:N \l__regex_balance_int +\cs_new_protected:Npn \__regex_break_true:w + #1 \__regex_break_point:TF #2 #3 {#2} +\cs_new_protected:Npn \__regex_break_point:TF #1 #2 { #2 } +\cs_new_protected:Npn \__regex_item_reverse:n #1 { #1 - \regex_break_point:TF { } \regex_break_true:w + \__regex_break_point:TF { } \__regex_break_true:w } -\cs_new_protected:Npn \regex_item_caseful_equal:n #1 +\cs_new_protected:Npn \__regex_item_caseful_equal:n #1 { - \if_num:w #1 = \l_regex_current_char_int - \exp_after:wN \regex_break_true:w + \if_int_compare:w #1 = \l__regex_current_char_int + \exp_after:wN \__regex_break_true:w \fi: } -\cs_new_protected:Npn \regex_item_caseful_range:nn #1 #2 +\cs_new_protected:Npn \__regex_item_caseful_range:nn #1 #2 { - \reverse_if:N \if_num:w #1 > \l_regex_current_char_int - \reverse_if:N \if_num:w #2 < \l_regex_current_char_int - \exp_after:wN \exp_after:wN \exp_after:wN \regex_break_true:w + \reverse_if:N \if_int_compare:w #1 > \l__regex_current_char_int + \reverse_if:N \if_int_compare:w #2 < \l__regex_current_char_int + \exp_after:wN \exp_after:wN \exp_after:wN \__regex_break_true:w \fi: \fi: } -\cs_new_protected:Npn \regex_item_caseless_equal:n #1 +\cs_new_protected:Npn \__regex_item_caseless_equal:n #1 { - \if_num:w #1 = \l_regex_current_char_int - \exp_after:wN \regex_break_true:w + \if_int_compare:w #1 = \l__regex_current_char_int + \exp_after:wN \__regex_break_true:w \fi: - \if_num:w \l_regex_case_changed_char_int = \c_max_int - \regex_compute_case_changed_char: + \if_int_compare:w \l__regex_case_changed_char_int = \c_max_int + \__regex_compute_case_changed_char: \fi: - \if_num:w #1 = \l_regex_case_changed_char_int - \exp_after:wN \regex_break_true:w + \if_int_compare:w #1 = \l__regex_case_changed_char_int + \exp_after:wN \__regex_break_true:w \fi: } -\cs_new_protected:Npn \regex_item_caseless_range:nn #1 #2 +\cs_new_protected:Npn \__regex_item_caseless_range:nn #1 #2 { - \reverse_if:N \if_num:w #1 > \l_regex_current_char_int - \reverse_if:N \if_num:w #2 < \l_regex_current_char_int - \exp_after:wN \exp_after:wN \exp_after:wN \regex_break_true:w + \reverse_if:N \if_int_compare:w #1 > \l__regex_current_char_int + \reverse_if:N \if_int_compare:w #2 < \l__regex_current_char_int + \exp_after:wN \exp_after:wN \exp_after:wN \__regex_break_true:w \fi: \fi: - \if_num:w \l_regex_case_changed_char_int = \c_max_int - \regex_compute_case_changed_char: + \if_int_compare:w \l__regex_case_changed_char_int = \c_max_int + \__regex_compute_case_changed_char: \fi: - \reverse_if:N \if_num:w #1 > \l_regex_case_changed_char_int - \reverse_if:N \if_num:w #2 < \l_regex_case_changed_char_int - \exp_after:wN \exp_after:wN \exp_after:wN \regex_break_true:w + \reverse_if:N \if_int_compare:w #1 > \l__regex_case_changed_char_int + \reverse_if:N \if_int_compare:w #2 < \l__regex_case_changed_char_int + \exp_after:wN \exp_after:wN \exp_after:wN \__regex_break_true:w \fi: \fi: } -\cs_new_protected_nopar:Npn \regex_compute_case_changed_char: +\cs_new_protected_nopar:Npn \__regex_compute_case_changed_char: { - \int_set_eq:NN \l_regex_case_changed_char_int \l_regex_current_char_int - \if_num:w \l_regex_current_char_int < \c_ninety_one - \if_num:w \l_regex_current_char_int < \c_sixty_five + \int_set_eq:NN \l__regex_case_changed_char_int \l__regex_current_char_int + \if_int_compare:w \l__regex_current_char_int < \c_ninety_one + \if_int_compare:w \l__regex_current_char_int < \c_sixty_five \else: - \int_add:Nn \l_regex_case_changed_char_int { \c_thirty_two } + \int_add:Nn \l__regex_case_changed_char_int { \c_thirty_two } \fi: \else: - \if_num:w \l_regex_current_char_int < \c_one_hundred_twenty_three - \if_num:w \l_regex_current_char_int < \c_ninety_seven + \if_int_compare:w \l__regex_current_char_int < \c_one_hundred_twenty_three + \if_int_compare:w \l__regex_current_char_int < \c_ninety_seven \else: - \int_sub:Nn \l_regex_case_changed_char_int { \c_thirty_two } + \int_sub:Nn \l__regex_case_changed_char_int { \c_thirty_two } \fi: \fi: \fi: } -\cs_new_eq:NN \regex_item_equal:n ? -\cs_new_eq:NN \regex_item_range:nn ? -\cs_new_protected:Npn \regex_item_catcode_aux: +\cs_new_eq:NN \__regex_item_equal:n ? +\cs_new_eq:NN \__regex_item_range:nn ? +\cs_new_protected:Npn \__regex_item_catcode: { " - \if_case:w \l_regex_current_catcode_int + \if_case:w \l__regex_current_catcode_int 1 \or: 4 \or: 10 \or: 40 \or: 100 \or: \or: 1000 \or: 4000 \or: 10000 \or: \or: 100000 \or: 400000 \or: 1000000 \or: 4000000 \else: 1*\c_zero \fi: } -\cs_new_protected:Npn \regex_item_catcode:nT #1 +\cs_new_protected:Npn \__regex_item_catcode:nT #1 { - \if_int_odd:w \int_eval:w #1 / \regex_item_catcode_aux: \int_eval_end: + \if_int_odd:w \__int_eval:w #1 / \__regex_item_catcode: \__int_eval_end: \exp_after:wN \use:n \else: \exp_after:wN \use_none:n \fi: } -\cs_new_protected:Npn \regex_item_catcode_reverse:nT #1#2 - { \regex_item_catcode:nT {#1} { \regex_item_reverse:n {#2} } } -\cs_new_protected:Npn \regex_item_exact:nn #1#2 +\cs_new_protected:Npn \__regex_item_catcode_reverse:nT #1#2 + { \__regex_item_catcode:nT {#1} { \__regex_item_reverse:n {#2} } } +\cs_new_protected:Npn \__regex_item_exact:nn #1#2 { - \if_num:w #1 = \l_regex_current_catcode_int - \if_num:w #2 = \l_regex_current_char_int - \exp_after:wN \exp_after:wN \exp_after:wN \regex_break_true:w + \if_int_compare:w #1 = \l__regex_current_catcode_int + \if_int_compare:w #2 = \l__regex_current_char_int + \exp_after:wN \exp_after:wN \exp_after:wN \__regex_break_true:w \fi: \fi: } -\cs_new_protected:Npn \regex_item_exact_cs:c #1 +\cs_new_protected:Npn \__regex_item_exact_cs:c #1 { - \int_compare:nNnTF \l_regex_current_catcode_int = \c_zero + \int_compare:nNnTF \l__regex_current_catcode_int = \c_zero { - \str_if_eq:xxTF + \str_if_eq_x:nnTF { \exp_after:wN \exp_after:wN \exp_after:wN \cs_to_str:N - \tex_the:D \tex_toks:D \l_regex_current_pos_int + \tex_the:D \tex_toks:D \l__regex_current_pos_int } { #1 } - { \regex_break_true:w } { } + { \__regex_break_true:w } { } } { } } -\cs_new_protected:Npn \regex_item_cs:n #1 +\cs_new_protected:Npn \__regex_item_cs:n #1 { - \int_compare:nNnT \l_regex_current_catcode_int = \c_zero + \int_compare:nNnT \l__regex_current_catcode_int = \c_zero { \group_begin: - \regex_single_match: - \regex_disable_submatches: - \regex_build_for_cs:n {#1} - \bool_set_eq:NN \l_regex_saved_success_bool \g_regex_success_bool - \exp_args:Nx \regex_match:n + \__regex_single_match: + \__regex_disable_submatches: + \__regex_build_for_cs:n {#1} + \bool_set_eq:NN \l__regex_saved_success_bool \g__regex_success_bool + \exp_args:Nx \__regex_match:n { \exp_after:wN \exp_after:wN \exp_after:wN \cs_to_str:N - \tex_the:D \tex_toks:D \l_regex_current_pos_int + \tex_the:D \tex_toks:D \l__regex_current_pos_int } - \if_meaning:w \c_true_bool \g_regex_success_bool - \group_insert_after:N \regex_break_true:w + \if_meaning:w \c_true_bool \g__regex_success_bool + \group_insert_after:N \__regex_break_true:w \fi: - \bool_gset_eq:NN \g_regex_success_bool \l_regex_saved_success_bool + \bool_gset_eq:NN \g__regex_success_bool \l__regex_saved_success_bool \group_end: } } -\cs_new_protected_nopar:Npn \regex_prop_d: - { \regex_item_caseful_range:nn \c_forty_eight { 57 } } % 0--9 -\cs_new_protected_nopar:Npn \regex_prop_h: - { - \regex_item_caseful_equal:n \c_thirty_two % space - \regex_item_caseful_equal:n \c_nine % tab - } -\cs_new_protected_nopar:Npn \regex_prop_s: - { - \regex_item_caseful_equal:n \c_thirty_two % space - \regex_item_caseful_equal:n \c_nine % tab - \regex_item_caseful_equal:n \c_ten % lf - \regex_item_caseful_equal:n \c_twelve % ff - \regex_item_caseful_equal:n \c_thirteen % cr - } -\cs_new_protected_nopar:Npn \regex_prop_v: - { \regex_item_caseful_range:nn \c_ten \c_thirteen } % lf, vtab, ff, cr -\cs_new_protected_nopar:Npn \regex_prop_w: - { - \regex_item_caseful_range:nn \c_ninety_seven { 122 } % a--z - \regex_item_caseful_range:nn \c_sixty_five { 90 } % A--Z - \regex_item_caseful_range:nn \c_forty_eight { 57 } % 0--9 - \regex_item_caseful_equal:n { 95 } % _ - } -\cs_new_protected_nopar:Npn \regex_prop_N: - { \regex_item_reverse:n { \regex_item_caseful_equal:n \c_ten } } -\cs_new_protected_nopar:Npn \regex_posix_alnum: - { \regex_posix_alpha: \regex_posix_digit: } -\cs_new_protected_nopar:Npn \regex_posix_alpha: - { \regex_posix_lower: \regex_posix_upper: } -\cs_new_protected_nopar:Npn \regex_posix_ascii: - { \regex_item_caseful_range:nn \c_zero \c_one_hundred_twenty_seven } -\cs_new_eq:NN \regex_posix_blank: \regex_prop_h: -\cs_new_protected_nopar:Npn \regex_posix_cntrl: - { - \regex_item_caseful_range:nn \c_zero { 31 } - \regex_item_caseful_equal:n \c_one_hundred_twenty_seven - } -\cs_new_eq:NN \regex_posix_digit: \regex_prop_d: -\cs_new_protected_nopar:Npn \regex_posix_graph: - { \regex_item_caseful_range:nn { 33 } { 126 } } -\cs_new_protected_nopar:Npn \regex_posix_lower: - { \regex_item_caseful_range:nn \c_ninety_seven { 122 } } -\cs_new_protected_nopar:Npn \regex_posix_print: - { \regex_item_caseful_range:nn \c_thirty_two { 126 } } -\cs_new_protected_nopar:Npn \regex_posix_punct: - { - \regex_item_caseful_range:nn { 33 } { 47 } - \regex_item_caseful_range:nn { 58 } { 64 } - \regex_item_caseful_range:nn { 91 } { 96 } - \regex_item_caseful_range:nn { 123 } { 126 } - } -\cs_new_protected_nopar:Npn \regex_posix_space: - { - \regex_item_caseful_equal:n \c_thirty_two - \regex_item_caseful_range:nn \c_nine \c_thirteen - } -\cs_new_protected_nopar:Npn \regex_posix_upper: - { \regex_item_caseful_range:nn \c_sixty_five { 90 } } -\cs_new_eq:NN \regex_posix_word: \regex_prop_w: -\cs_new_protected_nopar:Npn \regex_posix_xdigit: - { - \regex_posix_digit: - \regex_item_caseful_range:nn \c_sixty_five { 70 } - \regex_item_caseful_range:nn \c_ninety_seven { 102 } - } -\cs_new_protected:Npn \regex_escape_use:nnnn #1#2#3#4 - { - \trace_push:nnn { regex } { 1 } { regex_escape_use:nnnn } - \tl_set_build:Nw \l_regex_internal_a_tl - \cs_set_nopar:Npn \regex_escape_unescaped:N ##1 { #1 } - \cs_set_nopar:Npn \regex_escape_escaped:N ##1 { #2 } - \cs_set_nopar:Npn \regex_escape_raw:N ##1 { #3 } +\cs_new_protected_nopar:Npn \__regex_prop_d: + { \__regex_item_caseful_range:nn \c_forty_eight { 57 } } % 0--9 +\cs_new_protected_nopar:Npn \__regex_prop_h: + { + \__regex_item_caseful_equal:n \c_thirty_two % space + \__regex_item_caseful_equal:n \c_nine % tab + } +\cs_new_protected_nopar:Npn \__regex_prop_s: + { + \__regex_item_caseful_equal:n \c_thirty_two % space + \__regex_item_caseful_equal:n \c_nine % tab + \__regex_item_caseful_equal:n \c_ten % lf + \__regex_item_caseful_equal:n \c_twelve % ff + \__regex_item_caseful_equal:n \c_thirteen % cr + } +\cs_new_protected_nopar:Npn \__regex_prop_v: + { \__regex_item_caseful_range:nn \c_ten \c_thirteen } % lf, vtab, ff, cr +\cs_new_protected_nopar:Npn \__regex_prop_w: + { + \__regex_item_caseful_range:nn \c_ninety_seven { 122 } % a--z + \__regex_item_caseful_range:nn \c_sixty_five { 90 } % A--Z + \__regex_item_caseful_range:nn \c_forty_eight { 57 } % 0--9 + \__regex_item_caseful_equal:n { 95 } % _ + } +\cs_new_protected_nopar:Npn \__regex_prop_N: + { \__regex_item_reverse:n { \__regex_item_caseful_equal:n \c_ten } } +\cs_new_protected_nopar:Npn \__regex_posix_alnum: + { \__regex_posix_alpha: \__regex_posix_digit: } +\cs_new_protected_nopar:Npn \__regex_posix_alpha: + { \__regex_posix_lower: \__regex_posix_upper: } +\cs_new_protected_nopar:Npn \__regex_posix_ascii: + { \__regex_item_caseful_range:nn \c_zero \c_one_hundred_twenty_seven } +\cs_new_eq:NN \__regex_posix_blank: \__regex_prop_h: +\cs_new_protected_nopar:Npn \__regex_posix_cntrl: + { + \__regex_item_caseful_range:nn \c_zero { 31 } + \__regex_item_caseful_equal:n \c_one_hundred_twenty_seven + } +\cs_new_eq:NN \__regex_posix_digit: \__regex_prop_d: +\cs_new_protected_nopar:Npn \__regex_posix_graph: + { \__regex_item_caseful_range:nn { 33 } { 126 } } +\cs_new_protected_nopar:Npn \__regex_posix_lower: + { \__regex_item_caseful_range:nn \c_ninety_seven { 122 } } +\cs_new_protected_nopar:Npn \__regex_posix_print: + { \__regex_item_caseful_range:nn \c_thirty_two { 126 } } +\cs_new_protected_nopar:Npn \__regex_posix_punct: + { + \__regex_item_caseful_range:nn { 33 } { 47 } + \__regex_item_caseful_range:nn { 58 } { 64 } + \__regex_item_caseful_range:nn { 91 } { 96 } + \__regex_item_caseful_range:nn { 123 } { 126 } + } +\cs_new_protected_nopar:Npn \__regex_posix_space: + { + \__regex_item_caseful_equal:n \c_thirty_two + \__regex_item_caseful_range:nn \c_nine \c_thirteen + } +\cs_new_protected_nopar:Npn \__regex_posix_upper: + { \__regex_item_caseful_range:nn \c_sixty_five { 90 } } +\cs_new_eq:NN \__regex_posix_word: \__regex_prop_w: +\cs_new_protected_nopar:Npn \__regex_posix_xdigit: + { + \__regex_posix_digit: + \__regex_item_caseful_range:nn \c_sixty_five { 70 } + \__regex_item_caseful_range:nn \c_ninety_seven { 102 } + } +\cs_new_protected:Npn \__regex_escape_use:nnnn #1#2#3#4 + { + \trace_push:nnn { regex } { 1 } { @@_escape_use:nnnn } + \__tl_build:Nw \l__regex_internal_a_tl + \cs_set_nopar:Npn \__regex_escape_unescaped:N ##1 { #1 } + \cs_set_nopar:Npn \__regex_escape_escaped:N ##1 { #2 } + \cs_set_nopar:Npn \__regex_escape_raw:N ##1 { #3 } \int_set:Nn \tex_escapechar:D { 92 } - \str_gset_other:Nn \g_regex_internal_tl { #4 } - \tl_set:Nx \l_regex_internal_b_tl + \__str_gset_other:Nn \g__regex_internal_tl { #4 } + \tl_set:Nx \l__regex_internal_b_tl { - \exp_after:wN \regex_escape_loop:N \g_regex_internal_tl - { break } \prg_break_point:n { } + \exp_after:wN \__regex_escape_loop:N \g__regex_internal_tl + { break } \__prg_break_point: } - \tl_build_one:o \l_regex_internal_b_tl - \tl_build_end: - \trace_pop:nnn { regex } { 1 } { regex_escape_use:nnnn } - \l_regex_internal_a_tl + \__tl_build_one:o \l__regex_internal_b_tl + \__tl_build_end: + \trace_pop:nnn { regex } { 1 } { @@_escape_use:nnnn } + \l__regex_internal_a_tl } -\cs_new:Npn \regex_escape_loop:N #1 +\cs_new:Npn \__regex_escape_loop:N #1 { - \cs_if_exist_use:cF { regex_escape_\token_to_str:N #1:w } - { \regex_escape_unescaped:N #1 } - \regex_escape_loop:N + \cs_if_exist_use:cF { __regex_escape_\token_to_str:N #1:w } + { \__regex_escape_unescaped:N #1 } + \__regex_escape_loop:N } -\cs_new_nopar:cpn { regex_escape_ \c_backslash_str :w } - \regex_escape_loop:N #1 +\cs_new_nopar:cpn { __regex_escape_ \c_backslash_str :w } + \__regex_escape_loop:N #1 { - \cs_if_exist_use:cF { regex_escape_/\token_to_str:N #1:w } - { \regex_escape_escaped:N #1 } - \regex_escape_loop:N + \cs_if_exist_use:cF { __regex_escape_/\token_to_str:N #1:w } + { \__regex_escape_escaped:N #1 } + \__regex_escape_loop:N } -\cs_new_eq:NN \regex_escape_unescaped:N ? -\cs_new_eq:NN \regex_escape_escaped:N ? -\cs_new_eq:NN \regex_escape_raw:N ? -\cs_new_eq:NN \regex_escape_break:w \prg_map_break: -\cs_new_nopar:cpn { regex_escape_/break:w } +\cs_new_eq:NN \__regex_escape_unescaped:N ? +\cs_new_eq:NN \__regex_escape_escaped:N ? +\cs_new_eq:NN \__regex_escape_raw:N ? +\cs_new_eq:NN \__regex_escape_break:w \__prg_break: +\cs_new_nopar:cpn { __regex_escape_/break:w } { \if_false: { \fi: } - \msg_kernel_error:nn { regex } { trailing-backslash } + \__msg_kernel_error:nn { regex } { trailing-backslash } \exp_after:wN \use_none:n \exp_after:wN { \if_false: } \fi: } -\cs_new_nopar:cpn { regex_escape_~:w } { } -\cs_new_nopar:cpx { regex_escape_/a:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^G } -\cs_new_nopar:cpx { regex_escape_/t:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^I } -\cs_new_nopar:cpx { regex_escape_/n:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^J } -\cs_new_nopar:cpx { regex_escape_/f:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^L } -\cs_new_nopar:cpx { regex_escape_/r:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^M } -\cs_new_nopar:cpx { regex_escape_/e:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^[ } -\cs_new:cpn { regex_escape_/x:w } \regex_escape_loop:N - { - \exp_after:wN \regex_escape_x_end:w - \int_value:w "0 \regex_escape_x_test:N - } -\cs_new:Npn \regex_escape_x_end:w #1 ; +\cs_new_nopar:cpn { __regex_escape_~:w } { } +\cs_new_nopar:cpx { __regex_escape_/a:w } + { \exp_not:N \__regex_escape_raw:N \iow_char:N \^^G } +\cs_new_nopar:cpx { __regex_escape_/t:w } + { \exp_not:N \__regex_escape_raw:N \iow_char:N \^^I } +\cs_new_nopar:cpx { __regex_escape_/n:w } + { \exp_not:N \__regex_escape_raw:N \iow_char:N \^^J } +\cs_new_nopar:cpx { __regex_escape_/f:w } + { \exp_not:N \__regex_escape_raw:N \iow_char:N \^^L } +\cs_new_nopar:cpx { __regex_escape_/r:w } + { \exp_not:N \__regex_escape_raw:N \iow_char:N \^^M } +\cs_new_nopar:cpx { __regex_escape_/e:w } + { \exp_not:N \__regex_escape_raw:N \iow_char:N \^^[ } +\cs_new:cpn { __regex_escape_/x:w } \__regex_escape_loop:N + { + \exp_after:wN \__regex_escape_x_end:w + \__int_value:w "0 \__regex_escape_x_test:N + } +\cs_new:Npn \__regex_escape_x_end:w #1 ; { \int_compare:nNnTF {#1} < \c_two_hundred_fifty_six { - \exp_last_unbraced:Nf \regex_escape_raw:N - { \str_output_byte:n {#1} } + \exp_last_unbraced:Nf \__regex_escape_raw:N + { \__str_output_byte:n {#1} } } - { \regex_escape_x_large:n {#1} } + { \__regex_escape_x_large:n {#1} } } \group_begin: \char_set_catcode_other:n { 0 } - \cs_new:Npn \regex_escape_x_large:n #1 + \cs_new:Npn \__regex_escape_x_large:n #1 { \if_false: { \fi: } - \tl_build_one:o \l_regex_internal_b_tl + \__tl_build_one:o \l__regex_internal_b_tl \int_compare:nNnTF {#1} > \c_max_char_int { - \msg_kernel_error:nnx { regex } { x-overflow } {#1} - \tl_set:Nx \l_regex_internal_b_tl - { \if_false: } \fi: \regex_escape_loop:N + \__msg_kernel_error:nnx { regex } { x-overflow } {#1} + \tl_set:Nx \l__regex_internal_b_tl + { \if_false: } \fi: \__regex_escape_loop:N } { \char_set_lccode:nn { \c_zero } {#1} \tl_to_lowercase:n { - \tl_set:Nx \l_regex_internal_b_tl + \tl_set:Nx \l__regex_internal_b_tl { \if_false: } \fi: - \regex_escape_raw:N ^^@ - \regex_escape_loop:N + \__regex_escape_raw:N ^^@ + \__regex_escape_loop:N } } } \group_end: -\cs_new:Npn \regex_escape_x_test:N #1 +\cs_new:Npn \__regex_escape_x_test:N #1 { - \str_if_eq:xxTF {#1} { break } { ; } + \str_if_eq_x:nnTF {#1} { break } { ; } { \if_charcode:w \c_space_token #1 - \exp_after:wN \regex_escape_x_test:N + \exp_after:wN \__regex_escape_x_test:N \else: - \exp_after:wN \regex_escape_x_test_ii:N + \exp_after:wN \__regex_escape_x_test_ii:N \exp_after:wN #1 \fi: } } -\cs_new:Npn \regex_escape_x_test_ii:N #1 +\cs_new:Npn \__regex_escape_x_test_ii:N #1 { \if_charcode:w \c_lbrace_str #1 - \exp_after:wN \regex_escape_x_loop:N + \exp_after:wN \__regex_escape_x_loop:N \else: - \str_aux_hexadecimal_use:NTF #1 - { \exp_after:wN \regex_escape_x_ii:N } - { ; \exp_after:wN \regex_escape_loop:N \exp_after:wN #1 } + \__str_hexadecimal_use:NTF #1 + { \exp_after:wN \__regex_escape_x_ii:N } + { ; \exp_after:wN \__regex_escape_loop:N \exp_after:wN #1 } \fi: } -\cs_new:Npn \regex_escape_x_ii:N #1 +\cs_new:Npn \__regex_escape_x_ii:N #1 { - \str_if_eq:xxTF {#1} { break } { ; } + \str_if_eq_x:nnTF {#1} { break } { ; } { - \str_aux_hexadecimal_use:NTF #1 - { ; \regex_escape_loop:N } - { ; \regex_escape_loop:N #1 } + \__str_hexadecimal_use:NTF #1 + { ; \__regex_escape_loop:N } + { ; \__regex_escape_loop:N #1 } } } -\cs_new:Npn \regex_escape_x_loop:N #1 +\cs_new:Npn \__regex_escape_x_loop:N #1 { - \str_aux_hexadecimal_use:NTF #1 - { \regex_escape_x_loop:N } + \__str_hexadecimal_use:NTF #1 + { \__regex_escape_x_loop:N } { \token_if_eq_charcode:NNTF \c_space_token #1 - { \regex_escape_x_loop:N } + { \__regex_escape_x_loop:N } { ; \exp_after:wN \token_if_eq_charcode:NNTF \c_rbrace_str #1 - { \regex_escape_loop:N } + { \__regex_escape_loop:N } { \if_false: { \fi: } - \tl_build_one:o \l_regex_internal_b_tl - \msg_kernel_error:nn { regex } { x-missing-rbrace } {#1} - \tl_set:Nx \l_regex_internal_b_tl - { \if_false: } \fi: \regex_escape_loop:N #1 + \__tl_build_one:o \l__regex_internal_b_tl + \__msg_kernel_error:nn { regex } { x-missing-rbrace } {#1} + \tl_set:Nx \l__regex_internal_b_tl + { \if_false: } \fi: \__regex_escape_loop:N #1 } } } } -\prg_new_conditional:Npnn \regex_char_if_special:N #1 { TF } +\prg_new_conditional:Npnn \__regex_char_if_special:N #1 { TF } { - \if_num:w `#1 < \c_ninety_one - \if_num:w `#1 < \c_fifty_eight - \if_num:w `#1 < \c_forty_eight - \if_num:w `#1 < \c_thirty_two + \if_int_compare:w `#1 < \c_ninety_one + \if_int_compare:w `#1 < \c_fifty_eight + \if_int_compare:w `#1 < \c_forty_eight + \if_int_compare:w `#1 < \c_thirty_two \prg_return_false: \else: \prg_return_true: \fi: \else: \prg_return_false: \fi: \else: - \if_num:w `#1 < \c_sixty_five + \if_int_compare:w `#1 < \c_sixty_five \prg_return_true: \else: \prg_return_false: \fi: \fi: \else: - \if_num:w `#1 < \c_one_hundred_twenty_three - \if_num:w `#1 < \c_ninety_seven + \if_int_compare:w `#1 < \c_one_hundred_twenty_three + \if_int_compare:w `#1 < \c_ninety_seven \prg_return_true: \else: \prg_return_false: \fi: \else: - \if_num:w `#1 < \c_one_hundred_twenty_seven + \if_int_compare:w `#1 < \c_one_hundred_twenty_seven \prg_return_true: \else: \prg_return_false: \fi: \fi: \fi: } -\prg_new_conditional:Npnn \regex_char_if_alphanumeric:N #1 { TF } +\prg_new_conditional:Npnn \__regex_char_if_alphanumeric:N #1 { TF } { - \if_num:w `#1 < \c_ninety_one - \if_num:w `#1 < \c_fifty_eight - \if_num:w `#1 < \c_forty_eight + \if_int_compare:w `#1 < \c_ninety_one + \if_int_compare:w `#1 < \c_fifty_eight + \if_int_compare:w `#1 < \c_forty_eight \prg_return_false: \else: \prg_return_true: \fi: \else: - \if_num:w `#1 < \c_sixty_five + \if_int_compare:w `#1 < \c_sixty_five \prg_return_false: \else: \prg_return_true: \fi: \fi: \else: - \if_num:w `#1 < \c_one_hundred_twenty_three - \if_num:w `#1 < \c_ninety_seven + \if_int_compare:w `#1 < \c_one_hundred_twenty_three + \if_int_compare:w `#1 < \c_ninety_seven \prg_return_false: \else: \prg_return_true: \fi: \else: \prg_return_false: \fi: \fi: } -\int_new:N \l_regex_group_level_int -\int_new:N \l_regex_mode_int -\int_new:N \l_regex_catcodes_int -\int_new:N \l_regex_default_catcodes_int -\bool_new:N \l_regex_catcodes_bool -\int_const:Nn \c_regex_catcode_C_int { "1 } -\int_const:Nn \c_regex_catcode_B_int { "4 } -\int_const:Nn \c_regex_catcode_E_int { "10 } -\int_const:Nn \c_regex_catcode_M_int { "40 } -\int_const:Nn \c_regex_catcode_T_int { "100 } -\int_const:Nn \c_regex_catcode_P_int { "1000 } -\int_const:Nn \c_regex_catcode_U_int { "4000 } -\int_const:Nn \c_regex_catcode_D_int { "10000 } -\int_const:Nn \c_regex_catcode_S_int { "100000 } -\int_const:Nn \c_regex_catcode_L_int { "400000 } -\int_const:Nn \c_regex_catcode_O_int { "1000000 } -\int_const:Nn \c_regex_catcode_A_int { "4000000 } -\int_const:Nn \c_regex_all_catcodes_int { "5515155 } -\cs_new_eq:NN \l_regex_internal_regex \c_regex_no_match_regex -\seq_new:N \l_regex_show_prefix_seq -\int_new:N \l_regex_show_lines_int -\cs_new_protected:Npn \regex_get_digits:NTFw #1#2#3#4#5 - { - \regex_if_raw_digit:NNTF #4 #5 - { #1 = #5 \regex_get_digits_loop:nw {#2} } +\int_new:N \l__regex_group_level_int +\int_new:N \l__regex_mode_int +\int_new:N \l__regex_catcodes_int +\int_new:N \l__regex_default_catcodes_int +\bool_new:N \l__regex_catcodes_bool +\int_const:Nn \c__regex_catcode_C_int { "1 } +\int_const:Nn \c__regex_catcode_B_int { "4 } +\int_const:Nn \c__regex_catcode_E_int { "10 } +\int_const:Nn \c__regex_catcode_M_int { "40 } +\int_const:Nn \c__regex_catcode_T_int { "100 } +\int_const:Nn \c__regex_catcode_P_int { "1000 } +\int_const:Nn \c__regex_catcode_U_int { "4000 } +\int_const:Nn \c__regex_catcode_D_int { "10000 } +\int_const:Nn \c__regex_catcode_S_int { "100000 } +\int_const:Nn \c__regex_catcode_L_int { "400000 } +\int_const:Nn \c__regex_catcode_O_int { "1000000 } +\int_const:Nn \c__regex_catcode_A_int { "4000000 } +\int_const:Nn \c__regex_all_catcodes_int { "5515155 } +\cs_new_eq:NN \l__regex_internal_regex \c__regex_no_match_regex +\seq_new:N \l__regex_show_prefix_seq +\int_new:N \l__regex_show_lines_int +\cs_new_protected:Npn \__regex_get_digits:NTFw #1#2#3#4#5 + { + \__regex_if_raw_digit:NNTF #4 #5 + { #1 = #5 \__regex_get_digits_loop:nw {#2} } { #3 #4 #5 } } -\cs_new:Npn \regex_get_digits_loop:nw #1#2#3 +\cs_new:Npn \__regex_get_digits_loop:nw #1#2#3 { - \regex_if_raw_digit:NNTF #2 #3 - { #3 \regex_get_digits_loop:nw {#1} } + \__regex_if_raw_digit:NNTF #2 #3 + { #3 \__regex_get_digits_loop:nw {#1} } { \scan_stop: #1 #2 #3 } } -\prg_new_conditional:Npnn \regex_if_raw_digit:NN #1#2 { TF } +\prg_new_conditional:Npnn \__regex_if_raw_digit:NN #1#2 { TF } { - \if_meaning:w \regex_compile_raw:N #1 + \if_meaning:w \__regex_compile_raw:N #1 \if_int_compare:w \c_one < 1 #2 \exp_stop_f: \prg_return_true: \else: @@ -497,305 +497,305 @@ \prg_return_false: \fi: } -\cs_new_nopar:Npn \regex_if_in_class:TF +\cs_new_nopar:Npn \__regex_if_in_class:TF { - \if_int_odd:w \l_regex_mode_int + \if_int_odd:w \l__regex_mode_int \exp_after:wN \use_i:nn \else: \exp_after:wN \use_ii:nn \fi: } -\cs_new_nopar:Npn \regex_if_in_cs:TF +\cs_new_nopar:Npn \__regex_if_in_cs:TF { - \if_int_odd:w \l_regex_mode_int + \if_int_odd:w \l__regex_mode_int \exp_after:wN \use_ii:nn \else: - \if_int_compare:w \l_regex_mode_int < \c_zero + \if_int_compare:w \l__regex_mode_int < \c_zero \exp_after:wN \exp_after:wN \exp_after:wN \use_i:nn \else: \exp_after:wN \exp_after:wN \exp_after:wN \use_ii:nn \fi: \fi: } -\cs_new_nopar:Npn \regex_if_in_class_or_catcode:TF +\cs_new_nopar:Npn \__regex_if_in_class_or_catcode:TF { - \if_int_odd:w \l_regex_mode_int + \if_int_odd:w \l__regex_mode_int \exp_after:wN \use_i:nn \else: - \if_int_compare:w \l_regex_mode_int > \c_zero + \if_int_compare:w \l__regex_mode_int > \c_zero \exp_after:wN \exp_after:wN \exp_after:wN \use_i:nn \else: \exp_after:wN \exp_after:wN \exp_after:wN \use_ii:nn \fi: \fi: } -\cs_new_nopar:Npn \regex_if_within_catcode:TF +\cs_new_nopar:Npn \__regex_if_within_catcode:TF { - \if_int_compare:w \l_regex_mode_int > \c_zero + \if_int_compare:w \l__regex_mode_int > \c_zero \exp_after:wN \use_i:nn \else: \exp_after:wN \use_ii:nn \fi: } -\cs_new_protected:Npn \regex_chk_c_allowed:T +\cs_new_protected:Npn \__regex_chk_c_allowed:T { - \if_num:w \l_regex_mode_int = \c_zero + \if_int_compare:w \l__regex_mode_int = \c_zero \exp_after:wN \use:n \else: - \if_num:w \l_regex_mode_int = \c_three + \if_int_compare:w \l__regex_mode_int = \c_three \exp_after:wN \exp_after:wN \exp_after:wN \use:n \else: - \msg_kernel_error:nn { regex } { c-bad-mode } + \__msg_kernel_error:nn { regex } { c-bad-mode } \exp_after:wN \exp_after:wN \exp_after:wN \use_none:n \fi: \fi: } -\cs_new_protected:Npn \regex_mode_quit_c: +\cs_new_protected:Npn \__regex_mode_quit_c: { - \if_num:w \l_regex_mode_int = \c_two - \l_regex_mode_int = \c_zero + \if_int_compare:w \l__regex_mode_int = \c_two + \l__regex_mode_int = \c_zero \else: - \if_num:w \l_regex_mode_int = \c_six - \l_regex_mode_int = \c_three + \if_int_compare:w \l__regex_mode_int = \c_six + \l__regex_mode_int = \c_three \fi: \fi: } -\cs_new_protected_nopar:Npn \regex_compile:w +\cs_new_protected_nopar:Npn \__regex_compile:w { - \tl_set_build_x:Nw \l_regex_internal_regex - \int_zero:N \l_regex_group_level_int - \int_set_eq:NN \l_regex_default_catcodes_int \c_regex_all_catcodes_int - \int_set_eq:NN \l_regex_catcodes_int \l_regex_default_catcodes_int - \cs_set_nopar:Npn \regex_item_equal:n { \regex_item_caseful_equal:n } - \cs_set_nopar:Npn \regex_item_range:nn { \regex_item_caseful_range:nn } - \tl_build_one:n { \regex_branch:n { \if_false: } \fi: } + \__tl_build_x:Nw \l__regex_internal_regex + \int_zero:N \l__regex_group_level_int + \int_set_eq:NN \l__regex_default_catcodes_int \c__regex_all_catcodes_int + \int_set_eq:NN \l__regex_catcodes_int \l__regex_default_catcodes_int + \cs_set_nopar:Npn \__regex_item_equal:n { \__regex_item_caseful_equal:n } + \cs_set_nopar:Npn \__regex_item_range:nn { \__regex_item_caseful_range:nn } + \__tl_build_one:n { \__regex_branch:n { \if_false: } \fi: } } -\cs_new_protected_nopar:Npn \regex_compile_end: +\cs_new_protected_nopar:Npn \__regex_compile_end: { - \regex_if_in_class:TF + \__regex_if_in_class:TF { - \msg_kernel_error:nn { regex } { missing-rbrack } - \use:c { regex_compile_]: } + \__msg_kernel_error:nn { regex } { missing-rbrack } + \use:c { __regex_compile_]: } \prg_do_nothing: \prg_do_nothing: } { } - \if_num:w \l_regex_group_level_int > \c_zero - \msg_kernel_error:nnx { regex } { missing-rparen } - { \int_use:N \l_regex_group_level_int } + \if_int_compare:w \l__regex_group_level_int > \c_zero + \__msg_kernel_error:nnx { regex } { missing-rparen } + { \int_use:N \l__regex_group_level_int } \prg_replicate:nn - { \l_regex_group_level_int } + { \l__regex_group_level_int } { - \tl_build_one:n + \__tl_build_one:n { \if_false: { \fi: } \if_false: { \fi: } { 1 } { 0 } \c_true_bool } - \tl_build_end: - \tl_build_one:o \l_regex_internal_regex + \__tl_build_end: + \__tl_build_one:o \l__regex_internal_regex } \fi: - \tl_build_one:n { \if_false: { \fi: } } - \tl_build_end: + \__tl_build_one:n { \if_false: { \fi: } } + \__tl_build_end: } -\cs_new_protected:Npn \regex_compile:n #1 +\cs_new_protected:Npn \__regex_compile:n #1 { - \regex_compile:w + \__regex_compile:w \int_set:Nn \tex_escapechar:D { 92 } - \int_set_eq:NN \l_regex_mode_int \c_zero - \regex_escape_use:nnnn + \int_set_eq:NN \l__regex_mode_int \c_zero + \__regex_escape_use:nnnn { - \regex_char_if_special:NTF ##1 - \regex_compile_special:N \regex_compile_raw:N ##1 + \__regex_char_if_special:NTF ##1 + \__regex_compile_special:N \__regex_compile_raw:N ##1 } { - \regex_char_if_alphanumeric:NTF ##1 - \regex_compile_escaped:N \regex_compile_raw:N ##1 + \__regex_char_if_alphanumeric:NTF ##1 + \__regex_compile_escaped:N \__regex_compile_raw:N ##1 } - { \regex_compile_raw:N ##1 } + { \__regex_compile_raw:N ##1 } { #1 } \prg_do_nothing: \prg_do_nothing: \prg_do_nothing: \prg_do_nothing: - \int_compare:nNnT \l_regex_mode_int < \c_zero + \int_compare:nNnT \l__regex_mode_int < \c_zero { - \msg_kernel_error:nn { regex } { c-missing-rbrace } - \regex_compile_end: - \regex_compile_one:x - { \regex_item_cs:n { \exp_not:o \l_regex_internal_regex } } + \__msg_kernel_error:nn { regex } { c-missing-rbrace } + \__regex_compile_end: + \__regex_compile_one:x + { \__regex_item_cs:n { \exp_not:o \l__regex_internal_regex } } \prg_do_nothing: \prg_do_nothing: \prg_do_nothing: \prg_do_nothing: } - \regex_compile_end: + \__regex_compile_end: } -\cs_new_protected:Npn \regex_compile_special:N #1 +\cs_new_protected:Npn \__regex_compile_special:N #1 { - \cs_if_exist_use:cF { regex_compile_#1: } - { \regex_compile_raw:N #1 } + \cs_if_exist_use:cF { __regex_compile_#1: } + { \__regex_compile_raw:N #1 } } -\cs_new_protected:Npn \regex_compile_escaped:N #1 +\cs_new_protected:Npn \__regex_compile_escaped:N #1 { - \cs_if_exist_use:cF { regex_compile_/#1: } - { \regex_compile_raw:N #1 } + \cs_if_exist_use:cF { __regex_compile_/#1: } + { \__regex_compile_raw:N #1 } } -\cs_new_protected:Npn \regex_compile_one:x #1 +\cs_new_protected:Npn \__regex_compile_one:x #1 { - \regex_mode_quit_c: - \regex_if_in_class:TF { } + \__regex_mode_quit_c: + \__regex_if_in_class:TF { } { - \tl_build_one:n - { \regex_class:NnnnN \c_true_bool { \if_false: } \fi: } + \__tl_build_one:n + { \__regex_class:NnnnN \c_true_bool { \if_false: } \fi: } } - \tl_build_one:x + \__tl_build_one:x { - \if_num:w \l_regex_catcodes_int < \c_regex_all_catcodes_int - \regex_item_catcode:nT { \int_use:N \l_regex_catcodes_int } + \if_int_compare:w \l__regex_catcodes_int < \c__regex_all_catcodes_int + \__regex_item_catcode:nT { \int_use:N \l__regex_catcodes_int } { \exp_not:N \exp_not:n {#1} } \else: \exp_not:N \exp_not:n {#1} \fi: } - \int_set_eq:NN \l_regex_catcodes_int \l_regex_default_catcodes_int - \regex_if_in_class:TF { } { \regex_compile_quantifier:w } + \int_set_eq:NN \l__regex_catcodes_int \l__regex_default_catcodes_int + \__regex_if_in_class:TF { } { \__regex_compile_quantifier:w } } -\cs_new_protected:Npn \regex_compile_abort_tokens:n #1 +\cs_new_protected:Npn \__regex_compile_abort_tokens:n #1 { \use:x { \exp_args:No \tl_map_function:nN { \tl_to_str:n {#1} } - \regex_compile_raw:N + \__regex_compile_raw:N } } -\cs_generate_variant:Nn \regex_compile_abort_tokens:n { x } -\cs_new_protected:Npn \regex_compile_quantifier:w #1#2 +\cs_generate_variant:Nn \__regex_compile_abort_tokens:n { x } +\cs_new_protected:Npn \__regex_compile_quantifier:w #1#2 { - \token_if_eq_meaning:NNTF #1 \regex_compile_special:N + \token_if_eq_meaning:NNTF #1 \__regex_compile_special:N { - \cs_if_exist_use:cF { regex_compile_quantifier_#2:w } - { \regex_compile_quantifier_none: #1 #2 } + \cs_if_exist_use:cF { __regex_compile_quantifier_#2:w } + { \__regex_compile_quantifier_none: #1 #2 } } - { \regex_compile_quantifier_none: #1 #2 } + { \__regex_compile_quantifier_none: #1 #2 } } -\cs_new_protected:Npn \regex_compile_quantifier_none: - { \tl_build_one:n { \if_false: { \fi: } { 1 } { 0 } \c_false_bool } } -\cs_new_protected:Npn \regex_compile_quantifier_abort:xNN #1#2#3 +\cs_new_protected:Npn \__regex_compile_quantifier_none: + { \__tl_build_one:n { \if_false: { \fi: } { 1 } { 0 } \c_false_bool } } +\cs_new_protected:Npn \__regex_compile_quantifier_abort:xNN #1#2#3 { - \regex_compile_quantifier_none: - \msg_kernel_warning:nnxx { regex } { invalid-quantifier } {#1} {#3} - \regex_compile_abort_tokens:x {#1} + \__regex_compile_quantifier_none: + \__msg_kernel_warning:nnxx { regex } { invalid-quantifier } {#1} {#3} + \__regex_compile_abort_tokens:x {#1} #2 #3 } -\cs_new_protected:Npn \regex_compile_quantifier_lazyness:nnNN #1#2#3#4 +\cs_new_protected:Npn \__regex_compile_quantifier_lazyness:nnNN #1#2#3#4 { - \str_if_eq:nnTF { #3 #4 } { \regex_compile_special:N ? } - { \tl_build_one:n { \if_false: { \fi: } { #1 } { #2 } \c_true_bool } } + \str_if_eq:nnTF { #3 #4 } { \__regex_compile_special:N ? } + { \__tl_build_one:n { \if_false: { \fi: } { #1 } { #2 } \c_true_bool } } { - \tl_build_one:n { \if_false: { \fi: } { #1 } { #2 } \c_false_bool } + \__tl_build_one:n { \if_false: { \fi: } { #1 } { #2 } \c_false_bool } #3 #4 } } -\cs_new_protected_nopar:cpn { regex_compile_quantifier_?:w } - { \regex_compile_quantifier_lazyness:nnNN { 0 } { 1 } } -\cs_new_protected_nopar:cpn { regex_compile_quantifier_*:w } - { \regex_compile_quantifier_lazyness:nnNN { 0 } { -1 } } -\cs_new_protected_nopar:cpn { regex_compile_quantifier_+:w } - { \regex_compile_quantifier_lazyness:nnNN { 1 } { -1 } } -\cs_new_protected:cpn { regex_compile_quantifier_ \c_lbrace_str :w } +\cs_new_protected_nopar:cpn { __regex_compile_quantifier_?:w } + { \__regex_compile_quantifier_lazyness:nnNN { 0 } { 1 } } +\cs_new_protected_nopar:cpn { __regex_compile_quantifier_*:w } + { \__regex_compile_quantifier_lazyness:nnNN { 0 } { -1 } } +\cs_new_protected_nopar:cpn { __regex_compile_quantifier_+:w } + { \__regex_compile_quantifier_lazyness:nnNN { 1 } { -1 } } +\cs_new_protected:cpn { __regex_compile_quantifier_ \c_lbrace_str :w } { - \regex_get_digits:NTFw \l_regex_internal_a_int - { \regex_compile_quantifier_braced_i:w } - { \regex_compile_quantifier_abort:xNN { \c_lbrace_str } } + \__regex_get_digits:NTFw \l__regex_internal_a_int + { \__regex_compile_quantifier_braced_i:w } + { \__regex_compile_quantifier_abort:xNN { \c_lbrace_str } } } -\cs_new_protected:Npn \regex_compile_quantifier_braced_i:w #1#2 +\cs_new_protected:Npn \__regex_compile_quantifier_braced_i:w #1#2 { - \prg_case_str:xxn { #1 #2 } + \str_case_x:nnn { #1 #2 } { - { \regex_compile_special:N \c_rbrace_str } + { \__regex_compile_special:N \c_rbrace_str } { - \exp_args:No \regex_compile_quantifier_lazyness:nnNN - { \int_use:N \l_regex_internal_a_int } { 0 } + \exp_args:No \__regex_compile_quantifier_lazyness:nnNN + { \int_use:N \l__regex_internal_a_int } { 0 } } - { \regex_compile_special:N , } + { \__regex_compile_special:N , } { - \regex_get_digits:NTFw \l_regex_internal_b_int - { \regex_compile_quantifier_braced_iii:w } - { \regex_compile_quantifier_braced_ii:w } + \__regex_get_digits:NTFw \l__regex_internal_b_int + { \__regex_compile_quantifier_braced_iii:w } + { \__regex_compile_quantifier_braced_ii:w } } } { - \regex_compile_quantifier_abort:xNN - { \c_lbrace_str \int_use:N \l_regex_internal_a_int } + \__regex_compile_quantifier_abort:xNN + { \c_lbrace_str \int_use:N \l__regex_internal_a_int } #1 #2 } } -\cs_new_protected:Npn \regex_compile_quantifier_braced_ii:w #1#2 +\cs_new_protected:Npn \__regex_compile_quantifier_braced_ii:w #1#2 { - \str_if_eq:xxTF - { #1 #2 } { \regex_compile_special:N \c_rbrace_str } + \str_if_eq_x:nnTF + { #1 #2 } { \__regex_compile_special:N \c_rbrace_str } { - \exp_args:No \regex_compile_quantifier_lazyness:nnNN - { \int_use:N \l_regex_internal_a_int } { -1 } + \exp_args:No \__regex_compile_quantifier_lazyness:nnNN + { \int_use:N \l__regex_internal_a_int } { -1 } } { - \regex_compile_quantifier_abort:xNN - { \c_lbrace_str \int_use:N \l_regex_internal_a_int , } + \__regex_compile_quantifier_abort:xNN + { \c_lbrace_str \int_use:N \l__regex_internal_a_int , } #1 #2 } } -\cs_new_protected:Npn \regex_compile_quantifier_braced_iii:w #1#2 +\cs_new_protected:Npn \__regex_compile_quantifier_braced_iii:w #1#2 { - \str_if_eq:xxTF - { #1 #2 } { \regex_compile_special:N \c_rbrace_str } + \str_if_eq_x:nnTF + { #1 #2 } { \__regex_compile_special:N \c_rbrace_str } { - \if_num:w \l_regex_internal_a_int > \l_regex_internal_b_int - \msg_kernel_error:nnxx { regex } { backwards-quantifier } - { \int_use:N \l_regex_internal_a_int } - { \int_use:N \l_regex_internal_b_int } - \int_zero:N \l_regex_internal_b_int + \if_int_compare:w \l__regex_internal_a_int > \l__regex_internal_b_int + \__msg_kernel_error:nnxx { regex } { backwards-quantifier } + { \int_use:N \l__regex_internal_a_int } + { \int_use:N \l__regex_internal_b_int } + \int_zero:N \l__regex_internal_b_int \else: - \int_sub:Nn \l_regex_internal_b_int \l_regex_internal_a_int + \int_sub:Nn \l__regex_internal_b_int \l__regex_internal_a_int \fi: - \exp_args:Noo \regex_compile_quantifier_lazyness:nnNN - { \int_use:N \l_regex_internal_a_int } - { \int_use:N \l_regex_internal_b_int } + \exp_args:Noo \__regex_compile_quantifier_lazyness:nnNN + { \int_use:N \l__regex_internal_a_int } + { \int_use:N \l__regex_internal_b_int } } { - \regex_compile_quantifier_abort:xNN + \__regex_compile_quantifier_abort:xNN { \c_lbrace_str - \int_use:N \l_regex_internal_a_int , - \int_use:N \l_regex_internal_b_int + \int_use:N \l__regex_internal_a_int , + \int_use:N \l__regex_internal_b_int } #1 #2 } } -\cs_new_protected:Npn \regex_compile_raw_error:N #1 +\cs_new_protected:Npn \__regex_compile_raw_error:N #1 { - \msg_kernel_error:nnx { regex } { bad-escape } {#1} - \regex_compile_raw:N #1 + \__msg_kernel_error:nnx { regex } { bad-escape } {#1} + \__regex_compile_raw:N #1 } -\cs_new_protected:Npn \regex_compile_raw:N #1#2#3 +\cs_new_protected:Npn \__regex_compile_raw:N #1#2#3 { - \regex_if_in_class:TF + \__regex_if_in_class:TF { - \str_if_eq:nnTF {#2#3} { \regex_compile_special:N - } - { \regex_compile_range:Nw #1 } + \str_if_eq:nnTF {#2#3} { \__regex_compile_special:N - } + { \__regex_compile_range:Nw #1 } { - \regex_compile_one:x - { \regex_item_equal:n { \int_value:w `#1 ~ } } + \__regex_compile_one:x + { \__regex_item_equal:n { \__int_value:w `#1 ~ } } #2 #3 } } { - \regex_compile_one:x - { \regex_item_equal:n { \int_value:w `#1 ~ } } + \__regex_compile_one:x + { \__regex_item_equal:n { \__int_value:w `#1 ~ } } #2 #3 } } -\prg_new_protected_conditional:Npnn \regex_if_end_range:NN #1#2 { TF } +\prg_new_protected_conditional:Npnn \__regex_if_end_range:NN #1#2 { TF } { - \if_meaning:w \regex_compile_raw:N #1 + \if_meaning:w \__regex_compile_raw:N #1 \prg_return_true: \else: - \if_meaning:w \regex_compile_special:N #1 + \if_meaning:w \__regex_compile_special:N #1 \if_charcode:w ] #2 \prg_return_false: \else: @@ -806,1815 +806,1823 @@ \fi: \fi: } -\cs_new_protected:Npn \regex_compile_range:Nw #1#2#3 +\cs_new_protected:Npn \__regex_compile_range:Nw #1#2#3 { - \regex_if_end_range:NNTF #2 #3 + \__regex_if_end_range:NNTF #2 #3 { - \if_num:w `#1 > `#3 \exp_stop_f: - \msg_kernel_error:nnxx { regex } { range-backwards } {#1} {#3} + \if_int_compare:w `#1 > `#3 \exp_stop_f: + \__msg_kernel_error:nnxx { regex } { range-backwards } {#1} {#3} \else: - \tl_build_one:x + \__tl_build_one:x { - \if_num:w `#1 = `#3 \exp_stop_f: - \regex_item_equal:n + \if_int_compare:w `#1 = `#3 \exp_stop_f: + \__regex_item_equal:n \else: - \regex_item_range:nn { \int_value:w `#1 ~ } + \__regex_item_range:nn { \__int_value:w `#1 ~ } \fi: - { \int_value:w `#3 ~ } + { \__int_value:w `#3 ~ } } \fi: } { - \msg_kernel_warning:nnxx { regex } { range-missing-end } + \__msg_kernel_warning:nnxx { regex } { range-missing-end } {#1} { \c_backslash_str #3 } - \tl_build_one:x + \__tl_build_one:x { - \regex_item_equal:n { \int_value:w `#1 ~ } - \regex_item_equal:n { \int_value:w `- ~ } + \__regex_item_equal:n { \__int_value:w `#1 ~ } + \__regex_item_equal:n { \__int_value:w `- ~ } } #2#3 } } -\cs_new_protected_nopar:cpx { regex_compile_.: } +\cs_new_protected_nopar:cpx { __regex_compile_.: } { - \exp_not:N \regex_if_in_class:TF - { \regex_compile_raw:N . } - { \regex_compile_one:x \exp_not:c { regex_prop_.: } } + \exp_not:N \__regex_if_in_class:TF + { \__regex_compile_raw:N . } + { \__regex_compile_one:x \exp_not:c { __regex_prop_.: } } } -\cs_new_protected_nopar:cpn { regex_prop_.: } +\cs_new_protected_nopar:cpn { __regex_prop_.: } { - \if_num:w \l_regex_current_char_int > - \c_two - \exp_after:wN \regex_break_true:w + \if_int_compare:w \l__regex_current_char_int > - \c_two + \exp_after:wN \__regex_break_true:w \fi: } -\cs_set_protected:Npn \regex_tmp:w #1#2 +\cs_set_protected:Npn \__regex_tmp:w #1#2 { - \cs_new_protected_nopar:cpx { regex_compile_/#1: } - { \regex_compile_one:x \exp_not:c { regex_prop_#1: } } - \cs_new_protected_nopar:cpx { regex_compile_/#2: } + \cs_new_protected_nopar:cpx { __regex_compile_/#1: } + { \__regex_compile_one:x \exp_not:c { __regex_prop_#1: } } + \cs_new_protected_nopar:cpx { __regex_compile_/#2: } { - \regex_compile_one:x - { \regex_item_reverse:n \exp_not:c { regex_prop_#1: } } + \__regex_compile_one:x + { \__regex_item_reverse:n \exp_not:c { __regex_prop_#1: } } } } -\regex_tmp:w d D -\regex_tmp:w h H -\regex_tmp:w s S -\regex_tmp:w v V -\regex_tmp:w w W -\cs_new_protected_nopar:cpn { regex_compile_/N: } - { \regex_compile_one:x \regex_prop_N: } -\cs_new_protected:Npn \regex_compile_anchor:NF #1#2 +\__regex_tmp:w d D +\__regex_tmp:w h H +\__regex_tmp:w s S +\__regex_tmp:w v V +\__regex_tmp:w w W +\cs_new_protected_nopar:cpn { __regex_compile_/N: } + { \__regex_compile_one:x \__regex_prop_N: } +\cs_new_protected:Npn \__regex_compile_anchor:NF #1#2 { - \regex_if_in_class_or_catcode:TF {#2} + \__regex_if_in_class_or_catcode:TF {#2} { - \tl_build_one:n - { \regex_assertion:Nn \c_true_bool { \regex_anchor:N #1 } } + \__tl_build_one:n + { \__regex_assertion:Nn \c_true_bool { \__regex_anchor:N #1 } } } } -\cs_set_protected:Npn \regex_tmp:w #1#2 +\cs_set_protected:Npn \__regex_tmp:w #1#2 { - \cs_new_protected_nopar:cpn { regex_compile_/#1: } - { \regex_compile_anchor:NF #2 { \regex_compile_raw_error:N #1 } } + \cs_new_protected_nopar:cpn { __regex_compile_/#1: } + { \__regex_compile_anchor:NF #2 { \__regex_compile_raw_error:N #1 } } } -\regex_tmp:w A \l_regex_min_pos_int -\regex_tmp:w G \l_regex_start_pos_int -\regex_tmp:w Z \l_regex_max_pos_int -\regex_tmp:w z \l_regex_max_pos_int -\cs_set_protected:Npn \regex_tmp:w #1#2 +\__regex_tmp:w A \l__regex_min_pos_int +\__regex_tmp:w G \l__regex_start_pos_int +\__regex_tmp:w Z \l__regex_max_pos_int +\__regex_tmp:w z \l__regex_max_pos_int +\cs_set_protected:Npn \__regex_tmp:w #1#2 { - \cs_new_protected_nopar:cpn { regex_compile_#1: } - { \regex_compile_anchor:NF #2 { \regex_compile_raw:N #1 } } + \cs_new_protected_nopar:cpn { __regex_compile_#1: } + { \__regex_compile_anchor:NF #2 { \__regex_compile_raw:N #1 } } } -\exp_args:Nx \regex_tmp:w { \iow_char:N \^ } \l_regex_min_pos_int -\exp_args:Nx \regex_tmp:w { \iow_char:N \$ } \l_regex_max_pos_int -\cs_new_protected_nopar:cpn { regex_compile_/b: } +\exp_args:Nx \__regex_tmp:w { \iow_char:N \^ } \l__regex_min_pos_int +\exp_args:Nx \__regex_tmp:w { \iow_char:N \$ } \l__regex_max_pos_int +\cs_new_protected_nopar:cpn { __regex_compile_/b: } { - \regex_if_in_class_or_catcode:TF - { \regex_compile_raw_error:N b } + \__regex_if_in_class_or_catcode:TF + { \__regex_compile_raw_error:N b } { - \tl_build_one:n - { \regex_assertion:Nn \c_true_bool { \regex_b_test: } } + \__tl_build_one:n + { \__regex_assertion:Nn \c_true_bool { \__regex_b_test: } } } } -\cs_new_protected_nopar:cpn { regex_compile_/B: } +\cs_new_protected_nopar:cpn { __regex_compile_/B: } { - \regex_if_in_class_or_catcode:TF - { \regex_compile_raw_error:N B } + \__regex_if_in_class_or_catcode:TF + { \__regex_compile_raw_error:N B } { - \tl_build_one:n - { \regex_assertion:Nn \c_false_bool { \regex_b_test: } } + \__tl_build_one:n + { \__regex_assertion:Nn \c_false_bool { \__regex_b_test: } } } } -\cs_new_protected:cpn { regex_compile_]: } +\cs_new_protected:cpn { __regex_compile_]: } { - \regex_if_in_class:TF + \__regex_if_in_class:TF { - \if_num:w \l_regex_mode_int > \c_sixteen - \tl_build_one:n { \if_false: { \fi: } } + \if_int_compare:w \l__regex_mode_int > \c_sixteen + \__tl_build_one:n { \if_false: { \fi: } } \fi: - \tex_advance:D \l_regex_mode_int - \c_fifteen - \tex_divide:D \l_regex_mode_int \c_thirteen - \if_int_odd:w \l_regex_mode_int \else: - \exp_after:wN \regex_compile_quantifier:w + \tex_advance:D \l__regex_mode_int - \c_fifteen + \tex_divide:D \l__regex_mode_int \c_thirteen + \if_int_odd:w \l__regex_mode_int \else: + \exp_after:wN \__regex_compile_quantifier:w \fi: } - { \regex_compile_raw:N ] } + { \__regex_compile_raw:N ] } } -\cs_new_protected_nopar:cpn { regex_compile_[: } +\cs_new_protected_nopar:cpn { __regex_compile_[: } { - \regex_if_in_class:TF - { \regex_compile_class_posix_test:w } + \__regex_if_in_class:TF + { \__regex_compile_class_posix_test:w } { - \regex_if_within_catcode:TF + \__regex_if_within_catcode:TF { - \exp_after:wN \regex_compile_class_catcode:w - \int_use:N \l_regex_catcodes_int ; + \exp_after:wN \__regex_compile_class_catcode:w + \int_use:N \l__regex_catcodes_int ; } - { \regex_compile_class_normal:w } + { \__regex_compile_class_normal:w } } } -\cs_new_protected_nopar:Npn \regex_compile_class_normal:w +\cs_new_protected_nopar:Npn \__regex_compile_class_normal:w { - \regex_compile_class:TFNN - { \regex_class:NnnnN \c_true_bool } - { \regex_class:NnnnN \c_false_bool } + \__regex_compile_class:TFNN + { \__regex_class:NnnnN \c_true_bool } + { \__regex_class:NnnnN \c_false_bool } } -\cs_new_protected:Npn \regex_compile_class_catcode:w #1; +\cs_new_protected:Npn \__regex_compile_class_catcode:w #1; { - \if_int_compare:w \l_regex_mode_int = \c_two - \tl_build_one:n - { \regex_class:NnnnN \c_true_bool { \if_false: } \fi: } + \if_int_compare:w \l__regex_mode_int = \c_two + \__tl_build_one:n + { \__regex_class:NnnnN \c_true_bool { \if_false: } \fi: } \fi: - \int_set_eq:NN \l_regex_catcodes_int \l_regex_default_catcodes_int - \regex_compile_class:TFNN - { \regex_item_catcode:nT {#1} } - { \regex_item_catcode_reverse:nT {#1} } + \int_set_eq:NN \l__regex_catcodes_int \l__regex_default_catcodes_int + \__regex_compile_class:TFNN + { \__regex_item_catcode:nT {#1} } + { \__regex_item_catcode_reverse:nT {#1} } } -\cs_new_protected:Npn \regex_compile_class:TFNN #1#2#3#4 +\cs_new_protected:Npn \__regex_compile_class:TFNN #1#2#3#4 { - \l_regex_mode_int = \int_value:w \l_regex_mode_int 3 \exp_stop_f: - \str_if_eq:nnTF { #3 #4 } { \regex_compile_special:N ^ } + \l__regex_mode_int = \__int_value:w \l__regex_mode_int 3 \exp_stop_f: + \str_if_eq:nnTF { #3 #4 } { \__regex_compile_special:N ^ } { - \tl_build_one:n { #2 { \if_false: } \fi: } - \regex_compile_class_ii:NN + \__tl_build_one:n { #2 { \if_false: } \fi: } + \__regex_compile_class_ii:NN } { - \tl_build_one:n { #1 { \if_false: } \fi: } - \regex_compile_class_ii:NN #3 #4 + \__tl_build_one:n { #1 { \if_false: } \fi: } + \__regex_compile_class_ii:NN #3 #4 } } -\cs_new_protected:Npn \regex_compile_class_ii:NN #1#2 +\cs_new_protected:Npn \__regex_compile_class_ii:NN #1#2 { \token_if_eq_charcode:NNTF #2 ] - { \regex_compile_raw:N #2 } + { \__regex_compile_raw:N #2 } { #1 #2 } } -\cs_new_protected:Npn \regex_compile_class_posix_test:w #1#2 +\cs_new_protected:Npn \__regex_compile_class_posix_test:w #1#2 { - \token_if_eq_meaning:NNT \regex_compile_special:N #1 + \token_if_eq_meaning:NNT \__regex_compile_special:N #1 { - \prg_case_str:nnn { #2 } + \str_case:nnn { #2 } { - : { \regex_compile_class_posix:NNNNw } - = { \msg_kernel_warning:nnx { regex } { posix-unsupported } { = } } - . { \msg_kernel_warning:nnx { regex } { posix-unsupported } { . } } + : { \__regex_compile_class_posix:NNNNw } + = { \__msg_kernel_warning:nnx { regex } { posix-unsupported } { = } } + . { \__msg_kernel_warning:nnx { regex } { posix-unsupported } { . } } } { } } - \regex_compile_raw:N [ #1 #2 + \__regex_compile_raw:N [ #1 #2 } -\cs_new_protected:Npn \regex_compile_class_posix:NNNNw #1#2#3#4#5#6 +\cs_new_protected:Npn \__regex_compile_class_posix:NNNNw #1#2#3#4#5#6 { - \str_if_eq:nnTF { #5 #6 } { \regex_compile_special:N ^ } + \str_if_eq:nnTF { #5 #6 } { \__regex_compile_special:N ^ } { - \bool_set_false:N \l_regex_internal_bool - \tl_set:Nx \l_regex_internal_a_tl { \if_false: } \fi: - \regex_compile_class_posix_loop:w + \bool_set_false:N \l__regex_internal_bool + \tl_set:Nx \l__regex_internal_a_tl { \if_false: } \fi: + \__regex_compile_class_posix_loop:w } { - \bool_set_true:N \l_regex_internal_bool - \tl_set:Nx \l_regex_internal_a_tl { \if_false: } \fi: - \regex_compile_class_posix_loop:w #5 #6 + \bool_set_true:N \l__regex_internal_bool + \tl_set:Nx \l__regex_internal_a_tl { \if_false: } \fi: + \__regex_compile_class_posix_loop:w #5 #6 } } -\cs_new:Npn \regex_compile_class_posix_loop:w #1#2 +\cs_new:Npn \__regex_compile_class_posix_loop:w #1#2 { - \token_if_eq_meaning:NNTF \regex_compile_raw:N #1 - { #2 \regex_compile_class_posix_loop:w } - { \if_false: { \fi: } \regex_compile_class_posix_end:w #1 #2 } + \token_if_eq_meaning:NNTF \__regex_compile_raw:N #1 + { #2 \__regex_compile_class_posix_loop:w } + { \if_false: { \fi: } \__regex_compile_class_posix_end:w #1 #2 } } -\cs_new_protected:Npn \regex_compile_class_posix_end:w #1#2#3#4 +\cs_new_protected:Npn \__regex_compile_class_posix_end:w #1#2#3#4 { \str_if_eq:nnTF { #1 #2 #3 #4 } - { \regex_compile_special:N : \regex_compile_special:N ] } + { \__regex_compile_special:N : \__regex_compile_special:N ] } { - \cs_if_exist:cTF { regex_posix_ \l_regex_internal_a_tl : } + \cs_if_exist:cTF { __regex_posix_ \l__regex_internal_a_tl : } { - \regex_compile_one:x + \__regex_compile_one:x { - \bool_if:NF \l_regex_internal_bool \regex_item_reverse:n - \exp_not:c { regex_posix_ \l_regex_internal_a_tl : } + \bool_if:NF \l__regex_internal_bool \__regex_item_reverse:n + \exp_not:c { __regex_posix_ \l__regex_internal_a_tl : } } } { - \msg_kernel_warning:nnx { regex } { posix-unknown } - { \l_regex_internal_a_tl } - \regex_compile_abort_tokens:x + \__msg_kernel_warning:nnx { regex } { posix-unknown } + { \l__regex_internal_a_tl } + \__regex_compile_abort_tokens:x { - [: \bool_if:NF \l_regex_internal_bool { ^ } - \l_regex_internal_a_tl :] + [: \bool_if:NF \l__regex_internal_bool { ^ } + \l__regex_internal_a_tl :] } } } { - \msg_kernel_error:nnxx { regex } { posix-missing-close } - { [: \l_regex_internal_a_tl } { #2 #4 } - \regex_compile_abort_tokens:x { [: \l_regex_internal_a_tl } + \__msg_kernel_error:nnxx { regex } { posix-missing-close } + { [: \l__regex_internal_a_tl } { #2 #4 } + \__regex_compile_abort_tokens:x { [: \l__regex_internal_a_tl } #1 #2 #3 #4 } } -\cs_new_protected:Npn \regex_compile_group_begin:N #1 +\cs_new_protected:Npn \__regex_compile_group_begin:N #1 { - \tl_build_one:n { #1 { \if_false: } \fi: } - \regex_mode_quit_c: - \tl_set_build:Nw \l_regex_internal_regex - \int_set_eq:NN \l_regex_default_catcodes_int \l_regex_catcodes_int - \int_incr:N \l_regex_group_level_int - \tl_build_one:n { \regex_branch:n { \if_false: } \fi: } + \__tl_build_one:n { #1 { \if_false: } \fi: } + \__regex_mode_quit_c: + \__tl_build:Nw \l__regex_internal_regex + \int_set_eq:NN \l__regex_default_catcodes_int \l__regex_catcodes_int + \int_incr:N \l__regex_group_level_int + \__tl_build_one:n { \__regex_branch:n { \if_false: } \fi: } } -\cs_new_protected:Npn \regex_compile_group_end: +\cs_new_protected:Npn \__regex_compile_group_end: { - \if_num:w \l_regex_group_level_int > \c_zero - \tl_build_one:n { \if_false: { \fi: } } - \tl_build_end: - \int_set_eq:NN \l_regex_catcodes_int \l_regex_default_catcodes_int - \tl_build_one:o \l_regex_internal_regex - \exp_after:wN \regex_compile_quantifier:w + \if_int_compare:w \l__regex_group_level_int > \c_zero + \__tl_build_one:n { \if_false: { \fi: } } + \__tl_build_end: + \int_set_eq:NN \l__regex_catcodes_int \l__regex_default_catcodes_int + \__tl_build_one:o \l__regex_internal_regex + \exp_after:wN \__regex_compile_quantifier:w \else: - \msg_kernel_warning:nn { regex } { extra-rparen } - \exp_after:wN \regex_compile_raw:N \exp_after:wN ) + \__msg_kernel_warning:nn { regex } { extra-rparen } + \exp_after:wN \__regex_compile_raw:N \exp_after:wN ) \fi: } -\cs_new_protected_nopar:cpn { regex_compile_(: } +\cs_new_protected_nopar:cpn { __regex_compile_(: } { - \regex_if_in_class:TF { \regex_compile_raw:N ( } - { \regex_compile_lparen:w } + \__regex_if_in_class:TF { \__regex_compile_raw:N ( } + { \__regex_compile_lparen:w } } -\cs_new_protected:Npn \regex_compile_lparen:w #1#2#3#4 +\cs_new_protected:Npn \__regex_compile_lparen:w #1#2#3#4 { - \str_if_eq:nnTF { #1 #2 } { \regex_compile_special:N ? } + \str_if_eq:nnTF { #1 #2 } { \__regex_compile_special:N ? } { \cs_if_exist_use:cF - { regex_compile_special_group_\token_to_str:N #4 :w } + { __regex_compile_special_group_\token_to_str:N #4 :w } { - \msg_kernel_warning:nnx { regex } { special-group-unknown } + \__msg_kernel_warning:nnx { regex } { special-group-unknown } { (? \token_to_str:N #4 } - \regex_compile_group_begin:N \regex_group:nnnN - \regex_compile_raw:N ? #3 #4 + \__regex_compile_group_begin:N \__regex_group:nnnN + \__regex_compile_raw:N ? #3 #4 } } { - \regex_compile_group_begin:N \regex_group:nnnN + \__regex_compile_group_begin:N \__regex_group:nnnN #1 #2 #3 #4 } } -\cs_new_protected_nopar:cpn { regex_compile_|: } +\cs_new_protected_nopar:cpn { __regex_compile_|: } { - \regex_if_in_class:TF { \regex_compile_raw:N | } + \__regex_if_in_class:TF { \__regex_compile_raw:N | } { - \tl_build_one:n - { \if_false: { \fi: } \regex_branch:n { \if_false: } \fi: } + \__tl_build_one:n + { \if_false: { \fi: } \__regex_branch:n { \if_false: } \fi: } } } -\cs_new_protected_nopar:cpn { regex_compile_): } +\cs_new_protected_nopar:cpn { __regex_compile_): } { - \regex_if_in_class:TF { \regex_compile_raw:N ) } - { \regex_compile_group_end: } + \__regex_if_in_class:TF { \__regex_compile_raw:N ) } + { \__regex_compile_group_end: } } -\cs_new_protected_nopar:cpn { regex_compile_special_group_::w } - { \regex_compile_group_begin:N \regex_group_no_capture:nnnN } -\cs_new_protected_nopar:cpn { regex_compile_special_group_|:w } - { \regex_compile_group_begin:N \regex_group_resetting:nnnN } -\cs_new_protected:Npn \regex_compile_special_group_i:w #1#2 +\cs_new_protected_nopar:cpn { __regex_compile_special_group_::w } + { \__regex_compile_group_begin:N \__regex_group_no_capture:nnnN } +\cs_new_protected_nopar:cpn { __regex_compile_special_group_|:w } + { \__regex_compile_group_begin:N \__regex_group_resetting:nnnN } +\cs_new_protected:Npn \__regex_compile_special_group_i:w #1#2 { - \str_if_eq:nnTF { #1 #2 } { \regex_compile_special:N ) } + \str_if_eq:nnTF { #1 #2 } { \__regex_compile_special:N ) } { - \cs_set_nopar:Npn \regex_item_equal:n { \regex_item_caseless_equal:n } - \cs_set_nopar:Npn \regex_item_range:nn { \regex_item_caseless_range:nn } + \cs_set_nopar:Npn \__regex_item_equal:n { \__regex_item_caseless_equal:n } + \cs_set_nopar:Npn \__regex_item_range:nn { \__regex_item_caseless_range:nn } } { - \msg_kernel_warning:nnx { regex } { unknown-option } { (?i #2 } - \regex_compile_raw:N ( - \regex_compile_raw:N ? - \regex_compile_raw:N i + \__msg_kernel_warning:nnx { regex } { unknown-option } { (?i #2 } + \__regex_compile_raw:N ( + \__regex_compile_raw:N ? + \__regex_compile_raw:N i #1 #2 } } -\cs_new_protected_nopar:cpn { regex_compile_special_group_-:w } #1#2#3#4 +\cs_new_protected_nopar:cpn { __regex_compile_special_group_-:w } #1#2#3#4 { \str_if_eq:nnTF { #1 #2 #3 #4 } - { \regex_compile_raw:N i \regex_compile_special:N ) } + { \__regex_compile_raw:N i \__regex_compile_special:N ) } { - \cs_set_nopar:Npn \regex_item_equal:n { \regex_item_caseful_equal:n } - \cs_set_nopar:Npn \regex_item_range:nn { \regex_item_caseful_range:nn } + \cs_set_nopar:Npn \__regex_item_equal:n { \__regex_item_caseful_equal:n } + \cs_set_nopar:Npn \__regex_item_range:nn { \__regex_item_caseful_range:nn } } { - \msg_kernel_warning:nnx { regex } { unknown-option } { (?-#2#4 } - \regex_compile_raw:N ( - \regex_compile_raw:N ? - \regex_compile_raw:N - + \__msg_kernel_warning:nnx { regex } { unknown-option } { (?-#2#4 } + \__regex_compile_raw:N ( + \__regex_compile_raw:N ? + \__regex_compile_raw:N - #1 #2 #3 #4 } } -\cs_new_protected:cpn { regex_compile_/c: } - { \regex_chk_c_allowed:T { \regex_compile_c_test:NN } } -\cs_new_protected:Npn \regex_compile_c_test:NN #1#2 +\cs_new_protected:cpn { __regex_compile_/c: } + { \__regex_chk_c_allowed:T { \__regex_compile_c_test:NN } } +\cs_new_protected:Npn \__regex_compile_c_test:NN #1#2 { - \token_if_eq_meaning:NNTF #1 \regex_compile_raw:N + \token_if_eq_meaning:NNTF #1 \__regex_compile_raw:N { - \int_if_exist:cTF { c_regex_catcode_#2_int } + \int_if_exist:cTF { c__regex_catcode_#2_int } { - \int_set_eq:Nc \l_regex_catcodes_int { c_regex_catcode_#2_int } - \l_regex_mode_int - = \if_case:w \l_regex_mode_int \c_two \else: \c_six \fi: + \int_set_eq:Nc \l__regex_catcodes_int { c__regex_catcode_#2_int } + \l__regex_mode_int + = \if_case:w \l__regex_mode_int \c_two \else: \c_six \fi: } } - { \cs_if_exist_use:cF { regex_compile_c_#2:w } } + { \cs_if_exist_use:cF { __regex_compile_c_#2:w } } { - \msg_kernel_error:nnx { regex } { c-missing-category } {#2} + \__msg_kernel_error:nnx { regex } { c-missing-category } {#2} #1 #2 } } -\cs_new_protected:cpn { regex_compile_c_[:w } #1#2 +\cs_new_protected:cpn { __regex_compile_c_[:w } #1#2 { - \l_regex_mode_int - = \if_case:w \l_regex_mode_int \c_two \else: \c_six \fi: - \int_zero:N \l_regex_catcodes_int - \str_if_eq:nnTF { #1 #2 } { \regex_compile_special:N ^ } + \l__regex_mode_int + = \if_case:w \l__regex_mode_int \c_two \else: \c_six \fi: + \int_zero:N \l__regex_catcodes_int + \str_if_eq:nnTF { #1 #2 } { \__regex_compile_special:N ^ } { - \bool_set_false:N \l_regex_catcodes_bool - \regex_compile_c_lbrack_loop:NN + \bool_set_false:N \l__regex_catcodes_bool + \__regex_compile_c_lbrack_loop:NN } { - \bool_set_true:N \l_regex_catcodes_bool - \regex_compile_c_lbrack_loop:NN + \bool_set_true:N \l__regex_catcodes_bool + \__regex_compile_c_lbrack_loop:NN #1 #2 } } -\cs_new_protected:Npn \regex_compile_c_lbrack_loop:NN #1#2 +\cs_new_protected:Npn \__regex_compile_c_lbrack_loop:NN #1#2 { - \token_if_eq_meaning:NNTF #1 \regex_compile_raw:N + \token_if_eq_meaning:NNTF #1 \__regex_compile_raw:N { - \int_if_exist:cTF { c_regex_catcode_#2_int } + \int_if_exist:cTF { c__regex_catcode_#2_int } { - \exp_args:Nc \regex_compile_c_lbrack_add:N - { c_regex_catcode_#2_int } - \regex_compile_c_lbrack_loop:NN + \exp_args:Nc \__regex_compile_c_lbrack_add:N + { c__regex_catcode_#2_int } + \__regex_compile_c_lbrack_loop:NN } } { \token_if_eq_charcode:NNTF #2 ] - { \regex_compile_c_lbrack_end: } + { \__regex_compile_c_lbrack_end: } } { - \msg_kernel_error:nnx { regex } { c-missing-rbrack } {#2} - \regex_compile_c_lbrack_end: + \__msg_kernel_error:nnx { regex } { c-missing-rbrack } {#2} + \__regex_compile_c_lbrack_end: #1 #2 } } -\cs_new_protected:Npn \regex_compile_c_lbrack_add:N #1 +\cs_new_protected:Npn \__regex_compile_c_lbrack_add:N #1 { - \if_int_odd:w \int_eval:w \l_regex_catcodes_int / #1 \int_eval_end: + \if_int_odd:w \__int_eval:w \l__regex_catcodes_int / #1 \__int_eval_end: \else: - \tex_advance:D \l_regex_catcodes_int #1 + \tex_advance:D \l__regex_catcodes_int #1 \fi: } -\cs_new_protected_nopar:Npn \regex_compile_c_lbrack_end: +\cs_new_protected_nopar:Npn \__regex_compile_c_lbrack_end: { - \if_meaning:w \c_false_bool \l_regex_catcodes_bool - \int_set:Nn \l_regex_catcodes_int - { \c_regex_all_catcodes_int - \l_regex_catcodes_int } + \if_meaning:w \c_false_bool \l__regex_catcodes_bool + \int_set:Nn \l__regex_catcodes_int + { \c__regex_all_catcodes_int - \l__regex_catcodes_int } \fi: } -\cs_new_protected_nopar:cpn { regex_compile_c_ \c_lbrace_str :w } +\cs_new_protected_nopar:cpn { __regex_compile_c_ \c_lbrace_str :w } { - \regex_compile:w - \regex_disable_submatches: - \l_regex_mode_int - = - \if_case:w \l_regex_mode_int \c_two \else: \c_six \fi: + \__regex_compile:w + \__regex_disable_submatches: + \l__regex_mode_int + = - \if_case:w \l__regex_mode_int \c_two \else: \c_six \fi: } -\cs_new_protected:cpn { regex_compile_ \c_rbrace_str : } +\cs_new_protected:cpn { __regex_compile_ \c_rbrace_str : } { - \regex_if_in_cs:TF + \__regex_if_in_cs:TF { - \regex_compile_end: - \regex_compile_one:x - { \regex_item_cs:n { \exp_not:o \l_regex_internal_regex } } + \__regex_compile_end: + \__regex_compile_one:x + { \__regex_item_cs:n { \exp_not:o \l__regex_internal_regex } } } - { \exp_after:wN \regex_compile_raw:N \c_rbrace_str } + { \exp_after:wN \__regex_compile_raw:N \c_rbrace_str } } -\cs_new_protected:cpn { regex_compile_/u: } #1#2 +\cs_new_protected:cpn { __regex_compile_/u: } #1#2 { - \regex_if_in_class_or_catcode:TF - { \regex_compile_raw_error:N u #1 #2 } + \__regex_if_in_class_or_catcode:TF + { \__regex_compile_raw_error:N u #1 #2 } { - \str_if_eq:xxTF {#1#2} { \regex_compile_special:N \c_lbrace_str } + \str_if_eq_x:nnTF {#1#2} { \__regex_compile_special:N \c_lbrace_str } { - \tl_set:Nx \l_regex_internal_a_tl { \if_false: } \fi: - \regex_compile_u_loop:NN + \tl_set:Nx \l__regex_internal_a_tl { \if_false: } \fi: + \__regex_compile_u_loop:NN } { - \msg_kernel_error:nn { regex } { u-missing-lbrace } - \regex_compile_raw:N u #1 #2 + \__msg_kernel_error:nn { regex } { u-missing-lbrace } + \__regex_compile_raw:N u #1 #2 } } } -\cs_new:Npn \regex_compile_u_loop:NN #1#2 +\cs_new:Npn \__regex_compile_u_loop:NN #1#2 { - \token_if_eq_meaning:NNTF #1 \regex_compile_raw:N - { #2 \regex_compile_u_loop:NN } + \token_if_eq_meaning:NNTF #1 \__regex_compile_raw:N + { #2 \__regex_compile_u_loop:NN } { - \token_if_eq_meaning:NNTF #1 \regex_compile_special:N + \token_if_eq_meaning:NNTF #1 \__regex_compile_special:N { \exp_after:wN \token_if_eq_charcode:NNTF \c_rbrace_str #2 - { \if_false: { \fi: } \regex_compile_u_end: } - { #2 \regex_compile_u_loop:NN } + { \if_false: { \fi: } \__regex_compile_u_end: } + { #2 \__regex_compile_u_loop:NN } } { \if_false: { \fi: } - \msg_kernel_error:nnx { regex } { u-missing-rbrace } {#2} - \regex_compile_u_end: + \__msg_kernel_error:nnx { regex } { u-missing-rbrace } {#2} + \__regex_compile_u_end: #1 #2 } } } -\cs_new_protected:Npn \regex_compile_u_end: +\cs_new_protected:Npn \__regex_compile_u_end: { - \tl_set:Nv \l_regex_internal_a_tl { \l_regex_internal_a_tl } - \if_num:w \l_regex_mode_int = \c_zero - \regex_compile_u_not_cs: + \tl_set:Nv \l__regex_internal_a_tl { \l__regex_internal_a_tl } + \if_int_compare:w \l__regex_mode_int = \c_zero + \__regex_compile_u_not_cs: \else: - \regex_compile_u_in_cs: + \__regex_compile_u_in_cs: \fi: } -\cs_new_protected:Npn \regex_compile_u_in_cs: +\cs_new_protected:Npn \__regex_compile_u_in_cs: { - \exp_args:NNo \str_gset_other:Nn \g_regex_internal_tl - { \l_regex_internal_a_tl } - \tl_build_one:x + \exp_args:NNo \__str_gset_other:Nn \g__regex_internal_tl + { \l__regex_internal_a_tl } + \__tl_build_one:x { - \tl_map_function:NN \g_regex_internal_tl - \regex_compile_u_in_cs_aux:n + \tl_map_function:NN \g__regex_internal_tl + \__regex_compile_u_in_cs_aux:n } } -\cs_new:Npn \regex_compile_u_in_cs_aux:n #1 +\cs_new:Npn \__regex_compile_u_in_cs_aux:n #1 { - \regex_class:NnnnN \c_true_bool - { \regex_item_caseful_equal:n { \int_value:w `#1 } } + \__regex_class:NnnnN \c_true_bool + { \__regex_item_caseful_equal:n { \__int_value:w `#1 } } { 1 } { 0 } \c_false_bool } -\cs_new_protected:Npn \regex_compile_u_not_cs: +\cs_new_protected:Npn \__regex_compile_u_not_cs: { - \exp_args:No \tl_analysis_map_inline:nn { \l_regex_internal_a_tl } + \exp_args:No \__tl_analysis_map_inline:nn { \l__regex_internal_a_tl } { - \tl_build_one:n + \__tl_build_one:n { - \regex_class:NnnnN \c_true_bool + \__regex_class:NnnnN \c_true_bool { - \if_num:w "##2 = \c_zero - \regex_item_exact_cs:c { \exp_after:wN \cs_to_str:N ##1 } + \if_int_compare:w "##2 = \c_zero + \__regex_item_exact_cs:c { \exp_after:wN \cs_to_str:N ##1 } \else: - \regex_item_exact:nn { \int_value:w "##2 } { ##3 } + \__regex_item_exact:nn { \__int_value:w "##2 } { ##3 } \fi: } { 1 } { 0 } \c_false_bool } } } -\cs_new_protected_nopar:cpn { regex_compile_/K: } +\cs_new_protected_nopar:cpn { __regex_compile_/K: } { - \int_compare:nNnTF \l_regex_mode_int = \c_zero - { \tl_build_one:n { \regex_command_K: } } - { \regex_compile_raw_error:N K } + \int_compare:nNnTF \l__regex_mode_int = \c_zero + { \__tl_build_one:n { \__regex_command_K: } } + { \__regex_compile_raw_error:N K } } -\cs_new_protected:Npn \regex_show_aux:Nx #1#2 +\cs_new_protected:Npn \__regex_show:Nx #1#2 { - \tl_set_build:Nw \l_regex_internal_a_tl - \cs_set_protected_nopar:Npn \regex_branch:n + \__tl_build:Nw \l__regex_internal_a_tl + \cs_set_protected_nopar:Npn \__regex_branch:n { - \seq_pop_right:NN \l_regex_show_prefix_seq \l_regex_internal_a_tl - \regex_show_one:n { +-branch } - \seq_put_right:No \l_regex_show_prefix_seq \l_regex_internal_a_tl + \seq_pop_right:NN \l__regex_show_prefix_seq \l__regex_internal_a_tl + \__regex_show_one:n { +-branch } + \seq_put_right:No \l__regex_show_prefix_seq \l__regex_internal_a_tl \use:n } - \cs_set_protected_nopar:Npn \regex_group:nnnN - { \regex_show_group_aux:nnnnN { } } - \cs_set_protected_nopar:Npn \regex_group_no_capture:nnnN - { \regex_show_group_aux:nnnnN { ~(no~capture) } } - \cs_set_protected_nopar:Npn \regex_group_resetting:nnnN - { \regex_show_group_aux:nnnnN { ~(resetting) } } - \cs_set_eq:NN \regex_class:NnnnN \regex_show_class:NnnnN - \cs_set_protected_nopar:Npn \regex_command_K: - { \regex_show_one:n { reset~match~start~(\iow_char:N\\K) } } - \cs_set_protected:Npn \regex_assertion:Nn ##1##2 - { \regex_show_one:n { \bool_if:NF ##1 { negative~ } assertion:~##2 } } - \cs_set_nopar:Npn \regex_b_test: { word~boundary } - \cs_set_eq:NN \regex_anchor:N \regex_show_anchor_to_str:N - \cs_set_protected:Npn \regex_item_caseful_equal:n ##1 - { \regex_show_one:n { char~code~\int_eval:n{##1} } } - \cs_set_protected:Npn \regex_item_caseful_range:nn ##1##2 - { \regex_show_one:n { range~[\int_eval:n{##1}, \int_eval:n{##2}] } } - \cs_set_protected:Npn \regex_item_caseless_equal:n ##1 - { \regex_show_one:n { char~code~\int_eval:n{##1}~(caseless) } } - \cs_set_protected:Npn \regex_item_caseless_range:nn ##1##2 + \cs_set_protected_nopar:Npn \__regex_group:nnnN + { \__regex_show_group_aux:nnnnN { } } + \cs_set_protected_nopar:Npn \__regex_group_no_capture:nnnN + { \__regex_show_group_aux:nnnnN { ~(no~capture) } } + \cs_set_protected_nopar:Npn \__regex_group_resetting:nnnN + { \__regex_show_group_aux:nnnnN { ~(resetting) } } + \cs_set_eq:NN \__regex_class:NnnnN \__regex_show_class:NnnnN + \cs_set_protected_nopar:Npn \__regex_command_K: + { \__regex_show_one:n { reset~match~start~(\iow_char:N\\K) } } + \cs_set_protected:Npn \__regex_assertion:Nn ##1##2 + { \__regex_show_one:n { \bool_if:NF ##1 { negative~ } assertion:~##2 } } + \cs_set_nopar:Npn \__regex_b_test: { word~boundary } + \cs_set_eq:NN \__regex_anchor:N \__regex_show_anchor_to_str:N + \cs_set_protected:Npn \__regex_item_caseful_equal:n ##1 + { \__regex_show_one:n { char~code~\int_eval:n{##1} } } + \cs_set_protected:Npn \__regex_item_caseful_range:nn ##1##2 + { \__regex_show_one:n { range~[\int_eval:n{##1}, \int_eval:n{##2}] } } + \cs_set_protected:Npn \__regex_item_caseless_equal:n ##1 + { \__regex_show_one:n { char~code~\int_eval:n{##1}~(caseless) } } + \cs_set_protected:Npn \__regex_item_caseless_range:nn ##1##2 { - \regex_show_one:n + \__regex_show_one:n { Range~[\int_eval:n{##1}, \int_eval:n{##2}]~(caseless) } } - \cs_set_protected:Npn \regex_item_catcode:nT - { \regex_show_item_catcode:NnT \c_true_bool } - \cs_set_protected:Npn \regex_item_catcode_reverse:nT - { \regex_show_item_catcode:NnT \c_false_bool } - \cs_set_protected:Npn \regex_item_reverse:n - { \regex_show_scope:nn { Reversed~match } } - \cs_set_protected:Npn \regex_item_exact:nn ##1##2 - { \regex_show_one:n { char~##2,~catcode~##1 } } - \cs_set_protected:Npn \regex_item_exact_cs:c ##1 - { \regex_show_one:n { control~sequence~\iow_char:N\\##1 } } - \cs_set_protected:Npn \regex_item_cs:n - { \regex_show_scope:nn { control~sequence } } - \cs_set:cpn { regex_prop_.: } { \regex_show_one:n { any~token } } - \seq_clear:N \l_regex_show_prefix_seq - \regex_show_push:n { ~ } + \cs_set_protected:Npn \__regex_item_catcode:nT + { \__regex_show_item_catcode:NnT \c_true_bool } + \cs_set_protected:Npn \__regex_item_catcode_reverse:nT + { \__regex_show_item_catcode:NnT \c_false_bool } + \cs_set_protected:Npn \__regex_item_reverse:n + { \__regex_show_scope:nn { Reversed~match } } + \cs_set_protected:Npn \__regex_item_exact:nn ##1##2 + { \__regex_show_one:n { char~##2,~catcode~##1 } } + \cs_set_protected:Npn \__regex_item_exact_cs:c ##1 + { \__regex_show_one:n { control~sequence~\iow_char:N\\##1 } } + \cs_set_protected:Npn \__regex_item_cs:n + { \__regex_show_scope:nn { control~sequence } } + \cs_set:cpn { __regex_prop_.: } { \__regex_show_one:n { any~token } } + \seq_clear:N \l__regex_show_prefix_seq + \__regex_show_push:n { ~ } #1 - \tl_build_end: - \msg_aux_show:x { > Compiled~regex~#2: \l_regex_internal_a_tl } + \__tl_build_end: + \__msg_show_variable:x { > Compiled~regex~#2: \l__regex_internal_a_tl } } -\cs_new_protected:Npn \regex_show_one:n #1 +\cs_new_protected:Npn \__regex_show_one:n #1 { - \int_incr:N \l_regex_show_lines_int - \tl_build_one:x - { \iow_newline: \seq_use:N \l_regex_show_prefix_seq #1 } + \int_incr:N \l__regex_show_lines_int + \__tl_build_one:x + { + \iow_newline: + \seq_map_function:NN \l__regex_show_prefix_seq \use:n + #1 + } } -\cs_new_protected:Npn \regex_show_push:n #1 - { \seq_put_right:Nx \l_regex_show_prefix_seq { #1 ~ } } -\cs_new_protected:Npn \regex_show_pop: - { \seq_pop_right:NN \l_regex_show_prefix_seq \l_regex_internal_a_tl } -\cs_new_protected:Npn \regex_show_scope:nn #1#2 +\cs_new_protected:Npn \__regex_show_push:n #1 + { \seq_put_right:Nx \l__regex_show_prefix_seq { #1 ~ } } +\cs_new_protected:Npn \__regex_show_pop: + { \seq_pop_right:NN \l__regex_show_prefix_seq \l__regex_internal_a_tl } +\cs_new_protected:Npn \__regex_show_scope:nn #1#2 { - \regex_show_one:n {#1} - \regex_show_push:n { ~ } + \__regex_show_one:n {#1} + \__regex_show_push:n { ~ } #2 - \regex_show_pop: + \__regex_show_pop: } -\cs_new_protected:Npn \regex_show_group_aux:nnnnN #1#2#3#4#5 +\cs_new_protected:Npn \__regex_show_group_aux:nnnnN #1#2#3#4#5 { - \regex_show_one:n { ,-group~begin #1 } - \regex_show_push:n { | } + \__regex_show_one:n { ,-group~begin #1 } + \__regex_show_push:n { | } \use_ii:nn #2 - \regex_show_pop: - \regex_show_one:n - { `-group~end \regex_msg_repeated:nnN {#3} {#4} #5 } + \__regex_show_pop: + \__regex_show_one:n + { `-group~end \__regex_msg_repeated:nnN {#3} {#4} #5 } } -\cs_set:Npn \regex_show_class:NnnnN #1#2#3#4#5 +\cs_set:Npn \__regex_show_class:NnnnN #1#2#3#4#5 { - \tl_set_build:Nw \l_regex_internal_a_tl - \int_zero:N \l_regex_show_lines_int - \regex_show_push:n {~} + \__tl_build:Nw \l__regex_internal_a_tl + \int_zero:N \l__regex_show_lines_int + \__regex_show_push:n {~} #2 \exp_last_unbraced:Nf - \prg_case_int:nnn { \l_regex_show_lines_int } + \int_case:nnn { \l__regex_show_lines_int } { {0} { - \tl_build_end: - \regex_show_one:n { \bool_if:NTF #1 { Fail } { Pass } } + \__tl_build_end: + \__regex_show_one:n { \bool_if:NTF #1 { Fail } { Pass } } } {1} { - \tl_build_end: + \__tl_build_end: \bool_if:NTF #1 { #2 - \tl_build_one:n { \regex_msg_repeated:nnN {#3} {#4} #5 } + \__tl_build_one:n { \__regex_msg_repeated:nnN {#3} {#4} #5 } } { - \regex_show_one:n - { Don't~match~\regex_msg_repeated:nnN {#3} {#4} #5 } - \tl_build_one:o \l_regex_internal_a_tl + \__regex_show_one:n + { Don't~match~\__regex_msg_repeated:nnN {#3} {#4} #5 } + \__tl_build_one:o \l__regex_internal_a_tl } } } { - \tl_build_end: - \regex_show_one:n + \__tl_build_end: + \__regex_show_one:n { \bool_if:NTF #1 { M } { Don't~m } atch - \regex_msg_repeated:nnN {#3} {#4} #5 + \__regex_msg_repeated:nnN {#3} {#4} #5 } - \tl_build_one:o \l_regex_internal_a_tl + \__tl_build_one:o \l__regex_internal_a_tl } } -\cs_new:Npn \regex_show_anchor_to_str:N #1 +\cs_new:Npn \__regex_show_anchor_to_str:N #1 { anchor~at~ - \prg_case_str:nnn { #1 } + \str_case:nnn { #1 } { - { \l_regex_min_pos_int } { start~(\iow_char:N\\A) } - { \l_regex_start_pos_int } { start~of~match~(\iow_char:N\\G) } - { \l_regex_max_pos_int } { end~(\iow_char:N\\Z) } + { \l__regex_min_pos_int } { start~(\iow_char:N\\A) } + { \l__regex_start_pos_int } { start~of~match~(\iow_char:N\\G) } + { \l__regex_max_pos_int } { end~(\iow_char:N\\Z) } } { } } -\cs_new_protected:Npn \regex_show_item_catcode:NnT #1#2 +\cs_new_protected:Npn \__regex_show_item_catcode:NnT #1#2 { - \seq_set_split:Nnn \l_regex_internal_seq { } { CBEMTPUDSLOA } - \seq_set_filter:NNn \l_regex_internal_seq \l_regex_internal_seq - { \int_if_odd_p:n { #2 / \int_use:c { c_regex_catcode_##1_int } } } - \regex_show_scope:nn + \seq_set_split:Nnn \l__regex_internal_seq { } { CBEMTPUDSLOA } + \seq_set_filter:NNn \l__regex_internal_seq \l__regex_internal_seq + { \int_if_odd_p:n { #2 / \int_use:c { c__regex_catcode_##1_int } } } + \__regex_show_scope:nn { - categories~\seq_use:N \l_regex_internal_seq, ~ + categories~ + \seq_map_function:NN \l__regex_internal_seq \use:n + , ~ \bool_if:NF #1 { negative~ } class } } -\int_new:N \l_regex_min_state_int -\int_new:N \l_regex_max_state_int -\int_new:N \l_regex_left_state_int -\int_new:N \l_regex_right_state_int -\seq_new:N \l_regex_left_state_seq -\seq_new:N \l_regex_right_state_seq -\int_new:N \l_regex_capturing_group_int -\cs_new_protected:Npn \regex_build:n #1 +\int_new:N \l__regex_min_state_int +\int_new:N \l__regex_max_state_int +\int_new:N \l__regex_left_state_int +\int_new:N \l__regex_right_state_int +\seq_new:N \l__regex_left_state_seq +\seq_new:N \l__regex_right_state_seq +\int_new:N \l__regex_capturing_group_int +\cs_new_protected:Npn \__regex_build:n #1 { - \regex_compile:n {#1} - \regex_build:N \l_regex_internal_regex + \__regex_compile:n {#1} + \__regex_build:N \l__regex_internal_regex } -\cs_new_protected:Npn \regex_build:N #1 +\cs_new_protected:Npn \__regex_build:N #1 { - \trace_push:nnn { regex } { 1 } { regex_build } + \trace_push:nnn { regex } { 1 } { @@_build } \int_set:Nn \tex_escapechar:D { 92 } - \int_zero:N \l_regex_capturing_group_int - \int_set_eq:NN \l_regex_max_state_int \l_regex_min_state_int - \regex_build_new_state: - \regex_build_new_state: - \regex_toks_put_right:Nn \l_regex_left_state_int - { \regex_action_start_wildcard: } - \regex_group:nnnN {#1} { 1 } { 0 } \c_false_bool - \regex_toks_put_right:Nn \l_regex_right_state_int - { \regex_action_success: } - \regex_trace_states:n { 2 } - \trace_pop:nnn { regex } { 1 } { regex_build } - } -\cs_new_protected:Npn \regex_build_for_cs:n #1 - { - \trace_push:nnn { regex } { 1 } { regex_build_for_cs } - \int_set_eq:NN \l_regex_max_state_int \l_regex_min_state_int - \regex_build_new_state: - \regex_build_new_state: - \regex_push_lr_states: + \int_zero:N \l__regex_capturing_group_int + \int_set_eq:NN \l__regex_max_state_int \l__regex_min_state_int + \__regex_build_new_state: + \__regex_build_new_state: + \__regex_toks_put_right:Nn \l__regex_left_state_int + { \__regex_action_start_wildcard: } + \__regex_group:nnnN {#1} { 1 } { 0 } \c_false_bool + \__regex_toks_put_right:Nn \l__regex_right_state_int + { \__regex_action_success: } + \@@_trace_states:n { 2 } + \trace_pop:nnn { regex } { 1 } { @@_build } + } +\cs_new_protected:Npn \__regex_build_for_cs:n #1 + { + \trace_push:nnn { regex } { 1 } { @@_build_for_cs } + \int_set_eq:NN \l__regex_max_state_int \l__regex_min_state_int + \__regex_build_new_state: + \__regex_build_new_state: + \__regex_push_lr_states: #1 - \regex_pop_lr_states: - \regex_toks_put_right:Nn \l_regex_right_state_int + \__regex_pop_lr_states: + \__regex_toks_put_right:Nn \l__regex_right_state_int { - \if_num:w \l_regex_current_pos_int = \l_regex_max_pos_int - \exp_after:wN \regex_action_success: + \if_int_compare:w \l__regex_current_pos_int = \l__regex_max_pos_int + \exp_after:wN \__regex_action_success: \fi: } - \regex_trace_states:n { 2 } - \trace_pop:nnn { regex } { 1 } { regex_build_for_cs } + \@@_trace_states:n { 2 } + \trace_pop:nnn { regex } { 1 } { @@_build_for_cs } } -\cs_new_protected_nopar:Npn \regex_push_lr_states: +\cs_new_protected_nopar:Npn \__regex_push_lr_states: { - \seq_push:No \l_regex_left_state_seq - { \int_use:N \l_regex_left_state_int } - \seq_push:No \l_regex_right_state_seq - { \int_use:N \l_regex_right_state_int } + \seq_push:No \l__regex_left_state_seq + { \int_use:N \l__regex_left_state_int } + \seq_push:No \l__regex_right_state_seq + { \int_use:N \l__regex_right_state_int } } -\cs_new_protected_nopar:Npn \regex_pop_lr_states: +\cs_new_protected_nopar:Npn \__regex_pop_lr_states: { - \seq_pop:NN \l_regex_left_state_seq \l_regex_internal_a_tl - \int_set:Nn \l_regex_left_state_int \l_regex_internal_a_tl - \seq_pop:NN \l_regex_right_state_seq \l_regex_internal_a_tl - \int_set:Nn \l_regex_right_state_int \l_regex_internal_a_tl + \seq_pop:NN \l__regex_left_state_seq \l__regex_internal_a_tl + \int_set:Nn \l__regex_left_state_int \l__regex_internal_a_tl + \seq_pop:NN \l__regex_right_state_seq \l__regex_internal_a_tl + \int_set:Nn \l__regex_right_state_int \l__regex_internal_a_tl } -\cs_new_protected:Npn \regex_toks_put_left:Nx #1#2 +\cs_new_protected:Npn \__regex_toks_put_left:Nx #1#2 { - \cs_set_nopar:Npx \regex_tmp:w { #2 } + \cs_set_nopar:Npx \__regex_tmp:w { #2 } \tex_toks:D #1 \exp_after:wN \exp_after:wN \exp_after:wN - { \exp_after:wN \regex_tmp:w \tex_the:D \tex_toks:D #1 } + { \exp_after:wN \__regex_tmp:w \tex_the:D \tex_toks:D #1 } } -\cs_new_protected:Npn \regex_toks_put_right:Nx #1#2 +\cs_new_protected:Npn \__regex_toks_put_right:Nx #1#2 { - \cs_set_nopar:Npx \regex_tmp:w {#2} + \cs_set_nopar:Npx \__regex_tmp:w {#2} \tex_toks:D #1 \exp_after:wN - { \tex_the:D \tex_toks:D \exp_after:wN #1 \regex_tmp:w } + { \tex_the:D \tex_toks:D \exp_after:wN #1 \__regex_tmp:w } } -\cs_new_protected:Npn \regex_toks_put_right:Nn #1#2 +\cs_new_protected:Npn \__regex_toks_put_right:Nn #1#2 { \tex_toks:D #1 \exp_after:wN { \tex_the:D \tex_toks:D #1 #2 } } -\cs_new_protected:Npn \regex_build_transition_left:NNN #1#2#3 - { \regex_toks_put_left:Nx #2 { #1 { \int_eval:n { #3 - #2 } } } } -\cs_new_protected:Npn \regex_build_transition_right:nNn #1#2#3 - { \regex_toks_put_right:Nx #2 { #1 { \int_eval:n { #3 - #2 } } } } -\cs_new_protected_nopar:Npn \regex_build_new_state: +\cs_new_protected:Npn \__regex_build_transition_left:NNN #1#2#3 + { \__regex_toks_put_left:Nx #2 { #1 { \int_eval:n { #3 - #2 } } } } +\cs_new_protected:Npn \__regex_build_transition_right:nNn #1#2#3 + { \__regex_toks_put_right:Nx #2 { #1 { \int_eval:n { #3 - #2 } } } } +\cs_new_protected_nopar:Npn \__regex_build_new_state: { \trace:nnx { regex } { 2 } { regex~new~state~ - L=\int_use:N \l_regex_left_state_int ~ -> ~ - R=\int_use:N \l_regex_right_state_int ~ -> ~ - M=\int_use:N \l_regex_max_state_int ~ -> ~ - \int_eval:n { \l_regex_max_state_int + \c_one } + L=\int_use:N \l__regex_left_state_int ~ -> ~ + R=\int_use:N \l__regex_right_state_int ~ -> ~ + M=\int_use:N \l__regex_max_state_int ~ -> ~ + \int_eval:n { \l__regex_max_state_int + \c_one } } - \tex_toks:D \l_regex_max_state_int { } - \int_set_eq:NN \l_regex_left_state_int \l_regex_right_state_int - \int_set_eq:NN \l_regex_right_state_int \l_regex_max_state_int - \int_incr:N \l_regex_max_state_int + \tex_toks:D \l__regex_max_state_int { } + \int_set_eq:NN \l__regex_left_state_int \l__regex_right_state_int + \int_set_eq:NN \l__regex_right_state_int \l__regex_max_state_int + \int_incr:N \l__regex_max_state_int } -\cs_new_protected:Npn \regex_build_transitions_lazyness:NNNNN #1#2#3#4#5 +\cs_new_protected:Npn \__regex_build_transitions_lazyness:NNNNN #1#2#3#4#5 { - \regex_build_new_state: - \regex_toks_put_right:Nx \l_regex_left_state_int + \__regex_build_new_state: + \__regex_toks_put_right:Nx \l__regex_left_state_int { \if_meaning:w \c_true_bool #1 - #2 { \int_eval:n { #3 - \l_regex_left_state_int } } - #4 { \int_eval:n { #5 - \l_regex_left_state_int } } + #2 { \int_eval:n { #3 - \l__regex_left_state_int } } + #4 { \int_eval:n { #5 - \l__regex_left_state_int } } \else: - #4 { \int_eval:n { #5 - \l_regex_left_state_int } } - #2 { \int_eval:n { #3 - \l_regex_left_state_int } } + #4 { \int_eval:n { #5 - \l__regex_left_state_int } } + #2 { \int_eval:n { #3 - \l__regex_left_state_int } } \fi: } } -\cs_new_protected:Npn \regex_class:NnnnN #1#2#3#4#5 +\cs_new_protected:Npn \__regex_class:NnnnN #1#2#3#4#5 { - \cs_set_nopar:Npx \regex_tests_action_cost:n ##1 + \cs_set_nopar:Npx \__regex_tests_action_cost:n ##1 { \exp_not:n { \exp_not:n {#2} } \bool_if:NTF #1 - { \regex_break_point:TF { \regex_action_cost:n {##1} } { } } - { \regex_break_point:TF { } { \regex_action_cost:n {##1} } } + { \__regex_break_point:TF { \__regex_action_cost:n {##1} } { } } + { \__regex_break_point:TF { } { \__regex_action_cost:n {##1} } } } \if_case:w - #4 \exp_stop_f: - \regex_class_repeat:n {#3} - \or: \regex_class_repeat:nN {#3} #5 - \else: \regex_class_repeat:nnN {#3} {#4} #5 + \__regex_class_repeat:n {#3} + \or: \__regex_class_repeat:nN {#3} #5 + \else: \__regex_class_repeat:nnN {#3} {#4} #5 \fi: } -\cs_new:Npn \regex_tests_action_cost:n { \regex_action_cost:n } -\cs_new_protected:Npn \regex_class_repeat:n #1 +\cs_new:Npn \__regex_tests_action_cost:n { \__regex_action_cost:n } +\cs_new_protected:Npn \__regex_class_repeat:n #1 { \prg_replicate:nn {#1} { - \regex_build_new_state: - \regex_build_transition_right:nNn \regex_tests_action_cost:n - \l_regex_left_state_int \l_regex_right_state_int + \__regex_build_new_state: + \__regex_build_transition_right:nNn \__regex_tests_action_cost:n + \l__regex_left_state_int \l__regex_right_state_int } } -\cs_new_protected:Npn \regex_class_repeat:nN #1#2 +\cs_new_protected:Npn \__regex_class_repeat:nN #1#2 { - \if_num:w #1 = \c_zero - \regex_build_transitions_lazyness:NNNNN #2 - \regex_action_free:n \l_regex_right_state_int - \regex_tests_action_cost:n \l_regex_left_state_int + \if_int_compare:w #1 = \c_zero + \__regex_build_transitions_lazyness:NNNNN #2 + \__regex_action_free:n \l__regex_right_state_int + \__regex_tests_action_cost:n \l__regex_left_state_int \else: - \regex_class_repeat:n {#1} - \int_set_eq:NN \l_regex_internal_a_int \l_regex_left_state_int - \regex_build_transitions_lazyness:NNNNN #2 - \regex_action_free:n \l_regex_right_state_int - \regex_action_free:n \l_regex_internal_a_int + \__regex_class_repeat:n {#1} + \int_set_eq:NN \l__regex_internal_a_int \l__regex_left_state_int + \__regex_build_transitions_lazyness:NNNNN #2 + \__regex_action_free:n \l__regex_right_state_int + \__regex_action_free:n \l__regex_internal_a_int \fi: } -\cs_new_protected:Npn \regex_class_repeat:nnN #1#2#3 +\cs_new_protected:Npn \__regex_class_repeat:nnN #1#2#3 { - \regex_class_repeat:n {#1} - \int_set:Nn \l_regex_internal_a_int - { \l_regex_max_state_int + #2 - \c_one } + \__regex_class_repeat:n {#1} + \int_set:Nn \l__regex_internal_a_int + { \l__regex_max_state_int + #2 - \c_one } \prg_replicate:nn { #2 } { - \regex_build_transitions_lazyness:NNNNN #3 - \regex_action_free:n \l_regex_internal_a_int - \regex_tests_action_cost:n \l_regex_right_state_int + \__regex_build_transitions_lazyness:NNNNN #3 + \__regex_action_free:n \l__regex_internal_a_int + \__regex_tests_action_cost:n \l__regex_right_state_int } } -\cs_new_protected:Npn \regex_group_aux:nnnnN #1#2#3#4#5 +\cs_new_protected:Npn \__regex_group_aux:nnnnN #1#2#3#4#5 { - \trace_push:nnn { regex } { 1 } { regex_group } - \if_num:w #3 = \c_zero - \regex_build_new_state: - \regex_build_transition_right:nNn \regex_action_free_group:n - \l_regex_left_state_int \l_regex_right_state_int + \trace_push:nnn { regex } { 1 } { @@_group } + \if_int_compare:w #3 = \c_zero + \__regex_build_new_state: + \__regex_build_transition_right:nNn \__regex_action_free_group:n + \l__regex_left_state_int \l__regex_right_state_int \fi: - \regex_build_new_state: - \regex_push_lr_states: + \__regex_build_new_state: + \__regex_push_lr_states: #2 - \regex_pop_lr_states: + \__regex_pop_lr_states: \if_case:w - #4 \exp_stop_f: - \regex_group_repeat:nn {#1} {#3} - \or: \regex_group_repeat:nnN {#1} {#3} #5 - \else: \regex_group_repeat:nnnN {#1} {#3} {#4} #5 + \__regex_group_repeat:nn {#1} {#3} + \or: \__regex_group_repeat:nnN {#1} {#3} #5 + \else: \__regex_group_repeat:nnnN {#1} {#3} {#4} #5 \fi: - \trace_pop:nnn { regex } { 1 } { regex_group } + \trace_pop:nnn { regex } { 1 } { @@_group } } -\cs_new_protected:Npn \regex_group:nnnN #1 +\cs_new_protected:Npn \__regex_group:nnnN #1 { - \exp_args:No \regex_group_aux:nnnnN - { \int_use:N \l_regex_capturing_group_int } + \exp_args:No \__regex_group_aux:nnnnN + { \int_use:N \l__regex_capturing_group_int } { - \int_incr:N \l_regex_capturing_group_int + \int_incr:N \l__regex_capturing_group_int #1 } } -\cs_new_protected_nopar:Npn \regex_group_no_capture:nnnN - { \regex_group_aux:nnnnN { -1 } } -\cs_new_protected:Npn \regex_group_resetting:nnnN #1 +\cs_new_protected_nopar:Npn \__regex_group_no_capture:nnnN + { \__regex_group_aux:nnnnN { -1 } } +\cs_new_protected:Npn \__regex_group_resetting:nnnN #1 { - \regex_group_aux:nnnnN { -1 } + \__regex_group_aux:nnnnN { -1 } { - \exp_args:Noo \regex_group_resetting_loop:nnNn - { \int_use:N \l_regex_capturing_group_int } - { \int_use:N \l_regex_capturing_group_int } + \exp_args:Noo \__regex_group_resetting_loop:nnNn + { \int_use:N \l__regex_capturing_group_int } + { \int_use:N \l__regex_capturing_group_int } #1 - { ?? \prg_map_break:n } { } - \prg_break_point:n { } + { ?? \__prg_break:n } { } + \__prg_break_point: } } -\cs_new_protected:Npn \regex_group_resetting_loop:nnNn #1#2#3#4 +\cs_new_protected:Npn \__regex_group_resetting_loop:nnNn #1#2#3#4 { - \use_none:nn #3 { \int_set:Nn \l_regex_capturing_group_int {#1} } - \int_set:Nn \l_regex_capturing_group_int {#2} + \use_none:nn #3 { \int_set:Nn \l__regex_capturing_group_int {#1} } + \int_set:Nn \l__regex_capturing_group_int {#2} #3 {#4} - \exp_args:Nf \regex_group_resetting_loop:nnNn - { \int_max:nn {#1} { \l_regex_capturing_group_int } } + \exp_args:Nf \__regex_group_resetting_loop:nnNn + { \int_max:nn {#1} { \l__regex_capturing_group_int } } {#2} } -\cs_new_protected:Npn \regex_branch:n #1 +\cs_new_protected:Npn \__regex_branch:n #1 { - \trace_push:nnn { regex } { 1 } { regex_branch } - \regex_build_new_state: - \seq_get:NN \l_regex_left_state_seq \l_regex_internal_a_tl - \int_set:Nn \l_regex_left_state_int \l_regex_internal_a_tl - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_left_state_int \l_regex_right_state_int + \trace_push:nnn { regex } { 1 } { @@_branch } + \__regex_build_new_state: + \seq_get:NN \l__regex_left_state_seq \l__regex_internal_a_tl + \int_set:Nn \l__regex_left_state_int \l__regex_internal_a_tl + \__regex_build_transition_right:nNn \__regex_action_free:n + \l__regex_left_state_int \l__regex_right_state_int #1 - \seq_get:NN \l_regex_right_state_seq \l_regex_internal_a_tl - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_right_state_int \l_regex_internal_a_tl - \trace_pop:nnn { regex } { 1 } { regex_branch } + \seq_get:NN \l__regex_right_state_seq \l__regex_internal_a_tl + \__regex_build_transition_right:nNn \__regex_action_free:n + \l__regex_right_state_int \l__regex_internal_a_tl + \trace_pop:nnn { regex } { 1 } { @@_branch } } -\cs_new_protected:Npn \regex_group_repeat:nn #1#2 +\cs_new_protected:Npn \__regex_group_repeat:nn #1#2 { - \if_num:w #2 = \c_zero - \int_set:Nn \l_regex_max_state_int - { \l_regex_left_state_int - \c_one } - \regex_build_new_state: + \if_int_compare:w #2 = \c_zero + \int_set:Nn \l__regex_max_state_int + { \l__regex_left_state_int - \c_one } + \__regex_build_new_state: \else: - \regex_group_repeat_aux:n {#2} - \regex_group_submatches:nNN {#1} - \l_regex_internal_a_int \l_regex_right_state_int - \regex_build_new_state: + \__regex_group_repeat_aux:n {#2} + \__regex_group_submatches:nNN {#1} + \l__regex_internal_a_int \l__regex_right_state_int + \__regex_build_new_state: \fi: } -\cs_new_protected:Npn \regex_group_submatches:nNN #1#2#3 +\cs_new_protected:Npn \__regex_group_submatches:nNN #1#2#3 { - \if_num:w #1 > \c_minus_one - \regex_toks_put_left:Nx #2 { \regex_action_submatch:n { #1 < } } - \regex_toks_put_left:Nx #3 { \regex_action_submatch:n { #1 > } } + \if_int_compare:w #1 > \c_minus_one + \__regex_toks_put_left:Nx #2 { \__regex_action_submatch:n { #1 < } } + \__regex_toks_put_left:Nx #3 { \__regex_action_submatch:n { #1 > } } \fi: } -\cs_new_protected:Npn \regex_group_repeat_aux:n #1 +\cs_new_protected:Npn \__regex_group_repeat_aux:n #1 { - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_right_state_int \l_regex_max_state_int - \int_set_eq:NN \l_regex_internal_a_int \l_regex_left_state_int - \int_set_eq:NN \l_regex_internal_b_int \l_regex_max_state_int - \if_num:w \int_eval:w #1 > \c_one - \int_set:Nn \l_regex_internal_c_int + \__regex_build_transition_right:nNn \__regex_action_free:n + \l__regex_right_state_int \l__regex_max_state_int + \int_set_eq:NN \l__regex_internal_a_int \l__regex_left_state_int + \int_set_eq:NN \l__regex_internal_b_int \l__regex_max_state_int + \if_int_compare:w \__int_eval:w #1 > \c_one + \int_set:Nn \l__regex_internal_c_int { ( #1 - \c_one ) - * ( \l_regex_internal_b_int - \l_regex_internal_a_int ) + * ( \l__regex_internal_b_int - \l__regex_internal_a_int ) } - \tex_advance:D \l_regex_right_state_int \l_regex_internal_c_int - \tex_advance:D \l_regex_max_state_int \l_regex_internal_c_int - \prg_replicate:nn \l_regex_internal_c_int + \tex_advance:D \l__regex_right_state_int \l__regex_internal_c_int + \tex_advance:D \l__regex_max_state_int \l__regex_internal_c_int + \prg_replicate:nn \l__regex_internal_c_int { - \tex_toks:D \l_regex_internal_b_int - = \tex_toks:D \l_regex_internal_a_int - \tex_advance:D \l_regex_internal_a_int \c_one - \tex_advance:D \l_regex_internal_b_int \c_one + \tex_toks:D \l__regex_internal_b_int + = \tex_toks:D \l__regex_internal_a_int + \tex_advance:D \l__regex_internal_a_int \c_one + \tex_advance:D \l__regex_internal_b_int \c_one } \fi: } -\cs_new_protected:Npn \regex_group_repeat:nnN #1#2#3 +\cs_new_protected:Npn \__regex_group_repeat:nnN #1#2#3 { - \if_num:w #2 = \c_zero - \regex_group_submatches:nNN {#1} - \l_regex_left_state_int \l_regex_right_state_int - \int_set:Nn \l_regex_internal_a_int - { \l_regex_left_state_int - \c_one } - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_right_state_int \l_regex_internal_a_int - \regex_build_new_state: + \if_int_compare:w #2 = \c_zero + \__regex_group_submatches:nNN {#1} + \l__regex_left_state_int \l__regex_right_state_int + \int_set:Nn \l__regex_internal_a_int + { \l__regex_left_state_int - \c_one } + \__regex_build_transition_right:nNn \__regex_action_free:n + \l__regex_right_state_int \l__regex_internal_a_int + \__regex_build_new_state: \if_meaning:w \c_true_bool #3 - \regex_build_transition_left:NNN \regex_action_free:n - \l_regex_internal_a_int \l_regex_right_state_int + \__regex_build_transition_left:NNN \__regex_action_free:n + \l__regex_internal_a_int \l__regex_right_state_int \else: - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_internal_a_int \l_regex_right_state_int + \__regex_build_transition_right:nNn \__regex_action_free:n + \l__regex_internal_a_int \l__regex_right_state_int \fi: \else: - \regex_group_repeat_aux:n {#2} - \regex_group_submatches:nNN {#1} - \l_regex_internal_a_int \l_regex_right_state_int + \__regex_group_repeat_aux:n {#2} + \__regex_group_submatches:nNN {#1} + \l__regex_internal_a_int \l__regex_right_state_int \if_meaning:w \c_true_bool #3 - \regex_build_transition_right:nNn \regex_action_free_group:n - \l_regex_right_state_int \l_regex_internal_a_int + \__regex_build_transition_right:nNn \__regex_action_free_group:n + \l__regex_right_state_int \l__regex_internal_a_int \else: - \regex_build_transition_left:NNN \regex_action_free_group:n - \l_regex_right_state_int \l_regex_internal_a_int + \__regex_build_transition_left:NNN \__regex_action_free_group:n + \l__regex_right_state_int \l__regex_internal_a_int \fi: - \regex_build_new_state: + \__regex_build_new_state: \fi: } -\cs_new_protected:Npn \regex_group_repeat:nnnN #1#2#3#4 +\cs_new_protected:Npn \__regex_group_repeat:nnnN #1#2#3#4 { - \regex_group_submatches:nNN {#1} - \l_regex_left_state_int \l_regex_right_state_int - \regex_group_repeat_aux:n { #2 + #3 } + \__regex_group_submatches:nNN {#1} + \l__regex_left_state_int \l__regex_right_state_int + \__regex_group_repeat_aux:n { #2 + #3 } \if_meaning:w \c_true_bool #4 - \int_set_eq:NN \l_regex_left_state_int \l_regex_max_state_int + \int_set_eq:NN \l__regex_left_state_int \l__regex_max_state_int \prg_replicate:nn { #3 } { - \int_sub:Nn \l_regex_left_state_int - { \l_regex_internal_b_int - \l_regex_internal_a_int } - \regex_build_transition_left:NNN \regex_action_free:n - \l_regex_left_state_int \l_regex_max_state_int + \int_sub:Nn \l__regex_left_state_int + { \l__regex_internal_b_int - \l__regex_internal_a_int } + \__regex_build_transition_left:NNN \__regex_action_free:n + \l__regex_left_state_int \l__regex_max_state_int } \else: \prg_replicate:nn { #3 - \c_one } { - \int_sub:Nn \l_regex_right_state_int - { \l_regex_internal_b_int - \l_regex_internal_a_int } - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_right_state_int \l_regex_max_state_int + \int_sub:Nn \l__regex_right_state_int + { \l__regex_internal_b_int - \l__regex_internal_a_int } + \__regex_build_transition_right:nNn \__regex_action_free:n + \l__regex_right_state_int \l__regex_max_state_int } - \if_num:w #2 = \c_zero - \int_set:Nn \l_regex_right_state_int - { \l_regex_left_state_int - \c_one } + \if_int_compare:w #2 = \c_zero + \int_set:Nn \l__regex_right_state_int + { \l__regex_left_state_int - \c_one } \else: - \int_sub:Nn \l_regex_right_state_int - { \l_regex_internal_b_int - \l_regex_internal_a_int } + \int_sub:Nn \l__regex_right_state_int + { \l__regex_internal_b_int - \l__regex_internal_a_int } \fi: - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_right_state_int \l_regex_max_state_int + \__regex_build_transition_right:nNn \__regex_action_free:n + \l__regex_right_state_int \l__regex_max_state_int \fi: - \regex_build_new_state: + \__regex_build_new_state: } -\cs_new_protected:Npn \regex_assertion:Nn #1#2 +\cs_new_protected:Npn \__regex_assertion:Nn #1#2 { - \regex_build_new_state: - \regex_toks_put_right:Nx \l_regex_left_state_int + \__regex_build_new_state: + \__regex_toks_put_right:Nx \l__regex_left_state_int { \exp_not:n {#2} - \regex_break_point:TF + \__regex_break_point:TF \bool_if:NF #1 { { } } { - \regex_action_free:n + \__regex_action_free:n { \int_eval:n - { \l_regex_right_state_int - \l_regex_left_state_int } + { \l__regex_right_state_int - \l__regex_left_state_int } } } \bool_if:NT #1 { { } } } } -\cs_new_protected:Npn \regex_anchor:N #1 +\cs_new_protected:Npn \__regex_anchor:N #1 { - \if_num:w #1 = \l_regex_current_pos_int - \exp_after:wN \regex_break_true:w + \if_int_compare:w #1 = \l__regex_current_pos_int + \exp_after:wN \__regex_break_true:w \fi: } -\cs_new_protected_nopar:Npn \regex_b_test: +\cs_new_protected_nopar:Npn \__regex_b_test: { \group_begin: - \int_set_eq:NN \l_regex_current_char_int \l_regex_last_char_int - \regex_prop_w: - \regex_break_point:TF - { \group_end: \regex_item_reverse:n \regex_prop_w: } - { \group_end: \regex_prop_w: } - } -\cs_new_protected_nopar:Npn \regex_command_K: - { - \regex_build_new_state: - \regex_toks_put_right:Nx \l_regex_left_state_int - { - \regex_action_submatch:n { 0< } - \bool_set_true:N \l_regex_fresh_thread_bool - \regex_action_free:n - { \int_eval:n { \l_regex_right_state_int - \l_regex_left_state_int } } - \bool_set_false:N \l_regex_fresh_thread_bool - } - } -\int_new:N \l_regex_min_pos_int -\int_new:N \l_regex_max_pos_int -\int_new:N \l_regex_current_pos_int -\int_new:N \l_regex_start_pos_int -\int_new:N \l_regex_success_pos_int -\int_new:N \l_regex_current_char_int -\int_new:N \l_regex_current_catcode_int -\int_new:N \l_regex_last_char_int -\int_new:N \l_regex_case_changed_char_int -\int_new:N \l_regex_current_state_int -\prop_new:N \l_regex_current_submatches_prop -\prop_new:N \l_regex_success_submatches_prop -\int_new:N \l_regex_step_int -\int_new:N \l_regex_min_active_int -\int_new:N \l_regex_max_active_int -\tl_new:N \l_regex_every_match_tl -\bool_new:N \l_regex_fresh_thread_bool -\bool_new:N \l_regex_empty_success_bool -\cs_new_eq:NN \regex_if_two_empty_matches:F \use:n -\bool_new:N \g_regex_success_bool -\bool_new:N \l_regex_saved_success_bool -\bool_new:N \l_regex_match_success_bool -\cs_new_protected:Npn \regex_match:n #1 - { - \trace_push:nnx { regex } { 1 } { regex_match } + \int_set_eq:NN \l__regex_current_char_int \l__regex_last_char_int + \__regex_prop_w: + \__regex_break_point:TF + { \group_end: \__regex_item_reverse:n \__regex_prop_w: } + { \group_end: \__regex_prop_w: } + } +\cs_new_protected_nopar:Npn \__regex_command_K: + { + \__regex_build_new_state: + \__regex_toks_put_right:Nx \l__regex_left_state_int + { + \__regex_action_submatch:n { 0< } + \bool_set_true:N \l__regex_fresh_thread_bool + \__regex_action_free:n + { \int_eval:n { \l__regex_right_state_int - \l__regex_left_state_int } } + \bool_set_false:N \l__regex_fresh_thread_bool + } + } +\int_new:N \l__regex_min_pos_int +\int_new:N \l__regex_max_pos_int +\int_new:N \l__regex_current_pos_int +\int_new:N \l__regex_start_pos_int +\int_new:N \l__regex_success_pos_int +\int_new:N \l__regex_current_char_int +\int_new:N \l__regex_current_catcode_int +\int_new:N \l__regex_last_char_int +\int_new:N \l__regex_case_changed_char_int +\int_new:N \l__regex_current_state_int +\prop_new:N \l__regex_current_submatches_prop +\prop_new:N \l__regex_success_submatches_prop +\int_new:N \l__regex_step_int +\int_new:N \l__regex_min_active_int +\int_new:N \l__regex_max_active_int +\tl_new:N \l__regex_every_match_tl +\bool_new:N \l__regex_fresh_thread_bool +\bool_new:N \l__regex_empty_success_bool +\cs_new_eq:NN \__regex_if_two_empty_matches:F \use:n +\bool_new:N \g__regex_success_bool +\bool_new:N \l__regex_saved_success_bool +\bool_new:N \l__regex_match_success_bool +\cs_new_protected:Npn \__regex_match:n #1 + { + \trace_push:nnx { regex } { 1 } { @@_match } \trace:nnx { regex } { 1 } { analyzing~query~token~list } - \int_zero:N \l_regex_balance_int - \int_set:Nn \l_regex_current_pos_int { \c_two * \l_regex_max_state_int } - \regex_query_set:nnn { } { -1 } { -2 } - \int_set_eq:NN \l_regex_min_pos_int \l_regex_current_pos_int - \tl_analysis_map_inline:nn {#1} - { \regex_query_set:nnn {##1} {"##2} {##3} } - \int_set_eq:NN \l_regex_max_pos_int \l_regex_current_pos_int - \regex_query_set:nnn { } { -1 } { -2 } + \int_zero:N \l__regex_balance_int + \int_set:Nn \l__regex_current_pos_int { \c_two * \l__regex_max_state_int } + \__regex_query_set:nnn { } { -1 } { -2 } + \int_set_eq:NN \l__regex_min_pos_int \l__regex_current_pos_int + \__tl_analysis_map_inline:nn {#1} + { \__regex_query_set:nnn {##1} {"##2} {##3} } + \int_set_eq:NN \l__regex_max_pos_int \l__regex_current_pos_int + \__regex_query_set:nnn { } { -1 } { -2 } \trace:nnx { regex } { 1 } { initializing } - \bool_gset_false:N \g_regex_success_bool - \prg_stepwise_inline:nnnn - \l_regex_min_state_int \c_one { \l_regex_max_state_int - \c_one } + \bool_gset_false:N \g__regex_success_bool + \int_step_inline:nnnn + \l__regex_min_state_int \c_one { \l__regex_max_state_int - \c_one } { \tex_dimen:D ##1 \c_one sp \scan_stop: } - \int_set_eq:NN \l_regex_min_active_int \l_regex_max_state_int - \int_set_eq:NN \l_regex_step_int \c_zero - \int_set_eq:NN \l_regex_success_pos_int \l_regex_min_pos_int - \int_set:Nn \l_regex_submatch_int - { \c_two * \l_regex_max_state_int } - \bool_set_false:N \l_regex_empty_success_bool - \regex_match_once: - \trace_pop:nnx { regex } { 1 } { regex_match } - } -\cs_new_protected_nopar:Npn \regex_match_once: - { - \if_meaning:w \c_true_bool \l_regex_empty_success_bool - \cs_set_nopar:Npn \regex_if_two_empty_matches:F - { \int_compare:nNnF \l_regex_start_pos_int = \l_regex_current_pos_int } + \int_set_eq:NN \l__regex_min_active_int \l__regex_max_state_int + \int_set_eq:NN \l__regex_step_int \c_zero + \int_set_eq:NN \l__regex_success_pos_int \l__regex_min_pos_int + \int_set:Nn \l__regex_submatch_int + { \c_two * \l__regex_max_state_int } + \bool_set_false:N \l__regex_empty_success_bool + \__regex_match_once: + \trace_pop:nnx { regex } { 1 } { @@_match } + } +\cs_new_protected_nopar:Npn \__regex_match_once: + { + \if_meaning:w \c_true_bool \l__regex_empty_success_bool + \cs_set_nopar:Npn \__regex_if_two_empty_matches:F + { \int_compare:nNnF \l__regex_start_pos_int = \l__regex_current_pos_int } \else: - \cs_set_eq:NN \regex_if_two_empty_matches:F \use:n + \cs_set_eq:NN \__regex_if_two_empty_matches:F \use:n \fi: - \int_set_eq:NN \l_regex_start_pos_int \l_regex_success_pos_int - \bool_set_false:N \l_regex_match_success_bool - \prop_clear:N \l_regex_current_submatches_prop - \int_set_eq:NN \l_regex_max_active_int \l_regex_min_active_int - \regex_store_state:n { \l_regex_min_state_int } - \int_set:Nn \l_regex_current_pos_int - { \l_regex_start_pos_int - \c_one } - \regex_query_get: - \regex_match_loop: - \l_regex_every_match_tl + \int_set_eq:NN \l__regex_start_pos_int \l__regex_success_pos_int + \bool_set_false:N \l__regex_match_success_bool + \prop_clear:N \l__regex_current_submatches_prop + \int_set_eq:NN \l__regex_max_active_int \l__regex_min_active_int + \__regex_store_state:n { \l__regex_min_state_int } + \int_set:Nn \l__regex_current_pos_int + { \l__regex_start_pos_int - \c_one } + \__regex_query_get: + \__regex_match_loop: + \l__regex_every_match_tl } -\cs_new_protected_nopar:Npn \regex_single_match: +\cs_new_protected_nopar:Npn \__regex_single_match: { - \tl_set:Nn \l_regex_every_match_tl - { \bool_gset_eq:NN \g_regex_success_bool \l_regex_match_success_bool } + \tl_set:Nn \l__regex_every_match_tl + { \bool_gset_eq:NN \g__regex_success_bool \l__regex_match_success_bool } } -\cs_new_protected:Npn \regex_multi_match:n #1 +\cs_new_protected:Npn \__regex_multi_match:n #1 { - \tl_set:Nn \l_regex_every_match_tl + \tl_set:Nn \l__regex_every_match_tl { - \if_meaning:w \c_true_bool \l_regex_match_success_bool - \bool_gset_true:N \g_regex_success_bool + \if_meaning:w \c_true_bool \l__regex_match_success_bool + \bool_gset_true:N \g__regex_success_bool #1 - \exp_after:wN \regex_match_once: + \exp_after:wN \__regex_match_once: \fi: } } -\cs_new_protected_nopar:Npn \regex_match_loop: +\cs_new_protected_nopar:Npn \__regex_match_loop: { - \tex_advance:D \l_regex_step_int \c_two - \int_incr:N \l_regex_current_pos_int - \int_set_eq:NN \l_regex_last_char_int \l_regex_current_char_int - \int_set_eq:NN \l_regex_case_changed_char_int \c_max_int - \regex_query_get: + \tex_advance:D \l__regex_step_int \c_two + \int_incr:N \l__regex_current_pos_int + \int_set_eq:NN \l__regex_last_char_int \l__regex_current_char_int + \int_set_eq:NN \l__regex_case_changed_char_int \c_max_int + \__regex_query_get: \use:x { - \int_set_eq:NN \l_regex_max_active_int \l_regex_min_active_int - \exp_after:wN \regex_match_one_active:w - \int_use:N \l_regex_min_active_int ; + \int_set_eq:NN \l__regex_max_active_int \l__regex_min_active_int + \exp_after:wN \__regex_match_one_active:w + \int_use:N \l__regex_min_active_int ; } - \prg_break_point:n { \bool_set_false:N \l_regex_fresh_thread_bool } - \if_num:w \l_regex_max_active_int > \l_regex_min_active_int - \if_num:w \l_regex_current_pos_int < \l_regex_max_pos_int - \exp_after:wN \exp_after:wN \exp_after:wN \regex_match_loop: + \__prg_break_point: + \bool_set_false:N \l__regex_fresh_thread_bool %^^A was arg of break_point:n + \if_int_compare:w \l__regex_max_active_int > \l__regex_min_active_int + \if_int_compare:w \l__regex_current_pos_int < \l__regex_max_pos_int + \exp_after:wN \exp_after:wN \exp_after:wN \__regex_match_loop: \fi: \fi: } -\cs_new:Npn \regex_match_one_active:w #1; +\cs_new:Npn \__regex_match_one_active:w #1; { - \if_num:w #1 < \l_regex_max_active_int - \regex_use_state_and_submatches:nn - { \int_value:w \tex_skip:D #1 } + \if_int_compare:w #1 < \l__regex_max_active_int + \__regex_use_state_and_submatches:nn + { \__int_value:w \tex_skip:D #1 } { \tex_the:D \tex_toks:D #1 } - \exp_after:wN \regex_match_one_active:w - \int_use:N \int_eval:w #1 + \c_one \exp_after:wN ; + \exp_after:wN \__regex_match_one_active:w + \int_use:N \__int_eval:w #1 + \c_one \exp_after:wN ; \fi: } -\cs_new_protected:Npn \regex_query_set:nnn #1#2#3 +\cs_new_protected:Npn \__regex_query_set:nnn #1#2#3 { - \tex_muskip:D \l_regex_current_pos_int + \tex_muskip:D \l__regex_current_pos_int = \etex_gluetomu:D #3 sp plus #2 sp - minus \l_regex_balance_int sp + minus \l__regex_balance_int sp \scan_stop: - \tex_toks:D \l_regex_current_pos_int {#1} - \int_incr:N \l_regex_current_pos_int + \tex_toks:D \l__regex_current_pos_int {#1} + \int_incr:N \l__regex_current_pos_int \if_case:w #2 \exp_stop_f: - \or: \int_incr:N \l_regex_balance_int - \or: \int_decr:N \l_regex_balance_int + \or: \int_incr:N \l__regex_balance_int + \or: \int_decr:N \l__regex_balance_int \fi: } -\cs_new_protected_nopar:Npn \regex_query_get: +\cs_new_protected_nopar:Npn \__regex_query_get: { - \l_regex_current_char_int - = \etex_mutoglue:D \tex_muskip:D \l_regex_current_pos_int - \l_regex_current_catcode_int = \etex_gluestretch:D - \etex_mutoglue:D \tex_muskip:D \l_regex_current_pos_int + \l__regex_current_char_int + = \etex_mutoglue:D \tex_muskip:D \l__regex_current_pos_int + \l__regex_current_catcode_int = \etex_gluestretch:D + \etex_mutoglue:D \tex_muskip:D \l__regex_current_pos_int } -\cs_new_protected_nopar:Npn \regex_use_state: +\cs_new_protected_nopar:Npn \__regex_use_state: { - \trace:nnx { regex } { 2 } { state~\int_use:N \l_regex_current_state_int } - \tex_dimen:D \l_regex_current_state_int - = \l_regex_step_int sp \scan_stop: - \tex_the:D \tex_toks:D \l_regex_current_state_int - \tex_dimen:D \l_regex_current_state_int - = \int_eval:w \l_regex_step_int + \c_one \int_eval_end: sp \scan_stop: + \trace:nnx { regex } { 2 } { state~\int_use:N \l__regex_current_state_int } + \tex_dimen:D \l__regex_current_state_int + = \l__regex_step_int sp \scan_stop: + \tex_the:D \tex_toks:D \l__regex_current_state_int + \tex_dimen:D \l__regex_current_state_int + = \__int_eval:w \l__regex_step_int + \c_one \__int_eval_end: sp \scan_stop: } -\cs_new_protected:Npn \regex_use_state_and_submatches:nn #1 #2 +\cs_new_protected:Npn \__regex_use_state_and_submatches:nn #1 #2 { - \int_set:Nn \l_regex_current_state_int {#1} - \if_num:w \tex_dimen:D \l_regex_current_state_int < \l_regex_step_int - \tl_set:Nn \l_regex_current_submatches_prop {#2} - \exp_after:wN \regex_use_state: + \int_set:Nn \l__regex_current_state_int {#1} + \if_int_compare:w \tex_dimen:D \l__regex_current_state_int + < \l__regex_step_int + \tl_set:Nn \l__regex_current_submatches_prop {#2} + \exp_after:wN \__regex_use_state: \fi: \scan_stop: } -\cs_new_protected_nopar:Npn \regex_action_start_wildcard: +\cs_new_protected_nopar:Npn \__regex_action_start_wildcard: { - \bool_set_true:N \l_regex_fresh_thread_bool - \regex_action_free:n {1} - \bool_set_false:N \l_regex_fresh_thread_bool - \regex_action_cost:n {0} + \bool_set_true:N \l__regex_fresh_thread_bool + \__regex_action_free:n {1} + \bool_set_false:N \l__regex_fresh_thread_bool + \__regex_action_cost:n {0} } -\cs_new_protected_nopar:Npn \regex_action_free:n - { \regex_action_free_aux:nn { > \l_regex_step_int \else: } } -\cs_new_protected_nopar:Npn \regex_action_free_group:n - { \regex_action_free_aux:nn { < \l_regex_step_int } } -\cs_new_protected:Npn \regex_action_free_aux:nn #1#2 +\cs_new_protected_nopar:Npn \__regex_action_free:n + { \__regex_action_free_aux:nn { > \l__regex_step_int \else: } } +\cs_new_protected_nopar:Npn \__regex_action_free_group:n + { \__regex_action_free_aux:nn { < \l__regex_step_int } } +\cs_new_protected:Npn \__regex_action_free_aux:nn #1#2 { \use:x { - \int_add:Nn \l_regex_current_state_int {#2} + \int_add:Nn \l__regex_current_state_int {#2} \exp_not:n { - \if_num:w \tex_dimen:D \l_regex_current_state_int #1 - \exp_after:wN \regex_use_state: + \if_int_compare:w \tex_dimen:D \l__regex_current_state_int #1 + \exp_after:wN \__regex_use_state: \fi: } - \int_set:Nn \l_regex_current_state_int - { \int_use:N \l_regex_current_state_int } - \tl_set:Nn \exp_not:N \l_regex_current_submatches_prop - { \exp_not:o \l_regex_current_submatches_prop } + \int_set:Nn \l__regex_current_state_int + { \int_use:N \l__regex_current_state_int } + \tl_set:Nn \exp_not:N \l__regex_current_submatches_prop + { \exp_not:o \l__regex_current_submatches_prop } } } -\cs_new_protected:Npn \regex_action_cost:n #1 +\cs_new_protected:Npn \__regex_action_cost:n #1 { - \exp_args:No \regex_store_state:n - { \int_use:N \int_eval:w \l_regex_current_state_int + #1 } + \exp_args:No \__regex_store_state:n + { \int_use:N \__int_eval:w \l__regex_current_state_int + #1 } } -\cs_new_protected:Npn \regex_store_state:n #1 +\cs_new_protected:Npn \__regex_store_state:n #1 { - \regex_store_submatches: - \tex_skip:D \l_regex_max_active_int = #1 sp \scan_stop: - \int_incr:N \l_regex_max_active_int + \__regex_store_submatches: + \tex_skip:D \l__regex_max_active_int = #1 sp \scan_stop: + \int_incr:N \l__regex_max_active_int } -\cs_new_protected_nopar:Npn \regex_store_submatches: +\cs_new_protected_nopar:Npn \__regex_store_submatches: { - \tex_toks:D \l_regex_max_active_int \exp_after:wN - { \l_regex_current_submatches_prop } + \tex_toks:D \l__regex_max_active_int \exp_after:wN + { \l__regex_current_submatches_prop } } -\cs_new_protected_nopar:Npn \regex_disable_submatches: +\cs_new_protected_nopar:Npn \__regex_disable_submatches: { - \cs_set_protected_nopar:Npn \regex_store_submatches: { } - \cs_set_protected:Npn \regex_action_submatch:n ##1 { } + \cs_set_protected_nopar:Npn \__regex_store_submatches: { } + \cs_set_protected:Npn \__regex_action_submatch:n ##1 { } } -\cs_new_protected:Npn \regex_action_submatch:n #1 +\cs_new_protected:Npn \__regex_action_submatch:n #1 { - \prop_put:Nno \l_regex_current_submatches_prop {#1} - { \int_use:N \l_regex_current_pos_int } + \prop_put:Nno \l__regex_current_submatches_prop {#1} + { \int_use:N \l__regex_current_pos_int } } -\cs_new_protected_nopar:Npn \regex_action_success: +\cs_new_protected_nopar:Npn \__regex_action_success: { - \regex_if_two_empty_matches:F + \__regex_if_two_empty_matches:F { - \bool_set_true:N \l_regex_match_success_bool - \bool_set_eq:NN \l_regex_empty_success_bool - \l_regex_fresh_thread_bool - \int_set_eq:NN \l_regex_success_pos_int \l_regex_current_pos_int - \prop_set_eq:NN \l_regex_success_submatches_prop - \l_regex_current_submatches_prop - \prg_map_break: + \bool_set_true:N \l__regex_match_success_bool + \bool_set_eq:NN \l__regex_empty_success_bool + \l__regex_fresh_thread_bool + \int_set_eq:NN \l__regex_success_pos_int \l__regex_current_pos_int + \prop_set_eq:NN \l__regex_success_submatches_prop + \l__regex_current_submatches_prop + \__prg_break: } } -\int_new:N \l_regex_replacement_csnames_int -\tl_new:N \l_regex_balance_tl -\cs_new:Npn \regex_replacement_balance_one_match:n #1 - { - \regex_submatch_balance:n {#1} } -\cs_new:Npn \regex_replacement_do_one_match:n #1 +\int_new:N \l__regex_replacement_csnames_int +\tl_new:N \l__regex_balance_tl +\cs_new:Npn \__regex_replacement_balance_one_match:n #1 + { - \__regex_submatch_balance:n {#1} } +\cs_new:Npn \__regex_replacement_do_one_match:n #1 { - \regex_query_range:nn + \__regex_query_range:nn { \etex_glueshrink:D \tex_skip:D #1 } { \tex_skip:D #1 } } -\cs_new:Npn \regex_replacement_exp_not:N #1 { \exp_not:n {#1} } -\cs_new:Npn \regex_query_range:nn #1#2 +\cs_new:Npn \__regex_replacement_exp_not:N #1 { \exp_not:n {#1} } +\cs_new:Npn \__regex_query_range:nn #1#2 { - \exp_after:wN \regex_query_range_loop:ww - \int_use:N \int_eval:w #1 \exp_after:wN ; - \int_use:N \int_eval:w #2 ; - \prg_break_point:n { } + \exp_after:wN \__regex_query_range_loop:ww + \int_use:N \__int_eval:w #1 \exp_after:wN ; + \int_use:N \__int_eval:w #2 ; + \__prg_break_point: } -\cs_new:Npn \regex_query_range_loop:ww #1 ; #2 ; +\cs_new:Npn \__regex_query_range_loop:ww #1 ; #2 ; { - \if_num:w #1 < #2 \exp_stop_f: + \if_int_compare:w #1 < #2 \exp_stop_f: \else: - \exp_after:wN \prg_map_break: + \exp_after:wN \__prg_break: \fi: \tex_the:D \tex_toks:D #1 \exp_stop_f: - \exp_after:wN \regex_query_range_loop:ww - \int_use:N \int_eval:w #1 + \c_one ; #2 ; + \exp_after:wN \__regex_query_range_loop:ww + \int_use:N \__int_eval:w #1 + \c_one ; #2 ; } -\cs_new:Npn \regex_query_submatch:n #1 +\cs_new:Npn \__regex_query_submatch:n #1 { - \regex_query_range:nn - { \tex_skip:D \int_eval:w #1 } - { \etex_gluestretch:D \tex_skip:D \int_eval:w #1 } + \__regex_query_range:nn + { \tex_skip:D \__int_eval:w #1 } + { \etex_gluestretch:D \tex_skip:D \__int_eval:w #1 } } -\cs_new_protected:Npn \regex_submatch_balance:n #1 +\cs_new_protected:Npn \__regex_submatch_balance:n #1 { \etex_glueshrink:D \etex_mutoglue:D \etex_muexpr:D \tex_muskip:D \etex_gluestretch:D \tex_skip:D #1 - \tex_muskip:D \tex_skip:D #1 \scan_stop: } -\cs_new_protected:Npn \regex_replacement:n #1 +\cs_new_protected:Npn \__regex_replacement:n #1 { - \trace_push:nnn { regex } { 1 } { regex_replacement:n } - \tl_set_build:Nw \l_regex_internal_a_tl - \int_zero:N \l_regex_balance_int - \tl_clear:N \l_regex_balance_tl - \regex_escape_use:nnnn + \trace_push:nnn { regex } { 1 } { @@_replacement:n } + \__tl_build:Nw \l__regex_internal_a_tl + \int_zero:N \l__regex_balance_int + \tl_clear:N \l__regex_balance_tl + \__regex_escape_use:nnnn { \if_charcode:w \c_rbrace_str ##1 - \regex_replacement_rbrace:N \else: \tl_build_one:n \fi: ##1 + \__regex_replacement_rbrace:N \else: \__tl_build_one:n \fi: ##1 } - { \regex_replacement_escaped:N ##1 } - { \tl_build_one:n ##1 } + { \__regex_replacement_escaped:N ##1 } + { \__tl_build_one:n ##1 } {#1} \prg_do_nothing: \prg_do_nothing: - \if_int_compare:w \l_regex_replacement_csnames_int > \c_zero - \msg_kernel_error:nnx { regex } { replacement-missing-rbrace } - { \int_use:N \l_regex_replacement_csnames_int } - \tl_build_one:x - { \prg_replicate:nn \l_regex_replacement_csnames_int \cs_end: } + \if_int_compare:w \l__regex_replacement_csnames_int > \c_zero + \__msg_kernel_error:nnx { regex } { replacement-missing-rbrace } + { \int_use:N \l__regex_replacement_csnames_int } + \__tl_build_one:x + { \prg_replicate:nn \l__regex_replacement_csnames_int \cs_end: } \fi: - \cs_gset:Npx \regex_replacement_balance_one_match:n ##1 + \cs_gset:Npx \__regex_replacement_balance_one_match:n ##1 { - + \int_use:N \l_regex_balance_int - \l_regex_balance_tl - - \regex_submatch_balance:n {##1} + + \int_use:N \l__regex_balance_int + \l__regex_balance_tl + - \__regex_submatch_balance:n {##1} } - \tl_build_end: - \exp_args:No \regex_replacement_aux:n \l_regex_internal_a_tl - \trace_pop:nnn { regex } { 1 } { regex_replacement:n } + \__tl_build_end: + \exp_args:No \__regex_replacement_aux:n \l__regex_internal_a_tl + \trace_pop:nnn { regex } { 1 } { @@_replacement:n } } -\cs_new_protected:Npn \regex_replacement_aux:n #1 +\cs_new_protected:Npn \__regex_replacement_aux:n #1 { - \cs_set:Npn \regex_replacement_do_one_match:n ##1 + \cs_set:Npn \__regex_replacement_do_one_match:n ##1 { - \regex_query_range:nn + \__regex_query_range:nn { \etex_glueshrink:D \tex_skip:D ##1 } { \tex_skip:D ##1 } #1 } } -\cs_new_protected:Npn \regex_replacement_escaped:N #1 +\cs_new_protected:Npn \__regex_replacement_escaped:N #1 { - \cs_if_exist_use:cF { regex_replacement_#1:w } + \cs_if_exist_use:cF { __regex_replacement_#1:w } { - \if_num:w \c_one < 1#1 \exp_stop_f: - \regex_replacement_put_submatch:n {#1} + \if_int_compare:w \c_one < 1#1 \exp_stop_f: + \__regex_replacement_put_submatch:n {#1} \else: - \tl_build_one:n #1 + \__tl_build_one:n #1 \fi: } } -\cs_new_protected:Npn \regex_replacement_put_submatch:n #1 +\cs_new_protected:Npn \__regex_replacement_put_submatch:n #1 { - \if_num:w #1 < \l_regex_capturing_group_int - \tl_build_one:n { \regex_query_submatch:n { #1 + ##1 } } - \if_num:w \l_regex_replacement_csnames_int = \c_zero - \tl_put_right:Nn \l_regex_balance_tl - { + \regex_submatch_balance:n { \int_eval:w #1+##1 \int_eval_end: } } + \if_int_compare:w #1 < \l__regex_capturing_group_int + \__tl_build_one:n { \__regex_query_submatch:n { #1 + ##1 } } + \if_int_compare:w \l__regex_replacement_csnames_int = \c_zero + \tl_put_right:Nn \l__regex_balance_tl + { + \__regex_submatch_balance:n { \__int_eval:w #1+##1 \__int_eval_end: } } \fi: \fi: } -\cs_new_protected:Npn \regex_replacement_g:w #1#2 +\cs_new_protected:Npn \__regex_replacement_g:w #1#2 { - \str_if_eq:xxTF { #1#2 } { \tl_build_one:n \c_lbrace_str } + \str_if_eq_x:nnTF { #1#2 } { \__tl_build_one:n \c_lbrace_str } { - \int_zero:N \l_regex_internal_a_int - \regex_replacement_g_digits:NN + \int_zero:N \l__regex_internal_a_int + \__regex_replacement_g_digits:NN } - { \regex_replacement_error:NNN g #1 #2 } + { \__regex_replacement_error:NNN g #1 #2 } } -\cs_new_protected:Npn \regex_replacement_g_digits:NN #1#2 +\cs_new_protected:Npn \__regex_replacement_g_digits:NN #1#2 { - \token_if_eq_meaning:NNTF #1 \tl_build_one:n + \token_if_eq_meaning:NNTF #1 \__tl_build_one:n { - \if_num:w \c_one < 1#2 \exp_stop_f: - \int_set:Nn \l_regex_internal_a_int - { \c_ten * \l_regex_internal_a_int + #2 } + \if_int_compare:w \c_one < 1#2 \exp_stop_f: + \int_set:Nn \l__regex_internal_a_int + { \c_ten * \l__regex_internal_a_int + #2 } \exp_after:wN \use_i:nnn - \exp_after:wN \regex_replacement_g_digits:NN + \exp_after:wN \__regex_replacement_g_digits:NN \else: - \exp_after:wN \regex_replacement_error:NNN + \exp_after:wN \__regex_replacement_error:NNN \exp_after:wN g \fi: } { - \if_meaning:w \regex_replacement_rbrace:N #1 - \exp_args:No \regex_replacement_put_submatch:n - { \int_use:N \l_regex_internal_a_int } + \if_meaning:w \__regex_replacement_rbrace:N #1 + \exp_args:No \__regex_replacement_put_submatch:n + { \int_use:N \l__regex_internal_a_int } \exp_after:wN \use_none:nn \else: - \exp_after:wN \regex_replacement_error:NNN + \exp_after:wN \__regex_replacement_error:NNN \exp_after:wN g \fi: } #1 #2 } -\cs_new_protected:Npn \regex_replacement_c:w #1#2 +\cs_new_protected:Npn \__regex_replacement_c:w #1#2 { - \token_if_eq_meaning:NNTF #1 \tl_build_one:n + \token_if_eq_meaning:NNTF #1 \__tl_build_one:n { - \cs_if_exist_use:cF { regex_replacement_c_#2:w } - { \regex_replacement_error:NNN c #1#2 } + \cs_if_exist_use:cF { __regex_replacement_c_#2:w } + { \__regex_replacement_error:NNN c #1#2 } } - { \regex_replacement_error:NNN c #1#2 } + { \__regex_replacement_error:NNN c #1#2 } } -\cs_new_protected_nopar:cpn { regex_replacement_c_ \c_lbrace_str :w } +\cs_new_protected_nopar:cpn { __regex_replacement_c_ \c_lbrace_str :w } { - \if_case:w \l_regex_replacement_csnames_int - \tl_build_one:n - { \exp_not:n { \exp_after:wN \regex_replacement_exp_not:N \cs:w } } + \if_case:w \l__regex_replacement_csnames_int + \__tl_build_one:n + { \exp_not:n { \exp_after:wN \__regex_replacement_exp_not:N \cs:w } } \else: - \tl_build_one:n { \exp_not:n { \exp_after:wN \tl_to_str:N \cs:w } } + \__tl_build_one:n { \exp_not:n { \exp_after:wN \tl_to_str:N \cs:w } } \fi: - \int_incr:N \l_regex_replacement_csnames_int + \int_incr:N \l__regex_replacement_csnames_int } -\cs_new_protected:Npn \regex_replacement_u:w #1#2 +\cs_new_protected:Npn \__regex_replacement_u:w #1#2 { - \str_if_eq:xxTF { #1#2 } { \tl_build_one:n \c_lbrace_str } + \str_if_eq_x:nnTF { #1#2 } { \__tl_build_one:n \c_lbrace_str } { - \if_case:w \l_regex_replacement_csnames_int - \tl_build_one:n { \exp_not:n { \exp_after:wN \exp_not:V \cs:w } } + \if_case:w \l__regex_replacement_csnames_int + \__tl_build_one:n { \exp_not:n { \exp_after:wN \exp_not:V \cs:w } } \else: - \tl_build_one:n { \exp_not:n { \exp_after:wN \tl_to_str:V \cs:w } } + \__tl_build_one:n { \exp_not:n { \exp_after:wN \tl_to_str:V \cs:w } } \fi: - \int_incr:N \l_regex_replacement_csnames_int + \int_incr:N \l__regex_replacement_csnames_int } - { \regex_replacement_error:NNN u #1#2 } + { \__regex_replacement_error:NNN u #1#2 } } -\cs_new_protected:Npn \regex_replacement_rbrace:N #1 +\cs_new_protected:Npn \__regex_replacement_rbrace:N #1 { - \if_int_compare:w \l_regex_replacement_csnames_int > \c_zero - \tl_build_one:n \cs_end: - \int_decr:N \l_regex_replacement_csnames_int + \if_int_compare:w \l__regex_replacement_csnames_int > \c_zero + \__tl_build_one:n \cs_end: + \int_decr:N \l__regex_replacement_csnames_int \else: - \tl_build_one:n #1 + \__tl_build_one:n #1 \fi: } \group_begin: - \cs_new_protected:Npn \regex_replacement_char:nNN #1#2#3 + \cs_new_protected:Npn \__regex_replacement_char:nNN #1#2#3 { \if_meaning:w \prg_do_nothing: #3 - \msg_kernel_error:nn { regex } { replacement-catcode-end } + \__msg_kernel_error:nn { regex } { replacement-catcode-end } \else: \tex_lccode:D \c_zero = `#3 \scan_stop: - \tl_to_lowercase:n { \tl_build_one:n {#1} } + \tl_to_lowercase:n { \__tl_build_one:n {#1} } \fi: } \char_set_catcode_active:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_A:w - { \regex_replacement_char:nNN { \exp_not:n { \exp_not:N ^^@ } } } + \cs_new_protected_nopar:Npn \__regex_replacement_c_A:w + { \__regex_replacement_char:nNN { \exp_not:n { \exp_not:N ^^@ } } } \char_set_catcode_group_begin:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_B:w + \cs_new_protected_nopar:Npn \__regex_replacement_c_B:w { - \if_num:w \l_regex_replacement_csnames_int = \c_zero - \int_incr:N \l_regex_balance_int + \if_int_compare:w \l__regex_replacement_csnames_int = \c_zero + \int_incr:N \l__regex_balance_int \fi: - \regex_replacement_char:nNN + \__regex_replacement_char:nNN { \exp_not:n { \exp_after:wN ^^@ \if_false: } \fi: } } } - \cs_new_protected:Npn \regex_replacement_c_C:w #1#2 - { \tl_build_one:n { \exp_not:N \exp_not:N \exp_not:c {#2} } } + \cs_new_protected:Npn \__regex_replacement_c_C:w #1#2 + { \__tl_build_one:n { \exp_not:N \exp_not:N \exp_not:c {#2} } } \char_set_catcode_math_subscript:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_D:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \__regex_replacement_c_D:w + { \__regex_replacement_char:nNN { ^^@ } } \char_set_catcode_group_end:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_E:w + \cs_new_protected_nopar:Npn \__regex_replacement_c_E:w { - \if_num:w \l_regex_replacement_csnames_int = \c_zero - \int_decr:N \l_regex_balance_int + \if_int_compare:w \l__regex_replacement_csnames_int = \c_zero + \int_decr:N \l__regex_balance_int \fi: - \regex_replacement_char:nNN + \__regex_replacement_char:nNN { \exp_not:n { \if_false: { \fi: ^^@ } } } \char_set_catcode_letter:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_L:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \__regex_replacement_c_L:w + { \__regex_replacement_char:nNN { ^^@ } } \char_set_catcode_math_toggle:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_M:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \__regex_replacement_c_M:w + { \__regex_replacement_char:nNN { ^^@ } } \char_set_catcode_other:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_O:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \__regex_replacement_c_O:w + { \__regex_replacement_char:nNN { ^^@ } } \char_set_catcode_parameter:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_P:w + \cs_new_protected_nopar:Npn \__regex_replacement_c_P:w { - \regex_replacement_char:nNN + \__regex_replacement_char:nNN { \exp_not:n { \exp_not:n { ^^@^^@^^@^^@ } } } } - \cs_new_protected:Npn \regex_replacement_c_S:w #1#2 + \cs_new_protected:Npn \__regex_replacement_c_S:w #1#2 { \if_meaning:w \prg_do_nothing: #2 - \msg_kernel_error:nn { regex } { replacement-catcode-end } + \__msg_kernel_error:nn { regex } { replacement-catcode-end } \else: - \if_num:w `#2 = \c_zero - \msg_kernel_error:nn { regex } { replacement-null-space } + \if_int_compare:w `#2 = \c_zero + \__msg_kernel_error:nn { regex } { replacement-null-space } \fi: \tex_lccode:D 32 = `#2 \scan_stop: - \tl_to_lowercase:n { \tl_build_one:n {~} } + \tl_to_lowercase:n { \__tl_build_one:n {~} } \fi: } \char_set_catcode_alignment:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_T:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \__regex_replacement_c_T:w + { \__regex_replacement_char:nNN { ^^@ } } \char_set_catcode_math_superscript:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_U:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \__regex_replacement_c_U:w + { \__regex_replacement_char:nNN { ^^@ } } \group_end: -\cs_new_protected:Npn \regex_replacement_error:NNN #1#2#3 +\cs_new_protected:Npn \__regex_replacement_error:NNN #1#2#3 { - \msg_kernel_error:nnx { regex } { replacement-#1 } {#3} + \__msg_kernel_error:nnx { regex } { replacement-#1 } {#3} #2 #3 } \cs_new_protected:Npn \regex_new:N #1 - { \cs_new_eq:NN #1 \c_regex_no_match_regex } + { \cs_new_eq:NN #1 \c__regex_no_match_regex } \cs_new_protected_nopar:Npn \regex_set:Nn #1#2 { - \regex_compile:n {#2} - \tl_set_eq:NN #1 \l_regex_internal_regex + \__regex_compile:n {#2} + \tl_set_eq:NN #1 \l__regex_internal_regex } \cs_new_protected_nopar:Npn \regex_gset:Nn #1#2 { - \regex_compile:n {#2} - \tl_gset_eq:NN #1 \l_regex_internal_regex + \__regex_compile:n {#2} + \tl_gset_eq:NN #1 \l__regex_internal_regex } \cs_new_protected_nopar:Npn \regex_const:Nn #1#2 { - \regex_compile:n {#2} - \tl_const:Nx #1 { \exp_not:o \l_regex_internal_regex } + \__regex_compile:n {#2} + \tl_const:Nx #1 { \exp_not:o \l__regex_internal_regex } } \cs_new_protected:Npn \regex_show:n #1 { - \regex_compile:n {#1} - \regex_show_aux:Nx \l_regex_internal_regex + \__regex_compile:n {#1} + \__regex_show:Nx \l__regex_internal_regex { { \tl_to_str:n {#1} } } } \cs_new_protected:Npn \regex_show:N #1 - { \regex_show_aux:Nx #1 { variable~\token_to_str:N #1 } } + { \__regex_show:Nx #1 { variable~\token_to_str:N #1 } } \prg_new_protected_conditional:Npnn \regex_match:nn #1#2 { T , F , TF } { - \regex_match_aux:nn { \regex_build:n {#1} } {#2} - \regex_aux_return: + \__regex_if_match:nn { \__regex_build:n {#1} } {#2} + \__regex_return: } \prg_new_protected_conditional:Npnn \regex_match:Nn #1#2 { T , F , TF } { - \regex_match_aux:nn { \regex_build:N #1 } {#2} - \regex_aux_return: + \__regex_if_match:nn { \__regex_build:N #1 } {#2} + \__regex_return: } \cs_new_protected:Npn \regex_count:nnN #1 - { \regex_count_aux:nnN { \regex_build:n {#1} } } + { \__regex_count:nnN { \__regex_build:n {#1} } } \cs_new_protected:Npn \regex_count:NnN #1 - { \regex_count_aux:nnN { \regex_build:N #1 } } -\cs_set_protected:Npn \regex_tmp:w #1#2#3 + { \__regex_count:nnN { \__regex_build:N #1 } } +\cs_set_protected:Npn \__regex_tmp:w #1#2#3 { - \cs_new_protected:Npn #2 ##1 { #1 { \regex_build:n {##1} } } - \cs_new_protected:Npn #3 ##1 { #1 { \regex_build:N ##1 } } + \cs_new_protected:Npn #2 ##1 { #1 { \__regex_build:n {##1} } } + \cs_new_protected:Npn #3 ##1 { #1 { \__regex_build:N ##1 } } \prg_new_protected_conditional:Npnn #2 ##1##2##3 { T , F , TF } - { #1 { \regex_build:n {##1} } {##2} ##3 \regex_aux_return: } + { #1 { \__regex_build:n {##1} } {##2} ##3 \__regex_return: } \prg_new_protected_conditional:Npnn #3 ##1##2##3 { T , F , TF } - { #1 { \regex_build:N ##1 } {##2} ##3 \regex_aux_return: } + { #1 { \__regex_build:N ##1 } {##2} ##3 \__regex_return: } } -\regex_tmp:w \regex_extract_once_aux:nnN +\__regex_tmp:w \__regex_extract_once:nnN \regex_extract_once:nnN \regex_extract_once:NnN -\regex_tmp:w \regex_extract_all_aux:nnN +\__regex_tmp:w \__regex_extract_all:nnN \regex_extract_all:nnN \regex_extract_all:NnN -\regex_tmp:w \regex_replace_once_aux:nnN +\__regex_tmp:w \__regex_replace_once:nnN \regex_replace_once:nnN \regex_replace_once:NnN -\regex_tmp:w \regex_replace_all_aux:nnN +\__regex_tmp:w \__regex_replace_all:nnN \regex_replace_all:nnN \regex_replace_all:NnN -\regex_tmp:w \regex_split_aux:nnN \regex_split:nnN \regex_split:NnN -\int_new:N \l_regex_match_count_int -\flag_new:n { regex_begin } -\flag_new:n { regex_end } -\int_new:N \l_regex_submatch_int -\int_new:N \l_regex_zeroth_submatch_int -\cs_new_protected_nopar:Npn \regex_aux_return: - { - \if_meaning:w \c_true_bool \g_regex_success_bool +\__regex_tmp:w \__regex_split:nnN \regex_split:nnN \regex_split:NnN +\int_new:N \l__regex_match_count_int +\flag_new:n { __regex_begin } +\flag_new:n { __regex_end } +\int_new:N \l__regex_submatch_int +\int_new:N \l__regex_zeroth_submatch_int +\cs_new_protected_nopar:Npn \__regex_return: + { + \if_meaning:w \c_true_bool \g__regex_success_bool \prg_return_true: \else: \prg_return_false: \fi: } -\cs_new_protected:Npn \regex_match_aux:nn #1#2 +\cs_new_protected:Npn \__regex_if_match:nn #1#2 { \group_begin: - \regex_disable_submatches: - \regex_single_match: + \__regex_disable_submatches: + \__regex_single_match: #1 - \regex_match:n {#2} + \__regex_match:n {#2} \group_end: } -\cs_new_protected:Npn \regex_count_aux:nnN #1#2#3 +\cs_new_protected:Npn \__regex_count:nnN #1#2#3 { \group_begin: - \regex_disable_submatches: - \int_zero:N \l_regex_match_count_int - \regex_multi_match:n { \int_incr:N \l_regex_match_count_int } + \__regex_disable_submatches: + \int_zero:N \l__regex_match_count_int + \__regex_multi_match:n { \int_incr:N \l__regex_match_count_int } #1 - \regex_match:n {#2} + \__regex_match:n {#2} \exp_args:NNNo \group_end: - \int_set:Nn #3 { \int_use:N \l_regex_match_count_int } + \int_set:Nn #3 { \int_use:N \l__regex_match_count_int } } -\cs_new_protected:Npn \regex_extract_once_aux:nnN #1#2#3 +\cs_new_protected:Npn \__regex_extract_once:nnN #1#2#3 { \group_begin: - \regex_single_match: + \__regex_single_match: #1 - \regex_match:n {#2} - \regex_extract: - \regex_group_end_extract_seq:N #3 + \__regex_match:n {#2} + \__regex_extract: + \__regex_group_end_extract_seq:N #3 } -\cs_new_protected:Npn \regex_extract_all_aux:nnN #1#2#3 +\cs_new_protected:Npn \__regex_extract_all:nnN #1#2#3 { \group_begin: - \regex_multi_match:n { \regex_extract: } + \__regex_multi_match:n { \__regex_extract: } #1 - \regex_match:n {#2} - \regex_group_end_extract_seq:N #3 + \__regex_match:n {#2} + \__regex_group_end_extract_seq:N #3 } -\cs_new_protected:Npn \regex_split_aux:nnN #1#2#3 +\cs_new_protected:Npn \__regex_split:nnN #1#2#3 { \group_begin: - \regex_multi_match:n + \__regex_multi_match:n { - \if_num:w \l_regex_start_pos_int < \l_regex_success_pos_int - \regex_extract: - \tex_skip:D \l_regex_zeroth_submatch_int - = \l_regex_start_pos_int sp - plus \tex_skip:D \l_regex_zeroth_submatch_int \scan_stop: + \if_int_compare:w \l__regex_start_pos_int < \l__regex_success_pos_int + \__regex_extract: + \tex_skip:D \l__regex_zeroth_submatch_int + = \l__regex_start_pos_int sp + plus \tex_skip:D \l__regex_zeroth_submatch_int \scan_stop: \fi: } #1 - \regex_match:n {#2} - \tex_skip:D \l_regex_submatch_int - = \l_regex_start_pos_int sp plus \l_regex_max_pos_int sp \scan_stop: - \int_incr:N \l_regex_submatch_int - \if_meaning:w \c_true_bool \l_regex_empty_success_bool - \if_num:w \l_regex_start_pos_int = \l_regex_max_pos_int - \int_decr:N \l_regex_submatch_int + \__regex_match:n {#2} + \tex_skip:D \l__regex_submatch_int + = \l__regex_start_pos_int sp plus \l__regex_max_pos_int sp \scan_stop: + \int_incr:N \l__regex_submatch_int + \if_meaning:w \c_true_bool \l__regex_empty_success_bool + \if_int_compare:w \l__regex_start_pos_int = \l__regex_max_pos_int + \int_decr:N \l__regex_submatch_int \fi: \fi: - \regex_group_end_extract_seq:N #3 + \__regex_group_end_extract_seq:N #3 } -\cs_new_protected:Npn \regex_group_end_extract_seq:N #1 +\cs_new_protected:Npn \__regex_group_end_extract_seq:N #1 { - \cs_set_eq:NN \seq_item:n \scan_stop: - \flag_clear:n { regex_begin } - \flag_clear:n { regex_end } - \tl_set:Nx \l_regex_internal_a_tl + \cs_set_eq:NN \__seq_item:n \scan_stop: + \flag_clear:n { __regex_begin } + \flag_clear:n { __regex_end } + \tl_set:Nx \l__regex_internal_a_tl { - \prg_stepwise_function:nnnN - { \c_two * \l_regex_max_state_int } + \int_step_function:nnnN + { \c_two * \l__regex_max_state_int } \c_one - { \l_regex_submatch_int - \c_one } - \regex_extract_seq_aux:n + { \l__regex_submatch_int - \c_one } + \__regex_extract_seq_aux:n } \int_compare:nNnF - { \flag_height:n { regex_begin } + \flag_height:n { regex_end } } + { \flag_height:n { __regex_begin } + \flag_height:n { __regex_end } } = \c_zero { - \msg_kernel_error:nnxxx { regex } { result-unbalanced } + \__msg_kernel_error:nnxxx { regex } { result-unbalanced } { splitting~or~extracting~submatches } - { \flag_height:n { regex_end } } - { \flag_height:n { regex_begin } } + { \flag_height:n { __regex_end } } + { \flag_height:n { __regex_begin } } } \use:x { \group_end: - \tl_set:Nn \exp_not:N #1 { \l_regex_internal_a_tl } + \tl_set:Nn \exp_not:N #1 { \l__regex_internal_a_tl } } } -\cs_new:Npn \regex_extract_seq_aux:n #1 +\cs_new:Npn \__regex_extract_seq_aux:n #1 { - \seq_item:n + \__seq_item:n { - \exp_after:wN \regex_extract_seq_aux:ww - \int_value:w \regex_submatch_balance:n {#1} ; #1; + \exp_after:wN \__regex_extract_seq_aux:ww + \__int_value:w \__regex_submatch_balance:n {#1} ; #1; } } -\cs_new:Npn \regex_extract_seq_aux:ww #1; #2; +\cs_new:Npn \__regex_extract_seq_aux:ww #1; #2; { - \if_num:w #1 < \c_zero - \flag_raise:n { regex_end } + \if_int_compare:w #1 < \c_zero + \flag_raise:n { __regex_end } \prg_replicate:nn {-#1} { \exp_not:n { { \if_false: } \fi: } } \fi: - \regex_query_submatch:n {#2} - \if_num:w #1 > \c_zero - \flag_raise:n { regex_begin } + \__regex_query_submatch:n {#2} + \if_int_compare:w #1 > \c_zero + \flag_raise:n { __regex_begin } \prg_replicate:nn {#1} { \exp_not:n { \if_false: { \fi: } } } \fi: } -\cs_new_protected_nopar:Npn \regex_extract: +\cs_new_protected_nopar:Npn \__regex_extract: { - \if_meaning:w \c_true_bool \g_regex_success_bool - \int_set_eq:NN \l_regex_zeroth_submatch_int \l_regex_submatch_int - \prg_replicate:nn \l_regex_capturing_group_int + \if_meaning:w \c_true_bool \g__regex_success_bool + \int_set_eq:NN \l__regex_zeroth_submatch_int \l__regex_submatch_int + \prg_replicate:nn \l__regex_capturing_group_int { - \tex_skip:D \l_regex_submatch_int \c_zero sp \scan_stop: - \int_incr:N \l_regex_submatch_int + \tex_skip:D \l__regex_submatch_int \c_zero sp \scan_stop: + \int_incr:N \l__regex_submatch_int } - \prop_map_inline:Nn \l_regex_success_submatches_prop + \prop_map_inline:Nn \l__regex_success_submatches_prop { - \if_num:w ##1 \c_minus_one - \exp_after:wN \regex_extract_aux_e:wn \int_value:w + \if_int_compare:w ##1 \c_minus_one + \exp_after:wN \__regex_extract_e:wn \__int_value:w \else: - \exp_after:wN \regex_extract_aux_b:wn \int_value:w + \exp_after:wN \__regex_extract_b:wn \__int_value:w \fi: - \int_eval:w \l_regex_zeroth_submatch_int + ##1 {##2} + \__int_eval:w \l__regex_zeroth_submatch_int + ##1 {##2} } - \tex_skip:D \l_regex_zeroth_submatch_int - = \tex_the:D \tex_skip:D \l_regex_zeroth_submatch_int - minus \l_regex_start_pos_int sp \scan_stop: + \tex_skip:D \l__regex_zeroth_submatch_int + = \tex_the:D \tex_skip:D \l__regex_zeroth_submatch_int + minus \l__regex_start_pos_int sp \scan_stop: \fi: } -\cs_new_protected:Npn \regex_extract_aux_b:wn #1 < #2 +\cs_new_protected:Npn \__regex_extract_b:wn #1 < #2 { \tex_skip:D #1 = #2 sp plus \etex_gluestretch:D \tex_skip:D #1 \scan_stop: } -\cs_new_protected:Npn \regex_extract_aux_e:wn #1 > #2 +\cs_new_protected:Npn \__regex_extract_e:wn #1 > #2 { \tex_skip:D #1 = 1 \tex_skip:D #1 plus #2 sp \scan_stop: } -\cs_new_protected:Npn \regex_replace_once_aux:nnN #1#2#3 +\cs_new_protected:Npn \__regex_replace_once:nnN #1#2#3 { \group_begin: - \regex_single_match: + \__regex_single_match: #1 - \regex_replacement:n {#2} - \exp_args:No \regex_match:n { #3 } - \if_meaning:w \c_false_bool \g_regex_success_bool + \__regex_replacement:n {#2} + \exp_args:No \__regex_match:n { #3 } + \if_meaning:w \c_false_bool \g__regex_success_bool \group_end: \else: - \regex_extract: - \int_set:Nn \l_regex_balance_int + \__regex_extract: + \int_set:Nn \l__regex_balance_int { - \regex_replacement_balance_one_match:n - { \l_regex_zeroth_submatch_int } + \__regex_replacement_balance_one_match:n + { \l__regex_zeroth_submatch_int } } - \tl_set:Nx \l_regex_internal_a_tl + \tl_set:Nx \l__regex_internal_a_tl { - \regex_replacement_do_one_match:n { \l_regex_zeroth_submatch_int } - \regex_query_range:nn - { \etex_gluestretch:D \tex_skip:D \l_regex_zeroth_submatch_int } - { \l_regex_max_pos_int } + \__regex_replacement_do_one_match:n { \l__regex_zeroth_submatch_int } + \__regex_query_range:nn + { \etex_gluestretch:D \tex_skip:D \l__regex_zeroth_submatch_int } + { \l__regex_max_pos_int } } - \regex_group_end_replace:N #3 + \__regex_group_end_replace:N #3 \fi: } -\cs_new_protected:Npn \regex_replace_all_aux:nnN #1#2#3 +\cs_new_protected:Npn \__regex_replace_all:nnN #1#2#3 { \group_begin: - \regex_multi_match:n { \regex_extract: } + \__regex_multi_match:n { \__regex_extract: } #1 - \regex_replacement:n {#2} - \exp_args:No \regex_match:n {#3} - \int_set:Nn \l_regex_balance_int + \__regex_replacement:n {#2} + \exp_args:No \__regex_match:n {#3} + \int_set:Nn \l__regex_balance_int { 0 - \prg_stepwise_function:nnnN - { \c_two * \l_regex_max_state_int } - \l_regex_capturing_group_int - { \l_regex_submatch_int - \c_one } - \regex_replacement_balance_one_match:n + \int_step_function:nnnN + { \c_two * \l__regex_max_state_int } + \l__regex_capturing_group_int + { \l__regex_submatch_int - \c_one } + \__regex_replacement_balance_one_match:n } - \tl_set:Nx \l_regex_internal_a_tl + \tl_set:Nx \l__regex_internal_a_tl { - \prg_stepwise_function:nnnN - { \c_two * \l_regex_max_state_int } - \l_regex_capturing_group_int - { \l_regex_submatch_int - \c_one } - \regex_replacement_do_one_match:n - \regex_query_range:nn - \l_regex_start_pos_int \l_regex_max_pos_int + \int_step_function:nnnN + { \c_two * \l__regex_max_state_int } + \l__regex_capturing_group_int + { \l__regex_submatch_int - \c_one } + \__regex_replacement_do_one_match:n + \__regex_query_range:nn + \l__regex_start_pos_int \l__regex_max_pos_int } - \regex_group_end_replace:N #3 + \__regex_group_end_replace:N #3 } -\cs_new_protected_nopar:Npn \regex_group_end_replace:N #1 +\cs_new_protected_nopar:Npn \__regex_group_end_replace:N #1 { - \if_num:w \l_regex_balance_int = \c_zero + \if_int_compare:w \l__regex_balance_int = \c_zero \else: - \msg_kernel_error:nnxxx { regex } { result-unbalanced } + \__msg_kernel_error:nnxxx { regex } { result-unbalanced } { replacing } - { \int_max:nn { - \l_regex_balance_int } { \c_zero } } - { \int_max:nn { \l_regex_balance_int } { \c_zero } } + { \int_max:nn { - \l__regex_balance_int } { \c_zero } } + { \int_max:nn { \l__regex_balance_int } { \c_zero } } \fi: \use:x { \group_end: \tl_set:Nn \exp_not:N #1 { - \if_int_compare:w \l_regex_balance_int < \c_zero - \prg_replicate:nn { - \l_regex_balance_int } + \if_int_compare:w \l__regex_balance_int < \c_zero + \prg_replicate:nn { - \l__regex_balance_int } { { \if_false: } \fi: } \fi: - \l_regex_internal_a_tl - \if_int_compare:w \l_regex_balance_int > \c_zero - \prg_replicate:nn { \l_regex_balance_int } + \l__regex_internal_a_tl + \if_int_compare:w \l__regex_balance_int > \c_zero + \prg_replicate:nn { \l__regex_balance_int } { \if_false: { \fi: } } \fi: } } } -\msg_kernel_new:nnnn { regex } { trailing-backslash } +\__msg_kernel_new:nnnn { regex } { trailing-backslash } { Trailing~escape~character~\iow_char:N\\. } { A~regular~expression~or~its~replacement~text~ends~with~ the~escape~character~\iow_char:N\\.~It~will~be~ignored. } -\msg_kernel_new:nnnn { regex } { x-missing-rbrace } +\__msg_kernel_new:nnnn { regex } { x-missing-rbrace } { Missing~closing~brace~in~\iow_char:N\\x~hexadecimal~sequence. } { You~wrote~something~like~ `\iow_char:N\\x\{\int_to_hexadecimal:n{#1}'.~ The~closing~brace~is~missing. } -\msg_kernel_new:nnnn { regex } { x-overflow } +\__msg_kernel_new:nnnn { regex } { x-overflow } { Character~code~'#1'~too~large~in~\iow_char:N\\x~hexadecimal~sequence. } { You~wrote~something~like~ `\iow_char:N\\x\{\int_to_hexadecimal:n{#1}\}'.~ The~character~code~'#1'~is~larger~than~\int_use:N \c_max_char_int. } -\msg_kernel_new:nnnn { regex } { invalid-quantifier } +\__msg_kernel_new:nnnn { regex } { invalid-quantifier } { Braced~quantifier~'#1'~may~not~be~followed~by~'#2'. } { The~character~'#2'~is~invalid~in~the~braced~quantifier~'#1'.~ The~only~valid~quantifiers~are~'*',~'?',~'+',~'{}',~ '{,}'~and~'{,}',~followed~or~not~by~'?'. } -\msg_kernel_new:nnnn { regex } { missing-rbrack } +\__msg_kernel_new:nnnn { regex } { missing-rbrack } { Missing~right~bracket~inserted~in~regular~expression. } { LaTeX~was~given~a~regular~expression~where~a~character~class~ was~started~with~'[',~but~the~matching~']'~is~missing. } -\msg_kernel_new:nnnn { regex } { missing-rparen } +\__msg_kernel_new:nnnn { regex } { missing-rparen } { Missing~right~parenthes\int_compare:nTF{#1=1}{i}{e}s~ inserted~in~regular~expression. @@ -2624,44 +2632,44 @@ more~left~parenthes\int_compare:nTF{#1=1}{i}{e}s~than~right~ parenthes\int_compare:nTF{#1=1}{i}{e}s. } -\msg_kernel_new:nnnn { regex } { extra-rparen } +\__msg_kernel_new:nnnn { regex } { extra-rparen } { Extra~right~parenthesis~ignored~in~regular~expression. } { LaTeX~came~across~a~closing~parenthesis~when~no~submatch~group~ was~open.~The~parenthesis~will~be~ignored. } -\msg_kernel_new:nnnn { regex } { bad-escape } +\__msg_kernel_new:nnnn { regex } { bad-escape } { Invalid~escape~\c_backslash_str #1~ - \regex_if_in_cs:TF { within~a~control~sequence. } + \__regex_if_in_cs:TF { within~a~control~sequence. } { - \regex_if_in_class:TF + \__regex_if_in_class:TF { in~a~character~class. } { following~a~category~test. } } } { The~escape~sequence~\iow_char:N\\#1~may~not~appear~ - \regex_if_in_cs:TF + \__regex_if_in_cs:TF { within~a~control~sequence~test~introduced~by~ \iow_char:N\\c\iow_char:N\{. } { - \regex_if_in_class:TF + \__regex_if_in_class:TF { within~a~character~class~ } { following~a~category~test~such~as~\iow_char:N\\cL~ } because~it~does~not~match~exactly~one~character. } } -\msg_kernel_new:nnnn { regex } { range-missing-end } +\__msg_kernel_new:nnnn { regex } { range-missing-end } { Invalid~end-point~for~range~'#1-#2'~in~character~class. } { The~end-point~'#2'~of~the~range~'#1-#2'~may~not~serve~as~an~ end-point~for~a~range:~alphanumeric~characters~should~not~be~ escaped,~and~non-alphanumeric~characters~should~be~escaped. } -\msg_kernel_new:nnnn { regex } { range-backwards } +\__msg_kernel_new:nnnn { regex } { range-backwards } { Range~[#1-#2]~out~of~order~in~character~class. } { In~ranges~of~characters~[x-y]~appearing~in~character~classes,~ @@ -2669,27 +2677,27 @@ Here,~#1~has~character~code~\int_eval:n {`#1},~while~#2~has~ character~code~\int_eval:n {`#2}. } -\msg_kernel_new:nnnn { regex } { c-bad-mode } +\__msg_kernel_new:nnnn { regex } { c-bad-mode } { Invalid~nested~\iow_char:N\\c~escape~in~regular~expression. } { The~\iow_char:N\\c~escape~cannot~be~used~within~ a~control~sequence~test~'\iow_char:N\\c{...}'.~ To~combine~several~category~tests,~use~'\iow_char:N\\c[...]'. } -\msg_kernel_new:nnnn { regex } { c-missing-rbrace } +\__msg_kernel_new:nnnn { regex } { c-missing-rbrace } { Missing~right~brace~inserted~for~\iow_char:N\\c~escape. } { LaTeX~was~given~a~regular~expression~where~a~ '\iow_char:N\\c\iow_char:N\{...'~construction~was~not~ended~ with~a~closing~brace~'\iow_char:N\}'. } -\msg_kernel_new:nnnn { regex } { c-missing-rbrack } +\__msg_kernel_new:nnnn { regex } { c-missing-rbrack } { Missing~right~bracket~inserted~for~\iow_char:N\\c~escape. } { A~construction~'\iow_char:N\\c[...'~appears~in~a~ regular~expression,~but~the~closing~']'~is~not~present. } -\msg_kernel_new:nnnn { regex } { c-missing-category } +\__msg_kernel_new:nnnn { regex } { c-missing-category } { Invalid~character~'#1'~following~\iow_char:N\\c~escape. } { In~regular~expressions,~the~\iow_char:N\\c~escape~sequence~ @@ -2697,29 +2705,29 @@ capital~letter~representing~a~character~category,~namely~ one~of~ABCDELMOPSTU. } -\msg_kernel_new:nnnn { regex } { u-missing-lbrace } +\__msg_kernel_new:nnnn { regex } { u-missing-lbrace } { Missing~left~brace~following~\iow_char:N\\u~escape. } { The~\iow_char:N\\u~escape~sequence~must~be~followed~by~ a~brace~group~with~the~name~of~the~variable~to~use. } -\msg_kernel_new:nnnn { regex } { u-missing-rbrace } +\__msg_kernel_new:nnnn { regex } { u-missing-rbrace } { Missing~right~brace~inserted~for~\iow_char:N\\u~escape. } { LaTeX~ - \tl_if_empty:xTF {#2} + \str_if_eq_x:nnTF { } {#2} { reached~the~end~of~the~string~ } { encountered~an~escaped~alphanumeric~character '\iow_char:N\\#2'~ } when~parsing~the~argument~of~an~'\iow_char:N\\u\iow_char:N\{...\}'~escape. } -\msg_kernel_new:nnnn { regex } { posix-unsupported } +\__msg_kernel_new:nnnn { regex } { posix-unsupported } { POSIX~collating~element~'[#1 ~ #1]'~not~supported. } { The~[.foo.]~and~[=bar=]~syntaxes~have~a~special~meaning~in~POSIX~ regular~expressions.~This~is~not~supported~by~LaTeX.~Maybe~you~ forgot~to~escape~a~left~bracket~in~a~character~class? } -\msg_kernel_new:nnnn { regex } { posix-unknown } +\__msg_kernel_new:nnnn { regex } { posix-unknown } { POSIX~class~[:#1:]~unknown. } { [:#1:]~is~not~among~the~known~POSIX~classes~ @@ -2728,10 +2736,10 @@ [:print:],~[:punct:],~[:space:],~[:upper:],~ [:word:],~and~[:xdigit:]. } -\msg_kernel_new:nnnn { regex } { posix-missing-close } +\__msg_kernel_new:nnnn { regex } { posix-missing-close } { Missing~closing~':]'~for~POSIX~class. } { The~POSIX~syntax~'#1'~must~be~followed~by~':]',~not~'#2'. } -\msg_kernel_new:nnnn { regex } { result-unbalanced } +\__msg_kernel_new:nnnn { regex } { result-unbalanced } { Missing~brace~inserted~when~#1. } { LaTeX~was~asked~to~do~some~regular~expression~operation,~ @@ -2739,34 +2747,34 @@ of~begin-group~and~end-group~tokens.~Braces~were~inserted:~ #2~left,~#3~right. } -\msg_kernel_new:nnnn { regex } { unknown-option } +\__msg_kernel_new:nnnn { regex } { unknown-option } { Unknown~option~`#1'~for~regular~expressions. } { The~only~available~option~is~'case-insensitive',~toggled~by~ '(?i)'~and~'(?-i)'. } -\msg_kernel_new:nnnn { regex } { replacement-c } +\__msg_kernel_new:nnnn { regex } { replacement-c } { Misused~\iow_char:N\\c~command~in~a~replacement~text. } { In~a~replacement~text,~the~\iow_char:N\\c~escape~sequence~ can~be~followed~by~one~of~the~letters~ABCDELMOPSTU~ or~a~brace~group,~not~by~'#1'. } -\msg_kernel_new:nnnn { regex } { replacement-u } +\__msg_kernel_new:nnnn { regex } { replacement-u } { Misused~\iow_char:N\\u~command~in~a~replacement~text. } { In~a~replacement~text,~the~\iow_char:N\\u~escape~sequence~ must~be~~followed~by~a~brace~group~holding~the~name~of~the~ variable~to~use. } -\msg_kernel_new:nnnn { regex } { replacement-g } +\__msg_kernel_new:nnnn { regex } { replacement-g } { Missing~brace~for~the~\iow_char:N\\g~construction~in~a~replacement~text. } { In~the~replacement~text~for~a~regular~expression~search,~ submatches~are~represented~either~as~\iow_char:N \\g{dd..d},~ or~\\d,~where~`d'~are~single~digits.~Here,~a~brace~is~missing. } -\msg_kernel_new:nnnn { regex } { replacement-catcode-end } +\__msg_kernel_new:nnnn { regex } { replacement-catcode-end } { Missing~character~for~the~\iow_char:N\\c~ construction~in~a~replacement~text. @@ -2777,7 +2785,7 @@ the~character~category.~Then,~a~character~must~follow.~LaTeX~ reached~the~end~of~the~replacement~when~looking~for~that. } -\msg_kernel_new:nnnn { regex } { replacement-null-space } +\__msg_kernel_new:nnnn { regex } { replacement-null-space } { TeX~cannot~build~a~space~token~with~character~code~0. } { You~asked~for~a~character~token~with~category~'space',~ @@ -2786,18 +2794,18 @@ This~specific~case~is~impossible~and~will~be~replaced~ by~a~normal~space. } -\msg_kernel_new:nnnn { regex } { replacement-missing-rbrace } +\__msg_kernel_new:nnnn { regex } { replacement-missing-rbrace } { Missing~right~brace~inserted~in~replacement~text. } { - There~were~\int_use:N \l_regex_replacement_csnames_int \ + There~were~\int_use:N \l__regex_replacement_csnames_int \ missing~right~braces. } -\cs_new:Npn \regex_msg_repeated:nnN #1#2#3 +\cs_new:Npn \__regex_msg_repeated:nnN #1#2#3 { - \str_if_eq:xxF { #1 #2 } { 1 0 } + \str_if_eq_x:nnF { #1 #2 } { 1 0 } { , ~ repeated ~ - \prg_case_int:nnn {#2} + \int_case:nnn {#2} { { -1 } { #1~or~more~times,~\bool_if:NTF #3 { lazy } { greedy } } { 0 } { #1~times } @@ -2808,12 +2816,12 @@ } } } -\cs_new_protected:Npn \regex_trace_states:n #1 +\cs_new_protected:Npn \__regex_trace_states:n #1 { - \prg_stepwise_inline:nnnn - \l_regex_min_state_int + \int_step_inline:nnnn + \l__regex_min_state_int \c_one - { \l_regex_max_state_int - 1 } + { \l__regex_max_state_int - 1 } { \trace:nnx { regex } { #1 } { \iow_char:N \\toks ##1 = { \tex_the:D \tex_toks:D ##1 } } diff --git a/Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex.sty b/Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex.sty index d6babc2b1b7..2ba86a82c86 100644 --- a/Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex.sty +++ b/Master/texmf-dist/tex/latex/l3experimental/l3str/l3regex.sty @@ -46,446 +46,446 @@ %% %% ----------------------------------------------------------------------- \RequirePackage{expl3} -\GetIdInfo$Id: l3regex.dtx 3488 2012-03-03 19:49:03Z bruno $ - {L3 Experimental Regular Expressions} +\GetIdInfo$Id: l3regex.dtx 3940 2012-07-09 00:41:44Z bruno $ + {L3 Experimental regular expressions} \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} \RequirePackage{l3tl-build, l3tl-analysis, l3flag, l3str} \cs_generate_variant:Nn \tl_to_str:n { V } -\cs_new:Npn \regex_tmp:w { } -\tl_new:N \l_regex_internal_a_tl -\tl_new:N \l_regex_internal_b_tl -\int_new:N \l_regex_internal_a_int -\int_new:N \l_regex_internal_b_int -\int_new:N \l_regex_internal_c_int -\bool_new:N \l_regex_internal_bool -\seq_new:N \l_regex_internal_seq -\tl_new:N \g_regex_internal_tl -\tl_const:Nn \c_regex_no_match_regex - { - \regex_branch:n - { \regex_class:NnnnN \c_true_bool { } { 1 } { 0 } \c_true_bool } - } -\int_new:N \l_regex_balance_int -\cs_new_protected:Npn \regex_break_true:w - #1 \regex_break_point:TF #2 #3 {#2} -\cs_new_protected:Npn \regex_break_point:TF #1 #2 { #2 } -\cs_new_protected:Npn \regex_item_reverse:n #1 +\cs_new:Npn \__regex_tmp:w { } +\tl_new:N \l__regex_internal_a_tl +\tl_new:N \l__regex_internal_b_tl +\int_new:N \l__regex_internal_a_int +\int_new:N \l__regex_internal_b_int +\int_new:N \l__regex_internal_c_int +\bool_new:N \l__regex_internal_bool +\seq_new:N \l__regex_internal_seq +\tl_new:N \g__regex_internal_tl +\tl_const:Nn \c__regex_no_match_regex + { + \__regex_branch:n + { \__regex_class:NnnnN \c_true_bool { } { 1 } { 0 } \c_true_bool } + } +\int_new:N \l__regex_balance_int +\cs_new_protected:Npn \__regex_break_true:w + #1 \__regex_break_point:TF #2 #3 {#2} +\cs_new_protected:Npn \__regex_break_point:TF #1 #2 { #2 } +\cs_new_protected:Npn \__regex_item_reverse:n #1 { #1 - \regex_break_point:TF { } \regex_break_true:w + \__regex_break_point:TF { } \__regex_break_true:w } -\cs_new_protected:Npn \regex_item_caseful_equal:n #1 +\cs_new_protected:Npn \__regex_item_caseful_equal:n #1 { - \if_num:w #1 = \l_regex_current_char_int - \exp_after:wN \regex_break_true:w + \if_int_compare:w #1 = \l__regex_current_char_int + \exp_after:wN \__regex_break_true:w \fi: } -\cs_new_protected:Npn \regex_item_caseful_range:nn #1 #2 +\cs_new_protected:Npn \__regex_item_caseful_range:nn #1 #2 { - \reverse_if:N \if_num:w #1 > \l_regex_current_char_int - \reverse_if:N \if_num:w #2 < \l_regex_current_char_int - \exp_after:wN \exp_after:wN \exp_after:wN \regex_break_true:w + \reverse_if:N \if_int_compare:w #1 > \l__regex_current_char_int + \reverse_if:N \if_int_compare:w #2 < \l__regex_current_char_int + \exp_after:wN \exp_after:wN \exp_after:wN \__regex_break_true:w \fi: \fi: } -\cs_new_protected:Npn \regex_item_caseless_equal:n #1 +\cs_new_protected:Npn \__regex_item_caseless_equal:n #1 { - \if_num:w #1 = \l_regex_current_char_int - \exp_after:wN \regex_break_true:w + \if_int_compare:w #1 = \l__regex_current_char_int + \exp_after:wN \__regex_break_true:w \fi: - \if_num:w \l_regex_case_changed_char_int = \c_max_int - \regex_compute_case_changed_char: + \if_int_compare:w \l__regex_case_changed_char_int = \c_max_int + \__regex_compute_case_changed_char: \fi: - \if_num:w #1 = \l_regex_case_changed_char_int - \exp_after:wN \regex_break_true:w + \if_int_compare:w #1 = \l__regex_case_changed_char_int + \exp_after:wN \__regex_break_true:w \fi: } -\cs_new_protected:Npn \regex_item_caseless_range:nn #1 #2 +\cs_new_protected:Npn \__regex_item_caseless_range:nn #1 #2 { - \reverse_if:N \if_num:w #1 > \l_regex_current_char_int - \reverse_if:N \if_num:w #2 < \l_regex_current_char_int - \exp_after:wN \exp_after:wN \exp_after:wN \regex_break_true:w + \reverse_if:N \if_int_compare:w #1 > \l__regex_current_char_int + \reverse_if:N \if_int_compare:w #2 < \l__regex_current_char_int + \exp_after:wN \exp_after:wN \exp_after:wN \__regex_break_true:w \fi: \fi: - \if_num:w \l_regex_case_changed_char_int = \c_max_int - \regex_compute_case_changed_char: + \if_int_compare:w \l__regex_case_changed_char_int = \c_max_int + \__regex_compute_case_changed_char: \fi: - \reverse_if:N \if_num:w #1 > \l_regex_case_changed_char_int - \reverse_if:N \if_num:w #2 < \l_regex_case_changed_char_int - \exp_after:wN \exp_after:wN \exp_after:wN \regex_break_true:w + \reverse_if:N \if_int_compare:w #1 > \l__regex_case_changed_char_int + \reverse_if:N \if_int_compare:w #2 < \l__regex_case_changed_char_int + \exp_after:wN \exp_after:wN \exp_after:wN \__regex_break_true:w \fi: \fi: } -\cs_new_protected_nopar:Npn \regex_compute_case_changed_char: +\cs_new_protected_nopar:Npn \__regex_compute_case_changed_char: { - \int_set_eq:NN \l_regex_case_changed_char_int \l_regex_current_char_int - \if_num:w \l_regex_current_char_int < \c_ninety_one - \if_num:w \l_regex_current_char_int < \c_sixty_five + \int_set_eq:NN \l__regex_case_changed_char_int \l__regex_current_char_int + \if_int_compare:w \l__regex_current_char_int < \c_ninety_one + \if_int_compare:w \l__regex_current_char_int < \c_sixty_five \else: - \int_add:Nn \l_regex_case_changed_char_int { \c_thirty_two } + \int_add:Nn \l__regex_case_changed_char_int { \c_thirty_two } \fi: \else: - \if_num:w \l_regex_current_char_int < \c_one_hundred_twenty_three - \if_num:w \l_regex_current_char_int < \c_ninety_seven + \if_int_compare:w \l__regex_current_char_int < \c_one_hundred_twenty_three + \if_int_compare:w \l__regex_current_char_int < \c_ninety_seven \else: - \int_sub:Nn \l_regex_case_changed_char_int { \c_thirty_two } + \int_sub:Nn \l__regex_case_changed_char_int { \c_thirty_two } \fi: \fi: \fi: } -\cs_new_eq:NN \regex_item_equal:n ? -\cs_new_eq:NN \regex_item_range:nn ? -\cs_new_protected:Npn \regex_item_catcode_aux: +\cs_new_eq:NN \__regex_item_equal:n ? +\cs_new_eq:NN \__regex_item_range:nn ? +\cs_new_protected:Npn \__regex_item_catcode: { " - \if_case:w \l_regex_current_catcode_int + \if_case:w \l__regex_current_catcode_int 1 \or: 4 \or: 10 \or: 40 \or: 100 \or: \or: 1000 \or: 4000 \or: 10000 \or: \or: 100000 \or: 400000 \or: 1000000 \or: 4000000 \else: 1*\c_zero \fi: } -\cs_new_protected:Npn \regex_item_catcode:nT #1 +\cs_new_protected:Npn \__regex_item_catcode:nT #1 { - \if_int_odd:w \int_eval:w #1 / \regex_item_catcode_aux: \int_eval_end: + \if_int_odd:w \__int_eval:w #1 / \__regex_item_catcode: \__int_eval_end: \exp_after:wN \use:n \else: \exp_after:wN \use_none:n \fi: } -\cs_new_protected:Npn \regex_item_catcode_reverse:nT #1#2 - { \regex_item_catcode:nT {#1} { \regex_item_reverse:n {#2} } } -\cs_new_protected:Npn \regex_item_exact:nn #1#2 +\cs_new_protected:Npn \__regex_item_catcode_reverse:nT #1#2 + { \__regex_item_catcode:nT {#1} { \__regex_item_reverse:n {#2} } } +\cs_new_protected:Npn \__regex_item_exact:nn #1#2 { - \if_num:w #1 = \l_regex_current_catcode_int - \if_num:w #2 = \l_regex_current_char_int - \exp_after:wN \exp_after:wN \exp_after:wN \regex_break_true:w + \if_int_compare:w #1 = \l__regex_current_catcode_int + \if_int_compare:w #2 = \l__regex_current_char_int + \exp_after:wN \exp_after:wN \exp_after:wN \__regex_break_true:w \fi: \fi: } -\cs_new_protected:Npn \regex_item_exact_cs:c #1 +\cs_new_protected:Npn \__regex_item_exact_cs:c #1 { - \int_compare:nNnTF \l_regex_current_catcode_int = \c_zero + \int_compare:nNnTF \l__regex_current_catcode_int = \c_zero { - \str_if_eq:xxTF + \str_if_eq_x:nnTF { \exp_after:wN \exp_after:wN \exp_after:wN \cs_to_str:N - \tex_the:D \tex_toks:D \l_regex_current_pos_int + \tex_the:D \tex_toks:D \l__regex_current_pos_int } { #1 } - { \regex_break_true:w } { } + { \__regex_break_true:w } { } } { } } -\cs_new_protected:Npn \regex_item_cs:n #1 +\cs_new_protected:Npn \__regex_item_cs:n #1 { - \int_compare:nNnT \l_regex_current_catcode_int = \c_zero + \int_compare:nNnT \l__regex_current_catcode_int = \c_zero { \group_begin: - \regex_single_match: - \regex_disable_submatches: - \regex_build_for_cs:n {#1} - \bool_set_eq:NN \l_regex_saved_success_bool \g_regex_success_bool - \exp_args:Nx \regex_match:n + \__regex_single_match: + \__regex_disable_submatches: + \__regex_build_for_cs:n {#1} + \bool_set_eq:NN \l__regex_saved_success_bool \g__regex_success_bool + \exp_args:Nx \__regex_match:n { \exp_after:wN \exp_after:wN \exp_after:wN \cs_to_str:N - \tex_the:D \tex_toks:D \l_regex_current_pos_int + \tex_the:D \tex_toks:D \l__regex_current_pos_int } - \if_meaning:w \c_true_bool \g_regex_success_bool - \group_insert_after:N \regex_break_true:w + \if_meaning:w \c_true_bool \g__regex_success_bool + \group_insert_after:N \__regex_break_true:w \fi: - \bool_gset_eq:NN \g_regex_success_bool \l_regex_saved_success_bool + \bool_gset_eq:NN \g__regex_success_bool \l__regex_saved_success_bool \group_end: } } -\cs_new_protected_nopar:Npn \regex_prop_d: - { \regex_item_caseful_range:nn \c_forty_eight { 57 } } % 0--9 -\cs_new_protected_nopar:Npn \regex_prop_h: - { - \regex_item_caseful_equal:n \c_thirty_two % space - \regex_item_caseful_equal:n \c_nine % tab - } -\cs_new_protected_nopar:Npn \regex_prop_s: - { - \regex_item_caseful_equal:n \c_thirty_two % space - \regex_item_caseful_equal:n \c_nine % tab - \regex_item_caseful_equal:n \c_ten % lf - \regex_item_caseful_equal:n \c_twelve % ff - \regex_item_caseful_equal:n \c_thirteen % cr - } -\cs_new_protected_nopar:Npn \regex_prop_v: - { \regex_item_caseful_range:nn \c_ten \c_thirteen } % lf, vtab, ff, cr -\cs_new_protected_nopar:Npn \regex_prop_w: - { - \regex_item_caseful_range:nn \c_ninety_seven { 122 } % a--z - \regex_item_caseful_range:nn \c_sixty_five { 90 } % A--Z - \regex_item_caseful_range:nn \c_forty_eight { 57 } % 0--9 - \regex_item_caseful_equal:n { 95 } % _ - } -\cs_new_protected_nopar:Npn \regex_prop_N: - { \regex_item_reverse:n { \regex_item_caseful_equal:n \c_ten } } -\cs_new_protected_nopar:Npn \regex_posix_alnum: - { \regex_posix_alpha: \regex_posix_digit: } -\cs_new_protected_nopar:Npn \regex_posix_alpha: - { \regex_posix_lower: \regex_posix_upper: } -\cs_new_protected_nopar:Npn \regex_posix_ascii: - { \regex_item_caseful_range:nn \c_zero \c_one_hundred_twenty_seven } -\cs_new_eq:NN \regex_posix_blank: \regex_prop_h: -\cs_new_protected_nopar:Npn \regex_posix_cntrl: - { - \regex_item_caseful_range:nn \c_zero { 31 } - \regex_item_caseful_equal:n \c_one_hundred_twenty_seven - } -\cs_new_eq:NN \regex_posix_digit: \regex_prop_d: -\cs_new_protected_nopar:Npn \regex_posix_graph: - { \regex_item_caseful_range:nn { 33 } { 126 } } -\cs_new_protected_nopar:Npn \regex_posix_lower: - { \regex_item_caseful_range:nn \c_ninety_seven { 122 } } -\cs_new_protected_nopar:Npn \regex_posix_print: - { \regex_item_caseful_range:nn \c_thirty_two { 126 } } -\cs_new_protected_nopar:Npn \regex_posix_punct: - { - \regex_item_caseful_range:nn { 33 } { 47 } - \regex_item_caseful_range:nn { 58 } { 64 } - \regex_item_caseful_range:nn { 91 } { 96 } - \regex_item_caseful_range:nn { 123 } { 126 } - } -\cs_new_protected_nopar:Npn \regex_posix_space: - { - \regex_item_caseful_equal:n \c_thirty_two - \regex_item_caseful_range:nn \c_nine \c_thirteen - } -\cs_new_protected_nopar:Npn \regex_posix_upper: - { \regex_item_caseful_range:nn \c_sixty_five { 90 } } -\cs_new_eq:NN \regex_posix_word: \regex_prop_w: -\cs_new_protected_nopar:Npn \regex_posix_xdigit: - { - \regex_posix_digit: - \regex_item_caseful_range:nn \c_sixty_five { 70 } - \regex_item_caseful_range:nn \c_ninety_seven { 102 } - } -\cs_new_protected:Npn \regex_escape_use:nnnn #1#2#3#4 - { - \tl_set_build:Nw \l_regex_internal_a_tl - \cs_set_nopar:Npn \regex_escape_unescaped:N ##1 { #1 } - \cs_set_nopar:Npn \regex_escape_escaped:N ##1 { #2 } - \cs_set_nopar:Npn \regex_escape_raw:N ##1 { #3 } +\cs_new_protected_nopar:Npn \__regex_prop_d: + { \__regex_item_caseful_range:nn \c_forty_eight { 57 } } % 0--9 +\cs_new_protected_nopar:Npn \__regex_prop_h: + { + \__regex_item_caseful_equal:n \c_thirty_two % space + \__regex_item_caseful_equal:n \c_nine % tab + } +\cs_new_protected_nopar:Npn \__regex_prop_s: + { + \__regex_item_caseful_equal:n \c_thirty_two % space + \__regex_item_caseful_equal:n \c_nine % tab + \__regex_item_caseful_equal:n \c_ten % lf + \__regex_item_caseful_equal:n \c_twelve % ff + \__regex_item_caseful_equal:n \c_thirteen % cr + } +\cs_new_protected_nopar:Npn \__regex_prop_v: + { \__regex_item_caseful_range:nn \c_ten \c_thirteen } % lf, vtab, ff, cr +\cs_new_protected_nopar:Npn \__regex_prop_w: + { + \__regex_item_caseful_range:nn \c_ninety_seven { 122 } % a--z + \__regex_item_caseful_range:nn \c_sixty_five { 90 } % A--Z + \__regex_item_caseful_range:nn \c_forty_eight { 57 } % 0--9 + \__regex_item_caseful_equal:n { 95 } % _ + } +\cs_new_protected_nopar:Npn \__regex_prop_N: + { \__regex_item_reverse:n { \__regex_item_caseful_equal:n \c_ten } } +\cs_new_protected_nopar:Npn \__regex_posix_alnum: + { \__regex_posix_alpha: \__regex_posix_digit: } +\cs_new_protected_nopar:Npn \__regex_posix_alpha: + { \__regex_posix_lower: \__regex_posix_upper: } +\cs_new_protected_nopar:Npn \__regex_posix_ascii: + { \__regex_item_caseful_range:nn \c_zero \c_one_hundred_twenty_seven } +\cs_new_eq:NN \__regex_posix_blank: \__regex_prop_h: +\cs_new_protected_nopar:Npn \__regex_posix_cntrl: + { + \__regex_item_caseful_range:nn \c_zero { 31 } + \__regex_item_caseful_equal:n \c_one_hundred_twenty_seven + } +\cs_new_eq:NN \__regex_posix_digit: \__regex_prop_d: +\cs_new_protected_nopar:Npn \__regex_posix_graph: + { \__regex_item_caseful_range:nn { 33 } { 126 } } +\cs_new_protected_nopar:Npn \__regex_posix_lower: + { \__regex_item_caseful_range:nn \c_ninety_seven { 122 } } +\cs_new_protected_nopar:Npn \__regex_posix_print: + { \__regex_item_caseful_range:nn \c_thirty_two { 126 } } +\cs_new_protected_nopar:Npn \__regex_posix_punct: + { + \__regex_item_caseful_range:nn { 33 } { 47 } + \__regex_item_caseful_range:nn { 58 } { 64 } + \__regex_item_caseful_range:nn { 91 } { 96 } + \__regex_item_caseful_range:nn { 123 } { 126 } + } +\cs_new_protected_nopar:Npn \__regex_posix_space: + { + \__regex_item_caseful_equal:n \c_thirty_two + \__regex_item_caseful_range:nn \c_nine \c_thirteen + } +\cs_new_protected_nopar:Npn \__regex_posix_upper: + { \__regex_item_caseful_range:nn \c_sixty_five { 90 } } +\cs_new_eq:NN \__regex_posix_word: \__regex_prop_w: +\cs_new_protected_nopar:Npn \__regex_posix_xdigit: + { + \__regex_posix_digit: + \__regex_item_caseful_range:nn \c_sixty_five { 70 } + \__regex_item_caseful_range:nn \c_ninety_seven { 102 } + } +\cs_new_protected:Npn \__regex_escape_use:nnnn #1#2#3#4 + { + \__tl_build:Nw \l__regex_internal_a_tl + \cs_set_nopar:Npn \__regex_escape_unescaped:N ##1 { #1 } + \cs_set_nopar:Npn \__regex_escape_escaped:N ##1 { #2 } + \cs_set_nopar:Npn \__regex_escape_raw:N ##1 { #3 } \int_set:Nn \tex_escapechar:D { 92 } - \str_gset_other:Nn \g_regex_internal_tl { #4 } - \tl_set:Nx \l_regex_internal_b_tl + \__str_gset_other:Nn \g__regex_internal_tl { #4 } + \tl_set:Nx \l__regex_internal_b_tl { - \exp_after:wN \regex_escape_loop:N \g_regex_internal_tl - { break } \prg_break_point:n { } + \exp_after:wN \__regex_escape_loop:N \g__regex_internal_tl + { break } \__prg_break_point: } - \tl_build_one:o \l_regex_internal_b_tl - \tl_build_end: - \l_regex_internal_a_tl + \__tl_build_one:o \l__regex_internal_b_tl + \__tl_build_end: + \l__regex_internal_a_tl } -\cs_new:Npn \regex_escape_loop:N #1 +\cs_new:Npn \__regex_escape_loop:N #1 { - \cs_if_exist_use:cF { regex_escape_\token_to_str:N #1:w } - { \regex_escape_unescaped:N #1 } - \regex_escape_loop:N + \cs_if_exist_use:cF { __regex_escape_\token_to_str:N #1:w } + { \__regex_escape_unescaped:N #1 } + \__regex_escape_loop:N } -\cs_new_nopar:cpn { regex_escape_ \c_backslash_str :w } - \regex_escape_loop:N #1 +\cs_new_nopar:cpn { __regex_escape_ \c_backslash_str :w } + \__regex_escape_loop:N #1 { - \cs_if_exist_use:cF { regex_escape_/\token_to_str:N #1:w } - { \regex_escape_escaped:N #1 } - \regex_escape_loop:N + \cs_if_exist_use:cF { __regex_escape_/\token_to_str:N #1:w } + { \__regex_escape_escaped:N #1 } + \__regex_escape_loop:N } -\cs_new_eq:NN \regex_escape_unescaped:N ? -\cs_new_eq:NN \regex_escape_escaped:N ? -\cs_new_eq:NN \regex_escape_raw:N ? -\cs_new_eq:NN \regex_escape_break:w \prg_map_break: -\cs_new_nopar:cpn { regex_escape_/break:w } +\cs_new_eq:NN \__regex_escape_unescaped:N ? +\cs_new_eq:NN \__regex_escape_escaped:N ? +\cs_new_eq:NN \__regex_escape_raw:N ? +\cs_new_eq:NN \__regex_escape_break:w \__prg_break: +\cs_new_nopar:cpn { __regex_escape_/break:w } { \if_false: { \fi: } - \msg_kernel_error:nn { regex } { trailing-backslash } + \__msg_kernel_error:nn { regex } { trailing-backslash } \exp_after:wN \use_none:n \exp_after:wN { \if_false: } \fi: } -\cs_new_nopar:cpn { regex_escape_~:w } { } -\cs_new_nopar:cpx { regex_escape_/a:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^G } -\cs_new_nopar:cpx { regex_escape_/t:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^I } -\cs_new_nopar:cpx { regex_escape_/n:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^J } -\cs_new_nopar:cpx { regex_escape_/f:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^L } -\cs_new_nopar:cpx { regex_escape_/r:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^M } -\cs_new_nopar:cpx { regex_escape_/e:w } - { \exp_not:N \regex_escape_raw:N \iow_char:N \^^[ } -\cs_new:cpn { regex_escape_/x:w } \regex_escape_loop:N - { - \exp_after:wN \regex_escape_x_end:w - \int_value:w "0 \regex_escape_x_test:N - } -\cs_new:Npn \regex_escape_x_end:w #1 ; +\cs_new_nopar:cpn { __regex_escape_~:w } { } +\cs_new_nopar:cpx { __regex_escape_/a:w } + { \exp_not:N \__regex_escape_raw:N \iow_char:N \^^G } +\cs_new_nopar:cpx { __regex_escape_/t:w } + { \exp_not:N \__regex_escape_raw:N \iow_char:N \^^I } +\cs_new_nopar:cpx { __regex_escape_/n:w } + { \exp_not:N \__regex_escape_raw:N \iow_char:N \^^J } +\cs_new_nopar:cpx { __regex_escape_/f:w } + { \exp_not:N \__regex_escape_raw:N \iow_char:N \^^L } +\cs_new_nopar:cpx { __regex_escape_/r:w } + { \exp_not:N \__regex_escape_raw:N \iow_char:N \^^M } +\cs_new_nopar:cpx { __regex_escape_/e:w } + { \exp_not:N \__regex_escape_raw:N \iow_char:N \^^[ } +\cs_new:cpn { __regex_escape_/x:w } \__regex_escape_loop:N + { + \exp_after:wN \__regex_escape_x_end:w + \__int_value:w "0 \__regex_escape_x_test:N + } +\cs_new:Npn \__regex_escape_x_end:w #1 ; { \int_compare:nNnTF {#1} < \c_two_hundred_fifty_six { - \exp_last_unbraced:Nf \regex_escape_raw:N - { \str_output_byte:n {#1} } + \exp_last_unbraced:Nf \__regex_escape_raw:N + { \__str_output_byte:n {#1} } } - { \regex_escape_x_large:n {#1} } + { \__regex_escape_x_large:n {#1} } } \group_begin: \char_set_catcode_other:n { 0 } - \cs_new:Npn \regex_escape_x_large:n #1 + \cs_new:Npn \__regex_escape_x_large:n #1 { \if_false: { \fi: } - \tl_build_one:o \l_regex_internal_b_tl + \__tl_build_one:o \l__regex_internal_b_tl \int_compare:nNnTF {#1} > \c_max_char_int { - \msg_kernel_error:nnx { regex } { x-overflow } {#1} - \tl_set:Nx \l_regex_internal_b_tl - { \if_false: } \fi: \regex_escape_loop:N + \__msg_kernel_error:nnx { regex } { x-overflow } {#1} + \tl_set:Nx \l__regex_internal_b_tl + { \if_false: } \fi: \__regex_escape_loop:N } { \char_set_lccode:nn { \c_zero } {#1} \tl_to_lowercase:n { - \tl_set:Nx \l_regex_internal_b_tl + \tl_set:Nx \l__regex_internal_b_tl { \if_false: } \fi: - \regex_escape_raw:N ^^@ - \regex_escape_loop:N + \__regex_escape_raw:N ^^@ + \__regex_escape_loop:N } } } \group_end: -\cs_new:Npn \regex_escape_x_test:N #1 +\cs_new:Npn \__regex_escape_x_test:N #1 { - \str_if_eq:xxTF {#1} { break } { ; } + \str_if_eq_x:nnTF {#1} { break } { ; } { \if_charcode:w \c_space_token #1 - \exp_after:wN \regex_escape_x_test:N + \exp_after:wN \__regex_escape_x_test:N \else: - \exp_after:wN \regex_escape_x_test_ii:N + \exp_after:wN \__regex_escape_x_test_ii:N \exp_after:wN #1 \fi: } } -\cs_new:Npn \regex_escape_x_test_ii:N #1 +\cs_new:Npn \__regex_escape_x_test_ii:N #1 { \if_charcode:w \c_lbrace_str #1 - \exp_after:wN \regex_escape_x_loop:N + \exp_after:wN \__regex_escape_x_loop:N \else: - \str_aux_hexadecimal_use:NTF #1 - { \exp_after:wN \regex_escape_x_ii:N } - { ; \exp_after:wN \regex_escape_loop:N \exp_after:wN #1 } + \__str_hexadecimal_use:NTF #1 + { \exp_after:wN \__regex_escape_x_ii:N } + { ; \exp_after:wN \__regex_escape_loop:N \exp_after:wN #1 } \fi: } -\cs_new:Npn \regex_escape_x_ii:N #1 +\cs_new:Npn \__regex_escape_x_ii:N #1 { - \str_if_eq:xxTF {#1} { break } { ; } + \str_if_eq_x:nnTF {#1} { break } { ; } { - \str_aux_hexadecimal_use:NTF #1 - { ; \regex_escape_loop:N } - { ; \regex_escape_loop:N #1 } + \__str_hexadecimal_use:NTF #1 + { ; \__regex_escape_loop:N } + { ; \__regex_escape_loop:N #1 } } } -\cs_new:Npn \regex_escape_x_loop:N #1 +\cs_new:Npn \__regex_escape_x_loop:N #1 { - \str_aux_hexadecimal_use:NTF #1 - { \regex_escape_x_loop:N } + \__str_hexadecimal_use:NTF #1 + { \__regex_escape_x_loop:N } { \token_if_eq_charcode:NNTF \c_space_token #1 - { \regex_escape_x_loop:N } + { \__regex_escape_x_loop:N } { ; \exp_after:wN \token_if_eq_charcode:NNTF \c_rbrace_str #1 - { \regex_escape_loop:N } + { \__regex_escape_loop:N } { \if_false: { \fi: } - \tl_build_one:o \l_regex_internal_b_tl - \msg_kernel_error:nn { regex } { x-missing-rbrace } {#1} - \tl_set:Nx \l_regex_internal_b_tl - { \if_false: } \fi: \regex_escape_loop:N #1 + \__tl_build_one:o \l__regex_internal_b_tl + \__msg_kernel_error:nn { regex } { x-missing-rbrace } {#1} + \tl_set:Nx \l__regex_internal_b_tl + { \if_false: } \fi: \__regex_escape_loop:N #1 } } } } -\prg_new_conditional:Npnn \regex_char_if_special:N #1 { TF } +\prg_new_conditional:Npnn \__regex_char_if_special:N #1 { TF } { - \if_num:w `#1 < \c_ninety_one - \if_num:w `#1 < \c_fifty_eight - \if_num:w `#1 < \c_forty_eight - \if_num:w `#1 < \c_thirty_two + \if_int_compare:w `#1 < \c_ninety_one + \if_int_compare:w `#1 < \c_fifty_eight + \if_int_compare:w `#1 < \c_forty_eight + \if_int_compare:w `#1 < \c_thirty_two \prg_return_false: \else: \prg_return_true: \fi: \else: \prg_return_false: \fi: \else: - \if_num:w `#1 < \c_sixty_five + \if_int_compare:w `#1 < \c_sixty_five \prg_return_true: \else: \prg_return_false: \fi: \fi: \else: - \if_num:w `#1 < \c_one_hundred_twenty_three - \if_num:w `#1 < \c_ninety_seven + \if_int_compare:w `#1 < \c_one_hundred_twenty_three + \if_int_compare:w `#1 < \c_ninety_seven \prg_return_true: \else: \prg_return_false: \fi: \else: - \if_num:w `#1 < \c_one_hundred_twenty_seven + \if_int_compare:w `#1 < \c_one_hundred_twenty_seven \prg_return_true: \else: \prg_return_false: \fi: \fi: \fi: } -\prg_new_conditional:Npnn \regex_char_if_alphanumeric:N #1 { TF } +\prg_new_conditional:Npnn \__regex_char_if_alphanumeric:N #1 { TF } { - \if_num:w `#1 < \c_ninety_one - \if_num:w `#1 < \c_fifty_eight - \if_num:w `#1 < \c_forty_eight + \if_int_compare:w `#1 < \c_ninety_one + \if_int_compare:w `#1 < \c_fifty_eight + \if_int_compare:w `#1 < \c_forty_eight \prg_return_false: \else: \prg_return_true: \fi: \else: - \if_num:w `#1 < \c_sixty_five + \if_int_compare:w `#1 < \c_sixty_five \prg_return_false: \else: \prg_return_true: \fi: \fi: \else: - \if_num:w `#1 < \c_one_hundred_twenty_three - \if_num:w `#1 < \c_ninety_seven + \if_int_compare:w `#1 < \c_one_hundred_twenty_three + \if_int_compare:w `#1 < \c_ninety_seven \prg_return_false: \else: \prg_return_true: \fi: \else: \prg_return_false: \fi: \fi: } -\int_new:N \l_regex_group_level_int -\int_new:N \l_regex_mode_int -\int_new:N \l_regex_catcodes_int -\int_new:N \l_regex_default_catcodes_int -\bool_new:N \l_regex_catcodes_bool -\int_const:Nn \c_regex_catcode_C_int { "1 } -\int_const:Nn \c_regex_catcode_B_int { "4 } -\int_const:Nn \c_regex_catcode_E_int { "10 } -\int_const:Nn \c_regex_catcode_M_int { "40 } -\int_const:Nn \c_regex_catcode_T_int { "100 } -\int_const:Nn \c_regex_catcode_P_int { "1000 } -\int_const:Nn \c_regex_catcode_U_int { "4000 } -\int_const:Nn \c_regex_catcode_D_int { "10000 } -\int_const:Nn \c_regex_catcode_S_int { "100000 } -\int_const:Nn \c_regex_catcode_L_int { "400000 } -\int_const:Nn \c_regex_catcode_O_int { "1000000 } -\int_const:Nn \c_regex_catcode_A_int { "4000000 } -\int_const:Nn \c_regex_all_catcodes_int { "5515155 } -\cs_new_eq:NN \l_regex_internal_regex \c_regex_no_match_regex -\seq_new:N \l_regex_show_prefix_seq -\int_new:N \l_regex_show_lines_int -\cs_new_protected:Npn \regex_get_digits:NTFw #1#2#3#4#5 - { - \regex_if_raw_digit:NNTF #4 #5 - { #1 = #5 \regex_get_digits_loop:nw {#2} } +\int_new:N \l__regex_group_level_int +\int_new:N \l__regex_mode_int +\int_new:N \l__regex_catcodes_int +\int_new:N \l__regex_default_catcodes_int +\bool_new:N \l__regex_catcodes_bool +\int_const:Nn \c__regex_catcode_C_int { "1 } +\int_const:Nn \c__regex_catcode_B_int { "4 } +\int_const:Nn \c__regex_catcode_E_int { "10 } +\int_const:Nn \c__regex_catcode_M_int { "40 } +\int_const:Nn \c__regex_catcode_T_int { "100 } +\int_const:Nn \c__regex_catcode_P_int { "1000 } +\int_const:Nn \c__regex_catcode_U_int { "4000 } +\int_const:Nn \c__regex_catcode_D_int { "10000 } +\int_const:Nn \c__regex_catcode_S_int { "100000 } +\int_const:Nn \c__regex_catcode_L_int { "400000 } +\int_const:Nn \c__regex_catcode_O_int { "1000000 } +\int_const:Nn \c__regex_catcode_A_int { "4000000 } +\int_const:Nn \c__regex_all_catcodes_int { "5515155 } +\cs_new_eq:NN \l__regex_internal_regex \c__regex_no_match_regex +\seq_new:N \l__regex_show_prefix_seq +\int_new:N \l__regex_show_lines_int +\cs_new_protected:Npn \__regex_get_digits:NTFw #1#2#3#4#5 + { + \__regex_if_raw_digit:NNTF #4 #5 + { #1 = #5 \__regex_get_digits_loop:nw {#2} } { #3 #4 #5 } } -\cs_new:Npn \regex_get_digits_loop:nw #1#2#3 +\cs_new:Npn \__regex_get_digits_loop:nw #1#2#3 { - \regex_if_raw_digit:NNTF #2 #3 - { #3 \regex_get_digits_loop:nw {#1} } + \__regex_if_raw_digit:NNTF #2 #3 + { #3 \__regex_get_digits_loop:nw {#1} } { \scan_stop: #1 #2 #3 } } -\prg_new_conditional:Npnn \regex_if_raw_digit:NN #1#2 { TF } +\prg_new_conditional:Npnn \__regex_if_raw_digit:NN #1#2 { TF } { - \if_meaning:w \regex_compile_raw:N #1 + \if_meaning:w \__regex_compile_raw:N #1 \if_int_compare:w \c_one < 1 #2 \exp_stop_f: \prg_return_true: \else: @@ -495,305 +495,305 @@ \prg_return_false: \fi: } -\cs_new_nopar:Npn \regex_if_in_class:TF +\cs_new_nopar:Npn \__regex_if_in_class:TF { - \if_int_odd:w \l_regex_mode_int + \if_int_odd:w \l__regex_mode_int \exp_after:wN \use_i:nn \else: \exp_after:wN \use_ii:nn \fi: } -\cs_new_nopar:Npn \regex_if_in_cs:TF +\cs_new_nopar:Npn \__regex_if_in_cs:TF { - \if_int_odd:w \l_regex_mode_int + \if_int_odd:w \l__regex_mode_int \exp_after:wN \use_ii:nn \else: - \if_int_compare:w \l_regex_mode_int < \c_zero + \if_int_compare:w \l__regex_mode_int < \c_zero \exp_after:wN \exp_after:wN \exp_after:wN \use_i:nn \else: \exp_after:wN \exp_after:wN \exp_after:wN \use_ii:nn \fi: \fi: } -\cs_new_nopar:Npn \regex_if_in_class_or_catcode:TF +\cs_new_nopar:Npn \__regex_if_in_class_or_catcode:TF { - \if_int_odd:w \l_regex_mode_int + \if_int_odd:w \l__regex_mode_int \exp_after:wN \use_i:nn \else: - \if_int_compare:w \l_regex_mode_int > \c_zero + \if_int_compare:w \l__regex_mode_int > \c_zero \exp_after:wN \exp_after:wN \exp_after:wN \use_i:nn \else: \exp_after:wN \exp_after:wN \exp_after:wN \use_ii:nn \fi: \fi: } -\cs_new_nopar:Npn \regex_if_within_catcode:TF +\cs_new_nopar:Npn \__regex_if_within_catcode:TF { - \if_int_compare:w \l_regex_mode_int > \c_zero + \if_int_compare:w \l__regex_mode_int > \c_zero \exp_after:wN \use_i:nn \else: \exp_after:wN \use_ii:nn \fi: } -\cs_new_protected:Npn \regex_chk_c_allowed:T +\cs_new_protected:Npn \__regex_chk_c_allowed:T { - \if_num:w \l_regex_mode_int = \c_zero + \if_int_compare:w \l__regex_mode_int = \c_zero \exp_after:wN \use:n \else: - \if_num:w \l_regex_mode_int = \c_three + \if_int_compare:w \l__regex_mode_int = \c_three \exp_after:wN \exp_after:wN \exp_after:wN \use:n \else: - \msg_kernel_error:nn { regex } { c-bad-mode } + \__msg_kernel_error:nn { regex } { c-bad-mode } \exp_after:wN \exp_after:wN \exp_after:wN \use_none:n \fi: \fi: } -\cs_new_protected:Npn \regex_mode_quit_c: +\cs_new_protected:Npn \__regex_mode_quit_c: { - \if_num:w \l_regex_mode_int = \c_two - \l_regex_mode_int = \c_zero + \if_int_compare:w \l__regex_mode_int = \c_two + \l__regex_mode_int = \c_zero \else: - \if_num:w \l_regex_mode_int = \c_six - \l_regex_mode_int = \c_three + \if_int_compare:w \l__regex_mode_int = \c_six + \l__regex_mode_int = \c_three \fi: \fi: } -\cs_new_protected_nopar:Npn \regex_compile:w +\cs_new_protected_nopar:Npn \__regex_compile:w { - \tl_set_build_x:Nw \l_regex_internal_regex - \int_zero:N \l_regex_group_level_int - \int_set_eq:NN \l_regex_default_catcodes_int \c_regex_all_catcodes_int - \int_set_eq:NN \l_regex_catcodes_int \l_regex_default_catcodes_int - \cs_set_nopar:Npn \regex_item_equal:n { \regex_item_caseful_equal:n } - \cs_set_nopar:Npn \regex_item_range:nn { \regex_item_caseful_range:nn } - \tl_build_one:n { \regex_branch:n { \if_false: } \fi: } + \__tl_build_x:Nw \l__regex_internal_regex + \int_zero:N \l__regex_group_level_int + \int_set_eq:NN \l__regex_default_catcodes_int \c__regex_all_catcodes_int + \int_set_eq:NN \l__regex_catcodes_int \l__regex_default_catcodes_int + \cs_set_nopar:Npn \__regex_item_equal:n { \__regex_item_caseful_equal:n } + \cs_set_nopar:Npn \__regex_item_range:nn { \__regex_item_caseful_range:nn } + \__tl_build_one:n { \__regex_branch:n { \if_false: } \fi: } } -\cs_new_protected_nopar:Npn \regex_compile_end: +\cs_new_protected_nopar:Npn \__regex_compile_end: { - \regex_if_in_class:TF + \__regex_if_in_class:TF { - \msg_kernel_error:nn { regex } { missing-rbrack } - \use:c { regex_compile_]: } + \__msg_kernel_error:nn { regex } { missing-rbrack } + \use:c { __regex_compile_]: } \prg_do_nothing: \prg_do_nothing: } { } - \if_num:w \l_regex_group_level_int > \c_zero - \msg_kernel_error:nnx { regex } { missing-rparen } - { \int_use:N \l_regex_group_level_int } + \if_int_compare:w \l__regex_group_level_int > \c_zero + \__msg_kernel_error:nnx { regex } { missing-rparen } + { \int_use:N \l__regex_group_level_int } \prg_replicate:nn - { \l_regex_group_level_int } + { \l__regex_group_level_int } { - \tl_build_one:n + \__tl_build_one:n { \if_false: { \fi: } \if_false: { \fi: } { 1 } { 0 } \c_true_bool } - \tl_build_end: - \tl_build_one:o \l_regex_internal_regex + \__tl_build_end: + \__tl_build_one:o \l__regex_internal_regex } \fi: - \tl_build_one:n { \if_false: { \fi: } } - \tl_build_end: + \__tl_build_one:n { \if_false: { \fi: } } + \__tl_build_end: } -\cs_new_protected:Npn \regex_compile:n #1 +\cs_new_protected:Npn \__regex_compile:n #1 { - \regex_compile:w + \__regex_compile:w \int_set:Nn \tex_escapechar:D { 92 } - \int_set_eq:NN \l_regex_mode_int \c_zero - \regex_escape_use:nnnn + \int_set_eq:NN \l__regex_mode_int \c_zero + \__regex_escape_use:nnnn { - \regex_char_if_special:NTF ##1 - \regex_compile_special:N \regex_compile_raw:N ##1 + \__regex_char_if_special:NTF ##1 + \__regex_compile_special:N \__regex_compile_raw:N ##1 } { - \regex_char_if_alphanumeric:NTF ##1 - \regex_compile_escaped:N \regex_compile_raw:N ##1 + \__regex_char_if_alphanumeric:NTF ##1 + \__regex_compile_escaped:N \__regex_compile_raw:N ##1 } - { \regex_compile_raw:N ##1 } + { \__regex_compile_raw:N ##1 } { #1 } \prg_do_nothing: \prg_do_nothing: \prg_do_nothing: \prg_do_nothing: - \int_compare:nNnT \l_regex_mode_int < \c_zero + \int_compare:nNnT \l__regex_mode_int < \c_zero { - \msg_kernel_error:nn { regex } { c-missing-rbrace } - \regex_compile_end: - \regex_compile_one:x - { \regex_item_cs:n { \exp_not:o \l_regex_internal_regex } } + \__msg_kernel_error:nn { regex } { c-missing-rbrace } + \__regex_compile_end: + \__regex_compile_one:x + { \__regex_item_cs:n { \exp_not:o \l__regex_internal_regex } } \prg_do_nothing: \prg_do_nothing: \prg_do_nothing: \prg_do_nothing: } - \regex_compile_end: + \__regex_compile_end: } -\cs_new_protected:Npn \regex_compile_special:N #1 +\cs_new_protected:Npn \__regex_compile_special:N #1 { - \cs_if_exist_use:cF { regex_compile_#1: } - { \regex_compile_raw:N #1 } + \cs_if_exist_use:cF { __regex_compile_#1: } + { \__regex_compile_raw:N #1 } } -\cs_new_protected:Npn \regex_compile_escaped:N #1 +\cs_new_protected:Npn \__regex_compile_escaped:N #1 { - \cs_if_exist_use:cF { regex_compile_/#1: } - { \regex_compile_raw:N #1 } + \cs_if_exist_use:cF { __regex_compile_/#1: } + { \__regex_compile_raw:N #1 } } -\cs_new_protected:Npn \regex_compile_one:x #1 +\cs_new_protected:Npn \__regex_compile_one:x #1 { - \regex_mode_quit_c: - \regex_if_in_class:TF { } + \__regex_mode_quit_c: + \__regex_if_in_class:TF { } { - \tl_build_one:n - { \regex_class:NnnnN \c_true_bool { \if_false: } \fi: } + \__tl_build_one:n + { \__regex_class:NnnnN \c_true_bool { \if_false: } \fi: } } - \tl_build_one:x + \__tl_build_one:x { - \if_num:w \l_regex_catcodes_int < \c_regex_all_catcodes_int - \regex_item_catcode:nT { \int_use:N \l_regex_catcodes_int } + \if_int_compare:w \l__regex_catcodes_int < \c__regex_all_catcodes_int + \__regex_item_catcode:nT { \int_use:N \l__regex_catcodes_int } { \exp_not:N \exp_not:n {#1} } \else: \exp_not:N \exp_not:n {#1} \fi: } - \int_set_eq:NN \l_regex_catcodes_int \l_regex_default_catcodes_int - \regex_if_in_class:TF { } { \regex_compile_quantifier:w } + \int_set_eq:NN \l__regex_catcodes_int \l__regex_default_catcodes_int + \__regex_if_in_class:TF { } { \__regex_compile_quantifier:w } } -\cs_new_protected:Npn \regex_compile_abort_tokens:n #1 +\cs_new_protected:Npn \__regex_compile_abort_tokens:n #1 { \use:x { \exp_args:No \tl_map_function:nN { \tl_to_str:n {#1} } - \regex_compile_raw:N + \__regex_compile_raw:N } } -\cs_generate_variant:Nn \regex_compile_abort_tokens:n { x } -\cs_new_protected:Npn \regex_compile_quantifier:w #1#2 +\cs_generate_variant:Nn \__regex_compile_abort_tokens:n { x } +\cs_new_protected:Npn \__regex_compile_quantifier:w #1#2 { - \token_if_eq_meaning:NNTF #1 \regex_compile_special:N + \token_if_eq_meaning:NNTF #1 \__regex_compile_special:N { - \cs_if_exist_use:cF { regex_compile_quantifier_#2:w } - { \regex_compile_quantifier_none: #1 #2 } + \cs_if_exist_use:cF { __regex_compile_quantifier_#2:w } + { \__regex_compile_quantifier_none: #1 #2 } } - { \regex_compile_quantifier_none: #1 #2 } + { \__regex_compile_quantifier_none: #1 #2 } } -\cs_new_protected:Npn \regex_compile_quantifier_none: - { \tl_build_one:n { \if_false: { \fi: } { 1 } { 0 } \c_false_bool } } -\cs_new_protected:Npn \regex_compile_quantifier_abort:xNN #1#2#3 +\cs_new_protected:Npn \__regex_compile_quantifier_none: + { \__tl_build_one:n { \if_false: { \fi: } { 1 } { 0 } \c_false_bool } } +\cs_new_protected:Npn \__regex_compile_quantifier_abort:xNN #1#2#3 { - \regex_compile_quantifier_none: - \msg_kernel_warning:nnxx { regex } { invalid-quantifier } {#1} {#3} - \regex_compile_abort_tokens:x {#1} + \__regex_compile_quantifier_none: + \__msg_kernel_warning:nnxx { regex } { invalid-quantifier } {#1} {#3} + \__regex_compile_abort_tokens:x {#1} #2 #3 } -\cs_new_protected:Npn \regex_compile_quantifier_lazyness:nnNN #1#2#3#4 +\cs_new_protected:Npn \__regex_compile_quantifier_lazyness:nnNN #1#2#3#4 { - \str_if_eq:nnTF { #3 #4 } { \regex_compile_special:N ? } - { \tl_build_one:n { \if_false: { \fi: } { #1 } { #2 } \c_true_bool } } + \str_if_eq:nnTF { #3 #4 } { \__regex_compile_special:N ? } + { \__tl_build_one:n { \if_false: { \fi: } { #1 } { #2 } \c_true_bool } } { - \tl_build_one:n { \if_false: { \fi: } { #1 } { #2 } \c_false_bool } + \__tl_build_one:n { \if_false: { \fi: } { #1 } { #2 } \c_false_bool } #3 #4 } } -\cs_new_protected_nopar:cpn { regex_compile_quantifier_?:w } - { \regex_compile_quantifier_lazyness:nnNN { 0 } { 1 } } -\cs_new_protected_nopar:cpn { regex_compile_quantifier_*:w } - { \regex_compile_quantifier_lazyness:nnNN { 0 } { -1 } } -\cs_new_protected_nopar:cpn { regex_compile_quantifier_+:w } - { \regex_compile_quantifier_lazyness:nnNN { 1 } { -1 } } -\cs_new_protected:cpn { regex_compile_quantifier_ \c_lbrace_str :w } +\cs_new_protected_nopar:cpn { __regex_compile_quantifier_?:w } + { \__regex_compile_quantifier_lazyness:nnNN { 0 } { 1 } } +\cs_new_protected_nopar:cpn { __regex_compile_quantifier_*:w } + { \__regex_compile_quantifier_lazyness:nnNN { 0 } { -1 } } +\cs_new_protected_nopar:cpn { __regex_compile_quantifier_+:w } + { \__regex_compile_quantifier_lazyness:nnNN { 1 } { -1 } } +\cs_new_protected:cpn { __regex_compile_quantifier_ \c_lbrace_str :w } { - \regex_get_digits:NTFw \l_regex_internal_a_int - { \regex_compile_quantifier_braced_i:w } - { \regex_compile_quantifier_abort:xNN { \c_lbrace_str } } + \__regex_get_digits:NTFw \l__regex_internal_a_int + { \__regex_compile_quantifier_braced_i:w } + { \__regex_compile_quantifier_abort:xNN { \c_lbrace_str } } } -\cs_new_protected:Npn \regex_compile_quantifier_braced_i:w #1#2 +\cs_new_protected:Npn \__regex_compile_quantifier_braced_i:w #1#2 { - \prg_case_str:xxn { #1 #2 } + \str_case_x:nnn { #1 #2 } { - { \regex_compile_special:N \c_rbrace_str } + { \__regex_compile_special:N \c_rbrace_str } { - \exp_args:No \regex_compile_quantifier_lazyness:nnNN - { \int_use:N \l_regex_internal_a_int } { 0 } + \exp_args:No \__regex_compile_quantifier_lazyness:nnNN + { \int_use:N \l__regex_internal_a_int } { 0 } } - { \regex_compile_special:N , } + { \__regex_compile_special:N , } { - \regex_get_digits:NTFw \l_regex_internal_b_int - { \regex_compile_quantifier_braced_iii:w } - { \regex_compile_quantifier_braced_ii:w } + \__regex_get_digits:NTFw \l__regex_internal_b_int + { \__regex_compile_quantifier_braced_iii:w } + { \__regex_compile_quantifier_braced_ii:w } } } { - \regex_compile_quantifier_abort:xNN - { \c_lbrace_str \int_use:N \l_regex_internal_a_int } + \__regex_compile_quantifier_abort:xNN + { \c_lbrace_str \int_use:N \l__regex_internal_a_int } #1 #2 } } -\cs_new_protected:Npn \regex_compile_quantifier_braced_ii:w #1#2 +\cs_new_protected:Npn \__regex_compile_quantifier_braced_ii:w #1#2 { - \str_if_eq:xxTF - { #1 #2 } { \regex_compile_special:N \c_rbrace_str } + \str_if_eq_x:nnTF + { #1 #2 } { \__regex_compile_special:N \c_rbrace_str } { - \exp_args:No \regex_compile_quantifier_lazyness:nnNN - { \int_use:N \l_regex_internal_a_int } { -1 } + \exp_args:No \__regex_compile_quantifier_lazyness:nnNN + { \int_use:N \l__regex_internal_a_int } { -1 } } { - \regex_compile_quantifier_abort:xNN - { \c_lbrace_str \int_use:N \l_regex_internal_a_int , } + \__regex_compile_quantifier_abort:xNN + { \c_lbrace_str \int_use:N \l__regex_internal_a_int , } #1 #2 } } -\cs_new_protected:Npn \regex_compile_quantifier_braced_iii:w #1#2 +\cs_new_protected:Npn \__regex_compile_quantifier_braced_iii:w #1#2 { - \str_if_eq:xxTF - { #1 #2 } { \regex_compile_special:N \c_rbrace_str } + \str_if_eq_x:nnTF + { #1 #2 } { \__regex_compile_special:N \c_rbrace_str } { - \if_num:w \l_regex_internal_a_int > \l_regex_internal_b_int - \msg_kernel_error:nnxx { regex } { backwards-quantifier } - { \int_use:N \l_regex_internal_a_int } - { \int_use:N \l_regex_internal_b_int } - \int_zero:N \l_regex_internal_b_int + \if_int_compare:w \l__regex_internal_a_int > \l__regex_internal_b_int + \__msg_kernel_error:nnxx { regex } { backwards-quantifier } + { \int_use:N \l__regex_internal_a_int } + { \int_use:N \l__regex_internal_b_int } + \int_zero:N \l__regex_internal_b_int \else: - \int_sub:Nn \l_regex_internal_b_int \l_regex_internal_a_int + \int_sub:Nn \l__regex_internal_b_int \l__regex_internal_a_int \fi: - \exp_args:Noo \regex_compile_quantifier_lazyness:nnNN - { \int_use:N \l_regex_internal_a_int } - { \int_use:N \l_regex_internal_b_int } + \exp_args:Noo \__regex_compile_quantifier_lazyness:nnNN + { \int_use:N \l__regex_internal_a_int } + { \int_use:N \l__regex_internal_b_int } } { - \regex_compile_quantifier_abort:xNN + \__regex_compile_quantifier_abort:xNN { \c_lbrace_str - \int_use:N \l_regex_internal_a_int , - \int_use:N \l_regex_internal_b_int + \int_use:N \l__regex_internal_a_int , + \int_use:N \l__regex_internal_b_int } #1 #2 } } -\cs_new_protected:Npn \regex_compile_raw_error:N #1 +\cs_new_protected:Npn \__regex_compile_raw_error:N #1 { - \msg_kernel_error:nnx { regex } { bad-escape } {#1} - \regex_compile_raw:N #1 + \__msg_kernel_error:nnx { regex } { bad-escape } {#1} + \__regex_compile_raw:N #1 } -\cs_new_protected:Npn \regex_compile_raw:N #1#2#3 +\cs_new_protected:Npn \__regex_compile_raw:N #1#2#3 { - \regex_if_in_class:TF + \__regex_if_in_class:TF { - \str_if_eq:nnTF {#2#3} { \regex_compile_special:N - } - { \regex_compile_range:Nw #1 } + \str_if_eq:nnTF {#2#3} { \__regex_compile_special:N - } + { \__regex_compile_range:Nw #1 } { - \regex_compile_one:x - { \regex_item_equal:n { \int_value:w `#1 ~ } } + \__regex_compile_one:x + { \__regex_item_equal:n { \__int_value:w `#1 ~ } } #2 #3 } } { - \regex_compile_one:x - { \regex_item_equal:n { \int_value:w `#1 ~ } } + \__regex_compile_one:x + { \__regex_item_equal:n { \__int_value:w `#1 ~ } } #2 #3 } } -\prg_new_protected_conditional:Npnn \regex_if_end_range:NN #1#2 { TF } +\prg_new_protected_conditional:Npnn \__regex_if_end_range:NN #1#2 { TF } { - \if_meaning:w \regex_compile_raw:N #1 + \if_meaning:w \__regex_compile_raw:N #1 \prg_return_true: \else: - \if_meaning:w \regex_compile_special:N #1 + \if_meaning:w \__regex_compile_special:N #1 \if_charcode:w ] #2 \prg_return_false: \else: @@ -804,1790 +804,1798 @@ \fi: \fi: } -\cs_new_protected:Npn \regex_compile_range:Nw #1#2#3 +\cs_new_protected:Npn \__regex_compile_range:Nw #1#2#3 { - \regex_if_end_range:NNTF #2 #3 + \__regex_if_end_range:NNTF #2 #3 { - \if_num:w `#1 > `#3 \exp_stop_f: - \msg_kernel_error:nnxx { regex } { range-backwards } {#1} {#3} + \if_int_compare:w `#1 > `#3 \exp_stop_f: + \__msg_kernel_error:nnxx { regex } { range-backwards } {#1} {#3} \else: - \tl_build_one:x + \__tl_build_one:x { - \if_num:w `#1 = `#3 \exp_stop_f: - \regex_item_equal:n + \if_int_compare:w `#1 = `#3 \exp_stop_f: + \__regex_item_equal:n \else: - \regex_item_range:nn { \int_value:w `#1 ~ } + \__regex_item_range:nn { \__int_value:w `#1 ~ } \fi: - { \int_value:w `#3 ~ } + { \__int_value:w `#3 ~ } } \fi: } { - \msg_kernel_warning:nnxx { regex } { range-missing-end } + \__msg_kernel_warning:nnxx { regex } { range-missing-end } {#1} { \c_backslash_str #3 } - \tl_build_one:x + \__tl_build_one:x { - \regex_item_equal:n { \int_value:w `#1 ~ } - \regex_item_equal:n { \int_value:w `- ~ } + \__regex_item_equal:n { \__int_value:w `#1 ~ } + \__regex_item_equal:n { \__int_value:w `- ~ } } #2#3 } } -\cs_new_protected_nopar:cpx { regex_compile_.: } +\cs_new_protected_nopar:cpx { __regex_compile_.: } { - \exp_not:N \regex_if_in_class:TF - { \regex_compile_raw:N . } - { \regex_compile_one:x \exp_not:c { regex_prop_.: } } + \exp_not:N \__regex_if_in_class:TF + { \__regex_compile_raw:N . } + { \__regex_compile_one:x \exp_not:c { __regex_prop_.: } } } -\cs_new_protected_nopar:cpn { regex_prop_.: } +\cs_new_protected_nopar:cpn { __regex_prop_.: } { - \if_num:w \l_regex_current_char_int > - \c_two - \exp_after:wN \regex_break_true:w + \if_int_compare:w \l__regex_current_char_int > - \c_two + \exp_after:wN \__regex_break_true:w \fi: } -\cs_set_protected:Npn \regex_tmp:w #1#2 +\cs_set_protected:Npn \__regex_tmp:w #1#2 { - \cs_new_protected_nopar:cpx { regex_compile_/#1: } - { \regex_compile_one:x \exp_not:c { regex_prop_#1: } } - \cs_new_protected_nopar:cpx { regex_compile_/#2: } + \cs_new_protected_nopar:cpx { __regex_compile_/#1: } + { \__regex_compile_one:x \exp_not:c { __regex_prop_#1: } } + \cs_new_protected_nopar:cpx { __regex_compile_/#2: } { - \regex_compile_one:x - { \regex_item_reverse:n \exp_not:c { regex_prop_#1: } } + \__regex_compile_one:x + { \__regex_item_reverse:n \exp_not:c { __regex_prop_#1: } } } } -\regex_tmp:w d D -\regex_tmp:w h H -\regex_tmp:w s S -\regex_tmp:w v V -\regex_tmp:w w W -\cs_new_protected_nopar:cpn { regex_compile_/N: } - { \regex_compile_one:x \regex_prop_N: } -\cs_new_protected:Npn \regex_compile_anchor:NF #1#2 +\__regex_tmp:w d D +\__regex_tmp:w h H +\__regex_tmp:w s S +\__regex_tmp:w v V +\__regex_tmp:w w W +\cs_new_protected_nopar:cpn { __regex_compile_/N: } + { \__regex_compile_one:x \__regex_prop_N: } +\cs_new_protected:Npn \__regex_compile_anchor:NF #1#2 { - \regex_if_in_class_or_catcode:TF {#2} + \__regex_if_in_class_or_catcode:TF {#2} { - \tl_build_one:n - { \regex_assertion:Nn \c_true_bool { \regex_anchor:N #1 } } + \__tl_build_one:n + { \__regex_assertion:Nn \c_true_bool { \__regex_anchor:N #1 } } } } -\cs_set_protected:Npn \regex_tmp:w #1#2 +\cs_set_protected:Npn \__regex_tmp:w #1#2 { - \cs_new_protected_nopar:cpn { regex_compile_/#1: } - { \regex_compile_anchor:NF #2 { \regex_compile_raw_error:N #1 } } + \cs_new_protected_nopar:cpn { __regex_compile_/#1: } + { \__regex_compile_anchor:NF #2 { \__regex_compile_raw_error:N #1 } } } -\regex_tmp:w A \l_regex_min_pos_int -\regex_tmp:w G \l_regex_start_pos_int -\regex_tmp:w Z \l_regex_max_pos_int -\regex_tmp:w z \l_regex_max_pos_int -\cs_set_protected:Npn \regex_tmp:w #1#2 +\__regex_tmp:w A \l__regex_min_pos_int +\__regex_tmp:w G \l__regex_start_pos_int +\__regex_tmp:w Z \l__regex_max_pos_int +\__regex_tmp:w z \l__regex_max_pos_int +\cs_set_protected:Npn \__regex_tmp:w #1#2 { - \cs_new_protected_nopar:cpn { regex_compile_#1: } - { \regex_compile_anchor:NF #2 { \regex_compile_raw:N #1 } } + \cs_new_protected_nopar:cpn { __regex_compile_#1: } + { \__regex_compile_anchor:NF #2 { \__regex_compile_raw:N #1 } } } -\exp_args:Nx \regex_tmp:w { \iow_char:N \^ } \l_regex_min_pos_int -\exp_args:Nx \regex_tmp:w { \iow_char:N \$ } \l_regex_max_pos_int -\cs_new_protected_nopar:cpn { regex_compile_/b: } +\exp_args:Nx \__regex_tmp:w { \iow_char:N \^ } \l__regex_min_pos_int +\exp_args:Nx \__regex_tmp:w { \iow_char:N \$ } \l__regex_max_pos_int +\cs_new_protected_nopar:cpn { __regex_compile_/b: } { - \regex_if_in_class_or_catcode:TF - { \regex_compile_raw_error:N b } + \__regex_if_in_class_or_catcode:TF + { \__regex_compile_raw_error:N b } { - \tl_build_one:n - { \regex_assertion:Nn \c_true_bool { \regex_b_test: } } + \__tl_build_one:n + { \__regex_assertion:Nn \c_true_bool { \__regex_b_test: } } } } -\cs_new_protected_nopar:cpn { regex_compile_/B: } +\cs_new_protected_nopar:cpn { __regex_compile_/B: } { - \regex_if_in_class_or_catcode:TF - { \regex_compile_raw_error:N B } + \__regex_if_in_class_or_catcode:TF + { \__regex_compile_raw_error:N B } { - \tl_build_one:n - { \regex_assertion:Nn \c_false_bool { \regex_b_test: } } + \__tl_build_one:n + { \__regex_assertion:Nn \c_false_bool { \__regex_b_test: } } } } -\cs_new_protected:cpn { regex_compile_]: } +\cs_new_protected:cpn { __regex_compile_]: } { - \regex_if_in_class:TF + \__regex_if_in_class:TF { - \if_num:w \l_regex_mode_int > \c_sixteen - \tl_build_one:n { \if_false: { \fi: } } + \if_int_compare:w \l__regex_mode_int > \c_sixteen + \__tl_build_one:n { \if_false: { \fi: } } \fi: - \tex_advance:D \l_regex_mode_int - \c_fifteen - \tex_divide:D \l_regex_mode_int \c_thirteen - \if_int_odd:w \l_regex_mode_int \else: - \exp_after:wN \regex_compile_quantifier:w + \tex_advance:D \l__regex_mode_int - \c_fifteen + \tex_divide:D \l__regex_mode_int \c_thirteen + \if_int_odd:w \l__regex_mode_int \else: + \exp_after:wN \__regex_compile_quantifier:w \fi: } - { \regex_compile_raw:N ] } + { \__regex_compile_raw:N ] } } -\cs_new_protected_nopar:cpn { regex_compile_[: } +\cs_new_protected_nopar:cpn { __regex_compile_[: } { - \regex_if_in_class:TF - { \regex_compile_class_posix_test:w } + \__regex_if_in_class:TF + { \__regex_compile_class_posix_test:w } { - \regex_if_within_catcode:TF + \__regex_if_within_catcode:TF { - \exp_after:wN \regex_compile_class_catcode:w - \int_use:N \l_regex_catcodes_int ; + \exp_after:wN \__regex_compile_class_catcode:w + \int_use:N \l__regex_catcodes_int ; } - { \regex_compile_class_normal:w } + { \__regex_compile_class_normal:w } } } -\cs_new_protected_nopar:Npn \regex_compile_class_normal:w +\cs_new_protected_nopar:Npn \__regex_compile_class_normal:w { - \regex_compile_class:TFNN - { \regex_class:NnnnN \c_true_bool } - { \regex_class:NnnnN \c_false_bool } + \__regex_compile_class:TFNN + { \__regex_class:NnnnN \c_true_bool } + { \__regex_class:NnnnN \c_false_bool } } -\cs_new_protected:Npn \regex_compile_class_catcode:w #1; +\cs_new_protected:Npn \__regex_compile_class_catcode:w #1; { - \if_int_compare:w \l_regex_mode_int = \c_two - \tl_build_one:n - { \regex_class:NnnnN \c_true_bool { \if_false: } \fi: } + \if_int_compare:w \l__regex_mode_int = \c_two + \__tl_build_one:n + { \__regex_class:NnnnN \c_true_bool { \if_false: } \fi: } \fi: - \int_set_eq:NN \l_regex_catcodes_int \l_regex_default_catcodes_int - \regex_compile_class:TFNN - { \regex_item_catcode:nT {#1} } - { \regex_item_catcode_reverse:nT {#1} } + \int_set_eq:NN \l__regex_catcodes_int \l__regex_default_catcodes_int + \__regex_compile_class:TFNN + { \__regex_item_catcode:nT {#1} } + { \__regex_item_catcode_reverse:nT {#1} } } -\cs_new_protected:Npn \regex_compile_class:TFNN #1#2#3#4 +\cs_new_protected:Npn \__regex_compile_class:TFNN #1#2#3#4 { - \l_regex_mode_int = \int_value:w \l_regex_mode_int 3 \exp_stop_f: - \str_if_eq:nnTF { #3 #4 } { \regex_compile_special:N ^ } + \l__regex_mode_int = \__int_value:w \l__regex_mode_int 3 \exp_stop_f: + \str_if_eq:nnTF { #3 #4 } { \__regex_compile_special:N ^ } { - \tl_build_one:n { #2 { \if_false: } \fi: } - \regex_compile_class_ii:NN + \__tl_build_one:n { #2 { \if_false: } \fi: } + \__regex_compile_class_ii:NN } { - \tl_build_one:n { #1 { \if_false: } \fi: } - \regex_compile_class_ii:NN #3 #4 + \__tl_build_one:n { #1 { \if_false: } \fi: } + \__regex_compile_class_ii:NN #3 #4 } } -\cs_new_protected:Npn \regex_compile_class_ii:NN #1#2 +\cs_new_protected:Npn \__regex_compile_class_ii:NN #1#2 { \token_if_eq_charcode:NNTF #2 ] - { \regex_compile_raw:N #2 } + { \__regex_compile_raw:N #2 } { #1 #2 } } -\cs_new_protected:Npn \regex_compile_class_posix_test:w #1#2 +\cs_new_protected:Npn \__regex_compile_class_posix_test:w #1#2 { - \token_if_eq_meaning:NNT \regex_compile_special:N #1 + \token_if_eq_meaning:NNT \__regex_compile_special:N #1 { - \prg_case_str:nnn { #2 } + \str_case:nnn { #2 } { - : { \regex_compile_class_posix:NNNNw } - = { \msg_kernel_warning:nnx { regex } { posix-unsupported } { = } } - . { \msg_kernel_warning:nnx { regex } { posix-unsupported } { . } } + : { \__regex_compile_class_posix:NNNNw } + = { \__msg_kernel_warning:nnx { regex } { posix-unsupported } { = } } + . { \__msg_kernel_warning:nnx { regex } { posix-unsupported } { . } } } { } } - \regex_compile_raw:N [ #1 #2 + \__regex_compile_raw:N [ #1 #2 } -\cs_new_protected:Npn \regex_compile_class_posix:NNNNw #1#2#3#4#5#6 +\cs_new_protected:Npn \__regex_compile_class_posix:NNNNw #1#2#3#4#5#6 { - \str_if_eq:nnTF { #5 #6 } { \regex_compile_special:N ^ } + \str_if_eq:nnTF { #5 #6 } { \__regex_compile_special:N ^ } { - \bool_set_false:N \l_regex_internal_bool - \tl_set:Nx \l_regex_internal_a_tl { \if_false: } \fi: - \regex_compile_class_posix_loop:w + \bool_set_false:N \l__regex_internal_bool + \tl_set:Nx \l__regex_internal_a_tl { \if_false: } \fi: + \__regex_compile_class_posix_loop:w } { - \bool_set_true:N \l_regex_internal_bool - \tl_set:Nx \l_regex_internal_a_tl { \if_false: } \fi: - \regex_compile_class_posix_loop:w #5 #6 + \bool_set_true:N \l__regex_internal_bool + \tl_set:Nx \l__regex_internal_a_tl { \if_false: } \fi: + \__regex_compile_class_posix_loop:w #5 #6 } } -\cs_new:Npn \regex_compile_class_posix_loop:w #1#2 +\cs_new:Npn \__regex_compile_class_posix_loop:w #1#2 { - \token_if_eq_meaning:NNTF \regex_compile_raw:N #1 - { #2 \regex_compile_class_posix_loop:w } - { \if_false: { \fi: } \regex_compile_class_posix_end:w #1 #2 } + \token_if_eq_meaning:NNTF \__regex_compile_raw:N #1 + { #2 \__regex_compile_class_posix_loop:w } + { \if_false: { \fi: } \__regex_compile_class_posix_end:w #1 #2 } } -\cs_new_protected:Npn \regex_compile_class_posix_end:w #1#2#3#4 +\cs_new_protected:Npn \__regex_compile_class_posix_end:w #1#2#3#4 { \str_if_eq:nnTF { #1 #2 #3 #4 } - { \regex_compile_special:N : \regex_compile_special:N ] } + { \__regex_compile_special:N : \__regex_compile_special:N ] } { - \cs_if_exist:cTF { regex_posix_ \l_regex_internal_a_tl : } + \cs_if_exist:cTF { __regex_posix_ \l__regex_internal_a_tl : } { - \regex_compile_one:x + \__regex_compile_one:x { - \bool_if:NF \l_regex_internal_bool \regex_item_reverse:n - \exp_not:c { regex_posix_ \l_regex_internal_a_tl : } + \bool_if:NF \l__regex_internal_bool \__regex_item_reverse:n + \exp_not:c { __regex_posix_ \l__regex_internal_a_tl : } } } { - \msg_kernel_warning:nnx { regex } { posix-unknown } - { \l_regex_internal_a_tl } - \regex_compile_abort_tokens:x + \__msg_kernel_warning:nnx { regex } { posix-unknown } + { \l__regex_internal_a_tl } + \__regex_compile_abort_tokens:x { - [: \bool_if:NF \l_regex_internal_bool { ^ } - \l_regex_internal_a_tl :] + [: \bool_if:NF \l__regex_internal_bool { ^ } + \l__regex_internal_a_tl :] } } } { - \msg_kernel_error:nnxx { regex } { posix-missing-close } - { [: \l_regex_internal_a_tl } { #2 #4 } - \regex_compile_abort_tokens:x { [: \l_regex_internal_a_tl } + \__msg_kernel_error:nnxx { regex } { posix-missing-close } + { [: \l__regex_internal_a_tl } { #2 #4 } + \__regex_compile_abort_tokens:x { [: \l__regex_internal_a_tl } #1 #2 #3 #4 } } -\cs_new_protected:Npn \regex_compile_group_begin:N #1 +\cs_new_protected:Npn \__regex_compile_group_begin:N #1 { - \tl_build_one:n { #1 { \if_false: } \fi: } - \regex_mode_quit_c: - \tl_set_build:Nw \l_regex_internal_regex - \int_set_eq:NN \l_regex_default_catcodes_int \l_regex_catcodes_int - \int_incr:N \l_regex_group_level_int - \tl_build_one:n { \regex_branch:n { \if_false: } \fi: } + \__tl_build_one:n { #1 { \if_false: } \fi: } + \__regex_mode_quit_c: + \__tl_build:Nw \l__regex_internal_regex + \int_set_eq:NN \l__regex_default_catcodes_int \l__regex_catcodes_int + \int_incr:N \l__regex_group_level_int + \__tl_build_one:n { \__regex_branch:n { \if_false: } \fi: } } -\cs_new_protected:Npn \regex_compile_group_end: +\cs_new_protected:Npn \__regex_compile_group_end: { - \if_num:w \l_regex_group_level_int > \c_zero - \tl_build_one:n { \if_false: { \fi: } } - \tl_build_end: - \int_set_eq:NN \l_regex_catcodes_int \l_regex_default_catcodes_int - \tl_build_one:o \l_regex_internal_regex - \exp_after:wN \regex_compile_quantifier:w + \if_int_compare:w \l__regex_group_level_int > \c_zero + \__tl_build_one:n { \if_false: { \fi: } } + \__tl_build_end: + \int_set_eq:NN \l__regex_catcodes_int \l__regex_default_catcodes_int + \__tl_build_one:o \l__regex_internal_regex + \exp_after:wN \__regex_compile_quantifier:w \else: - \msg_kernel_warning:nn { regex } { extra-rparen } - \exp_after:wN \regex_compile_raw:N \exp_after:wN ) + \__msg_kernel_warning:nn { regex } { extra-rparen } + \exp_after:wN \__regex_compile_raw:N \exp_after:wN ) \fi: } -\cs_new_protected_nopar:cpn { regex_compile_(: } +\cs_new_protected_nopar:cpn { __regex_compile_(: } { - \regex_if_in_class:TF { \regex_compile_raw:N ( } - { \regex_compile_lparen:w } + \__regex_if_in_class:TF { \__regex_compile_raw:N ( } + { \__regex_compile_lparen:w } } -\cs_new_protected:Npn \regex_compile_lparen:w #1#2#3#4 +\cs_new_protected:Npn \__regex_compile_lparen:w #1#2#3#4 { - \str_if_eq:nnTF { #1 #2 } { \regex_compile_special:N ? } + \str_if_eq:nnTF { #1 #2 } { \__regex_compile_special:N ? } { \cs_if_exist_use:cF - { regex_compile_special_group_\token_to_str:N #4 :w } + { __regex_compile_special_group_\token_to_str:N #4 :w } { - \msg_kernel_warning:nnx { regex } { special-group-unknown } + \__msg_kernel_warning:nnx { regex } { special-group-unknown } { (? \token_to_str:N #4 } - \regex_compile_group_begin:N \regex_group:nnnN - \regex_compile_raw:N ? #3 #4 + \__regex_compile_group_begin:N \__regex_group:nnnN + \__regex_compile_raw:N ? #3 #4 } } { - \regex_compile_group_begin:N \regex_group:nnnN + \__regex_compile_group_begin:N \__regex_group:nnnN #1 #2 #3 #4 } } -\cs_new_protected_nopar:cpn { regex_compile_|: } +\cs_new_protected_nopar:cpn { __regex_compile_|: } { - \regex_if_in_class:TF { \regex_compile_raw:N | } + \__regex_if_in_class:TF { \__regex_compile_raw:N | } { - \tl_build_one:n - { \if_false: { \fi: } \regex_branch:n { \if_false: } \fi: } + \__tl_build_one:n + { \if_false: { \fi: } \__regex_branch:n { \if_false: } \fi: } } } -\cs_new_protected_nopar:cpn { regex_compile_): } +\cs_new_protected_nopar:cpn { __regex_compile_): } { - \regex_if_in_class:TF { \regex_compile_raw:N ) } - { \regex_compile_group_end: } + \__regex_if_in_class:TF { \__regex_compile_raw:N ) } + { \__regex_compile_group_end: } } -\cs_new_protected_nopar:cpn { regex_compile_special_group_::w } - { \regex_compile_group_begin:N \regex_group_no_capture:nnnN } -\cs_new_protected_nopar:cpn { regex_compile_special_group_|:w } - { \regex_compile_group_begin:N \regex_group_resetting:nnnN } -\cs_new_protected:Npn \regex_compile_special_group_i:w #1#2 +\cs_new_protected_nopar:cpn { __regex_compile_special_group_::w } + { \__regex_compile_group_begin:N \__regex_group_no_capture:nnnN } +\cs_new_protected_nopar:cpn { __regex_compile_special_group_|:w } + { \__regex_compile_group_begin:N \__regex_group_resetting:nnnN } +\cs_new_protected:Npn \__regex_compile_special_group_i:w #1#2 { - \str_if_eq:nnTF { #1 #2 } { \regex_compile_special:N ) } + \str_if_eq:nnTF { #1 #2 } { \__regex_compile_special:N ) } { - \cs_set_nopar:Npn \regex_item_equal:n { \regex_item_caseless_equal:n } - \cs_set_nopar:Npn \regex_item_range:nn { \regex_item_caseless_range:nn } + \cs_set_nopar:Npn \__regex_item_equal:n { \__regex_item_caseless_equal:n } + \cs_set_nopar:Npn \__regex_item_range:nn { \__regex_item_caseless_range:nn } } { - \msg_kernel_warning:nnx { regex } { unknown-option } { (?i #2 } - \regex_compile_raw:N ( - \regex_compile_raw:N ? - \regex_compile_raw:N i + \__msg_kernel_warning:nnx { regex } { unknown-option } { (?i #2 } + \__regex_compile_raw:N ( + \__regex_compile_raw:N ? + \__regex_compile_raw:N i #1 #2 } } -\cs_new_protected_nopar:cpn { regex_compile_special_group_-:w } #1#2#3#4 +\cs_new_protected_nopar:cpn { __regex_compile_special_group_-:w } #1#2#3#4 { \str_if_eq:nnTF { #1 #2 #3 #4 } - { \regex_compile_raw:N i \regex_compile_special:N ) } + { \__regex_compile_raw:N i \__regex_compile_special:N ) } { - \cs_set_nopar:Npn \regex_item_equal:n { \regex_item_caseful_equal:n } - \cs_set_nopar:Npn \regex_item_range:nn { \regex_item_caseful_range:nn } + \cs_set_nopar:Npn \__regex_item_equal:n { \__regex_item_caseful_equal:n } + \cs_set_nopar:Npn \__regex_item_range:nn { \__regex_item_caseful_range:nn } } { - \msg_kernel_warning:nnx { regex } { unknown-option } { (?-#2#4 } - \regex_compile_raw:N ( - \regex_compile_raw:N ? - \regex_compile_raw:N - + \__msg_kernel_warning:nnx { regex } { unknown-option } { (?-#2#4 } + \__regex_compile_raw:N ( + \__regex_compile_raw:N ? + \__regex_compile_raw:N - #1 #2 #3 #4 } } -\cs_new_protected:cpn { regex_compile_/c: } - { \regex_chk_c_allowed:T { \regex_compile_c_test:NN } } -\cs_new_protected:Npn \regex_compile_c_test:NN #1#2 +\cs_new_protected:cpn { __regex_compile_/c: } + { \__regex_chk_c_allowed:T { \__regex_compile_c_test:NN } } +\cs_new_protected:Npn \__regex_compile_c_test:NN #1#2 { - \token_if_eq_meaning:NNTF #1 \regex_compile_raw:N + \token_if_eq_meaning:NNTF #1 \__regex_compile_raw:N { - \int_if_exist:cTF { c_regex_catcode_#2_int } + \int_if_exist:cTF { c__regex_catcode_#2_int } { - \int_set_eq:Nc \l_regex_catcodes_int { c_regex_catcode_#2_int } - \l_regex_mode_int - = \if_case:w \l_regex_mode_int \c_two \else: \c_six \fi: + \int_set_eq:Nc \l__regex_catcodes_int { c__regex_catcode_#2_int } + \l__regex_mode_int + = \if_case:w \l__regex_mode_int \c_two \else: \c_six \fi: } } - { \cs_if_exist_use:cF { regex_compile_c_#2:w } } + { \cs_if_exist_use:cF { __regex_compile_c_#2:w } } { - \msg_kernel_error:nnx { regex } { c-missing-category } {#2} + \__msg_kernel_error:nnx { regex } { c-missing-category } {#2} #1 #2 } } -\cs_new_protected:cpn { regex_compile_c_[:w } #1#2 +\cs_new_protected:cpn { __regex_compile_c_[:w } #1#2 { - \l_regex_mode_int - = \if_case:w \l_regex_mode_int \c_two \else: \c_six \fi: - \int_zero:N \l_regex_catcodes_int - \str_if_eq:nnTF { #1 #2 } { \regex_compile_special:N ^ } + \l__regex_mode_int + = \if_case:w \l__regex_mode_int \c_two \else: \c_six \fi: + \int_zero:N \l__regex_catcodes_int + \str_if_eq:nnTF { #1 #2 } { \__regex_compile_special:N ^ } { - \bool_set_false:N \l_regex_catcodes_bool - \regex_compile_c_lbrack_loop:NN + \bool_set_false:N \l__regex_catcodes_bool + \__regex_compile_c_lbrack_loop:NN } { - \bool_set_true:N \l_regex_catcodes_bool - \regex_compile_c_lbrack_loop:NN + \bool_set_true:N \l__regex_catcodes_bool + \__regex_compile_c_lbrack_loop:NN #1 #2 } } -\cs_new_protected:Npn \regex_compile_c_lbrack_loop:NN #1#2 +\cs_new_protected:Npn \__regex_compile_c_lbrack_loop:NN #1#2 { - \token_if_eq_meaning:NNTF #1 \regex_compile_raw:N + \token_if_eq_meaning:NNTF #1 \__regex_compile_raw:N { - \int_if_exist:cTF { c_regex_catcode_#2_int } + \int_if_exist:cTF { c__regex_catcode_#2_int } { - \exp_args:Nc \regex_compile_c_lbrack_add:N - { c_regex_catcode_#2_int } - \regex_compile_c_lbrack_loop:NN + \exp_args:Nc \__regex_compile_c_lbrack_add:N + { c__regex_catcode_#2_int } + \__regex_compile_c_lbrack_loop:NN } } { \token_if_eq_charcode:NNTF #2 ] - { \regex_compile_c_lbrack_end: } + { \__regex_compile_c_lbrack_end: } } { - \msg_kernel_error:nnx { regex } { c-missing-rbrack } {#2} - \regex_compile_c_lbrack_end: + \__msg_kernel_error:nnx { regex } { c-missing-rbrack } {#2} + \__regex_compile_c_lbrack_end: #1 #2 } } -\cs_new_protected:Npn \regex_compile_c_lbrack_add:N #1 +\cs_new_protected:Npn \__regex_compile_c_lbrack_add:N #1 { - \if_int_odd:w \int_eval:w \l_regex_catcodes_int / #1 \int_eval_end: + \if_int_odd:w \__int_eval:w \l__regex_catcodes_int / #1 \__int_eval_end: \else: - \tex_advance:D \l_regex_catcodes_int #1 + \tex_advance:D \l__regex_catcodes_int #1 \fi: } -\cs_new_protected_nopar:Npn \regex_compile_c_lbrack_end: +\cs_new_protected_nopar:Npn \__regex_compile_c_lbrack_end: { - \if_meaning:w \c_false_bool \l_regex_catcodes_bool - \int_set:Nn \l_regex_catcodes_int - { \c_regex_all_catcodes_int - \l_regex_catcodes_int } + \if_meaning:w \c_false_bool \l__regex_catcodes_bool + \int_set:Nn \l__regex_catcodes_int + { \c__regex_all_catcodes_int - \l__regex_catcodes_int } \fi: } -\cs_new_protected_nopar:cpn { regex_compile_c_ \c_lbrace_str :w } +\cs_new_protected_nopar:cpn { __regex_compile_c_ \c_lbrace_str :w } { - \regex_compile:w - \regex_disable_submatches: - \l_regex_mode_int - = - \if_case:w \l_regex_mode_int \c_two \else: \c_six \fi: + \__regex_compile:w + \__regex_disable_submatches: + \l__regex_mode_int + = - \if_case:w \l__regex_mode_int \c_two \else: \c_six \fi: } -\cs_new_protected:cpn { regex_compile_ \c_rbrace_str : } +\cs_new_protected:cpn { __regex_compile_ \c_rbrace_str : } { - \regex_if_in_cs:TF + \__regex_if_in_cs:TF { - \regex_compile_end: - \regex_compile_one:x - { \regex_item_cs:n { \exp_not:o \l_regex_internal_regex } } + \__regex_compile_end: + \__regex_compile_one:x + { \__regex_item_cs:n { \exp_not:o \l__regex_internal_regex } } } - { \exp_after:wN \regex_compile_raw:N \c_rbrace_str } + { \exp_after:wN \__regex_compile_raw:N \c_rbrace_str } } -\cs_new_protected:cpn { regex_compile_/u: } #1#2 +\cs_new_protected:cpn { __regex_compile_/u: } #1#2 { - \regex_if_in_class_or_catcode:TF - { \regex_compile_raw_error:N u #1 #2 } + \__regex_if_in_class_or_catcode:TF + { \__regex_compile_raw_error:N u #1 #2 } { - \str_if_eq:xxTF {#1#2} { \regex_compile_special:N \c_lbrace_str } + \str_if_eq_x:nnTF {#1#2} { \__regex_compile_special:N \c_lbrace_str } { - \tl_set:Nx \l_regex_internal_a_tl { \if_false: } \fi: - \regex_compile_u_loop:NN + \tl_set:Nx \l__regex_internal_a_tl { \if_false: } \fi: + \__regex_compile_u_loop:NN } { - \msg_kernel_error:nn { regex } { u-missing-lbrace } - \regex_compile_raw:N u #1 #2 + \__msg_kernel_error:nn { regex } { u-missing-lbrace } + \__regex_compile_raw:N u #1 #2 } } } -\cs_new:Npn \regex_compile_u_loop:NN #1#2 +\cs_new:Npn \__regex_compile_u_loop:NN #1#2 { - \token_if_eq_meaning:NNTF #1 \regex_compile_raw:N - { #2 \regex_compile_u_loop:NN } + \token_if_eq_meaning:NNTF #1 \__regex_compile_raw:N + { #2 \__regex_compile_u_loop:NN } { - \token_if_eq_meaning:NNTF #1 \regex_compile_special:N + \token_if_eq_meaning:NNTF #1 \__regex_compile_special:N { \exp_after:wN \token_if_eq_charcode:NNTF \c_rbrace_str #2 - { \if_false: { \fi: } \regex_compile_u_end: } - { #2 \regex_compile_u_loop:NN } + { \if_false: { \fi: } \__regex_compile_u_end: } + { #2 \__regex_compile_u_loop:NN } } { \if_false: { \fi: } - \msg_kernel_error:nnx { regex } { u-missing-rbrace } {#2} - \regex_compile_u_end: + \__msg_kernel_error:nnx { regex } { u-missing-rbrace } {#2} + \__regex_compile_u_end: #1 #2 } } } -\cs_new_protected:Npn \regex_compile_u_end: +\cs_new_protected:Npn \__regex_compile_u_end: { - \tl_set:Nv \l_regex_internal_a_tl { \l_regex_internal_a_tl } - \if_num:w \l_regex_mode_int = \c_zero - \regex_compile_u_not_cs: + \tl_set:Nv \l__regex_internal_a_tl { \l__regex_internal_a_tl } + \if_int_compare:w \l__regex_mode_int = \c_zero + \__regex_compile_u_not_cs: \else: - \regex_compile_u_in_cs: + \__regex_compile_u_in_cs: \fi: } -\cs_new_protected:Npn \regex_compile_u_in_cs: +\cs_new_protected:Npn \__regex_compile_u_in_cs: { - \exp_args:NNo \str_gset_other:Nn \g_regex_internal_tl - { \l_regex_internal_a_tl } - \tl_build_one:x + \exp_args:NNo \__str_gset_other:Nn \g__regex_internal_tl + { \l__regex_internal_a_tl } + \__tl_build_one:x { - \tl_map_function:NN \g_regex_internal_tl - \regex_compile_u_in_cs_aux:n + \tl_map_function:NN \g__regex_internal_tl + \__regex_compile_u_in_cs_aux:n } } -\cs_new:Npn \regex_compile_u_in_cs_aux:n #1 +\cs_new:Npn \__regex_compile_u_in_cs_aux:n #1 { - \regex_class:NnnnN \c_true_bool - { \regex_item_caseful_equal:n { \int_value:w `#1 } } + \__regex_class:NnnnN \c_true_bool + { \__regex_item_caseful_equal:n { \__int_value:w `#1 } } { 1 } { 0 } \c_false_bool } -\cs_new_protected:Npn \regex_compile_u_not_cs: +\cs_new_protected:Npn \__regex_compile_u_not_cs: { - \exp_args:No \tl_analysis_map_inline:nn { \l_regex_internal_a_tl } + \exp_args:No \__tl_analysis_map_inline:nn { \l__regex_internal_a_tl } { - \tl_build_one:n + \__tl_build_one:n { - \regex_class:NnnnN \c_true_bool + \__regex_class:NnnnN \c_true_bool { - \if_num:w "##2 = \c_zero - \regex_item_exact_cs:c { \exp_after:wN \cs_to_str:N ##1 } + \if_int_compare:w "##2 = \c_zero + \__regex_item_exact_cs:c { \exp_after:wN \cs_to_str:N ##1 } \else: - \regex_item_exact:nn { \int_value:w "##2 } { ##3 } + \__regex_item_exact:nn { \__int_value:w "##2 } { ##3 } \fi: } { 1 } { 0 } \c_false_bool } } } -\cs_new_protected_nopar:cpn { regex_compile_/K: } +\cs_new_protected_nopar:cpn { __regex_compile_/K: } { - \int_compare:nNnTF \l_regex_mode_int = \c_zero - { \tl_build_one:n { \regex_command_K: } } - { \regex_compile_raw_error:N K } + \int_compare:nNnTF \l__regex_mode_int = \c_zero + { \__tl_build_one:n { \__regex_command_K: } } + { \__regex_compile_raw_error:N K } } -\cs_new_protected:Npn \regex_show_aux:Nx #1#2 +\cs_new_protected:Npn \__regex_show:Nx #1#2 { - \tl_set_build:Nw \l_regex_internal_a_tl - \cs_set_protected_nopar:Npn \regex_branch:n + \__tl_build:Nw \l__regex_internal_a_tl + \cs_set_protected_nopar:Npn \__regex_branch:n { - \seq_pop_right:NN \l_regex_show_prefix_seq \l_regex_internal_a_tl - \regex_show_one:n { +-branch } - \seq_put_right:No \l_regex_show_prefix_seq \l_regex_internal_a_tl + \seq_pop_right:NN \l__regex_show_prefix_seq \l__regex_internal_a_tl + \__regex_show_one:n { +-branch } + \seq_put_right:No \l__regex_show_prefix_seq \l__regex_internal_a_tl \use:n } - \cs_set_protected_nopar:Npn \regex_group:nnnN - { \regex_show_group_aux:nnnnN { } } - \cs_set_protected_nopar:Npn \regex_group_no_capture:nnnN - { \regex_show_group_aux:nnnnN { ~(no~capture) } } - \cs_set_protected_nopar:Npn \regex_group_resetting:nnnN - { \regex_show_group_aux:nnnnN { ~(resetting) } } - \cs_set_eq:NN \regex_class:NnnnN \regex_show_class:NnnnN - \cs_set_protected_nopar:Npn \regex_command_K: - { \regex_show_one:n { reset~match~start~(\iow_char:N\\K) } } - \cs_set_protected:Npn \regex_assertion:Nn ##1##2 - { \regex_show_one:n { \bool_if:NF ##1 { negative~ } assertion:~##2 } } - \cs_set_nopar:Npn \regex_b_test: { word~boundary } - \cs_set_eq:NN \regex_anchor:N \regex_show_anchor_to_str:N - \cs_set_protected:Npn \regex_item_caseful_equal:n ##1 - { \regex_show_one:n { char~code~\int_eval:n{##1} } } - \cs_set_protected:Npn \regex_item_caseful_range:nn ##1##2 - { \regex_show_one:n { range~[\int_eval:n{##1}, \int_eval:n{##2}] } } - \cs_set_protected:Npn \regex_item_caseless_equal:n ##1 - { \regex_show_one:n { char~code~\int_eval:n{##1}~(caseless) } } - \cs_set_protected:Npn \regex_item_caseless_range:nn ##1##2 + \cs_set_protected_nopar:Npn \__regex_group:nnnN + { \__regex_show_group_aux:nnnnN { } } + \cs_set_protected_nopar:Npn \__regex_group_no_capture:nnnN + { \__regex_show_group_aux:nnnnN { ~(no~capture) } } + \cs_set_protected_nopar:Npn \__regex_group_resetting:nnnN + { \__regex_show_group_aux:nnnnN { ~(resetting) } } + \cs_set_eq:NN \__regex_class:NnnnN \__regex_show_class:NnnnN + \cs_set_protected_nopar:Npn \__regex_command_K: + { \__regex_show_one:n { reset~match~start~(\iow_char:N\\K) } } + \cs_set_protected:Npn \__regex_assertion:Nn ##1##2 + { \__regex_show_one:n { \bool_if:NF ##1 { negative~ } assertion:~##2 } } + \cs_set_nopar:Npn \__regex_b_test: { word~boundary } + \cs_set_eq:NN \__regex_anchor:N \__regex_show_anchor_to_str:N + \cs_set_protected:Npn \__regex_item_caseful_equal:n ##1 + { \__regex_show_one:n { char~code~\int_eval:n{##1} } } + \cs_set_protected:Npn \__regex_item_caseful_range:nn ##1##2 + { \__regex_show_one:n { range~[\int_eval:n{##1}, \int_eval:n{##2}] } } + \cs_set_protected:Npn \__regex_item_caseless_equal:n ##1 + { \__regex_show_one:n { char~code~\int_eval:n{##1}~(caseless) } } + \cs_set_protected:Npn \__regex_item_caseless_range:nn ##1##2 { - \regex_show_one:n + \__regex_show_one:n { Range~[\int_eval:n{##1}, \int_eval:n{##2}]~(caseless) } } - \cs_set_protected:Npn \regex_item_catcode:nT - { \regex_show_item_catcode:NnT \c_true_bool } - \cs_set_protected:Npn \regex_item_catcode_reverse:nT - { \regex_show_item_catcode:NnT \c_false_bool } - \cs_set_protected:Npn \regex_item_reverse:n - { \regex_show_scope:nn { Reversed~match } } - \cs_set_protected:Npn \regex_item_exact:nn ##1##2 - { \regex_show_one:n { char~##2,~catcode~##1 } } - \cs_set_protected:Npn \regex_item_exact_cs:c ##1 - { \regex_show_one:n { control~sequence~\iow_char:N\\##1 } } - \cs_set_protected:Npn \regex_item_cs:n - { \regex_show_scope:nn { control~sequence } } - \cs_set:cpn { regex_prop_.: } { \regex_show_one:n { any~token } } - \seq_clear:N \l_regex_show_prefix_seq - \regex_show_push:n { ~ } + \cs_set_protected:Npn \__regex_item_catcode:nT + { \__regex_show_item_catcode:NnT \c_true_bool } + \cs_set_protected:Npn \__regex_item_catcode_reverse:nT + { \__regex_show_item_catcode:NnT \c_false_bool } + \cs_set_protected:Npn \__regex_item_reverse:n + { \__regex_show_scope:nn { Reversed~match } } + \cs_set_protected:Npn \__regex_item_exact:nn ##1##2 + { \__regex_show_one:n { char~##2,~catcode~##1 } } + \cs_set_protected:Npn \__regex_item_exact_cs:c ##1 + { \__regex_show_one:n { control~sequence~\iow_char:N\\##1 } } + \cs_set_protected:Npn \__regex_item_cs:n + { \__regex_show_scope:nn { control~sequence } } + \cs_set:cpn { __regex_prop_.: } { \__regex_show_one:n { any~token } } + \seq_clear:N \l__regex_show_prefix_seq + \__regex_show_push:n { ~ } #1 - \tl_build_end: - \msg_aux_show:x { > Compiled~regex~#2: \l_regex_internal_a_tl } + \__tl_build_end: + \__msg_show_variable:x { > Compiled~regex~#2: \l__regex_internal_a_tl } } -\cs_new_protected:Npn \regex_show_one:n #1 +\cs_new_protected:Npn \__regex_show_one:n #1 { - \int_incr:N \l_regex_show_lines_int - \tl_build_one:x - { \iow_newline: \seq_use:N \l_regex_show_prefix_seq #1 } + \int_incr:N \l__regex_show_lines_int + \__tl_build_one:x + { + \iow_newline: + \seq_map_function:NN \l__regex_show_prefix_seq \use:n + #1 + } } -\cs_new_protected:Npn \regex_show_push:n #1 - { \seq_put_right:Nx \l_regex_show_prefix_seq { #1 ~ } } -\cs_new_protected:Npn \regex_show_pop: - { \seq_pop_right:NN \l_regex_show_prefix_seq \l_regex_internal_a_tl } -\cs_new_protected:Npn \regex_show_scope:nn #1#2 +\cs_new_protected:Npn \__regex_show_push:n #1 + { \seq_put_right:Nx \l__regex_show_prefix_seq { #1 ~ } } +\cs_new_protected:Npn \__regex_show_pop: + { \seq_pop_right:NN \l__regex_show_prefix_seq \l__regex_internal_a_tl } +\cs_new_protected:Npn \__regex_show_scope:nn #1#2 { - \regex_show_one:n {#1} - \regex_show_push:n { ~ } + \__regex_show_one:n {#1} + \__regex_show_push:n { ~ } #2 - \regex_show_pop: + \__regex_show_pop: } -\cs_new_protected:Npn \regex_show_group_aux:nnnnN #1#2#3#4#5 +\cs_new_protected:Npn \__regex_show_group_aux:nnnnN #1#2#3#4#5 { - \regex_show_one:n { ,-group~begin #1 } - \regex_show_push:n { | } + \__regex_show_one:n { ,-group~begin #1 } + \__regex_show_push:n { | } \use_ii:nn #2 - \regex_show_pop: - \regex_show_one:n - { `-group~end \regex_msg_repeated:nnN {#3} {#4} #5 } + \__regex_show_pop: + \__regex_show_one:n + { `-group~end \__regex_msg_repeated:nnN {#3} {#4} #5 } } -\cs_set:Npn \regex_show_class:NnnnN #1#2#3#4#5 +\cs_set:Npn \__regex_show_class:NnnnN #1#2#3#4#5 { - \tl_set_build:Nw \l_regex_internal_a_tl - \int_zero:N \l_regex_show_lines_int - \regex_show_push:n {~} + \__tl_build:Nw \l__regex_internal_a_tl + \int_zero:N \l__regex_show_lines_int + \__regex_show_push:n {~} #2 \exp_last_unbraced:Nf - \prg_case_int:nnn { \l_regex_show_lines_int } + \int_case:nnn { \l__regex_show_lines_int } { {0} { - \tl_build_end: - \regex_show_one:n { \bool_if:NTF #1 { Fail } { Pass } } + \__tl_build_end: + \__regex_show_one:n { \bool_if:NTF #1 { Fail } { Pass } } } {1} { - \tl_build_end: + \__tl_build_end: \bool_if:NTF #1 { #2 - \tl_build_one:n { \regex_msg_repeated:nnN {#3} {#4} #5 } + \__tl_build_one:n { \__regex_msg_repeated:nnN {#3} {#4} #5 } } { - \regex_show_one:n - { Don't~match~\regex_msg_repeated:nnN {#3} {#4} #5 } - \tl_build_one:o \l_regex_internal_a_tl + \__regex_show_one:n + { Don't~match~\__regex_msg_repeated:nnN {#3} {#4} #5 } + \__tl_build_one:o \l__regex_internal_a_tl } } } { - \tl_build_end: - \regex_show_one:n + \__tl_build_end: + \__regex_show_one:n { \bool_if:NTF #1 { M } { Don't~m } atch - \regex_msg_repeated:nnN {#3} {#4} #5 + \__regex_msg_repeated:nnN {#3} {#4} #5 } - \tl_build_one:o \l_regex_internal_a_tl + \__tl_build_one:o \l__regex_internal_a_tl } } -\cs_new:Npn \regex_show_anchor_to_str:N #1 +\cs_new:Npn \__regex_show_anchor_to_str:N #1 { anchor~at~ - \prg_case_str:nnn { #1 } + \str_case:nnn { #1 } { - { \l_regex_min_pos_int } { start~(\iow_char:N\\A) } - { \l_regex_start_pos_int } { start~of~match~(\iow_char:N\\G) } - { \l_regex_max_pos_int } { end~(\iow_char:N\\Z) } + { \l__regex_min_pos_int } { start~(\iow_char:N\\A) } + { \l__regex_start_pos_int } { start~of~match~(\iow_char:N\\G) } + { \l__regex_max_pos_int } { end~(\iow_char:N\\Z) } } { } } -\cs_new_protected:Npn \regex_show_item_catcode:NnT #1#2 +\cs_new_protected:Npn \__regex_show_item_catcode:NnT #1#2 { - \seq_set_split:Nnn \l_regex_internal_seq { } { CBEMTPUDSLOA } - \seq_set_filter:NNn \l_regex_internal_seq \l_regex_internal_seq - { \int_if_odd_p:n { #2 / \int_use:c { c_regex_catcode_##1_int } } } - \regex_show_scope:nn + \seq_set_split:Nnn \l__regex_internal_seq { } { CBEMTPUDSLOA } + \seq_set_filter:NNn \l__regex_internal_seq \l__regex_internal_seq + { \int_if_odd_p:n { #2 / \int_use:c { c__regex_catcode_##1_int } } } + \__regex_show_scope:nn { - categories~\seq_use:N \l_regex_internal_seq, ~ + categories~ + \seq_map_function:NN \l__regex_internal_seq \use:n + , ~ \bool_if:NF #1 { negative~ } class } } -\int_new:N \l_regex_min_state_int -\int_new:N \l_regex_max_state_int -\int_new:N \l_regex_left_state_int -\int_new:N \l_regex_right_state_int -\seq_new:N \l_regex_left_state_seq -\seq_new:N \l_regex_right_state_seq -\int_new:N \l_regex_capturing_group_int -\cs_new_protected:Npn \regex_build:n #1 +\int_new:N \l__regex_min_state_int +\int_new:N \l__regex_max_state_int +\int_new:N \l__regex_left_state_int +\int_new:N \l__regex_right_state_int +\seq_new:N \l__regex_left_state_seq +\seq_new:N \l__regex_right_state_seq +\int_new:N \l__regex_capturing_group_int +\cs_new_protected:Npn \__regex_build:n #1 { - \regex_compile:n {#1} - \regex_build:N \l_regex_internal_regex + \__regex_compile:n {#1} + \__regex_build:N \l__regex_internal_regex } -\cs_new_protected:Npn \regex_build:N #1 +\cs_new_protected:Npn \__regex_build:N #1 { \int_set:Nn \tex_escapechar:D { 92 } - \int_zero:N \l_regex_capturing_group_int - \int_set_eq:NN \l_regex_max_state_int \l_regex_min_state_int - \regex_build_new_state: - \regex_build_new_state: - \regex_toks_put_right:Nn \l_regex_left_state_int - { \regex_action_start_wildcard: } - \regex_group:nnnN {#1} { 1 } { 0 } \c_false_bool - \regex_toks_put_right:Nn \l_regex_right_state_int - { \regex_action_success: } - } -\cs_new_protected:Npn \regex_build_for_cs:n #1 - { - \int_set_eq:NN \l_regex_max_state_int \l_regex_min_state_int - \regex_build_new_state: - \regex_build_new_state: - \regex_push_lr_states: + \int_zero:N \l__regex_capturing_group_int + \int_set_eq:NN \l__regex_max_state_int \l__regex_min_state_int + \__regex_build_new_state: + \__regex_build_new_state: + \__regex_toks_put_right:Nn \l__regex_left_state_int + { \__regex_action_start_wildcard: } + \__regex_group:nnnN {#1} { 1 } { 0 } \c_false_bool + \__regex_toks_put_right:Nn \l__regex_right_state_int + { \__regex_action_success: } + } +\cs_new_protected:Npn \__regex_build_for_cs:n #1 + { + \int_set_eq:NN \l__regex_max_state_int \l__regex_min_state_int + \__regex_build_new_state: + \__regex_build_new_state: + \__regex_push_lr_states: #1 - \regex_pop_lr_states: - \regex_toks_put_right:Nn \l_regex_right_state_int + \__regex_pop_lr_states: + \__regex_toks_put_right:Nn \l__regex_right_state_int { - \if_num:w \l_regex_current_pos_int = \l_regex_max_pos_int - \exp_after:wN \regex_action_success: + \if_int_compare:w \l__regex_current_pos_int = \l__regex_max_pos_int + \exp_after:wN \__regex_action_success: \fi: } } -\cs_new_protected_nopar:Npn \regex_push_lr_states: +\cs_new_protected_nopar:Npn \__regex_push_lr_states: { - \seq_push:No \l_regex_left_state_seq - { \int_use:N \l_regex_left_state_int } - \seq_push:No \l_regex_right_state_seq - { \int_use:N \l_regex_right_state_int } + \seq_push:No \l__regex_left_state_seq + { \int_use:N \l__regex_left_state_int } + \seq_push:No \l__regex_right_state_seq + { \int_use:N \l__regex_right_state_int } } -\cs_new_protected_nopar:Npn \regex_pop_lr_states: +\cs_new_protected_nopar:Npn \__regex_pop_lr_states: { - \seq_pop:NN \l_regex_left_state_seq \l_regex_internal_a_tl - \int_set:Nn \l_regex_left_state_int \l_regex_internal_a_tl - \seq_pop:NN \l_regex_right_state_seq \l_regex_internal_a_tl - \int_set:Nn \l_regex_right_state_int \l_regex_internal_a_tl + \seq_pop:NN \l__regex_left_state_seq \l__regex_internal_a_tl + \int_set:Nn \l__regex_left_state_int \l__regex_internal_a_tl + \seq_pop:NN \l__regex_right_state_seq \l__regex_internal_a_tl + \int_set:Nn \l__regex_right_state_int \l__regex_internal_a_tl } -\cs_new_protected:Npn \regex_toks_put_left:Nx #1#2 +\cs_new_protected:Npn \__regex_toks_put_left:Nx #1#2 { - \cs_set_nopar:Npx \regex_tmp:w { #2 } + \cs_set_nopar:Npx \__regex_tmp:w { #2 } \tex_toks:D #1 \exp_after:wN \exp_after:wN \exp_after:wN - { \exp_after:wN \regex_tmp:w \tex_the:D \tex_toks:D #1 } + { \exp_after:wN \__regex_tmp:w \tex_the:D \tex_toks:D #1 } } -\cs_new_protected:Npn \regex_toks_put_right:Nx #1#2 +\cs_new_protected:Npn \__regex_toks_put_right:Nx #1#2 { - \cs_set_nopar:Npx \regex_tmp:w {#2} + \cs_set_nopar:Npx \__regex_tmp:w {#2} \tex_toks:D #1 \exp_after:wN - { \tex_the:D \tex_toks:D \exp_after:wN #1 \regex_tmp:w } + { \tex_the:D \tex_toks:D \exp_after:wN #1 \__regex_tmp:w } } -\cs_new_protected:Npn \regex_toks_put_right:Nn #1#2 +\cs_new_protected:Npn \__regex_toks_put_right:Nn #1#2 { \tex_toks:D #1 \exp_after:wN { \tex_the:D \tex_toks:D #1 #2 } } -\cs_new_protected:Npn \regex_build_transition_left:NNN #1#2#3 - { \regex_toks_put_left:Nx #2 { #1 { \int_eval:n { #3 - #2 } } } } -\cs_new_protected:Npn \regex_build_transition_right:nNn #1#2#3 - { \regex_toks_put_right:Nx #2 { #1 { \int_eval:n { #3 - #2 } } } } -\cs_new_protected_nopar:Npn \regex_build_new_state: +\cs_new_protected:Npn \__regex_build_transition_left:NNN #1#2#3 + { \__regex_toks_put_left:Nx #2 { #1 { \int_eval:n { #3 - #2 } } } } +\cs_new_protected:Npn \__regex_build_transition_right:nNn #1#2#3 + { \__regex_toks_put_right:Nx #2 { #1 { \int_eval:n { #3 - #2 } } } } +\cs_new_protected_nopar:Npn \__regex_build_new_state: { - \tex_toks:D \l_regex_max_state_int { } - \int_set_eq:NN \l_regex_left_state_int \l_regex_right_state_int - \int_set_eq:NN \l_regex_right_state_int \l_regex_max_state_int - \int_incr:N \l_regex_max_state_int + \tex_toks:D \l__regex_max_state_int { } + \int_set_eq:NN \l__regex_left_state_int \l__regex_right_state_int + \int_set_eq:NN \l__regex_right_state_int \l__regex_max_state_int + \int_incr:N \l__regex_max_state_int } -\cs_new_protected:Npn \regex_build_transitions_lazyness:NNNNN #1#2#3#4#5 +\cs_new_protected:Npn \__regex_build_transitions_lazyness:NNNNN #1#2#3#4#5 { - \regex_build_new_state: - \regex_toks_put_right:Nx \l_regex_left_state_int + \__regex_build_new_state: + \__regex_toks_put_right:Nx \l__regex_left_state_int { \if_meaning:w \c_true_bool #1 - #2 { \int_eval:n { #3 - \l_regex_left_state_int } } - #4 { \int_eval:n { #5 - \l_regex_left_state_int } } + #2 { \int_eval:n { #3 - \l__regex_left_state_int } } + #4 { \int_eval:n { #5 - \l__regex_left_state_int } } \else: - #4 { \int_eval:n { #5 - \l_regex_left_state_int } } - #2 { \int_eval:n { #3 - \l_regex_left_state_int } } + #4 { \int_eval:n { #5 - \l__regex_left_state_int } } + #2 { \int_eval:n { #3 - \l__regex_left_state_int } } \fi: } } -\cs_new_protected:Npn \regex_class:NnnnN #1#2#3#4#5 +\cs_new_protected:Npn \__regex_class:NnnnN #1#2#3#4#5 { - \cs_set_nopar:Npx \regex_tests_action_cost:n ##1 + \cs_set_nopar:Npx \__regex_tests_action_cost:n ##1 { \exp_not:n { \exp_not:n {#2} } \bool_if:NTF #1 - { \regex_break_point:TF { \regex_action_cost:n {##1} } { } } - { \regex_break_point:TF { } { \regex_action_cost:n {##1} } } + { \__regex_break_point:TF { \__regex_action_cost:n {##1} } { } } + { \__regex_break_point:TF { } { \__regex_action_cost:n {##1} } } } \if_case:w - #4 \exp_stop_f: - \regex_class_repeat:n {#3} - \or: \regex_class_repeat:nN {#3} #5 - \else: \regex_class_repeat:nnN {#3} {#4} #5 + \__regex_class_repeat:n {#3} + \or: \__regex_class_repeat:nN {#3} #5 + \else: \__regex_class_repeat:nnN {#3} {#4} #5 \fi: } -\cs_new:Npn \regex_tests_action_cost:n { \regex_action_cost:n } -\cs_new_protected:Npn \regex_class_repeat:n #1 +\cs_new:Npn \__regex_tests_action_cost:n { \__regex_action_cost:n } +\cs_new_protected:Npn \__regex_class_repeat:n #1 { \prg_replicate:nn {#1} { - \regex_build_new_state: - \regex_build_transition_right:nNn \regex_tests_action_cost:n - \l_regex_left_state_int \l_regex_right_state_int + \__regex_build_new_state: + \__regex_build_transition_right:nNn \__regex_tests_action_cost:n + \l__regex_left_state_int \l__regex_right_state_int } } -\cs_new_protected:Npn \regex_class_repeat:nN #1#2 +\cs_new_protected:Npn \__regex_class_repeat:nN #1#2 { - \if_num:w #1 = \c_zero - \regex_build_transitions_lazyness:NNNNN #2 - \regex_action_free:n \l_regex_right_state_int - \regex_tests_action_cost:n \l_regex_left_state_int + \if_int_compare:w #1 = \c_zero + \__regex_build_transitions_lazyness:NNNNN #2 + \__regex_action_free:n \l__regex_right_state_int + \__regex_tests_action_cost:n \l__regex_left_state_int \else: - \regex_class_repeat:n {#1} - \int_set_eq:NN \l_regex_internal_a_int \l_regex_left_state_int - \regex_build_transitions_lazyness:NNNNN #2 - \regex_action_free:n \l_regex_right_state_int - \regex_action_free:n \l_regex_internal_a_int + \__regex_class_repeat:n {#1} + \int_set_eq:NN \l__regex_internal_a_int \l__regex_left_state_int + \__regex_build_transitions_lazyness:NNNNN #2 + \__regex_action_free:n \l__regex_right_state_int + \__regex_action_free:n \l__regex_internal_a_int \fi: } -\cs_new_protected:Npn \regex_class_repeat:nnN #1#2#3 +\cs_new_protected:Npn \__regex_class_repeat:nnN #1#2#3 { - \regex_class_repeat:n {#1} - \int_set:Nn \l_regex_internal_a_int - { \l_regex_max_state_int + #2 - \c_one } + \__regex_class_repeat:n {#1} + \int_set:Nn \l__regex_internal_a_int + { \l__regex_max_state_int + #2 - \c_one } \prg_replicate:nn { #2 } { - \regex_build_transitions_lazyness:NNNNN #3 - \regex_action_free:n \l_regex_internal_a_int - \regex_tests_action_cost:n \l_regex_right_state_int + \__regex_build_transitions_lazyness:NNNNN #3 + \__regex_action_free:n \l__regex_internal_a_int + \__regex_tests_action_cost:n \l__regex_right_state_int } } -\cs_new_protected:Npn \regex_group_aux:nnnnN #1#2#3#4#5 +\cs_new_protected:Npn \__regex_group_aux:nnnnN #1#2#3#4#5 { - \if_num:w #3 = \c_zero - \regex_build_new_state: - \regex_build_transition_right:nNn \regex_action_free_group:n - \l_regex_left_state_int \l_regex_right_state_int + \if_int_compare:w #3 = \c_zero + \__regex_build_new_state: + \__regex_build_transition_right:nNn \__regex_action_free_group:n + \l__regex_left_state_int \l__regex_right_state_int \fi: - \regex_build_new_state: - \regex_push_lr_states: + \__regex_build_new_state: + \__regex_push_lr_states: #2 - \regex_pop_lr_states: + \__regex_pop_lr_states: \if_case:w - #4 \exp_stop_f: - \regex_group_repeat:nn {#1} {#3} - \or: \regex_group_repeat:nnN {#1} {#3} #5 - \else: \regex_group_repeat:nnnN {#1} {#3} {#4} #5 + \__regex_group_repeat:nn {#1} {#3} + \or: \__regex_group_repeat:nnN {#1} {#3} #5 + \else: \__regex_group_repeat:nnnN {#1} {#3} {#4} #5 \fi: } -\cs_new_protected:Npn \regex_group:nnnN #1 +\cs_new_protected:Npn \__regex_group:nnnN #1 { - \exp_args:No \regex_group_aux:nnnnN - { \int_use:N \l_regex_capturing_group_int } + \exp_args:No \__regex_group_aux:nnnnN + { \int_use:N \l__regex_capturing_group_int } { - \int_incr:N \l_regex_capturing_group_int + \int_incr:N \l__regex_capturing_group_int #1 } } -\cs_new_protected_nopar:Npn \regex_group_no_capture:nnnN - { \regex_group_aux:nnnnN { -1 } } -\cs_new_protected:Npn \regex_group_resetting:nnnN #1 +\cs_new_protected_nopar:Npn \__regex_group_no_capture:nnnN + { \__regex_group_aux:nnnnN { -1 } } +\cs_new_protected:Npn \__regex_group_resetting:nnnN #1 { - \regex_group_aux:nnnnN { -1 } + \__regex_group_aux:nnnnN { -1 } { - \exp_args:Noo \regex_group_resetting_loop:nnNn - { \int_use:N \l_regex_capturing_group_int } - { \int_use:N \l_regex_capturing_group_int } + \exp_args:Noo \__regex_group_resetting_loop:nnNn + { \int_use:N \l__regex_capturing_group_int } + { \int_use:N \l__regex_capturing_group_int } #1 - { ?? \prg_map_break:n } { } - \prg_break_point:n { } + { ?? \__prg_break:n } { } + \__prg_break_point: } } -\cs_new_protected:Npn \regex_group_resetting_loop:nnNn #1#2#3#4 +\cs_new_protected:Npn \__regex_group_resetting_loop:nnNn #1#2#3#4 { - \use_none:nn #3 { \int_set:Nn \l_regex_capturing_group_int {#1} } - \int_set:Nn \l_regex_capturing_group_int {#2} + \use_none:nn #3 { \int_set:Nn \l__regex_capturing_group_int {#1} } + \int_set:Nn \l__regex_capturing_group_int {#2} #3 {#4} - \exp_args:Nf \regex_group_resetting_loop:nnNn - { \int_max:nn {#1} { \l_regex_capturing_group_int } } + \exp_args:Nf \__regex_group_resetting_loop:nnNn + { \int_max:nn {#1} { \l__regex_capturing_group_int } } {#2} } -\cs_new_protected:Npn \regex_branch:n #1 +\cs_new_protected:Npn \__regex_branch:n #1 { - \regex_build_new_state: - \seq_get:NN \l_regex_left_state_seq \l_regex_internal_a_tl - \int_set:Nn \l_regex_left_state_int \l_regex_internal_a_tl - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_left_state_int \l_regex_right_state_int + \__regex_build_new_state: + \seq_get:NN \l__regex_left_state_seq \l__regex_internal_a_tl + \int_set:Nn \l__regex_left_state_int \l__regex_internal_a_tl + \__regex_build_transition_right:nNn \__regex_action_free:n + \l__regex_left_state_int \l__regex_right_state_int #1 - \seq_get:NN \l_regex_right_state_seq \l_regex_internal_a_tl - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_right_state_int \l_regex_internal_a_tl + \seq_get:NN \l__regex_right_state_seq \l__regex_internal_a_tl + \__regex_build_transition_right:nNn \__regex_action_free:n + \l__regex_right_state_int \l__regex_internal_a_tl } -\cs_new_protected:Npn \regex_group_repeat:nn #1#2 +\cs_new_protected:Npn \__regex_group_repeat:nn #1#2 { - \if_num:w #2 = \c_zero - \int_set:Nn \l_regex_max_state_int - { \l_regex_left_state_int - \c_one } - \regex_build_new_state: + \if_int_compare:w #2 = \c_zero + \int_set:Nn \l__regex_max_state_int + { \l__regex_left_state_int - \c_one } + \__regex_build_new_state: \else: - \regex_group_repeat_aux:n {#2} - \regex_group_submatches:nNN {#1} - \l_regex_internal_a_int \l_regex_right_state_int - \regex_build_new_state: + \__regex_group_repeat_aux:n {#2} + \__regex_group_submatches:nNN {#1} + \l__regex_internal_a_int \l__regex_right_state_int + \__regex_build_new_state: \fi: } -\cs_new_protected:Npn \regex_group_submatches:nNN #1#2#3 +\cs_new_protected:Npn \__regex_group_submatches:nNN #1#2#3 { - \if_num:w #1 > \c_minus_one - \regex_toks_put_left:Nx #2 { \regex_action_submatch:n { #1 < } } - \regex_toks_put_left:Nx #3 { \regex_action_submatch:n { #1 > } } + \if_int_compare:w #1 > \c_minus_one + \__regex_toks_put_left:Nx #2 { \__regex_action_submatch:n { #1 < } } + \__regex_toks_put_left:Nx #3 { \__regex_action_submatch:n { #1 > } } \fi: } -\cs_new_protected:Npn \regex_group_repeat_aux:n #1 +\cs_new_protected:Npn \__regex_group_repeat_aux:n #1 { - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_right_state_int \l_regex_max_state_int - \int_set_eq:NN \l_regex_internal_a_int \l_regex_left_state_int - \int_set_eq:NN \l_regex_internal_b_int \l_regex_max_state_int - \if_num:w \int_eval:w #1 > \c_one - \int_set:Nn \l_regex_internal_c_int + \__regex_build_transition_right:nNn \__regex_action_free:n + \l__regex_right_state_int \l__regex_max_state_int + \int_set_eq:NN \l__regex_internal_a_int \l__regex_left_state_int + \int_set_eq:NN \l__regex_internal_b_int \l__regex_max_state_int + \if_int_compare:w \__int_eval:w #1 > \c_one + \int_set:Nn \l__regex_internal_c_int { ( #1 - \c_one ) - * ( \l_regex_internal_b_int - \l_regex_internal_a_int ) + * ( \l__regex_internal_b_int - \l__regex_internal_a_int ) } - \tex_advance:D \l_regex_right_state_int \l_regex_internal_c_int - \tex_advance:D \l_regex_max_state_int \l_regex_internal_c_int - \prg_replicate:nn \l_regex_internal_c_int + \tex_advance:D \l__regex_right_state_int \l__regex_internal_c_int + \tex_advance:D \l__regex_max_state_int \l__regex_internal_c_int + \prg_replicate:nn \l__regex_internal_c_int { - \tex_toks:D \l_regex_internal_b_int - = \tex_toks:D \l_regex_internal_a_int - \tex_advance:D \l_regex_internal_a_int \c_one - \tex_advance:D \l_regex_internal_b_int \c_one + \tex_toks:D \l__regex_internal_b_int + = \tex_toks:D \l__regex_internal_a_int + \tex_advance:D \l__regex_internal_a_int \c_one + \tex_advance:D \l__regex_internal_b_int \c_one } \fi: } -\cs_new_protected:Npn \regex_group_repeat:nnN #1#2#3 +\cs_new_protected:Npn \__regex_group_repeat:nnN #1#2#3 { - \if_num:w #2 = \c_zero - \regex_group_submatches:nNN {#1} - \l_regex_left_state_int \l_regex_right_state_int - \int_set:Nn \l_regex_internal_a_int - { \l_regex_left_state_int - \c_one } - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_right_state_int \l_regex_internal_a_int - \regex_build_new_state: + \if_int_compare:w #2 = \c_zero + \__regex_group_submatches:nNN {#1} + \l__regex_left_state_int \l__regex_right_state_int + \int_set:Nn \l__regex_internal_a_int + { \l__regex_left_state_int - \c_one } + \__regex_build_transition_right:nNn \__regex_action_free:n + \l__regex_right_state_int \l__regex_internal_a_int + \__regex_build_new_state: \if_meaning:w \c_true_bool #3 - \regex_build_transition_left:NNN \regex_action_free:n - \l_regex_internal_a_int \l_regex_right_state_int + \__regex_build_transition_left:NNN \__regex_action_free:n + \l__regex_internal_a_int \l__regex_right_state_int \else: - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_internal_a_int \l_regex_right_state_int + \__regex_build_transition_right:nNn \__regex_action_free:n + \l__regex_internal_a_int \l__regex_right_state_int \fi: \else: - \regex_group_repeat_aux:n {#2} - \regex_group_submatches:nNN {#1} - \l_regex_internal_a_int \l_regex_right_state_int + \__regex_group_repeat_aux:n {#2} + \__regex_group_submatches:nNN {#1} + \l__regex_internal_a_int \l__regex_right_state_int \if_meaning:w \c_true_bool #3 - \regex_build_transition_right:nNn \regex_action_free_group:n - \l_regex_right_state_int \l_regex_internal_a_int + \__regex_build_transition_right:nNn \__regex_action_free_group:n + \l__regex_right_state_int \l__regex_internal_a_int \else: - \regex_build_transition_left:NNN \regex_action_free_group:n - \l_regex_right_state_int \l_regex_internal_a_int + \__regex_build_transition_left:NNN \__regex_action_free_group:n + \l__regex_right_state_int \l__regex_internal_a_int \fi: - \regex_build_new_state: + \__regex_build_new_state: \fi: } -\cs_new_protected:Npn \regex_group_repeat:nnnN #1#2#3#4 +\cs_new_protected:Npn \__regex_group_repeat:nnnN #1#2#3#4 { - \regex_group_submatches:nNN {#1} - \l_regex_left_state_int \l_regex_right_state_int - \regex_group_repeat_aux:n { #2 + #3 } + \__regex_group_submatches:nNN {#1} + \l__regex_left_state_int \l__regex_right_state_int + \__regex_group_repeat_aux:n { #2 + #3 } \if_meaning:w \c_true_bool #4 - \int_set_eq:NN \l_regex_left_state_int \l_regex_max_state_int + \int_set_eq:NN \l__regex_left_state_int \l__regex_max_state_int \prg_replicate:nn { #3 } { - \int_sub:Nn \l_regex_left_state_int - { \l_regex_internal_b_int - \l_regex_internal_a_int } - \regex_build_transition_left:NNN \regex_action_free:n - \l_regex_left_state_int \l_regex_max_state_int + \int_sub:Nn \l__regex_left_state_int + { \l__regex_internal_b_int - \l__regex_internal_a_int } + \__regex_build_transition_left:NNN \__regex_action_free:n + \l__regex_left_state_int \l__regex_max_state_int } \else: \prg_replicate:nn { #3 - \c_one } { - \int_sub:Nn \l_regex_right_state_int - { \l_regex_internal_b_int - \l_regex_internal_a_int } - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_right_state_int \l_regex_max_state_int + \int_sub:Nn \l__regex_right_state_int + { \l__regex_internal_b_int - \l__regex_internal_a_int } + \__regex_build_transition_right:nNn \__regex_action_free:n + \l__regex_right_state_int \l__regex_max_state_int } - \if_num:w #2 = \c_zero - \int_set:Nn \l_regex_right_state_int - { \l_regex_left_state_int - \c_one } + \if_int_compare:w #2 = \c_zero + \int_set:Nn \l__regex_right_state_int + { \l__regex_left_state_int - \c_one } \else: - \int_sub:Nn \l_regex_right_state_int - { \l_regex_internal_b_int - \l_regex_internal_a_int } + \int_sub:Nn \l__regex_right_state_int + { \l__regex_internal_b_int - \l__regex_internal_a_int } \fi: - \regex_build_transition_right:nNn \regex_action_free:n - \l_regex_right_state_int \l_regex_max_state_int + \__regex_build_transition_right:nNn \__regex_action_free:n + \l__regex_right_state_int \l__regex_max_state_int \fi: - \regex_build_new_state: + \__regex_build_new_state: } -\cs_new_protected:Npn \regex_assertion:Nn #1#2 +\cs_new_protected:Npn \__regex_assertion:Nn #1#2 { - \regex_build_new_state: - \regex_toks_put_right:Nx \l_regex_left_state_int + \__regex_build_new_state: + \__regex_toks_put_right:Nx \l__regex_left_state_int { \exp_not:n {#2} - \regex_break_point:TF + \__regex_break_point:TF \bool_if:NF #1 { { } } { - \regex_action_free:n + \__regex_action_free:n { \int_eval:n - { \l_regex_right_state_int - \l_regex_left_state_int } + { \l__regex_right_state_int - \l__regex_left_state_int } } } \bool_if:NT #1 { { } } } } -\cs_new_protected:Npn \regex_anchor:N #1 +\cs_new_protected:Npn \__regex_anchor:N #1 { - \if_num:w #1 = \l_regex_current_pos_int - \exp_after:wN \regex_break_true:w + \if_int_compare:w #1 = \l__regex_current_pos_int + \exp_after:wN \__regex_break_true:w \fi: } -\cs_new_protected_nopar:Npn \regex_b_test: +\cs_new_protected_nopar:Npn \__regex_b_test: { \group_begin: - \int_set_eq:NN \l_regex_current_char_int \l_regex_last_char_int - \regex_prop_w: - \regex_break_point:TF - { \group_end: \regex_item_reverse:n \regex_prop_w: } - { \group_end: \regex_prop_w: } - } -\cs_new_protected_nopar:Npn \regex_command_K: - { - \regex_build_new_state: - \regex_toks_put_right:Nx \l_regex_left_state_int - { - \regex_action_submatch:n { 0< } - \bool_set_true:N \l_regex_fresh_thread_bool - \regex_action_free:n - { \int_eval:n { \l_regex_right_state_int - \l_regex_left_state_int } } - \bool_set_false:N \l_regex_fresh_thread_bool - } - } -\int_new:N \l_regex_min_pos_int -\int_new:N \l_regex_max_pos_int -\int_new:N \l_regex_current_pos_int -\int_new:N \l_regex_start_pos_int -\int_new:N \l_regex_success_pos_int -\int_new:N \l_regex_current_char_int -\int_new:N \l_regex_current_catcode_int -\int_new:N \l_regex_last_char_int -\int_new:N \l_regex_case_changed_char_int -\int_new:N \l_regex_current_state_int -\prop_new:N \l_regex_current_submatches_prop -\prop_new:N \l_regex_success_submatches_prop -\int_new:N \l_regex_step_int -\int_new:N \l_regex_min_active_int -\int_new:N \l_regex_max_active_int -\tl_new:N \l_regex_every_match_tl -\bool_new:N \l_regex_fresh_thread_bool -\bool_new:N \l_regex_empty_success_bool -\cs_new_eq:NN \regex_if_two_empty_matches:F \use:n -\bool_new:N \g_regex_success_bool -\bool_new:N \l_regex_saved_success_bool -\bool_new:N \l_regex_match_success_bool -\cs_new_protected:Npn \regex_match:n #1 - { - \int_zero:N \l_regex_balance_int - \int_set:Nn \l_regex_current_pos_int { \c_two * \l_regex_max_state_int } - \regex_query_set:nnn { } { -1 } { -2 } - \int_set_eq:NN \l_regex_min_pos_int \l_regex_current_pos_int - \tl_analysis_map_inline:nn {#1} - { \regex_query_set:nnn {##1} {"##2} {##3} } - \int_set_eq:NN \l_regex_max_pos_int \l_regex_current_pos_int - \regex_query_set:nnn { } { -1 } { -2 } - \bool_gset_false:N \g_regex_success_bool - \prg_stepwise_inline:nnnn - \l_regex_min_state_int \c_one { \l_regex_max_state_int - \c_one } + \int_set_eq:NN \l__regex_current_char_int \l__regex_last_char_int + \__regex_prop_w: + \__regex_break_point:TF + { \group_end: \__regex_item_reverse:n \__regex_prop_w: } + { \group_end: \__regex_prop_w: } + } +\cs_new_protected_nopar:Npn \__regex_command_K: + { + \__regex_build_new_state: + \__regex_toks_put_right:Nx \l__regex_left_state_int + { + \__regex_action_submatch:n { 0< } + \bool_set_true:N \l__regex_fresh_thread_bool + \__regex_action_free:n + { \int_eval:n { \l__regex_right_state_int - \l__regex_left_state_int } } + \bool_set_false:N \l__regex_fresh_thread_bool + } + } +\int_new:N \l__regex_min_pos_int +\int_new:N \l__regex_max_pos_int +\int_new:N \l__regex_current_pos_int +\int_new:N \l__regex_start_pos_int +\int_new:N \l__regex_success_pos_int +\int_new:N \l__regex_current_char_int +\int_new:N \l__regex_current_catcode_int +\int_new:N \l__regex_last_char_int +\int_new:N \l__regex_case_changed_char_int +\int_new:N \l__regex_current_state_int +\prop_new:N \l__regex_current_submatches_prop +\prop_new:N \l__regex_success_submatches_prop +\int_new:N \l__regex_step_int +\int_new:N \l__regex_min_active_int +\int_new:N \l__regex_max_active_int +\tl_new:N \l__regex_every_match_tl +\bool_new:N \l__regex_fresh_thread_bool +\bool_new:N \l__regex_empty_success_bool +\cs_new_eq:NN \__regex_if_two_empty_matches:F \use:n +\bool_new:N \g__regex_success_bool +\bool_new:N \l__regex_saved_success_bool +\bool_new:N \l__regex_match_success_bool +\cs_new_protected:Npn \__regex_match:n #1 + { + \int_zero:N \l__regex_balance_int + \int_set:Nn \l__regex_current_pos_int { \c_two * \l__regex_max_state_int } + \__regex_query_set:nnn { } { -1 } { -2 } + \int_set_eq:NN \l__regex_min_pos_int \l__regex_current_pos_int + \__tl_analysis_map_inline:nn {#1} + { \__regex_query_set:nnn {##1} {"##2} {##3} } + \int_set_eq:NN \l__regex_max_pos_int \l__regex_current_pos_int + \__regex_query_set:nnn { } { -1 } { -2 } + \bool_gset_false:N \g__regex_success_bool + \int_step_inline:nnnn + \l__regex_min_state_int \c_one { \l__regex_max_state_int - \c_one } { \tex_dimen:D ##1 \c_one sp \scan_stop: } - \int_set_eq:NN \l_regex_min_active_int \l_regex_max_state_int - \int_set_eq:NN \l_regex_step_int \c_zero - \int_set_eq:NN \l_regex_success_pos_int \l_regex_min_pos_int - \int_set:Nn \l_regex_submatch_int - { \c_two * \l_regex_max_state_int } - \bool_set_false:N \l_regex_empty_success_bool - \regex_match_once: - } -\cs_new_protected_nopar:Npn \regex_match_once: - { - \if_meaning:w \c_true_bool \l_regex_empty_success_bool - \cs_set_nopar:Npn \regex_if_two_empty_matches:F - { \int_compare:nNnF \l_regex_start_pos_int = \l_regex_current_pos_int } + \int_set_eq:NN \l__regex_min_active_int \l__regex_max_state_int + \int_set_eq:NN \l__regex_step_int \c_zero + \int_set_eq:NN \l__regex_success_pos_int \l__regex_min_pos_int + \int_set:Nn \l__regex_submatch_int + { \c_two * \l__regex_max_state_int } + \bool_set_false:N \l__regex_empty_success_bool + \__regex_match_once: + } +\cs_new_protected_nopar:Npn \__regex_match_once: + { + \if_meaning:w \c_true_bool \l__regex_empty_success_bool + \cs_set_nopar:Npn \__regex_if_two_empty_matches:F + { \int_compare:nNnF \l__regex_start_pos_int = \l__regex_current_pos_int } \else: - \cs_set_eq:NN \regex_if_two_empty_matches:F \use:n + \cs_set_eq:NN \__regex_if_two_empty_matches:F \use:n \fi: - \int_set_eq:NN \l_regex_start_pos_int \l_regex_success_pos_int - \bool_set_false:N \l_regex_match_success_bool - \prop_clear:N \l_regex_current_submatches_prop - \int_set_eq:NN \l_regex_max_active_int \l_regex_min_active_int - \regex_store_state:n { \l_regex_min_state_int } - \int_set:Nn \l_regex_current_pos_int - { \l_regex_start_pos_int - \c_one } - \regex_query_get: - \regex_match_loop: - \l_regex_every_match_tl + \int_set_eq:NN \l__regex_start_pos_int \l__regex_success_pos_int + \bool_set_false:N \l__regex_match_success_bool + \prop_clear:N \l__regex_current_submatches_prop + \int_set_eq:NN \l__regex_max_active_int \l__regex_min_active_int + \__regex_store_state:n { \l__regex_min_state_int } + \int_set:Nn \l__regex_current_pos_int + { \l__regex_start_pos_int - \c_one } + \__regex_query_get: + \__regex_match_loop: + \l__regex_every_match_tl } -\cs_new_protected_nopar:Npn \regex_single_match: +\cs_new_protected_nopar:Npn \__regex_single_match: { - \tl_set:Nn \l_regex_every_match_tl - { \bool_gset_eq:NN \g_regex_success_bool \l_regex_match_success_bool } + \tl_set:Nn \l__regex_every_match_tl + { \bool_gset_eq:NN \g__regex_success_bool \l__regex_match_success_bool } } -\cs_new_protected:Npn \regex_multi_match:n #1 +\cs_new_protected:Npn \__regex_multi_match:n #1 { - \tl_set:Nn \l_regex_every_match_tl + \tl_set:Nn \l__regex_every_match_tl { - \if_meaning:w \c_true_bool \l_regex_match_success_bool - \bool_gset_true:N \g_regex_success_bool + \if_meaning:w \c_true_bool \l__regex_match_success_bool + \bool_gset_true:N \g__regex_success_bool #1 - \exp_after:wN \regex_match_once: + \exp_after:wN \__regex_match_once: \fi: } } -\cs_new_protected_nopar:Npn \regex_match_loop: +\cs_new_protected_nopar:Npn \__regex_match_loop: { - \tex_advance:D \l_regex_step_int \c_two - \int_incr:N \l_regex_current_pos_int - \int_set_eq:NN \l_regex_last_char_int \l_regex_current_char_int - \int_set_eq:NN \l_regex_case_changed_char_int \c_max_int - \regex_query_get: + \tex_advance:D \l__regex_step_int \c_two + \int_incr:N \l__regex_current_pos_int + \int_set_eq:NN \l__regex_last_char_int \l__regex_current_char_int + \int_set_eq:NN \l__regex_case_changed_char_int \c_max_int + \__regex_query_get: \use:x { - \int_set_eq:NN \l_regex_max_active_int \l_regex_min_active_int - \exp_after:wN \regex_match_one_active:w - \int_use:N \l_regex_min_active_int ; + \int_set_eq:NN \l__regex_max_active_int \l__regex_min_active_int + \exp_after:wN \__regex_match_one_active:w + \int_use:N \l__regex_min_active_int ; } - \prg_break_point:n { \bool_set_false:N \l_regex_fresh_thread_bool } - \if_num:w \l_regex_max_active_int > \l_regex_min_active_int - \if_num:w \l_regex_current_pos_int < \l_regex_max_pos_int - \exp_after:wN \exp_after:wN \exp_after:wN \regex_match_loop: + \__prg_break_point: + \bool_set_false:N \l__regex_fresh_thread_bool %^^A was arg of break_point:n + \if_int_compare:w \l__regex_max_active_int > \l__regex_min_active_int + \if_int_compare:w \l__regex_current_pos_int < \l__regex_max_pos_int + \exp_after:wN \exp_after:wN \exp_after:wN \__regex_match_loop: \fi: \fi: } -\cs_new:Npn \regex_match_one_active:w #1; +\cs_new:Npn \__regex_match_one_active:w #1; { - \if_num:w #1 < \l_regex_max_active_int - \regex_use_state_and_submatches:nn - { \int_value:w \tex_skip:D #1 } + \if_int_compare:w #1 < \l__regex_max_active_int + \__regex_use_state_and_submatches:nn + { \__int_value:w \tex_skip:D #1 } { \tex_the:D \tex_toks:D #1 } - \exp_after:wN \regex_match_one_active:w - \int_use:N \int_eval:w #1 + \c_one \exp_after:wN ; + \exp_after:wN \__regex_match_one_active:w + \int_use:N \__int_eval:w #1 + \c_one \exp_after:wN ; \fi: } -\cs_new_protected:Npn \regex_query_set:nnn #1#2#3 +\cs_new_protected:Npn \__regex_query_set:nnn #1#2#3 { - \tex_muskip:D \l_regex_current_pos_int + \tex_muskip:D \l__regex_current_pos_int = \etex_gluetomu:D #3 sp plus #2 sp - minus \l_regex_balance_int sp + minus \l__regex_balance_int sp \scan_stop: - \tex_toks:D \l_regex_current_pos_int {#1} - \int_incr:N \l_regex_current_pos_int + \tex_toks:D \l__regex_current_pos_int {#1} + \int_incr:N \l__regex_current_pos_int \if_case:w #2 \exp_stop_f: - \or: \int_incr:N \l_regex_balance_int - \or: \int_decr:N \l_regex_balance_int + \or: \int_incr:N \l__regex_balance_int + \or: \int_decr:N \l__regex_balance_int \fi: } -\cs_new_protected_nopar:Npn \regex_query_get: +\cs_new_protected_nopar:Npn \__regex_query_get: { - \l_regex_current_char_int - = \etex_mutoglue:D \tex_muskip:D \l_regex_current_pos_int - \l_regex_current_catcode_int = \etex_gluestretch:D - \etex_mutoglue:D \tex_muskip:D \l_regex_current_pos_int + \l__regex_current_char_int + = \etex_mutoglue:D \tex_muskip:D \l__regex_current_pos_int + \l__regex_current_catcode_int = \etex_gluestretch:D + \etex_mutoglue:D \tex_muskip:D \l__regex_current_pos_int } -\cs_new_protected_nopar:Npn \regex_use_state: +\cs_new_protected_nopar:Npn \__regex_use_state: { - \tex_dimen:D \l_regex_current_state_int - = \l_regex_step_int sp \scan_stop: - \tex_the:D \tex_toks:D \l_regex_current_state_int - \tex_dimen:D \l_regex_current_state_int - = \int_eval:w \l_regex_step_int + \c_one \int_eval_end: sp \scan_stop: + \tex_dimen:D \l__regex_current_state_int + = \l__regex_step_int sp \scan_stop: + \tex_the:D \tex_toks:D \l__regex_current_state_int + \tex_dimen:D \l__regex_current_state_int + = \__int_eval:w \l__regex_step_int + \c_one \__int_eval_end: sp \scan_stop: } -\cs_new_protected:Npn \regex_use_state_and_submatches:nn #1 #2 +\cs_new_protected:Npn \__regex_use_state_and_submatches:nn #1 #2 { - \int_set:Nn \l_regex_current_state_int {#1} - \if_num:w \tex_dimen:D \l_regex_current_state_int < \l_regex_step_int - \tl_set:Nn \l_regex_current_submatches_prop {#2} - \exp_after:wN \regex_use_state: + \int_set:Nn \l__regex_current_state_int {#1} + \if_int_compare:w \tex_dimen:D \l__regex_current_state_int + < \l__regex_step_int + \tl_set:Nn \l__regex_current_submatches_prop {#2} + \exp_after:wN \__regex_use_state: \fi: \scan_stop: } -\cs_new_protected_nopar:Npn \regex_action_start_wildcard: +\cs_new_protected_nopar:Npn \__regex_action_start_wildcard: { - \bool_set_true:N \l_regex_fresh_thread_bool - \regex_action_free:n {1} - \bool_set_false:N \l_regex_fresh_thread_bool - \regex_action_cost:n {0} + \bool_set_true:N \l__regex_fresh_thread_bool + \__regex_action_free:n {1} + \bool_set_false:N \l__regex_fresh_thread_bool + \__regex_action_cost:n {0} } -\cs_new_protected_nopar:Npn \regex_action_free:n - { \regex_action_free_aux:nn { > \l_regex_step_int \else: } } -\cs_new_protected_nopar:Npn \regex_action_free_group:n - { \regex_action_free_aux:nn { < \l_regex_step_int } } -\cs_new_protected:Npn \regex_action_free_aux:nn #1#2 +\cs_new_protected_nopar:Npn \__regex_action_free:n + { \__regex_action_free_aux:nn { > \l__regex_step_int \else: } } +\cs_new_protected_nopar:Npn \__regex_action_free_group:n + { \__regex_action_free_aux:nn { < \l__regex_step_int } } +\cs_new_protected:Npn \__regex_action_free_aux:nn #1#2 { \use:x { - \int_add:Nn \l_regex_current_state_int {#2} + \int_add:Nn \l__regex_current_state_int {#2} \exp_not:n { - \if_num:w \tex_dimen:D \l_regex_current_state_int #1 - \exp_after:wN \regex_use_state: + \if_int_compare:w \tex_dimen:D \l__regex_current_state_int #1 + \exp_after:wN \__regex_use_state: \fi: } - \int_set:Nn \l_regex_current_state_int - { \int_use:N \l_regex_current_state_int } - \tl_set:Nn \exp_not:N \l_regex_current_submatches_prop - { \exp_not:o \l_regex_current_submatches_prop } + \int_set:Nn \l__regex_current_state_int + { \int_use:N \l__regex_current_state_int } + \tl_set:Nn \exp_not:N \l__regex_current_submatches_prop + { \exp_not:o \l__regex_current_submatches_prop } } } -\cs_new_protected:Npn \regex_action_cost:n #1 +\cs_new_protected:Npn \__regex_action_cost:n #1 { - \exp_args:No \regex_store_state:n - { \int_use:N \int_eval:w \l_regex_current_state_int + #1 } + \exp_args:No \__regex_store_state:n + { \int_use:N \__int_eval:w \l__regex_current_state_int + #1 } } -\cs_new_protected:Npn \regex_store_state:n #1 +\cs_new_protected:Npn \__regex_store_state:n #1 { - \regex_store_submatches: - \tex_skip:D \l_regex_max_active_int = #1 sp \scan_stop: - \int_incr:N \l_regex_max_active_int + \__regex_store_submatches: + \tex_skip:D \l__regex_max_active_int = #1 sp \scan_stop: + \int_incr:N \l__regex_max_active_int } -\cs_new_protected_nopar:Npn \regex_store_submatches: +\cs_new_protected_nopar:Npn \__regex_store_submatches: { - \tex_toks:D \l_regex_max_active_int \exp_after:wN - { \l_regex_current_submatches_prop } + \tex_toks:D \l__regex_max_active_int \exp_after:wN + { \l__regex_current_submatches_prop } } -\cs_new_protected_nopar:Npn \regex_disable_submatches: +\cs_new_protected_nopar:Npn \__regex_disable_submatches: { - \cs_set_protected_nopar:Npn \regex_store_submatches: { } - \cs_set_protected:Npn \regex_action_submatch:n ##1 { } + \cs_set_protected_nopar:Npn \__regex_store_submatches: { } + \cs_set_protected:Npn \__regex_action_submatch:n ##1 { } } -\cs_new_protected:Npn \regex_action_submatch:n #1 +\cs_new_protected:Npn \__regex_action_submatch:n #1 { - \prop_put:Nno \l_regex_current_submatches_prop {#1} - { \int_use:N \l_regex_current_pos_int } + \prop_put:Nno \l__regex_current_submatches_prop {#1} + { \int_use:N \l__regex_current_pos_int } } -\cs_new_protected_nopar:Npn \regex_action_success: +\cs_new_protected_nopar:Npn \__regex_action_success: { - \regex_if_two_empty_matches:F + \__regex_if_two_empty_matches:F { - \bool_set_true:N \l_regex_match_success_bool - \bool_set_eq:NN \l_regex_empty_success_bool - \l_regex_fresh_thread_bool - \int_set_eq:NN \l_regex_success_pos_int \l_regex_current_pos_int - \prop_set_eq:NN \l_regex_success_submatches_prop - \l_regex_current_submatches_prop - \prg_map_break: + \bool_set_true:N \l__regex_match_success_bool + \bool_set_eq:NN \l__regex_empty_success_bool + \l__regex_fresh_thread_bool + \int_set_eq:NN \l__regex_success_pos_int \l__regex_current_pos_int + \prop_set_eq:NN \l__regex_success_submatches_prop + \l__regex_current_submatches_prop + \__prg_break: } } -\int_new:N \l_regex_replacement_csnames_int -\tl_new:N \l_regex_balance_tl -\cs_new:Npn \regex_replacement_balance_one_match:n #1 - { - \regex_submatch_balance:n {#1} } -\cs_new:Npn \regex_replacement_do_one_match:n #1 +\int_new:N \l__regex_replacement_csnames_int +\tl_new:N \l__regex_balance_tl +\cs_new:Npn \__regex_replacement_balance_one_match:n #1 + { - \__regex_submatch_balance:n {#1} } +\cs_new:Npn \__regex_replacement_do_one_match:n #1 { - \regex_query_range:nn + \__regex_query_range:nn { \etex_glueshrink:D \tex_skip:D #1 } { \tex_skip:D #1 } } -\cs_new:Npn \regex_replacement_exp_not:N #1 { \exp_not:n {#1} } -\cs_new:Npn \regex_query_range:nn #1#2 +\cs_new:Npn \__regex_replacement_exp_not:N #1 { \exp_not:n {#1} } +\cs_new:Npn \__regex_query_range:nn #1#2 { - \exp_after:wN \regex_query_range_loop:ww - \int_use:N \int_eval:w #1 \exp_after:wN ; - \int_use:N \int_eval:w #2 ; - \prg_break_point:n { } + \exp_after:wN \__regex_query_range_loop:ww + \int_use:N \__int_eval:w #1 \exp_after:wN ; + \int_use:N \__int_eval:w #2 ; + \__prg_break_point: } -\cs_new:Npn \regex_query_range_loop:ww #1 ; #2 ; +\cs_new:Npn \__regex_query_range_loop:ww #1 ; #2 ; { - \if_num:w #1 < #2 \exp_stop_f: + \if_int_compare:w #1 < #2 \exp_stop_f: \else: - \exp_after:wN \prg_map_break: + \exp_after:wN \__prg_break: \fi: \tex_the:D \tex_toks:D #1 \exp_stop_f: - \exp_after:wN \regex_query_range_loop:ww - \int_use:N \int_eval:w #1 + \c_one ; #2 ; + \exp_after:wN \__regex_query_range_loop:ww + \int_use:N \__int_eval:w #1 + \c_one ; #2 ; } -\cs_new:Npn \regex_query_submatch:n #1 +\cs_new:Npn \__regex_query_submatch:n #1 { - \regex_query_range:nn - { \tex_skip:D \int_eval:w #1 } - { \etex_gluestretch:D \tex_skip:D \int_eval:w #1 } + \__regex_query_range:nn + { \tex_skip:D \__int_eval:w #1 } + { \etex_gluestretch:D \tex_skip:D \__int_eval:w #1 } } -\cs_new_protected:Npn \regex_submatch_balance:n #1 +\cs_new_protected:Npn \__regex_submatch_balance:n #1 { \etex_glueshrink:D \etex_mutoglue:D \etex_muexpr:D \tex_muskip:D \etex_gluestretch:D \tex_skip:D #1 - \tex_muskip:D \tex_skip:D #1 \scan_stop: } -\cs_new_protected:Npn \regex_replacement:n #1 +\cs_new_protected:Npn \__regex_replacement:n #1 { - \tl_set_build:Nw \l_regex_internal_a_tl - \int_zero:N \l_regex_balance_int - \tl_clear:N \l_regex_balance_tl - \regex_escape_use:nnnn + \__tl_build:Nw \l__regex_internal_a_tl + \int_zero:N \l__regex_balance_int + \tl_clear:N \l__regex_balance_tl + \__regex_escape_use:nnnn { \if_charcode:w \c_rbrace_str ##1 - \regex_replacement_rbrace:N \else: \tl_build_one:n \fi: ##1 + \__regex_replacement_rbrace:N \else: \__tl_build_one:n \fi: ##1 } - { \regex_replacement_escaped:N ##1 } - { \tl_build_one:n ##1 } + { \__regex_replacement_escaped:N ##1 } + { \__tl_build_one:n ##1 } {#1} \prg_do_nothing: \prg_do_nothing: - \if_int_compare:w \l_regex_replacement_csnames_int > \c_zero - \msg_kernel_error:nnx { regex } { replacement-missing-rbrace } - { \int_use:N \l_regex_replacement_csnames_int } - \tl_build_one:x - { \prg_replicate:nn \l_regex_replacement_csnames_int \cs_end: } + \if_int_compare:w \l__regex_replacement_csnames_int > \c_zero + \__msg_kernel_error:nnx { regex } { replacement-missing-rbrace } + { \int_use:N \l__regex_replacement_csnames_int } + \__tl_build_one:x + { \prg_replicate:nn \l__regex_replacement_csnames_int \cs_end: } \fi: - \cs_gset:Npx \regex_replacement_balance_one_match:n ##1 + \cs_gset:Npx \__regex_replacement_balance_one_match:n ##1 { - + \int_use:N \l_regex_balance_int - \l_regex_balance_tl - - \regex_submatch_balance:n {##1} + + \int_use:N \l__regex_balance_int + \l__regex_balance_tl + - \__regex_submatch_balance:n {##1} } - \tl_build_end: - \exp_args:No \regex_replacement_aux:n \l_regex_internal_a_tl + \__tl_build_end: + \exp_args:No \__regex_replacement_aux:n \l__regex_internal_a_tl } -\cs_new_protected:Npn \regex_replacement_aux:n #1 +\cs_new_protected:Npn \__regex_replacement_aux:n #1 { - \cs_set:Npn \regex_replacement_do_one_match:n ##1 + \cs_set:Npn \__regex_replacement_do_one_match:n ##1 { - \regex_query_range:nn + \__regex_query_range:nn { \etex_glueshrink:D \tex_skip:D ##1 } { \tex_skip:D ##1 } #1 } } -\cs_new_protected:Npn \regex_replacement_escaped:N #1 +\cs_new_protected:Npn \__regex_replacement_escaped:N #1 { - \cs_if_exist_use:cF { regex_replacement_#1:w } + \cs_if_exist_use:cF { __regex_replacement_#1:w } { - \if_num:w \c_one < 1#1 \exp_stop_f: - \regex_replacement_put_submatch:n {#1} + \if_int_compare:w \c_one < 1#1 \exp_stop_f: + \__regex_replacement_put_submatch:n {#1} \else: - \tl_build_one:n #1 + \__tl_build_one:n #1 \fi: } } -\cs_new_protected:Npn \regex_replacement_put_submatch:n #1 +\cs_new_protected:Npn \__regex_replacement_put_submatch:n #1 { - \if_num:w #1 < \l_regex_capturing_group_int - \tl_build_one:n { \regex_query_submatch:n { #1 + ##1 } } - \if_num:w \l_regex_replacement_csnames_int = \c_zero - \tl_put_right:Nn \l_regex_balance_tl - { + \regex_submatch_balance:n { \int_eval:w #1+##1 \int_eval_end: } } + \if_int_compare:w #1 < \l__regex_capturing_group_int + \__tl_build_one:n { \__regex_query_submatch:n { #1 + ##1 } } + \if_int_compare:w \l__regex_replacement_csnames_int = \c_zero + \tl_put_right:Nn \l__regex_balance_tl + { + \__regex_submatch_balance:n { \__int_eval:w #1+##1 \__int_eval_end: } } \fi: \fi: } -\cs_new_protected:Npn \regex_replacement_g:w #1#2 +\cs_new_protected:Npn \__regex_replacement_g:w #1#2 { - \str_if_eq:xxTF { #1#2 } { \tl_build_one:n \c_lbrace_str } + \str_if_eq_x:nnTF { #1#2 } { \__tl_build_one:n \c_lbrace_str } { - \int_zero:N \l_regex_internal_a_int - \regex_replacement_g_digits:NN + \int_zero:N \l__regex_internal_a_int + \__regex_replacement_g_digits:NN } - { \regex_replacement_error:NNN g #1 #2 } + { \__regex_replacement_error:NNN g #1 #2 } } -\cs_new_protected:Npn \regex_replacement_g_digits:NN #1#2 +\cs_new_protected:Npn \__regex_replacement_g_digits:NN #1#2 { - \token_if_eq_meaning:NNTF #1 \tl_build_one:n + \token_if_eq_meaning:NNTF #1 \__tl_build_one:n { - \if_num:w \c_one < 1#2 \exp_stop_f: - \int_set:Nn \l_regex_internal_a_int - { \c_ten * \l_regex_internal_a_int + #2 } + \if_int_compare:w \c_one < 1#2 \exp_stop_f: + \int_set:Nn \l__regex_internal_a_int + { \c_ten * \l__regex_internal_a_int + #2 } \exp_after:wN \use_i:nnn - \exp_after:wN \regex_replacement_g_digits:NN + \exp_after:wN \__regex_replacement_g_digits:NN \else: - \exp_after:wN \regex_replacement_error:NNN + \exp_after:wN \__regex_replacement_error:NNN \exp_after:wN g \fi: } { - \if_meaning:w \regex_replacement_rbrace:N #1 - \exp_args:No \regex_replacement_put_submatch:n - { \int_use:N \l_regex_internal_a_int } + \if_meaning:w \__regex_replacement_rbrace:N #1 + \exp_args:No \__regex_replacement_put_submatch:n + { \int_use:N \l__regex_internal_a_int } \exp_after:wN \use_none:nn \else: - \exp_after:wN \regex_replacement_error:NNN + \exp_after:wN \__regex_replacement_error:NNN \exp_after:wN g \fi: } #1 #2 } -\cs_new_protected:Npn \regex_replacement_c:w #1#2 +\cs_new_protected:Npn \__regex_replacement_c:w #1#2 { - \token_if_eq_meaning:NNTF #1 \tl_build_one:n + \token_if_eq_meaning:NNTF #1 \__tl_build_one:n { - \cs_if_exist_use:cF { regex_replacement_c_#2:w } - { \regex_replacement_error:NNN c #1#2 } + \cs_if_exist_use:cF { __regex_replacement_c_#2:w } + { \__regex_replacement_error:NNN c #1#2 } } - { \regex_replacement_error:NNN c #1#2 } + { \__regex_replacement_error:NNN c #1#2 } } -\cs_new_protected_nopar:cpn { regex_replacement_c_ \c_lbrace_str :w } +\cs_new_protected_nopar:cpn { __regex_replacement_c_ \c_lbrace_str :w } { - \if_case:w \l_regex_replacement_csnames_int - \tl_build_one:n - { \exp_not:n { \exp_after:wN \regex_replacement_exp_not:N \cs:w } } + \if_case:w \l__regex_replacement_csnames_int + \__tl_build_one:n + { \exp_not:n { \exp_after:wN \__regex_replacement_exp_not:N \cs:w } } \else: - \tl_build_one:n { \exp_not:n { \exp_after:wN \tl_to_str:N \cs:w } } + \__tl_build_one:n { \exp_not:n { \exp_after:wN \tl_to_str:N \cs:w } } \fi: - \int_incr:N \l_regex_replacement_csnames_int + \int_incr:N \l__regex_replacement_csnames_int } -\cs_new_protected:Npn \regex_replacement_u:w #1#2 +\cs_new_protected:Npn \__regex_replacement_u:w #1#2 { - \str_if_eq:xxTF { #1#2 } { \tl_build_one:n \c_lbrace_str } + \str_if_eq_x:nnTF { #1#2 } { \__tl_build_one:n \c_lbrace_str } { - \if_case:w \l_regex_replacement_csnames_int - \tl_build_one:n { \exp_not:n { \exp_after:wN \exp_not:V \cs:w } } + \if_case:w \l__regex_replacement_csnames_int + \__tl_build_one:n { \exp_not:n { \exp_after:wN \exp_not:V \cs:w } } \else: - \tl_build_one:n { \exp_not:n { \exp_after:wN \tl_to_str:V \cs:w } } + \__tl_build_one:n { \exp_not:n { \exp_after:wN \tl_to_str:V \cs:w } } \fi: - \int_incr:N \l_regex_replacement_csnames_int + \int_incr:N \l__regex_replacement_csnames_int } - { \regex_replacement_error:NNN u #1#2 } + { \__regex_replacement_error:NNN u #1#2 } } -\cs_new_protected:Npn \regex_replacement_rbrace:N #1 +\cs_new_protected:Npn \__regex_replacement_rbrace:N #1 { - \if_int_compare:w \l_regex_replacement_csnames_int > \c_zero - \tl_build_one:n \cs_end: - \int_decr:N \l_regex_replacement_csnames_int + \if_int_compare:w \l__regex_replacement_csnames_int > \c_zero + \__tl_build_one:n \cs_end: + \int_decr:N \l__regex_replacement_csnames_int \else: - \tl_build_one:n #1 + \__tl_build_one:n #1 \fi: } \group_begin: - \cs_new_protected:Npn \regex_replacement_char:nNN #1#2#3 + \cs_new_protected:Npn \__regex_replacement_char:nNN #1#2#3 { \if_meaning:w \prg_do_nothing: #3 - \msg_kernel_error:nn { regex } { replacement-catcode-end } + \__msg_kernel_error:nn { regex } { replacement-catcode-end } \else: \tex_lccode:D \c_zero = `#3 \scan_stop: - \tl_to_lowercase:n { \tl_build_one:n {#1} } + \tl_to_lowercase:n { \__tl_build_one:n {#1} } \fi: } \char_set_catcode_active:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_A:w - { \regex_replacement_char:nNN { \exp_not:n { \exp_not:N ^^@ } } } + \cs_new_protected_nopar:Npn \__regex_replacement_c_A:w + { \__regex_replacement_char:nNN { \exp_not:n { \exp_not:N ^^@ } } } \char_set_catcode_group_begin:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_B:w + \cs_new_protected_nopar:Npn \__regex_replacement_c_B:w { - \if_num:w \l_regex_replacement_csnames_int = \c_zero - \int_incr:N \l_regex_balance_int + \if_int_compare:w \l__regex_replacement_csnames_int = \c_zero + \int_incr:N \l__regex_balance_int \fi: - \regex_replacement_char:nNN + \__regex_replacement_char:nNN { \exp_not:n { \exp_after:wN ^^@ \if_false: } \fi: } } } - \cs_new_protected:Npn \regex_replacement_c_C:w #1#2 - { \tl_build_one:n { \exp_not:N \exp_not:N \exp_not:c {#2} } } + \cs_new_protected:Npn \__regex_replacement_c_C:w #1#2 + { \__tl_build_one:n { \exp_not:N \exp_not:N \exp_not:c {#2} } } \char_set_catcode_math_subscript:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_D:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \__regex_replacement_c_D:w + { \__regex_replacement_char:nNN { ^^@ } } \char_set_catcode_group_end:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_E:w + \cs_new_protected_nopar:Npn \__regex_replacement_c_E:w { - \if_num:w \l_regex_replacement_csnames_int = \c_zero - \int_decr:N \l_regex_balance_int + \if_int_compare:w \l__regex_replacement_csnames_int = \c_zero + \int_decr:N \l__regex_balance_int \fi: - \regex_replacement_char:nNN + \__regex_replacement_char:nNN { \exp_not:n { \if_false: { \fi: ^^@ } } } \char_set_catcode_letter:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_L:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \__regex_replacement_c_L:w + { \__regex_replacement_char:nNN { ^^@ } } \char_set_catcode_math_toggle:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_M:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \__regex_replacement_c_M:w + { \__regex_replacement_char:nNN { ^^@ } } \char_set_catcode_other:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_O:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \__regex_replacement_c_O:w + { \__regex_replacement_char:nNN { ^^@ } } \char_set_catcode_parameter:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_P:w + \cs_new_protected_nopar:Npn \__regex_replacement_c_P:w { - \regex_replacement_char:nNN + \__regex_replacement_char:nNN { \exp_not:n { \exp_not:n { ^^@^^@^^@^^@ } } } } - \cs_new_protected:Npn \regex_replacement_c_S:w #1#2 + \cs_new_protected:Npn \__regex_replacement_c_S:w #1#2 { \if_meaning:w \prg_do_nothing: #2 - \msg_kernel_error:nn { regex } { replacement-catcode-end } + \__msg_kernel_error:nn { regex } { replacement-catcode-end } \else: - \if_num:w `#2 = \c_zero - \msg_kernel_error:nn { regex } { replacement-null-space } + \if_int_compare:w `#2 = \c_zero + \__msg_kernel_error:nn { regex } { replacement-null-space } \fi: \tex_lccode:D 32 = `#2 \scan_stop: - \tl_to_lowercase:n { \tl_build_one:n {~} } + \tl_to_lowercase:n { \__tl_build_one:n {~} } \fi: } \char_set_catcode_alignment:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_T:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \__regex_replacement_c_T:w + { \__regex_replacement_char:nNN { ^^@ } } \char_set_catcode_math_superscript:N \^^@ - \cs_new_protected_nopar:Npn \regex_replacement_c_U:w - { \regex_replacement_char:nNN { ^^@ } } + \cs_new_protected_nopar:Npn \__regex_replacement_c_U:w + { \__regex_replacement_char:nNN { ^^@ } } \group_end: -\cs_new_protected:Npn \regex_replacement_error:NNN #1#2#3 +\cs_new_protected:Npn \__regex_replacement_error:NNN #1#2#3 { - \msg_kernel_error:nnx { regex } { replacement-#1 } {#3} + \__msg_kernel_error:nnx { regex } { replacement-#1 } {#3} #2 #3 } \cs_new_protected:Npn \regex_new:N #1 - { \cs_new_eq:NN #1 \c_regex_no_match_regex } + { \cs_new_eq:NN #1 \c__regex_no_match_regex } \cs_new_protected_nopar:Npn \regex_set:Nn #1#2 { - \regex_compile:n {#2} - \tl_set_eq:NN #1 \l_regex_internal_regex + \__regex_compile:n {#2} + \tl_set_eq:NN #1 \l__regex_internal_regex } \cs_new_protected_nopar:Npn \regex_gset:Nn #1#2 { - \regex_compile:n {#2} - \tl_gset_eq:NN #1 \l_regex_internal_regex + \__regex_compile:n {#2} + \tl_gset_eq:NN #1 \l__regex_internal_regex } \cs_new_protected_nopar:Npn \regex_const:Nn #1#2 { - \regex_compile:n {#2} - \tl_const:Nx #1 { \exp_not:o \l_regex_internal_regex } + \__regex_compile:n {#2} + \tl_const:Nx #1 { \exp_not:o \l__regex_internal_regex } } \cs_new_protected:Npn \regex_show:n #1 { - \regex_compile:n {#1} - \regex_show_aux:Nx \l_regex_internal_regex + \__regex_compile:n {#1} + \__regex_show:Nx \l__regex_internal_regex { { \tl_to_str:n {#1} } } } \cs_new_protected:Npn \regex_show:N #1 - { \regex_show_aux:Nx #1 { variable~\token_to_str:N #1 } } + { \__regex_show:Nx #1 { variable~\token_to_str:N #1 } } \prg_new_protected_conditional:Npnn \regex_match:nn #1#2 { T , F , TF } { - \regex_match_aux:nn { \regex_build:n {#1} } {#2} - \regex_aux_return: + \__regex_if_match:nn { \__regex_build:n {#1} } {#2} + \__regex_return: } \prg_new_protected_conditional:Npnn \regex_match:Nn #1#2 { T , F , TF } { - \regex_match_aux:nn { \regex_build:N #1 } {#2} - \regex_aux_return: + \__regex_if_match:nn { \__regex_build:N #1 } {#2} + \__regex_return: } \cs_new_protected:Npn \regex_count:nnN #1 - { \regex_count_aux:nnN { \regex_build:n {#1} } } + { \__regex_count:nnN { \__regex_build:n {#1} } } \cs_new_protected:Npn \regex_count:NnN #1 - { \regex_count_aux:nnN { \regex_build:N #1 } } -\cs_set_protected:Npn \regex_tmp:w #1#2#3 + { \__regex_count:nnN { \__regex_build:N #1 } } +\cs_set_protected:Npn \__regex_tmp:w #1#2#3 { - \cs_new_protected:Npn #2 ##1 { #1 { \regex_build:n {##1} } } - \cs_new_protected:Npn #3 ##1 { #1 { \regex_build:N ##1 } } + \cs_new_protected:Npn #2 ##1 { #1 { \__regex_build:n {##1} } } + \cs_new_protected:Npn #3 ##1 { #1 { \__regex_build:N ##1 } } \prg_new_protected_conditional:Npnn #2 ##1##2##3 { T , F , TF } - { #1 { \regex_build:n {##1} } {##2} ##3 \regex_aux_return: } + { #1 { \__regex_build:n {##1} } {##2} ##3 \__regex_return: } \prg_new_protected_conditional:Npnn #3 ##1##2##3 { T , F , TF } - { #1 { \regex_build:N ##1 } {##2} ##3 \regex_aux_return: } + { #1 { \__regex_build:N ##1 } {##2} ##3 \__regex_return: } } -\regex_tmp:w \regex_extract_once_aux:nnN +\__regex_tmp:w \__regex_extract_once:nnN \regex_extract_once:nnN \regex_extract_once:NnN -\regex_tmp:w \regex_extract_all_aux:nnN +\__regex_tmp:w \__regex_extract_all:nnN \regex_extract_all:nnN \regex_extract_all:NnN -\regex_tmp:w \regex_replace_once_aux:nnN +\__regex_tmp:w \__regex_replace_once:nnN \regex_replace_once:nnN \regex_replace_once:NnN -\regex_tmp:w \regex_replace_all_aux:nnN +\__regex_tmp:w \__regex_replace_all:nnN \regex_replace_all:nnN \regex_replace_all:NnN -\regex_tmp:w \regex_split_aux:nnN \regex_split:nnN \regex_split:NnN -\int_new:N \l_regex_match_count_int -\flag_new:n { regex_begin } -\flag_new:n { regex_end } -\int_new:N \l_regex_submatch_int -\int_new:N \l_regex_zeroth_submatch_int -\cs_new_protected_nopar:Npn \regex_aux_return: - { - \if_meaning:w \c_true_bool \g_regex_success_bool +\__regex_tmp:w \__regex_split:nnN \regex_split:nnN \regex_split:NnN +\int_new:N \l__regex_match_count_int +\flag_new:n { __regex_begin } +\flag_new:n { __regex_end } +\int_new:N \l__regex_submatch_int +\int_new:N \l__regex_zeroth_submatch_int +\cs_new_protected_nopar:Npn \__regex_return: + { + \if_meaning:w \c_true_bool \g__regex_success_bool \prg_return_true: \else: \prg_return_false: \fi: } -\cs_new_protected:Npn \regex_match_aux:nn #1#2 +\cs_new_protected:Npn \__regex_if_match:nn #1#2 { \group_begin: - \regex_disable_submatches: - \regex_single_match: + \__regex_disable_submatches: + \__regex_single_match: #1 - \regex_match:n {#2} + \__regex_match:n {#2} \group_end: } -\cs_new_protected:Npn \regex_count_aux:nnN #1#2#3 +\cs_new_protected:Npn \__regex_count:nnN #1#2#3 { \group_begin: - \regex_disable_submatches: - \int_zero:N \l_regex_match_count_int - \regex_multi_match:n { \int_incr:N \l_regex_match_count_int } + \__regex_disable_submatches: + \int_zero:N \l__regex_match_count_int + \__regex_multi_match:n { \int_incr:N \l__regex_match_count_int } #1 - \regex_match:n {#2} + \__regex_match:n {#2} \exp_args:NNNo \group_end: - \int_set:Nn #3 { \int_use:N \l_regex_match_count_int } + \int_set:Nn #3 { \int_use:N \l__regex_match_count_int } } -\cs_new_protected:Npn \regex_extract_once_aux:nnN #1#2#3 +\cs_new_protected:Npn \__regex_extract_once:nnN #1#2#3 { \group_begin: - \regex_single_match: + \__regex_single_match: #1 - \regex_match:n {#2} - \regex_extract: - \regex_group_end_extract_seq:N #3 + \__regex_match:n {#2} + \__regex_extract: + \__regex_group_end_extract_seq:N #3 } -\cs_new_protected:Npn \regex_extract_all_aux:nnN #1#2#3 +\cs_new_protected:Npn \__regex_extract_all:nnN #1#2#3 { \group_begin: - \regex_multi_match:n { \regex_extract: } + \__regex_multi_match:n { \__regex_extract: } #1 - \regex_match:n {#2} - \regex_group_end_extract_seq:N #3 + \__regex_match:n {#2} + \__regex_group_end_extract_seq:N #3 } -\cs_new_protected:Npn \regex_split_aux:nnN #1#2#3 +\cs_new_protected:Npn \__regex_split:nnN #1#2#3 { \group_begin: - \regex_multi_match:n + \__regex_multi_match:n { - \if_num:w \l_regex_start_pos_int < \l_regex_success_pos_int - \regex_extract: - \tex_skip:D \l_regex_zeroth_submatch_int - = \l_regex_start_pos_int sp - plus \tex_skip:D \l_regex_zeroth_submatch_int \scan_stop: + \if_int_compare:w \l__regex_start_pos_int < \l__regex_success_pos_int + \__regex_extract: + \tex_skip:D \l__regex_zeroth_submatch_int + = \l__regex_start_pos_int sp + plus \tex_skip:D \l__regex_zeroth_submatch_int \scan_stop: \fi: } #1 - \regex_match:n {#2} - \tex_skip:D \l_regex_submatch_int - = \l_regex_start_pos_int sp plus \l_regex_max_pos_int sp \scan_stop: - \int_incr:N \l_regex_submatch_int - \if_meaning:w \c_true_bool \l_regex_empty_success_bool - \if_num:w \l_regex_start_pos_int = \l_regex_max_pos_int - \int_decr:N \l_regex_submatch_int + \__regex_match:n {#2} + \tex_skip:D \l__regex_submatch_int + = \l__regex_start_pos_int sp plus \l__regex_max_pos_int sp \scan_stop: + \int_incr:N \l__regex_submatch_int + \if_meaning:w \c_true_bool \l__regex_empty_success_bool + \if_int_compare:w \l__regex_start_pos_int = \l__regex_max_pos_int + \int_decr:N \l__regex_submatch_int \fi: \fi: - \regex_group_end_extract_seq:N #3 + \__regex_group_end_extract_seq:N #3 } -\cs_new_protected:Npn \regex_group_end_extract_seq:N #1 +\cs_new_protected:Npn \__regex_group_end_extract_seq:N #1 { - \cs_set_eq:NN \seq_item:n \scan_stop: - \flag_clear:n { regex_begin } - \flag_clear:n { regex_end } - \tl_set:Nx \l_regex_internal_a_tl + \cs_set_eq:NN \__seq_item:n \scan_stop: + \flag_clear:n { __regex_begin } + \flag_clear:n { __regex_end } + \tl_set:Nx \l__regex_internal_a_tl { - \prg_stepwise_function:nnnN - { \c_two * \l_regex_max_state_int } + \int_step_function:nnnN + { \c_two * \l__regex_max_state_int } \c_one - { \l_regex_submatch_int - \c_one } - \regex_extract_seq_aux:n + { \l__regex_submatch_int - \c_one } + \__regex_extract_seq_aux:n } \int_compare:nNnF - { \flag_height:n { regex_begin } + \flag_height:n { regex_end } } + { \flag_height:n { __regex_begin } + \flag_height:n { __regex_end } } = \c_zero { - \msg_kernel_error:nnxxx { regex } { result-unbalanced } + \__msg_kernel_error:nnxxx { regex } { result-unbalanced } { splitting~or~extracting~submatches } - { \flag_height:n { regex_end } } - { \flag_height:n { regex_begin } } + { \flag_height:n { __regex_end } } + { \flag_height:n { __regex_begin } } } \use:x { \group_end: - \tl_set:Nn \exp_not:N #1 { \l_regex_internal_a_tl } + \tl_set:Nn \exp_not:N #1 { \l__regex_internal_a_tl } } } -\cs_new:Npn \regex_extract_seq_aux:n #1 +\cs_new:Npn \__regex_extract_seq_aux:n #1 { - \seq_item:n + \__seq_item:n { - \exp_after:wN \regex_extract_seq_aux:ww - \int_value:w \regex_submatch_balance:n {#1} ; #1; + \exp_after:wN \__regex_extract_seq_aux:ww + \__int_value:w \__regex_submatch_balance:n {#1} ; #1; } } -\cs_new:Npn \regex_extract_seq_aux:ww #1; #2; +\cs_new:Npn \__regex_extract_seq_aux:ww #1; #2; { - \if_num:w #1 < \c_zero - \flag_raise:n { regex_end } + \if_int_compare:w #1 < \c_zero + \flag_raise:n { __regex_end } \prg_replicate:nn {-#1} { \exp_not:n { { \if_false: } \fi: } } \fi: - \regex_query_submatch:n {#2} - \if_num:w #1 > \c_zero - \flag_raise:n { regex_begin } + \__regex_query_submatch:n {#2} + \if_int_compare:w #1 > \c_zero + \flag_raise:n { __regex_begin } \prg_replicate:nn {#1} { \exp_not:n { \if_false: { \fi: } } } \fi: } -\cs_new_protected_nopar:Npn \regex_extract: +\cs_new_protected_nopar:Npn \__regex_extract: { - \if_meaning:w \c_true_bool \g_regex_success_bool - \int_set_eq:NN \l_regex_zeroth_submatch_int \l_regex_submatch_int - \prg_replicate:nn \l_regex_capturing_group_int + \if_meaning:w \c_true_bool \g__regex_success_bool + \int_set_eq:NN \l__regex_zeroth_submatch_int \l__regex_submatch_int + \prg_replicate:nn \l__regex_capturing_group_int { - \tex_skip:D \l_regex_submatch_int \c_zero sp \scan_stop: - \int_incr:N \l_regex_submatch_int + \tex_skip:D \l__regex_submatch_int \c_zero sp \scan_stop: + \int_incr:N \l__regex_submatch_int } - \prop_map_inline:Nn \l_regex_success_submatches_prop + \prop_map_inline:Nn \l__regex_success_submatches_prop { - \if_num:w ##1 \c_minus_one - \exp_after:wN \regex_extract_aux_e:wn \int_value:w + \if_int_compare:w ##1 \c_minus_one + \exp_after:wN \__regex_extract_e:wn \__int_value:w \else: - \exp_after:wN \regex_extract_aux_b:wn \int_value:w + \exp_after:wN \__regex_extract_b:wn \__int_value:w \fi: - \int_eval:w \l_regex_zeroth_submatch_int + ##1 {##2} + \__int_eval:w \l__regex_zeroth_submatch_int + ##1 {##2} } - \tex_skip:D \l_regex_zeroth_submatch_int - = \tex_the:D \tex_skip:D \l_regex_zeroth_submatch_int - minus \l_regex_start_pos_int sp \scan_stop: + \tex_skip:D \l__regex_zeroth_submatch_int + = \tex_the:D \tex_skip:D \l__regex_zeroth_submatch_int + minus \l__regex_start_pos_int sp \scan_stop: \fi: } -\cs_new_protected:Npn \regex_extract_aux_b:wn #1 < #2 +\cs_new_protected:Npn \__regex_extract_b:wn #1 < #2 { \tex_skip:D #1 = #2 sp plus \etex_gluestretch:D \tex_skip:D #1 \scan_stop: } -\cs_new_protected:Npn \regex_extract_aux_e:wn #1 > #2 +\cs_new_protected:Npn \__regex_extract_e:wn #1 > #2 { \tex_skip:D #1 = 1 \tex_skip:D #1 plus #2 sp \scan_stop: } -\cs_new_protected:Npn \regex_replace_once_aux:nnN #1#2#3 +\cs_new_protected:Npn \__regex_replace_once:nnN #1#2#3 { \group_begin: - \regex_single_match: + \__regex_single_match: #1 - \regex_replacement:n {#2} - \exp_args:No \regex_match:n { #3 } - \if_meaning:w \c_false_bool \g_regex_success_bool + \__regex_replacement:n {#2} + \exp_args:No \__regex_match:n { #3 } + \if_meaning:w \c_false_bool \g__regex_success_bool \group_end: \else: - \regex_extract: - \int_set:Nn \l_regex_balance_int + \__regex_extract: + \int_set:Nn \l__regex_balance_int { - \regex_replacement_balance_one_match:n - { \l_regex_zeroth_submatch_int } + \__regex_replacement_balance_one_match:n + { \l__regex_zeroth_submatch_int } } - \tl_set:Nx \l_regex_internal_a_tl + \tl_set:Nx \l__regex_internal_a_tl { - \regex_replacement_do_one_match:n { \l_regex_zeroth_submatch_int } - \regex_query_range:nn - { \etex_gluestretch:D \tex_skip:D \l_regex_zeroth_submatch_int } - { \l_regex_max_pos_int } + \__regex_replacement_do_one_match:n { \l__regex_zeroth_submatch_int } + \__regex_query_range:nn + { \etex_gluestretch:D \tex_skip:D \l__regex_zeroth_submatch_int } + { \l__regex_max_pos_int } } - \regex_group_end_replace:N #3 + \__regex_group_end_replace:N #3 \fi: } -\cs_new_protected:Npn \regex_replace_all_aux:nnN #1#2#3 +\cs_new_protected:Npn \__regex_replace_all:nnN #1#2#3 { \group_begin: - \regex_multi_match:n { \regex_extract: } + \__regex_multi_match:n { \__regex_extract: } #1 - \regex_replacement:n {#2} - \exp_args:No \regex_match:n {#3} - \int_set:Nn \l_regex_balance_int + \__regex_replacement:n {#2} + \exp_args:No \__regex_match:n {#3} + \int_set:Nn \l__regex_balance_int { 0 - \prg_stepwise_function:nnnN - { \c_two * \l_regex_max_state_int } - \l_regex_capturing_group_int - { \l_regex_submatch_int - \c_one } - \regex_replacement_balance_one_match:n + \int_step_function:nnnN + { \c_two * \l__regex_max_state_int } + \l__regex_capturing_group_int + { \l__regex_submatch_int - \c_one } + \__regex_replacement_balance_one_match:n } - \tl_set:Nx \l_regex_internal_a_tl + \tl_set:Nx \l__regex_internal_a_tl { - \prg_stepwise_function:nnnN - { \c_two * \l_regex_max_state_int } - \l_regex_capturing_group_int - { \l_regex_submatch_int - \c_one } - \regex_replacement_do_one_match:n - \regex_query_range:nn - \l_regex_start_pos_int \l_regex_max_pos_int + \int_step_function:nnnN + { \c_two * \l__regex_max_state_int } + \l__regex_capturing_group_int + { \l__regex_submatch_int - \c_one } + \__regex_replacement_do_one_match:n + \__regex_query_range:nn + \l__regex_start_pos_int \l__regex_max_pos_int } - \regex_group_end_replace:N #3 + \__regex_group_end_replace:N #3 } -\cs_new_protected_nopar:Npn \regex_group_end_replace:N #1 +\cs_new_protected_nopar:Npn \__regex_group_end_replace:N #1 { - \if_num:w \l_regex_balance_int = \c_zero + \if_int_compare:w \l__regex_balance_int = \c_zero \else: - \msg_kernel_error:nnxxx { regex } { result-unbalanced } + \__msg_kernel_error:nnxxx { regex } { result-unbalanced } { replacing } - { \int_max:nn { - \l_regex_balance_int } { \c_zero } } - { \int_max:nn { \l_regex_balance_int } { \c_zero } } + { \int_max:nn { - \l__regex_balance_int } { \c_zero } } + { \int_max:nn { \l__regex_balance_int } { \c_zero } } \fi: \use:x { \group_end: \tl_set:Nn \exp_not:N #1 { - \if_int_compare:w \l_regex_balance_int < \c_zero - \prg_replicate:nn { - \l_regex_balance_int } + \if_int_compare:w \l__regex_balance_int < \c_zero + \prg_replicate:nn { - \l__regex_balance_int } { { \if_false: } \fi: } \fi: - \l_regex_internal_a_tl - \if_int_compare:w \l_regex_balance_int > \c_zero - \prg_replicate:nn { \l_regex_balance_int } + \l__regex_internal_a_tl + \if_int_compare:w \l__regex_balance_int > \c_zero + \prg_replicate:nn { \l__regex_balance_int } { \if_false: { \fi: } } \fi: } } } -\msg_kernel_new:nnnn { regex } { trailing-backslash } +\__msg_kernel_new:nnnn { regex } { trailing-backslash } { Trailing~escape~character~\iow_char:N\\. } { A~regular~expression~or~its~replacement~text~ends~with~ the~escape~character~\iow_char:N\\.~It~will~be~ignored. } -\msg_kernel_new:nnnn { regex } { x-missing-rbrace } +\__msg_kernel_new:nnnn { regex } { x-missing-rbrace } { Missing~closing~brace~in~\iow_char:N\\x~hexadecimal~sequence. } { You~wrote~something~like~ `\iow_char:N\\x\{\int_to_hexadecimal:n{#1}'.~ The~closing~brace~is~missing. } -\msg_kernel_new:nnnn { regex } { x-overflow } +\__msg_kernel_new:nnnn { regex } { x-overflow } { Character~code~'#1'~too~large~in~\iow_char:N\\x~hexadecimal~sequence. } { You~wrote~something~like~ `\iow_char:N\\x\{\int_to_hexadecimal:n{#1}\}'.~ The~character~code~'#1'~is~larger~than~\int_use:N \c_max_char_int. } -\msg_kernel_new:nnnn { regex } { invalid-quantifier } +\__msg_kernel_new:nnnn { regex } { invalid-quantifier } { Braced~quantifier~'#1'~may~not~be~followed~by~'#2'. } { The~character~'#2'~is~invalid~in~the~braced~quantifier~'#1'.~ The~only~valid~quantifiers~are~'*',~'?',~'+',~'{}',~ '{,}'~and~'{,}',~followed~or~not~by~'?'. } -\msg_kernel_new:nnnn { regex } { missing-rbrack } +\__msg_kernel_new:nnnn { regex } { missing-rbrack } { Missing~right~bracket~inserted~in~regular~expression. } { LaTeX~was~given~a~regular~expression~where~a~character~class~ was~started~with~'[',~but~the~matching~']'~is~missing. } -\msg_kernel_new:nnnn { regex } { missing-rparen } +\__msg_kernel_new:nnnn { regex } { missing-rparen } { Missing~right~parenthes\int_compare:nTF{#1=1}{i}{e}s~ inserted~in~regular~expression. @@ -2597,44 +2605,44 @@ more~left~parenthes\int_compare:nTF{#1=1}{i}{e}s~than~right~ parenthes\int_compare:nTF{#1=1}{i}{e}s. } -\msg_kernel_new:nnnn { regex } { extra-rparen } +\__msg_kernel_new:nnnn { regex } { extra-rparen } { Extra~right~parenthesis~ignored~in~regular~expression. } { LaTeX~came~across~a~closing~parenthesis~when~no~submatch~group~ was~open.~The~parenthesis~will~be~ignored. } -\msg_kernel_new:nnnn { regex } { bad-escape } +\__msg_kernel_new:nnnn { regex } { bad-escape } { Invalid~escape~\c_backslash_str #1~ - \regex_if_in_cs:TF { within~a~control~sequence. } + \__regex_if_in_cs:TF { within~a~control~sequence. } { - \regex_if_in_class:TF + \__regex_if_in_class:TF { in~a~character~class. } { following~a~category~test. } } } { The~escape~sequence~\iow_char:N\\#1~may~not~appear~ - \regex_if_in_cs:TF + \__regex_if_in_cs:TF { within~a~control~sequence~test~introduced~by~ \iow_char:N\\c\iow_char:N\{. } { - \regex_if_in_class:TF + \__regex_if_in_class:TF { within~a~character~class~ } { following~a~category~test~such~as~\iow_char:N\\cL~ } because~it~does~not~match~exactly~one~character. } } -\msg_kernel_new:nnnn { regex } { range-missing-end } +\__msg_kernel_new:nnnn { regex } { range-missing-end } { Invalid~end-point~for~range~'#1-#2'~in~character~class. } { The~end-point~'#2'~of~the~range~'#1-#2'~may~not~serve~as~an~ end-point~for~a~range:~alphanumeric~characters~should~not~be~ escaped,~and~non-alphanumeric~characters~should~be~escaped. } -\msg_kernel_new:nnnn { regex } { range-backwards } +\__msg_kernel_new:nnnn { regex } { range-backwards } { Range~[#1-#2]~out~of~order~in~character~class. } { In~ranges~of~characters~[x-y]~appearing~in~character~classes,~ @@ -2642,27 +2650,27 @@ Here,~#1~has~character~code~\int_eval:n {`#1},~while~#2~has~ character~code~\int_eval:n {`#2}. } -\msg_kernel_new:nnnn { regex } { c-bad-mode } +\__msg_kernel_new:nnnn { regex } { c-bad-mode } { Invalid~nested~\iow_char:N\\c~escape~in~regular~expression. } { The~\iow_char:N\\c~escape~cannot~be~used~within~ a~control~sequence~test~'\iow_char:N\\c{...}'.~ To~combine~several~category~tests,~use~'\iow_char:N\\c[...]'. } -\msg_kernel_new:nnnn { regex } { c-missing-rbrace } +\__msg_kernel_new:nnnn { regex } { c-missing-rbrace } { Missing~right~brace~inserted~for~\iow_char:N\\c~escape. } { LaTeX~was~given~a~regular~expression~where~a~ '\iow_char:N\\c\iow_char:N\{...'~construction~was~not~ended~ with~a~closing~brace~'\iow_char:N\}'. } -\msg_kernel_new:nnnn { regex } { c-missing-rbrack } +\__msg_kernel_new:nnnn { regex } { c-missing-rbrack } { Missing~right~bracket~inserted~for~\iow_char:N\\c~escape. } { A~construction~'\iow_char:N\\c[...'~appears~in~a~ regular~expression,~but~the~closing~']'~is~not~present. } -\msg_kernel_new:nnnn { regex } { c-missing-category } +\__msg_kernel_new:nnnn { regex } { c-missing-category } { Invalid~character~'#1'~following~\iow_char:N\\c~escape. } { In~regular~expressions,~the~\iow_char:N\\c~escape~sequence~ @@ -2670,29 +2678,29 @@ capital~letter~representing~a~character~category,~namely~ one~of~ABCDELMOPSTU. } -\msg_kernel_new:nnnn { regex } { u-missing-lbrace } +\__msg_kernel_new:nnnn { regex } { u-missing-lbrace } { Missing~left~brace~following~\iow_char:N\\u~escape. } { The~\iow_char:N\\u~escape~sequence~must~be~followed~by~ a~brace~group~with~the~name~of~the~variable~to~use. } -\msg_kernel_new:nnnn { regex } { u-missing-rbrace } +\__msg_kernel_new:nnnn { regex } { u-missing-rbrace } { Missing~right~brace~inserted~for~\iow_char:N\\u~escape. } { LaTeX~ - \tl_if_empty:xTF {#2} + \str_if_eq_x:nnTF { } {#2} { reached~the~end~of~the~string~ } { encountered~an~escaped~alphanumeric~character '\iow_char:N\\#2'~ } when~parsing~the~argument~of~an~'\iow_char:N\\u\iow_char:N\{...\}'~escape. } -\msg_kernel_new:nnnn { regex } { posix-unsupported } +\__msg_kernel_new:nnnn { regex } { posix-unsupported } { POSIX~collating~element~'[#1 ~ #1]'~not~supported. } { The~[.foo.]~and~[=bar=]~syntaxes~have~a~special~meaning~in~POSIX~ regular~expressions.~This~is~not~supported~by~LaTeX.~Maybe~you~ forgot~to~escape~a~left~bracket~in~a~character~class? } -\msg_kernel_new:nnnn { regex } { posix-unknown } +\__msg_kernel_new:nnnn { regex } { posix-unknown } { POSIX~class~[:#1:]~unknown. } { [:#1:]~is~not~among~the~known~POSIX~classes~ @@ -2701,10 +2709,10 @@ [:print:],~[:punct:],~[:space:],~[:upper:],~ [:word:],~and~[:xdigit:]. } -\msg_kernel_new:nnnn { regex } { posix-missing-close } +\__msg_kernel_new:nnnn { regex } { posix-missing-close } { Missing~closing~':]'~for~POSIX~class. } { The~POSIX~syntax~'#1'~must~be~followed~by~':]',~not~'#2'. } -\msg_kernel_new:nnnn { regex } { result-unbalanced } +\__msg_kernel_new:nnnn { regex } { result-unbalanced } { Missing~brace~inserted~when~#1. } { LaTeX~was~asked~to~do~some~regular~expression~operation,~ @@ -2712,34 +2720,34 @@ of~begin-group~and~end-group~tokens.~Braces~were~inserted:~ #2~left,~#3~right. } -\msg_kernel_new:nnnn { regex } { unknown-option } +\__msg_kernel_new:nnnn { regex } { unknown-option } { Unknown~option~`#1'~for~regular~expressions. } { The~only~available~option~is~'case-insensitive',~toggled~by~ '(?i)'~and~'(?-i)'. } -\msg_kernel_new:nnnn { regex } { replacement-c } +\__msg_kernel_new:nnnn { regex } { replacement-c } { Misused~\iow_char:N\\c~command~in~a~replacement~text. } { In~a~replacement~text,~the~\iow_char:N\\c~escape~sequence~ can~be~followed~by~one~of~the~letters~ABCDELMOPSTU~ or~a~brace~group,~not~by~'#1'. } -\msg_kernel_new:nnnn { regex } { replacement-u } +\__msg_kernel_new:nnnn { regex } { replacement-u } { Misused~\iow_char:N\\u~command~in~a~replacement~text. } { In~a~replacement~text,~the~\iow_char:N\\u~escape~sequence~ must~be~~followed~by~a~brace~group~holding~the~name~of~the~ variable~to~use. } -\msg_kernel_new:nnnn { regex } { replacement-g } +\__msg_kernel_new:nnnn { regex } { replacement-g } { Missing~brace~for~the~\iow_char:N\\g~construction~in~a~replacement~text. } { In~the~replacement~text~for~a~regular~expression~search,~ submatches~are~represented~either~as~\iow_char:N \\g{dd..d},~ or~\\d,~where~`d'~are~single~digits.~Here,~a~brace~is~missing. } -\msg_kernel_new:nnnn { regex } { replacement-catcode-end } +\__msg_kernel_new:nnnn { regex } { replacement-catcode-end } { Missing~character~for~the~\iow_char:N\\c~ construction~in~a~replacement~text. @@ -2750,7 +2758,7 @@ the~character~category.~Then,~a~character~must~follow.~LaTeX~ reached~the~end~of~the~replacement~when~looking~for~that. } -\msg_kernel_new:nnnn { regex } { replacement-null-space } +\__msg_kernel_new:nnnn { regex } { replacement-null-space } { TeX~cannot~build~a~space~token~with~character~code~0. } { You~asked~for~a~character~token~with~category~'space',~ @@ -2759,18 +2767,18 @@ This~specific~case~is~impossible~and~will~be~replaced~ by~a~normal~space. } -\msg_kernel_new:nnnn { regex } { replacement-missing-rbrace } +\__msg_kernel_new:nnnn { regex } { replacement-missing-rbrace } { Missing~right~brace~inserted~in~replacement~text. } { - There~were~\int_use:N \l_regex_replacement_csnames_int \ + There~were~\int_use:N \l__regex_replacement_csnames_int \ missing~right~braces. } -\cs_new:Npn \regex_msg_repeated:nnN #1#2#3 +\cs_new:Npn \__regex_msg_repeated:nnN #1#2#3 { - \str_if_eq:xxF { #1 #2 } { 1 0 } + \str_if_eq_x:nnF { #1 #2 } { 1 0 } { , ~ repeated ~ - \prg_case_int:nnn {#2} + \int_case:nnn {#2} { { -1 } { #1~or~more~times,~\bool_if:NTF #3 { lazy } { greedy } } { 0 } { #1~times } diff --git a/Master/texmf-dist/tex/latex/l3experimental/l3str/l3str.sty b/Master/texmf-dist/tex/latex/l3experimental/l3str/l3str.sty index 88042a5337d..3ad97283e3b 100644 --- a/Master/texmf-dist/tex/latex/l3experimental/l3str/l3str.sty +++ b/Master/texmf-dist/tex/latex/l3experimental/l3str/l3str.sty @@ -46,8 +46,8 @@ %% %% ----------------------------------------------------------------------- \RequirePackage{expl3} -\GetIdInfo$Id: l3str.dtx 3460 2012-02-26 12:16:41Z will $ - {L3 Experimental Strings} +\GetIdInfo$Id: l3str.dtx 3940 2012-07-09 00:41:44Z bruno $ + {L3 Experimental strings} \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} \RequirePackage{l3tl-analysis,l3tl-build,l3flag} @@ -70,10 +70,10 @@ { \exp_not:c { tl_ #1 :Nx } ##1 { \exp_not:N \tl_to_str:n {##2} } } \exp_args:Nc \cs_generate_variant:Nn { str_ #1 :Nn } { Nx , cn , cx } } -\cs_new_protected_nopar:Npn \str_tmp:w { } -\tl_new:N \l_str_internal_tl -\int_new:N \l_str_internal_int -\tl_new:N \g_str_result_tl +\cs_new_protected_nopar:Npn \__str_tmp:w { } +\tl_new:N \l__str_internal_tl +\int_new:N \l__str_internal_int +\tl_new:N \g__str_result_tl \int_const:Nn \c_forty_eight { 48 } \int_const:Nn \c_fifty_eight { 58 } \int_const:Nn \c_sixty_five { 65 } @@ -83,30 +83,30 @@ \int_const:Nn \c_one_hundred_twenty_seven { 127 } \int_const:Nn \c_max_char_int { \pdftex_if_engine:TF { "FF } { "10FFFF } } -\int_const:Nn \c_str_replacement_char_int { "FFFD } +\int_const:Nn \c__str_replacement_char_int { "FFFD } \tl_const:Nx \c_backslash_str { \cs_to_str:N \\ } \tl_const:Nx \c_lbrace_str { \cs_to_str:N \{ } \tl_const:Nx \c_rbrace_str { \cs_to_str:N \} } \tl_const:Nx \c_hash_str { \cs_to_str:N \# } \tl_const:Nx \c_tilde_str { \cs_to_str:N \~ } \tl_const:Nx \c_percent_str { \cs_to_str:N \% } -\prop_new:N \g_str_file_prop -\prop_gput:Nnn \g_str_file_prop { latin1 } { iso88591 } -\prop_gput:Nnn \g_str_file_prop { latin2 } { iso88592 } -\prop_gput:Nnn \g_str_file_prop { latin3 } { iso88593 } -\prop_gput:Nnn \g_str_file_prop { latin4 } { iso88594 } -\prop_gput:Nnn \g_str_file_prop { latin5 } { iso88599 } -\prop_gput:Nnn \g_str_file_prop { latin6 } { iso885910 } -\prop_gput:Nnn \g_str_file_prop { latin7 } { iso885913 } -\prop_gput:Nnn \g_str_file_prop { latin8 } { iso885914 } -\prop_gput:Nnn \g_str_file_prop { latin9 } { iso885915 } -\prop_gput:Nnn \g_str_file_prop { latin10 } { iso885916 } -\prop_gput:Nnn \g_str_file_prop { utf16le } { utf16 } -\prop_gput:Nnn \g_str_file_prop { utf16be } { utf16 } -\prop_gput:Nnn \g_str_file_prop { utf32le } { utf32 } -\prop_gput:Nnn \g_str_file_prop { utf32be } { utf32 } -\prop_gput:Nnn \g_str_file_prop { hexadecimal } { hex } -\bool_new:N \g_str_error_bool +\prop_new:N \g__str_alias_prop +\prop_gput:Nnn \g__str_alias_prop { latin1 } { iso88591 } +\prop_gput:Nnn \g__str_alias_prop { latin2 } { iso88592 } +\prop_gput:Nnn \g__str_alias_prop { latin3 } { iso88593 } +\prop_gput:Nnn \g__str_alias_prop { latin4 } { iso88594 } +\prop_gput:Nnn \g__str_alias_prop { latin5 } { iso88599 } +\prop_gput:Nnn \g__str_alias_prop { latin6 } { iso885910 } +\prop_gput:Nnn \g__str_alias_prop { latin7 } { iso885913 } +\prop_gput:Nnn \g__str_alias_prop { latin8 } { iso885914 } +\prop_gput:Nnn \g__str_alias_prop { latin9 } { iso885915 } +\prop_gput:Nnn \g__str_alias_prop { latin10 } { iso885916 } +\prop_gput:Nnn \g__str_alias_prop { utf16le } { utf16 } +\prop_gput:Nnn \g__str_alias_prop { utf16be } { utf16 } +\prop_gput:Nnn \g__str_alias_prop { utf32le } { utf32 } +\prop_gput:Nnn \g__str_alias_prop { utf32be } { utf32 } +\prop_gput:Nnn \g__str_alias_prop { hexadecimal } { hex } +\bool_new:N \g__str_error_bool \flag_new:n { str_byte } \flag_new:n { str_error } \group_begin: @@ -115,21 +115,21 @@ \tl_to_lowercase:n { \group_end: - \cs_new:Npn \tl_to_other_str:n #1 + \cs_new:Npn \__str_to_other:n #1 { - \exp_after:wN \tl_to_other_str_loop:w \tl_to_str:n {#1} ~ % + \exp_after:wN \__str_to_other_loop:w \tl_to_str:n {#1} ~ % A ~ A ~ A ~ A ~ A ~ A ~ A ~ A ~ \q_mark \q_stop } - \cs_new:Npn \tl_to_other_str_loop:w + \cs_new:Npn \__str_to_other_loop:w #1 ~ #2 ~ #3 ~ #4 ~ #5 ~ #6 ~ #7 ~ #8 ~ #9 \q_stop { \if_meaning:w A #8 - \tl_to_other_str_end:w + \__str_to_other_end:w \fi: - \tl_to_other_str_loop:w + \__str_to_other_loop:w #9 #1 * #2 * #3 * #4 * #5 * #6 * #7 * #8 * \q_stop } - \cs_new:Npn \tl_to_other_str_end:w \fi: #1 \q_mark #2 * A #3 \q_stop + \cs_new:Npn \__str_to_other_end:w \fi: #1 \q_mark #2 * A #3 \q_stop { \fi: #2 } } \group_begin: @@ -138,24 +138,24 @@ \tl_to_lowercase:n { \group_end: - \cs_new_protected:Npn \str_gset_other:Nn #1#2 + \cs_new_protected:Npn \__str_gset_other:Nn #1#2 { \tl_gset:Nx #1 { - \exp_after:wN \str_gset_other_loop:w \tl_to_str:n {#2} ~ % + \exp_after:wN \__str_gset_other_loop:w \tl_to_str:n {#2} ~ % A ~ A ~ A ~ A ~ A ~ A ~ A ~ A ~ A ~ \q_stop } } - \cs_new:Npn \str_gset_other_loop:w + \cs_new:Npn \__str_gset_other_loop:w #1 ~ #2 ~ #3 ~ #4 ~ #5 ~ #6 ~ #7 ~ #8 ~ #9 ~ { \if_meaning:w A #9 - \str_gset_other_end:w + \__str_gset_other_end:w \fi: #1 * #2 * #3 * #4 * #5 * #6 * #7 * #8 * #9 - \str_gset_other_loop:w * + \__str_gset_other_loop:w * } - \cs_new:Npn \str_gset_other_end:w \fi: #1 * A #2 \q_stop + \cs_new:Npn \__str_gset_other_end:w \fi: #1 * A #2 \q_stop { \fi: #1 } } \cs_new:Npn \str_count_spaces:N @@ -164,39 +164,39 @@ { \int_eval:n { - \exp_after:wN \str_count_spaces_loop:wwwwwwwww + \exp_after:wN \__str_count_spaces_loop:wwwwwwwww \tl_to_str:n {#1} ~ X 7 ~ X 6 ~ X 5 ~ X 4 ~ X 3 ~ X 2 ~ X 1 ~ X 0 ~ X -1 ~ \q_stop } } -\cs_new:Npn \str_count_spaces_loop:wwwwwwwww #1~#2~#3~#4~#5~#6~#7~#8~#9~ +\cs_new:Npn \__str_count_spaces_loop:wwwwwwwww #1~#2~#3~#4~#5~#6~#7~#8~#9~ { \if_meaning:w X #9 \exp_after:wN \use_none_delimit_by_q_stop:w \fi: - \c_nine + \str_count_spaces_loop:wwwwwwwww + \c_nine + \__str_count_spaces_loop:wwwwwwwww } -\cs_new_nopar:Npn \str_length:N { \exp_args:No \str_length:n } -\cs_new:Npn \str_length:n #1 +\cs_new_nopar:Npn \str_count:N { \exp_args:No \str_count:n } +\cs_new:Npn \str_count:n #1 { - \str_length_aux:n + \__str_count:n { \str_count_spaces:n {#1} - + \exp_after:wN \str_length_loop:NNNNNNNNN \tl_to_str:n {#1} + + \exp_after:wN \__str_count_loop:NNNNNNNNN \tl_to_str:n {#1} } } -\cs_new:Npn \str_length_unsafe:n #1 +\cs_new:Npn \__str_count_unsafe:n #1 { - \str_length_aux:n - { \str_length_loop:NNNNNNNNN #1 } + \__str_count:n + { \__str_count_loop:NNNNNNNNN #1 } } -\cs_new:Npn \str_length_ignore_spaces:n #1 +\cs_new:Npn \str_count_ignore_spaces:n #1 { - \str_length_aux:n - { \exp_after:wN \str_length_loop:NNNNNNNNN \tl_to_str:n {#1} } + \__str_count:n + { \exp_after:wN \__str_count_loop:NNNNNNNNN \tl_to_str:n {#1} } } -\cs_new:Npn \str_length_aux:n #1 +\cs_new:Npn \__str_count:n #1 { \int_eval:n { @@ -207,21 +207,21 @@ \q_stop } } -\cs_set:Npn \str_length_loop:NNNNNNNNN #1#2#3#4#5#6#7#8#9 +\cs_set:Npn \__str_count_loop:NNNNNNNNN #1#2#3#4#5#6#7#8#9 { \if_meaning:w X #9 \exp_after:wN \use_none_delimit_by_q_stop:w \fi: - \c_nine + \str_length_loop:NNNNNNNNN + \c_nine + \__str_count_loop:NNNNNNNNN } \cs_new_nopar:Npn \str_head:N { \exp_args:No \str_head:n } \cs_set:Npn \str_head:n #1 { - \exp_after:wN \str_head_aux:w + \exp_after:wN \__str_head:w \tl_to_str:n {#1} { { } } ~ \q_stop } -\cs_set:Npn \str_head_aux:w #1 ~ % +\cs_set:Npn \__str_head:w #1 ~ % { \use_i_delimit_by_q_stop:nw #1 { ~ } } \cs_new:Npn \str_head_ignore_spaces:n #1 { @@ -231,82 +231,85 @@ \cs_new_nopar:Npn \str_tail:N { \exp_args:No \str_tail:n } \cs_set:Npn \str_tail:n #1 { - \exp_after:wN \str_tail_aux:w + \exp_after:wN \__str_tail:w \reverse_if:N \if_charcode:w \scan_stop: \tl_to_str:n {#1} X X \q_stop } -\cs_set:Npn \str_tail_aux:w #1 X #2 \q_stop { \fi: #1 } +\cs_set:Npn \__str_tail:w #1 X #2 \q_stop { \fi: #1 } \cs_new:Npn \str_tail_ignore_spaces:n #1 { - \exp_after:wN \str_tail_aux_ii:w + \exp_after:wN \__str_tail_ii:w \tl_to_str:n {#1} X X \q_stop } -\cs_new:Npn \str_tail_aux_ii:w #1 #2 X #3 \q_stop { #2 } -\cs_new:Npn \str_skip_c_zero:w #1; +\cs_new:Npn \__str_tail_ii:w #1 #2 X #3 \q_stop { #2 } +\cs_new:Npn \__str_skip_c_zero:w #1; { - \if_num:w \int_eval:w #1 > \c_eight - \exp_after:wN \str_skip_aux:wNNNNNNNN + \if_int_compare:w \__int_eval:w #1 > \c_eight + \exp_after:wN \__str_skip_loop:wNNNNNNNN \else: - \exp_after:wN \str_skip_end:w - \int_use:N \int_eval:w + \exp_after:wN \__str_skip_end:w + \int_use:N \__int_eval:w \fi: #1 ; } -\cs_new:Npn \str_skip_aux:wNNNNNNNN #1; #2#3#4#5#6#7#8#9 - { \exp_after:wN \str_skip_c_zero:w \int_use:N \int_eval:w #1 - \c_eight ; } -\cs_new:Npn \str_skip_end:w #1 ; +\cs_new:Npn \__str_skip_loop:wNNNNNNNN #1; #2#3#4#5#6#7#8#9 + { \exp_after:wN \__str_skip_c_zero:w \int_use:N \__int_eval:w #1 - \c_eight ; } +\cs_new:Npn \__str_skip_end:w #1 ; { - \exp_after:wN \str_skip_end_ii:NNNNNNNN - \if_case:w \if_num:w #1 > \c_zero #1 \else: 0 \fi: \exp_stop_f: + \exp_after:wN \__str_skip_end_ii:NNNNNNNN + \if_case:w \if_int_compare:w #1 > \c_zero #1 \else: 0 \fi: \exp_stop_f: \or: \or: \or: \or: \or: \or: \or: \or: } -\cs_new:Npn \str_skip_end_ii:NNNNNNNN #1#2#3#4#5#6#7#8 { \fi: \c_zero } -\cs_new:Npn \str_collect_delimit_by_q_stop:w #1; +\cs_new:Npn \__str_skip_end_ii:NNNNNNNN #1#2#3#4#5#6#7#8 { \fi: \c_zero } +\cs_new:Npn \__str_collect_delimit_by_q_stop:w #1; { - \exp_after:wN \str_collect_aux:wn - \int_use:N \int_eval:w #1 ; + \exp_after:wN \__str_collect_loop:wn + \int_use:N \__int_eval:w #1 ; { } } -\cs_new:Npn \str_collect_aux:wn #1 ; +\cs_new:Npn \__str_collect_loop:wn #1 ; { - \if_num:w #1 > \c_seven - \exp_after:wN \str_collect_aux:wnNNNNNNN + \if_int_compare:w #1 > \c_seven + \exp_after:wN \__str_collect_loop_ii:wnNNNNNNN \else: - \exp_after:wN \str_collect_end:wn + \exp_after:wN \__str_collect_end:wn \fi: #1 ; } -\cs_new:Npn \str_collect_aux:wnNNNNNNN #1; #2 #3#4#5#6#7#8#9 +\cs_new:Npn \__str_collect_loop_ii:wnNNNNNNN #1; #2 #3#4#5#6#7#8#9 { - \exp_after:wN \str_collect_aux:wn - \int_use:N \int_eval:w #1 - \c_seven ; + \exp_after:wN \__str_collect_loop:wn + \int_use:N \__int_eval:w #1 - \c_seven ; { #2 #3#4#5#6#7#8#9 } } -\cs_new:Npn \str_collect_end:wn #1 ; +\cs_new:Npn \__str_collect_end:wn #1 ; { - \exp_after:wN \str_collect_end_ii:nnnnnnnnw - \if_case:w \if_num:w #1 > \c_zero #1 \else: 0 \fi: \exp_stop_f: + \exp_after:wN \__str_collect_end_ii:nnnnnnnnw + \if_case:w \if_int_compare:w #1 > \c_zero #1 \else: 0 \fi: \exp_stop_f: \or: \or: \or: \or: \or: \or: \fi: } -\cs_new:Npn \str_collect_end_ii:nnnnnnnnw #1#2#3#4#5#6#7#8 #9 \q_stop +\cs_new:Npn \__str_collect_end_ii:nnnnnnnnw #1#2#3#4#5#6#7#8 #9 \q_stop { #1#2#3#4#5#6#7#8 } \cs_new_nopar:Npn \str_item:Nn { \exp_args:No \str_item:nn } -\cs_new:Npn \str_item:nn #1 +\cs_new:Npn \str_item:nn #1#2 { - \exp_last_unbraced:Nf \token_to_str:N - \exp_args:Nf \str_item_unsafe:nn { \tl_to_other_str:n {#1} } + \exp_args:Nf \tl_to_str:n + { + \exp_args:Nf \__str_item_unsafe:nn + { \__str_to_other:n {#1} } {#2} + } } \cs_new:Npn \str_item_ignore_spaces:nn #1 - { \exp_args:No \str_item_unsafe:nn { \tl_to_str:n {#1} } } -\cs_new:Npn \str_item_unsafe:nn #1#2 + { \exp_args:No \__str_item_unsafe:nn { \tl_to_str:n {#1} } } +\cs_new:Npn \__str_item_unsafe:nn #1#2 { - \exp_after:wN \str_item_aux:ww - \int_use:N \int_eval:w #2 \exp_after:wN ; - \int_value:w \str_length_unsafe:n {#1} ; - #1 + \exp_after:wN \__str_item:ww + \int_use:N \__int_eval:w #2 \exp_after:wN ; + \__int_value:w \__str_count_unsafe:n {#1} ; + { } #1 \q_stop } -\cs_new:Npn \str_item_aux:ww #1; #2; +\cs_new:Npn \__str_item:ww #1; #2; { \int_compare:nNnTF {#1} < \c_zero { @@ -314,16 +317,16 @@ { \use_none_delimit_by_q_stop:w } { \exp_after:wN \use_i_delimit_by_q_stop:nw - \tex_romannumeral:D \str_skip_c_zero:w #1 + #2 ; + \tex_romannumeral:D \__str_skip_c_zero:w #1 + #2 + \c_one ; } } { - \int_compare:nNnTF {#1} < {#2} + \int_compare:nNnTF {#1} > {#2} + { \use_none_delimit_by_q_stop:w } { \exp_after:wN \use_i_delimit_by_q_stop:nw - \tex_romannumeral:D \str_skip_c_zero:w #1 ; + \tex_romannumeral:D \__str_skip_c_zero:w #1 ; } - { \use_none_delimit_by_q_stop:w } } } \cs_new_nopar:Npn \str_substr:Nnn { \exp_args:No \str_substr:nnn } @@ -331,51 +334,51 @@ { \exp_args:Nf \tl_to_str:n { - \exp_args:Nf \str_substr_unsafe:nnn - { \tl_to_other_str:n {#1} } {#2} {#3} + \exp_args:Nf \__str_substr_unsafe:nnn + { \__str_to_other:n {#1} } {#2} {#3} } } \cs_new:Npn \str_substr_ignore_spaces:nnn #1 - { \exp_args:No \str_substr_unsafe:nnn { \tl_to_str:n {#1} } } -\cs_new:Npn \str_substr_unsafe:nnn #1#2#3 - { - \exp_after:wN \str_substr_aux:www - \int_value:w \str_length_unsafe:n {#1} \exp_after:wN ; - \int_use:N \int_eval:w #2 + \c_zero \exp_after:wN ; - \int_use:N \int_eval:w - \exp_args:Nf \str_substr_aux:nN {#3} \c_max_int ; - #1 + { \exp_args:No \__str_substr_unsafe:nnn { \tl_to_str:n {#1} } } +\cs_new:Npn \__str_substr_unsafe:nnn #1#2#3 + { + \exp_after:wN \__str_substr_ii:www + \__int_value:w \__str_count_unsafe:n {#1} \exp_after:wN ; + \int_use:N \__int_eval:w #2 + \c_zero \exp_after:wN ; + \int_use:N \__int_eval:w + \exp_args:Nf \__str_substr_i:nN {#3} \c_max_int ; + { } #1 \q_stop } -\cs_new:Npn \str_substr_aux:nN #1 #2 +\cs_new:Npn \__str_substr_i:nN #1 #2 { \tl_if_empty:nTF {#1} {#2} {#1} } -\cs_new:Npn \str_substr_aux:www #1; #2; #3; +\cs_new:Npn \__str_substr_ii:www #1; #2; #3; { - \exp_args:Nf \str_substr_aux:nnw - { \str_substr_normalize_range:nn {#2} {#1} } - { \str_substr_normalize_range:nn {#3} {#1} } + \exp_args:Nf \__str_substr_iii:nnw + { \__str_substr_normalize_range:nn {#2} {#1} } + { \__str_substr_normalize_range:nn {#3} {#1} } } -\cs_new:Npn \str_substr_aux:nnw #1#2 +\cs_new:Npn \__str_substr_iii:nnw #1#2 { - \exp_after:wN \str_collect_delimit_by_q_stop:w - \int_use:N \int_eval:w #2 - #1 \exp_after:wN ; - \tex_romannumeral:D \str_skip_c_zero:w #1 ; + \exp_after:wN \__str_collect_delimit_by_q_stop:w + \int_use:N \__int_eval:w #2 + \c_one - #1 \exp_after:wN ; + \tex_romannumeral:D \__str_skip_c_zero:w #1 ; } -\cs_new:Npn \str_substr_normalize_range:nn #1#2 +\cs_new:Npn \__str_substr_normalize_range:nn #1#2 { \int_eval:n { - \if_num:w #1 < \c_zero - \if_num:w #1 < - #2 \exp_stop_f: + \if_int_compare:w #1 < \c_zero + \if_int_compare:w #1 < - #2 \exp_stop_f: \c_zero \else: - #1 + #2 + #1 + #2 + \c_one \fi: \else: - \if_num:w #1 < #2 \exp_stop_f: - #1 - \else: + \if_int_compare:w #1 > #2 \exp_stop_f: #2 + \else: + #1 \fi: \fi: } @@ -385,45 +388,45 @@ \if_int_compare:w \pdftex_strcmp:D { \tl_to_str:N #1 } { \tl_to_str:N #2 } = \c_zero \prg_return_true: \else: \prg_return_false: \fi: } -\prg_new_conditional:Npnn \str_if_contains_char:NN #1#2 { T , TF } +\prg_new_conditional:Npnn \__str_if_contains_char:NN #1#2 { T , TF } { - \exp_after:wN \str_if_contains_char_aux:NN \exp_after:wN #2 - \tl_to_str:n \exp_after:wN {#1} { \prg_map_break:n { ? \fi: } } - \prg_break_point:n { } + \exp_after:wN \__str_if_contains_char_aux:NN \exp_after:wN #2 + \etex_detokenize:D \exp_after:wN {#1} { \__prg_break:n { ? \fi: } } + \__prg_break_point: \prg_return_false: } -\prg_new_conditional:Npnn \str_if_contains_char:nN #1#2 { TF } +\prg_new_conditional:Npnn \__str_if_contains_char:nN #1#2 { TF } { - \exp_after:wN \str_if_contains_char_aux:NN \exp_after:wN #2 - \tl_to_str:n {#1} { \prg_map_break:n { ? \fi: } } - \prg_break_point:n { } + \exp_after:wN \__str_if_contains_char_aux:NN \exp_after:wN #2 + \tl_to_str:n {#1} { \__prg_break:n { ? \fi: } } + \__prg_break_point: \prg_return_false: } -\cs_new:Npn \str_if_contains_char_aux:NN #1#2 +\cs_new:Npn \__str_if_contains_char_aux:NN #1#2 { \if_charcode:w #1 #2 - \exp_after:wN \str_if_contains_char_true: + \exp_after:wN \__str_if_contains_char_true: \fi: - \str_if_contains_char_aux:NN #1 + \__str_if_contains_char_aux:NN #1 } -\cs_new_nopar:Npn \str_if_contains_char_true: - { \prg_map_break:n { \prg_return_true: \use_none:n } } -\prg_new_conditional:Npnn \str_aux_octal_use:N #1 { TF } +\cs_new_nopar:Npn \__str_if_contains_char_true: + { \__prg_break:n { \prg_return_true: \use_none:n } } +\prg_new_conditional:Npnn \__str_octal_use:N #1 { TF } { - \if_num:w \c_one < '1 \token_to_str:N #1 \exp_stop_f: + \if_int_compare:w \c_one < '1 \token_to_str:N #1 \exp_stop_f: #1 \prg_return_true: \else: \prg_return_false: \fi: } -\prg_new_conditional:Npnn \str_aux_hexadecimal_use:N #1 { TF } +\prg_new_conditional:Npnn \__str_hexadecimal_use:N #1 { TF } { - \if_num:w \c_two < "1 \token_to_str:N #1 \exp_stop_f: + \if_int_compare:w \c_two < "1 \token_to_str:N #1 \exp_stop_f: #1 \prg_return_true: \else: - \if_case:w \int_eval:w + \if_case:w \__int_eval:w \exp_after:wN ` \token_to_str:N #1 - `a - \int_eval_end: + \__int_eval_end: A \or: B \or: C @@ -442,223 +445,223 @@ \cs_generate_variant:Nn \str_show:N { c } \group_begin: \char_set_catcode_other:n { \c_zero } - \tl_set:Nx \l_str_internal_tl { \tl_to_str:n { 0123456789ABCDEF } } - \exp_args:No \tl_map_inline:nn { \l_str_internal_tl " } + \tl_set:Nx \l__str_internal_tl { \tl_to_str:n { 0123456789ABCDEF } } + \exp_args:No \tl_map_inline:nn { \l__str_internal_tl " } { \char_set_lccode:nn {`#1} { \c_zero } } - \tl_map_inline:Nn \l_str_internal_tl + \tl_map_inline:Nn \l__str_internal_tl { - \tl_map_inline:Nn \l_str_internal_tl + \tl_map_inline:Nn \l__str_internal_tl { \char_set_lccode:nn { \c_zero } {"#1##1} \tl_to_lowercase:n { \tl_const:cx - { c_str_byte_ \int_eval:n {"#1##1} _tl } + { c__str_byte_ \int_eval:n {"#1##1} _tl } { ^^@ #1 ##1 } } } } \group_end: -\tl_const:cn { c_str_byte_-1_tl } { { } \use_none:n { } } -\cs_new:Npn \str_output_byte:n #1 - { \str_output_byte:w #1 \str_output_end: } -\cs_new_nopar:Npn \str_output_byte:w +\tl_const:cn { c__str_byte_-1_tl } { { } \use_none:n { } } +\cs_new:Npn \__str_output_byte:n #1 + { \__str_output_byte:w #1 \__str_output_end: } +\cs_new_nopar:Npn \__str_output_byte:w { \exp_after:wN \exp_after:wN \exp_after:wN \use_i:nnn - \cs:w c_str_byte_ \int_use:N \int_eval:w + \cs:w c__str_byte_ \int_use:N \__int_eval:w } -\cs_new:Npn \str_output_hexadecimal:n #1 - { \str_output_hexadecimal:w #1 \str_output_end: } -\cs_new_nopar:Npn \str_output_hexadecimal:w +\cs_new:Npn \__str_output_hexadecimal:n #1 + { \__str_output_hexadecimal:w #1 \__str_output_end: } +\cs_new_nopar:Npn \__str_output_hexadecimal:w { \exp_after:wN \exp_after:wN \exp_after:wN \use_none:n - \cs:w c_str_byte_ \int_use:N \int_eval:w + \cs:w c__str_byte_ \int_use:N \__int_eval:w } -\cs_new_nopar:Npn \str_output_end: - { \int_eval_end: _tl \cs_end: } -\cs_new:Npn \str_output_byte_pair_be:n #1 +\cs_new_nopar:Npn \__str_output_end: + { \__int_eval_end: _tl \cs_end: } +\cs_new:Npn \__str_output_byte_pair_be:n #1 { - \exp_args:Nf \str_output_byte_pair_aux:nnN + \exp_args:Nf \__str_output_byte_pair:nnN { \int_div_truncate:nn { #1 } { "100 } } {#1} \use:nn } -\cs_new:Npn \str_output_byte_pair_le:n #1 +\cs_new:Npn \__str_output_byte_pair_le:n #1 { - \exp_args:Nf \str_output_byte_pair_aux:nnN + \exp_args:Nf \__str_output_byte_pair:nnN { \int_div_truncate:nn { #1 } { "100 } } {#1} \use_ii_i:nn } -\cs_new:Npn \str_output_byte_pair_aux:nnN #1#2#3 +\cs_new:Npn \__str_output_byte_pair:nnN #1#2#3 { #3 - { \str_output_byte:n { #1 } } - { \str_output_byte:n { #2 - #1 * "100 } } + { \__str_output_byte:n { #1 } } + { \__str_output_byte:n { #2 - #1 * "100 } } } -\cs_new_protected:Npn \str_convert_gmap:N #1 +\cs_new_protected:Npn \__str_convert_gmap:N #1 { - \tl_gset:Nx \g_str_result_tl + \tl_gset:Nx \g__str_result_tl { - \exp_after:wN \str_convert_gmap_loop:NN + \exp_after:wN \__str_convert_gmap_loop:NN \exp_after:wN #1 - \g_str_result_tl { ? \prg_map_break: } - \prg_break_point:n { } + \g__str_result_tl { ? \__prg_break: } + \__prg_break_point: } } -\cs_new:Npn \str_convert_gmap_loop:NN #1#2 +\cs_new:Npn \__str_convert_gmap_loop:NN #1#2 { \use_none:n #2 #1#2 - \str_convert_gmap_loop:NN #1 + \__str_convert_gmap_loop:NN #1 } -\cs_new_protected:Npn \str_convert_gmap_internal:N #1 +\cs_new_protected:Npn \__str_convert_gmap_internal:N #1 { - \tl_gset:Nx \g_str_result_tl + \tl_gset:Nx \g__str_result_tl { - \exp_after:wN \str_convert_gmap_internal_loop:Nww + \exp_after:wN \__str_convert_gmap_internal_loop:Nww \exp_after:wN #1 - \g_str_result_tl \s_tl \q_stop \prg_map_break: \s_tl - \prg_break_point:n { } + \g__str_result_tl \s__tl \q_stop \__prg_break: \s__tl + \__prg_break_point: } } -\cs_new:Npn \str_convert_gmap_internal_loop:Nww #1 #2 \s_tl #3 \s_tl +\cs_new:Npn \__str_convert_gmap_internal_loop:Nww #1 #2 \s__tl #3 \s__tl { \use_none_delimit_by_q_stop:w #3 \q_stop #1 {#3} - \str_convert_gmap_internal_loop:Nww #1 + \__str_convert_gmap_internal_loop:Nww #1 } -\cs_new_protected:Npn \str_if_flag_error:nnx #1 +\cs_new_protected:Npn \__str_if_flag_error:nnx #1 { \flag_if_raised:nTF {#1} - { \msg_kernel_error:nnx { str } } + { \__msg_kernel_error:nnx { str } } { \use_none:nn } } -\cs_new_protected:Npn \str_if_flag_no_error:nnx #1#2#3 - { \flag_if_raised:nT {#1} { \bool_gset_true:N \g_str_error_bool } } -\cs_new_protected:Npn \str_if_flag_times:nT #1#2 +\cs_new_protected:Npn \__str_if_flag_no_error:nnx #1#2#3 + { \flag_if_raised:nT {#1} { \bool_gset_true:N \g__str_error_bool } } +\cs_new_protected:Npn \__str_if_flag_times:nT #1#2 { \flag_if_raised:nT {#1} { #2~(x \flag_height:n {#1} ) } } \cs_new_protected_nopar:Npn \str_set_convert:Nnnn - { \str_convert_aux_i:nNNnnn { } \tl_set_eq:NN } + { \__str_convert_i:nNNnnn { } \tl_set_eq:NN } \cs_new_protected_nopar:Npn \str_gset_convert:Nnnn - { \str_convert_aux_i:nNNnnn { } \tl_gset_eq:NN } + { \__str_convert_i:nNNnnn { } \tl_gset_eq:NN } \prg_new_protected_conditional:Npnn \str_set_convert:Nnnn #1#2#3#4 { T , F , TF } { - \bool_gset_false:N \g_str_error_bool - \str_convert_aux_i:nNNnnn - { \cs_set_eq:NN \str_if_flag_error:nnx \str_if_flag_no_error:nnx } + \bool_gset_false:N \g__str_error_bool + \__str_convert_i:nNNnnn + { \cs_set_eq:NN \__str_if_flag_error:nnx \__str_if_flag_no_error:nnx } \tl_set_eq:NN #1 {#2} {#3} {#4} - \bool_if:NTF \g_str_error_bool \prg_return_false: \prg_return_true: + \bool_if:NTF \g__str_error_bool \prg_return_false: \prg_return_true: } \prg_new_protected_conditional:Npnn \str_gset_convert:Nnnn #1#2#3#4 { T , F , TF } { - \bool_gset_false:N \g_str_error_bool - \str_convert_aux_i:nNNnnn - { \cs_set_eq:NN \str_if_flag_error:nnx \str_if_flag_no_error:nnx } + \bool_gset_false:N \g__str_error_bool + \__str_convert_i:nNNnnn + { \cs_set_eq:NN \__str_if_flag_error:nnx \__str_if_flag_no_error:nnx } \tl_gset_eq:NN #1 {#2} {#3} {#4} - \bool_if:NTF \g_str_error_bool \prg_return_false: \prg_return_true: + \bool_if:NTF \g__str_error_bool \prg_return_false: \prg_return_true: } -\cs_new_protected:Npn \str_convert_aux_i:nNNnnn #1#2#3#4#5#6 +\cs_new_protected:Npn \__str_convert_i:nNNnnn #1#2#3#4#5#6 { \group_begin: #1 - \str_gset_other:Nn \g_str_result_tl {#4} - \exp_after:wN \str_convert_aux_ii:wwwnn + \__str_gset_other:Nn \g__str_result_tl {#4} + \exp_after:wN \__str_convert_ii:wwwnn \tl_to_str:n {#5} /// \q_stop { decode } { unescape } \prg_do_nothing: - \str_convert_decode_: - \exp_after:wN \str_convert_aux_ii:wwwnn + \__str_convert_decode_: + \exp_after:wN \__str_convert_ii:wwwnn \tl_to_str:n {#6} /// \q_stop { encode } { escape } \use_ii_i:nn - \str_convert_encode_: + \__str_convert_encode_: \group_end: - #2 #3 \g_str_result_tl + #2 #3 \g__str_result_tl } -\cs_new_protected:Npn \str_convert_aux_ii:wwwnn +\cs_new_protected:Npn \__str_convert_ii:wwwnn #1 / #2 // #3 \q_stop #4#5 { - \str_convert_aux_iv:nnn {enc} {#4} {#1} - \str_convert_aux_iv:nnn {esc} {#5} {#2} - \exp_args:Ncc \str_convert_aux_iii:NNnNN - { str_convert_#4_#1: } { str_convert_#5_#2: } {#2} + \__str_convert_iv:nnn {enc} {#4} {#1} + \__str_convert_iv:nnn {esc} {#5} {#2} + \exp_args:Ncc \__str_convert_iii:NNnNN + { __str_convert_#4_#1: } { __str_convert_#5_#2: } {#2} } -\cs_new_protected:Npn \str_convert_aux_iii:NNnNN #1#2#3#4#5 +\cs_new_protected:Npn \__str_convert_iii:NNnNN #1#2#3#4#5 { \if_meaning:w #1 #5 \tl_if_empty:nF {#3} - { \msg_kernel_error:nnx { str } { native-escaping } {#3} } + { \__msg_kernel_error:nnx { str } { native-escaping } {#3} } #1 \else: #4 #2 #1 \fi: } -\cs_new_protected:Npn \str_convert_aux_iv:nnn #1#2#3 +\cs_new_protected:Npn \__str_convert_iv:nnn #1#2#3 { - \cs_if_exist:cF { str_convert_#2_#3: } + \cs_if_exist:cF { __str_convert_#2_#3: } { - \exp_args:Nx \str_convert_aux_v:nnnn - { \str_convert_lowercase_alphanum:n {#3} } + \exp_args:Nx \__str_convert_v:nnnn + { \__str_convert_lowercase_alphanum:n {#3} } {#1} {#2} {#3} } } -\cs_new_protected:Npn \str_convert_aux_v:nnnn #1#2#3#4 +\cs_new_protected:Npn \__str_convert_v:nnnn #1#2#3#4 { - \cs_if_exist:cF { str_convert_#3_#1: } + \cs_if_exist:cF { __str_convert_#3_#1: } { - \prop_get:NnNF \g_str_file_prop {#1} \l_str_internal_tl - { \tl_set:Nn \l_str_internal_tl {#1} } - \cs_if_exist:cF { str_convert_#3_ \l_str_internal_tl : } + \prop_get:NnNF \g__str_alias_prop {#1} \l__str_internal_tl + { \tl_set:Nn \l__str_internal_tl {#1} } + \cs_if_exist:cF { __str_convert_#3_ \l__str_internal_tl : } { - \file_if_exist:nTF { l3str-#2- \l_str_internal_tl .def } + \file_if_exist:nTF { l3str-#2- \l__str_internal_tl .def } { \group_begin: - \str_load_catcodes: - \file_input:n { l3str-#2- \l_str_internal_tl .def } + \__str_load_catcodes: + \file_input:n { l3str-#2- \l__str_internal_tl .def } \group_end: } { - \tl_clear:N \l_str_internal_tl - \msg_kernel_error:nnxx { str } { unknown-#2 } {#4} {#1} + \tl_clear:N \l__str_internal_tl + \__msg_kernel_error:nnxx { str } { unknown-#2 } {#4} {#1} } } - \cs_if_exist:cF { str_convert_#3_#1: } + \cs_if_exist:cF { __str_convert_#3_#1: } { - \cs_gset_eq:cc { str_convert_#3_#1: } - { str_convert_#3_ \l_str_internal_tl : } + \cs_gset_eq:cc { __str_convert_#3_#1: } + { __str_convert_#3_ \l__str_internal_tl : } } } - \cs_gset_eq:cc { str_convert_#3_#4: } { str_convert_#3_#1: } + \cs_gset_eq:cc { __str_convert_#3_#4: } { __str_convert_#3_#1: } } -\cs_new:Npn \str_convert_lowercase_alphanum:n #1 +\cs_new:Npn \__str_convert_lowercase_alphanum:n #1 { - \exp_after:wN \str_convert_lowercase_alphanum_loop:N - \tl_to_str:n {#1} { ? \prg_map_break: } - \prg_break_point:n { } + \exp_after:wN \__str_convert_lowercase_alphanum_loop:N + \tl_to_str:n {#1} { ? \__prg_break: } + \__prg_break_point: } -\cs_new:Npn \str_convert_lowercase_alphanum_loop:N #1 +\cs_new:Npn \__str_convert_lowercase_alphanum_loop:N #1 { \use_none:n #1 - \if_num:w `#1 < \c_ninety_one - \if_num:w `#1 < \c_sixty_five - \if_num:w \c_one < 1#1 \exp_stop_f: + \if_int_compare:w `#1 < \c_ninety_one + \if_int_compare:w `#1 < \c_sixty_five + \if_int_compare:w \c_one < 1#1 \exp_stop_f: #1 \fi: \else: - \str_output_byte:n { `#1 + \c_thirty_two } + \__str_output_byte:n { `#1 + \c_thirty_two } \fi: \else: - \if_num:w `#1 < \c_one_hundred_twenty_three - \if_num:w `#1 < \c_ninety_seven + \if_int_compare:w `#1 < \c_one_hundred_twenty_three + \if_int_compare:w `#1 < \c_ninety_seven \else: #1 \fi: \fi: \fi: - \str_convert_lowercase_alphanum_loop:N + \__str_convert_lowercase_alphanum_loop:N } -\cs_new_protected:Npn \str_load_catcodes: +\cs_new_protected:Npn \__str_load_catcodes: { \char_set_catcode_escape:N \\ \char_set_catcode_group_begin:N \{ @@ -677,64 +680,64 @@ \int_set:Nn \tex_endlinechar:D {32} } \pdftex_if_engine:TF - { \cs_new_eq:NN \str_filter_bytes:n \use:n } + { \cs_new_eq:NN \__str_filter_bytes:n \use:n } { - \cs_new:Npn \str_filter_bytes:n #1 + \cs_new:Npn \__str_filter_bytes:n #1 { - \str_filter_bytes_aux:N #1 - { ? \prg_map_break: } - \prg_break_point:n { } + \__str_filter_bytes_aux:N #1 + { ? \__prg_break: } + \__prg_break_point: } - \cs_new:Npn \str_filter_bytes_aux:N #1 + \cs_new:Npn \__str_filter_bytes_aux:N #1 { \use_none:n #1 - \if_num:w `#1 < 256 \exp_stop_f: + \if_int_compare:w `#1 < 256 \exp_stop_f: #1 \else: \flag_raise:n { str_byte } \fi: - \str_filter_bytes_aux:N + \__str_filter_bytes_aux:N } } \pdftex_if_engine:TF - { \cs_new_protected_nopar:Npn \str_convert_unescape_: { } } + { \cs_new_protected_nopar:Npn \__str_convert_unescape_: { } } { - \cs_new_protected_nopar:Npn \str_convert_unescape_: + \cs_new_protected_nopar:Npn \__str_convert_unescape_: { \flag_clear:n { str_byte } - \tl_gset:Nx \g_str_result_tl - { \exp_args:No \str_filter_bytes:n \g_str_result_tl } - \str_if_flag_error:nnx { str_byte } { non-byte } { bytes } + \tl_gset:Nx \g__str_result_tl + { \exp_args:No \__str_filter_bytes:n \g__str_result_tl } + \__str_if_flag_error:nnx { str_byte } { non-byte } { bytes } } } -\cs_new_eq:NN \str_convert_unescape_bytes: \str_convert_unescape_: -\cs_new_protected_nopar:Npn \str_convert_escape_: { } -\cs_new_eq:NN \str_convert_escape_bytes: \str_convert_escape_: -\cs_new_protected_nopar:Npn \str_convert_decode_: - { \str_convert_gmap:N \str_decode_native_char:N } -\cs_new:Npn \str_decode_native_char:N #1 - { #1 \s_tl \int_value:w `#1 \s_tl } +\cs_new_eq:NN \__str_convert_unescape_bytes: \__str_convert_unescape_: +\cs_new_protected_nopar:Npn \__str_convert_escape_: { } +\cs_new_eq:NN \__str_convert_escape_bytes: \__str_convert_escape_: +\cs_new_protected_nopar:Npn \__str_convert_decode_: + { \__str_convert_gmap:N \__str_decode_native_char:N } +\cs_new:Npn \__str_decode_native_char:N #1 + { #1 \s__tl \__int_value:w `#1 \s__tl } \group_begin: \char_set_catcode_other:n { 0 } \pdftex_if_engine:TF { - \cs_new_protected_nopar:Npn \str_convert_encode_: + \cs_new_protected_nopar:Npn \__str_convert_encode_: { \flag_clear:n { str_error } - \str_convert_gmap_internal:N \str_encode_native_char:n - \str_if_flag_error:nnx { str_error } + \__str_convert_gmap_internal:N \__str_encode_native_char:n + \__str_if_flag_error:nnx { str_error } { pdfTeX-native-overflow } { } } - \cs_new:Npn \str_encode_native_char:n #1 + \cs_new:Npn \__str_encode_native_char:n #1 { - \if_num:w #1 < \c_two_hundred_fifty_six - \str_output_byte:n {#1} + \if_int_compare:w #1 < \c_two_hundred_fifty_six + \__str_output_byte:n {#1} \else: \flag_raise:n { str_error } ? \fi: } - \msg_kernel_new:nnnn { str } { pdfTeX-native-overflow } + \__msg_kernel_new:nnnn { str } { pdfTeX-native-overflow } { Character~code~too~large~for~pdfTeX. } { The~pdfTeX~engine~only~supports~8-bit~characters:~ @@ -743,141 +746,141 @@ } } { - \cs_new_protected_nopar:Npn \str_convert_encode_: + \cs_new_protected_nopar:Npn \__str_convert_encode_: { - \int_zero:N \l_tl_build_offset_int - \tl_gset_build_x:Nw \g_str_result_tl - \exp_after:wN \str_encode_native_loop:w - \g_str_result_tl \s_tl { \q_stop \prg_map_break: } \s_tl - \prg_break_point:n { } - \tl_build_end: + \int_zero:N \l__tl_build_offset_int + \__tl_gbuild_x:Nw \g__str_result_tl + \exp_after:wN \__str_encode_native_loop:w + \g__str_result_tl \s__tl { \q_stop \__prg_break: } \s__tl + \__prg_break_point: + \__tl_build_end: } - \cs_new_protected:Npn \str_encode_native_loop:w #1 \s_tl #2 \s_tl + \cs_new_protected:Npn \__str_encode_native_loop:w #1 \s__tl #2 \s__tl { \use_none_delimit_by_q_stop:w #2 \q_stop - \tex_lccode:D \l_str_internal_int \int_eval:w #2 \int_eval_end: - \tl_to_lowercase:n { \tl_build_one:n { ^^@ } } - \str_encode_native_loop:w + \tex_lccode:D \l__str_internal_int \__int_eval:w #2 \__int_eval_end: + \tl_to_lowercase:n { \__tl_build_one:n { ^^@ } } + \__str_encode_native_loop:w } } \group_end: -\cs_new_protected:Npn \str_declare_eight_bit_encoding:nnn #1#2#3 +\cs_new_protected:Npn \__str_declare_eight_bit_encoding:nnn #1#2#3 { - \tl_set:Nn \l_str_internal_tl {#1} - \cs_new_protected_nopar:cpn { str_convert_decode_#1: } - { \str_convert_decode_eight_bit:n {#1} } - \cs_new_protected_nopar:cpn { str_convert_encode_#1: } - { \str_convert_encode_eight_bit:n {#1} } - \tl_const:cn { c_str_encoding_#1_tl } {#2} - \tl_const:cn { c_str_encoding_#1_missing_tl } {#3} + \tl_set:Nn \l__str_internal_tl {#1} + \cs_new_protected_nopar:cpn { __str_convert_decode_#1: } + { \__str_convert_decode_eight_bit:n {#1} } + \cs_new_protected_nopar:cpn { __str_convert_encode_#1: } + { \__str_convert_encode_eight_bit:n {#1} } + \tl_const:cn { c__str_encoding_#1_tl } {#2} + \tl_const:cn { c__str_encoding_#1_missing_tl } {#3} } -\cs_new_protected:Npn \str_convert_decode_eight_bit:n #1 +\cs_new_protected:Npn \__str_convert_decode_eight_bit:n #1 { \group_begin: - \int_zero:N \l_str_internal_int - \exp_last_unbraced:Nx \str_decode_eight_bit_load:nn - { \tl_use:c { c_str_encoding_#1_tl } } - { \q_stop \prg_map_break: } { } - \prg_break_point:n { } - \exp_last_unbraced:Nx \str_decode_eight_bit_load_missing:n - { \tl_use:c { c_str_encoding_#1_missing_tl } } - { \q_stop \prg_map_break: } - \prg_break_point:n { } + \int_zero:N \l__str_internal_int + \exp_last_unbraced:Nx \__str_decode_eight_bit_load:nn + { \tl_use:c { c__str_encoding_#1_tl } } + { \q_stop \__prg_break: } { } + \__prg_break_point: + \exp_last_unbraced:Nx \__str_decode_eight_bit_load_missing:n + { \tl_use:c { c__str_encoding_#1_missing_tl } } + { \q_stop \__prg_break: } + \__prg_break_point: \flag_clear:n { str_error } - \str_convert_gmap:N \str_decode_eight_bit_char:N - \str_if_flag_error:nnx { str_error } { decode-8-bit } {#1} + \__str_convert_gmap:N \__str_decode_eight_bit_char:N + \__str_if_flag_error:nnx { str_error } { decode-8-bit } {#1} \group_end: } -\cs_new_protected:Npn \str_decode_eight_bit_load:nn #1#2 +\cs_new_protected:Npn \__str_decode_eight_bit_load:nn #1#2 { \use_none_delimit_by_q_stop:w #1 \q_stop - \tex_dimen:D "#1 = \l_str_internal_int sp \scan_stop: - \tex_skip:D \l_str_internal_int = "#1 sp \scan_stop: - \tex_toks:D \l_str_internal_int \exp_after:wN { \int_value:w "#2 } - \tex_advance:D \l_str_internal_int \c_one - \str_decode_eight_bit_load:nn + \tex_dimen:D "#1 = \l__str_internal_int sp \scan_stop: + \tex_skip:D \l__str_internal_int = "#1 sp \scan_stop: + \tex_toks:D \l__str_internal_int \exp_after:wN { \__int_value:w "#2 } + \tex_advance:D \l__str_internal_int \c_one + \__str_decode_eight_bit_load:nn } -\cs_new_protected:Npn \str_decode_eight_bit_load_missing:n #1 +\cs_new_protected:Npn \__str_decode_eight_bit_load_missing:n #1 { \use_none_delimit_by_q_stop:w #1 \q_stop - \tex_dimen:D "#1 = \l_str_internal_int sp \scan_stop: - \tex_skip:D \l_str_internal_int = "#1 sp \scan_stop: - \tex_toks:D \l_str_internal_int \exp_after:wN - { \int_use:N \c_str_replacement_char_int } - \tex_advance:D \l_str_internal_int \c_one - \str_decode_eight_bit_load_missing:n - } -\cs_new:Npn \str_decode_eight_bit_char:N #1 - { - #1 \s_tl - \if_num:w \tex_dimen:D `#1 < \l_str_internal_int - \if_num:w \tex_skip:D \tex_dimen:D `#1 = `#1 \exp_stop_f: + \tex_dimen:D "#1 = \l__str_internal_int sp \scan_stop: + \tex_skip:D \l__str_internal_int = "#1 sp \scan_stop: + \tex_toks:D \l__str_internal_int \exp_after:wN + { \int_use:N \c__str_replacement_char_int } + \tex_advance:D \l__str_internal_int \c_one + \__str_decode_eight_bit_load_missing:n + } +\cs_new:Npn \__str_decode_eight_bit_char:N #1 + { + #1 \s__tl + \if_int_compare:w \tex_dimen:D `#1 < \l__str_internal_int + \if_int_compare:w \tex_skip:D \tex_dimen:D `#1 = `#1 \exp_stop_f: \tex_the:D \tex_toks:D \tex_dimen:D \fi: \fi: - \int_value:w `#1 \s_tl + \__int_value:w `#1 \s__tl } -\cs_new_protected:Npn \str_convert_encode_eight_bit:n #1 +\cs_new_protected:Npn \__str_convert_encode_eight_bit:n #1 { \group_begin: - \int_zero:N \l_str_internal_int - \exp_last_unbraced:Nx \str_encode_eight_bit_load:nn - { \tl_use:c { c_str_encoding_#1_tl } } - { \q_stop \prg_map_break: } { } - \prg_break_point:n { } + \int_zero:N \l__str_internal_int + \exp_last_unbraced:Nx \__str_encode_eight_bit_load:nn + { \tl_use:c { c__str_encoding_#1_tl } } + { \q_stop \__prg_break: } { } + \__prg_break_point: \flag_clear:n { str_error } - \str_convert_gmap_internal:N \str_encode_eight_bit_char:n - \str_if_flag_error:nnx { str_error } { encode-8-bit } {#1} + \__str_convert_gmap_internal:N \__str_encode_eight_bit_char:n + \__str_if_flag_error:nnx { str_error } { encode-8-bit } {#1} \group_end: } -\cs_new_protected:Npn \str_encode_eight_bit_load:nn #1#2 +\cs_new_protected:Npn \__str_encode_eight_bit_load:nn #1#2 { \use_none_delimit_by_q_stop:w #1 \q_stop - \tex_dimen:D "#2 = \l_str_internal_int sp \scan_stop: - \tex_skip:D \l_str_internal_int = "#2 sp \scan_stop: - \exp_args:NNf \tex_toks:D \l_str_internal_int - { \str_output_byte:n { "#1 } } - \tex_advance:D \l_str_internal_int \c_one - \str_encode_eight_bit_load:nn + \tex_dimen:D "#2 = \l__str_internal_int sp \scan_stop: + \tex_skip:D \l__str_internal_int = "#2 sp \scan_stop: + \exp_args:NNf \tex_toks:D \l__str_internal_int + { \__str_output_byte:n { "#1 } } + \tex_advance:D \l__str_internal_int \c_one + \__str_encode_eight_bit_load:nn } -\cs_new:Npn \str_encode_eight_bit_char:n #1 +\cs_new:Npn \__str_encode_eight_bit_char:n #1 { - \if_num:w #1 > \c_max_register_int + \if_int_compare:w #1 > \c_max_register_int \flag_raise:n { str_error } \else: - \if_num:w \tex_dimen:D #1 < \l_str_internal_int - \if_num:w \tex_skip:D \tex_dimen:D #1 = #1 \exp_stop_f: + \if_int_compare:w \tex_dimen:D #1 < \l__str_internal_int + \if_int_compare:w \tex_skip:D \tex_dimen:D #1 = #1 \exp_stop_f: \tex_the:D \tex_toks:D \tex_dimen:D #1 \exp_stop_f: \exp_after:wN \exp_after:wN \exp_after:wN \use_none:nn \fi: \fi: - \str_encode_eight_bit_char_aux:n {#1} + \__str_encode_eight_bit_char_aux:n {#1} \fi: } -\cs_new:Npn \str_encode_eight_bit_char_aux:n #1 +\cs_new:Npn \__str_encode_eight_bit_char_aux:n #1 { - \if_num:w #1 < \c_two_hundred_fifty_six - \str_output_byte:n {#1} + \if_int_compare:w #1 < \c_two_hundred_fifty_six + \__str_output_byte:n {#1} \else: \flag_raise:n { str_error } \fi: } -\msg_kernel_new:nnn { str } { unknown-esc } +\__msg_kernel_new:nnn { str } { unknown-esc } { Escaping~scheme~'#1'~(filtered:~'#2')~unknown. } -\msg_kernel_new:nnn { str } { unknown-enc } +\__msg_kernel_new:nnn { str } { unknown-enc } { Encoding~scheme~'#1'~(filtered:~'#2')~unknown. } -\msg_kernel_new:nnnn { str } { native-escaping } +\__msg_kernel_new:nnnn { str } { native-escaping } { The~'native'~encoding~scheme~does~not~support~any~escaping. } { Since~native~strings~do~not~consist~in~bytes,~ none~of~the~escaping~methods~make~sense.~ The~specified~escaping,~'#1',~will be ignored. } -\msg_kernel_new:nnn { str } { file-not-found } +\__msg_kernel_new:nnn { str } { file-not-found } { File~'l3str-#1.def'~not~found. } \pdftex_if_engine:F { - \msg_kernel_new:nnnn { str } { non-byte } + \__msg_kernel_new:nnnn { str } { non-byte } { String~invalid~in~escaping~'#1':~it~may~only~contain~bytes. } { Some~characters~in~the~string~you~asked~to~convert~are~not~ @@ -891,22 +894,22 @@ } } } -\msg_kernel_new:nnnn { str } { decode-8-bit } +\__msg_kernel_new:nnnn { str } { decode-8-bit } { Invalid~string~in~encoding~'#1'. } { LaTeX~came~across~a~byte~which~is~not~defined~to~represent~ any~character~in~the~encoding~'#1'. } -\msg_kernel_new:nnnn { str } { encode-8-bit } +\__msg_kernel_new:nnnn { str } { encode-8-bit } { Unicode~string~cannot~be~converted~to~encoding~'#1'. } { The~encoding~'#1'~only~contains~a~subset~of~all~Unicode~characters.~ LaTeX~was~asked~to~convert~a~string~to~that~encoding,~but~that~ string~contains~a~character~that~'#1'~does~not~support. } -\cs_set:Npn \str_length_skip_spaces:N - { \exp_args:No \str_length_skip_spaces:n } -\cs_set_eq:NN \str_length_skip_spaces:n \str_length_ignore_spaces:n +\cs_new_eq:NN \str_length:N \str_count:N +\cs_new_eq:NN \str_length:n \str_count:n +\cs_new_eq:NN \str_length_ignore_spaces:n \str_count_ignore_spaces:n %% %% %% End of file `l3str.sty'. diff --git a/Master/texmf-dist/tex/latex/l3experimental/l3str/l3tl-analysis.sty b/Master/texmf-dist/tex/latex/l3experimental/l3str/l3tl-analysis.sty index 545d7ce7069..8b90db9c751 100644 --- a/Master/texmf-dist/tex/latex/l3experimental/l3str/l3tl-analysis.sty +++ b/Master/texmf-dist/tex/latex/l3experimental/l3str/l3tl-analysis.sty @@ -47,216 +47,216 @@ %% ----------------------------------------------------------------------- \RequirePackage{expl3} \GetIdInfo$Id: l3tl-analysis.dtx 3039 2011-12-08 09:22:35Z bruno $ - {L3 Experimental Analysis of Token lists} + {L3 Experimental token lists analysis} \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} \RequirePackage{l3str} -\scan_new:N \s_tl -\tl_new:N \l_tl_analysis_internal_tl -\cs_new_eq:NN \l_tl_analysis_token ? -\cs_new_eq:NN \l_tl_analysis_char_token ? -\int_new:N \l_tl_analysis_normal_int -\int_new:N \l_tl_analysis_index_int -\int_new:N \l_tl_analysis_nesting_int -\int_new:N \l_tl_analysis_type_int -\tl_new:N \g_tl_analysis_result_tl -\cs_new_nopar:Npn \tl_analysis_extract_charcode: - { - \exp_after:wN \tl_analysis_extract_charcode_aux:w - \token_to_meaning:N \l_tl_analysis_token - } -\cs_new:Npn \tl_analysis_extract_charcode_aux:w #1 ~ #2 ~ { ` } -\cs_new:Npn \tl_analysis_cs_space_count:NN #1 #2 +\__scan_new:N \s__tl +\tl_new:N \l__tl_analysis_internal_tl +\cs_new_eq:NN \l__tl_analysis_token ? +\cs_new_eq:NN \l__tl_analysis_char_token ? +\int_new:N \l__tl_analysis_normal_int +\int_new:N \l__tl_analysis_index_int +\int_new:N \l__tl_analysis_nesting_int +\int_new:N \l__tl_analysis_type_int +\tl_new:N \g__tl_analysis_result_tl +\cs_new_nopar:Npn \__tl_analysis_extract_charcode: + { + \exp_after:wN \__tl_analysis_extract_charcode_aux:w + \token_to_meaning:N \l__tl_analysis_token + } +\cs_new:Npn \__tl_analysis_extract_charcode_aux:w #1 ~ #2 ~ { ` } +\cs_new:Npn \__tl_analysis_cs_space_count:NN #1 #2 { \exp_after:wN #1 - \int_value:w \int_eval:w \c_zero - \exp_after:wN \tl_analysis_cs_space_count:w + \__int_value:w \__int_eval:w \c_zero + \exp_after:wN \__tl_analysis_cs_space_count:w \token_to_str:N #2 - \fi: \tl_analysis_cs_space_count_end:w ; ~ ! + \fi: \__tl_analysis_cs_space_count_end:w ; ~ ! } -\cs_new:Npn \tl_analysis_cs_space_count:w #1 ~ +\cs_new:Npn \__tl_analysis_cs_space_count:w #1 ~ { \if_false: #1 #1 \fi: + \c_one - \tl_analysis_cs_space_count:w + \__tl_analysis_cs_space_count:w } -\cs_new:Npn \tl_analysis_cs_space_count_end:w ; #1 \fi: #2 ! - { \exp_after:wN ; \int_value:w \str_length_ignore_spaces:n {#1} ; } -\cs_new_protected:Npn \tl_analysis:n #1 +\cs_new:Npn \__tl_analysis_cs_space_count_end:w ; #1 \fi: #2 ! + { \exp_after:wN ; \__int_value:w \str_count_ignore_spaces:n {#1} ; } +\cs_new_protected:Npn \__tl_analysis:n #1 { \group_begin: \group_align_safe_begin: - \tl_analysis_setup:n {#1} - \tl_analysis_i:n {#1} - \tl_analysis_ii:n {#1} + \__tl_analysis_setup:n {#1} + \__tl_analysis_i:n {#1} + \__tl_analysis_ii:n {#1} \group_align_safe_end: \group_end: } -\cs_new_protected:Npn \tl_analysis_setup:n #1 +\cs_new_protected:Npn \__tl_analysis_setup:n #1 { \int_set_eq:NN \tex_escapechar:D \c_minus_one - \exp_after:wN \tl_analysis_disable_loop:N - \tl_to_str:n {#1} { ~ } { ? ~ \prg_map_break: } - \prg_break_point:n { } + \exp_after:wN \__tl_analysis_disable_loop:N + \tl_to_str:n {#1} { ~ } { ? ~ \__prg_break: } + \__prg_break_point: } \group_begin: \char_set_catcode_active:N \^^@ - \cs_new_protected:Npn \tl_analysis_disable_loop:N #1 + \cs_new_protected:Npn \__tl_analysis_disable_loop:N #1 { \tex_lccode:D \c_zero `#1 ~ \tl_to_lowercase:n { \tex_let:D ^^@ } \c_undefined:D - \tl_analysis_disable_loop:N + \__tl_analysis_disable_loop:N } \group_end: -\cs_new_protected:Npn \tl_analysis_i:n #1 +\cs_new_protected:Npn \__tl_analysis_i:n #1 { \int_set:Nn \tex_escapechar:D { 92 } - \int_zero:N \l_tl_analysis_normal_int - \int_zero:N \l_tl_analysis_index_int - \int_zero:N \l_tl_analysis_nesting_int - \if_false: { \fi: \tl_analysis_i_loop:w #1 } - \int_decr:N \l_tl_analysis_index_int - } -\cs_new_protected_nopar:Npn \tl_analysis_i_loop:w - { \tex_futurelet:D \l_tl_analysis_token \tl_analysis_i_type:w } -\cs_new_protected_nopar:Npn \tl_analysis_i_type:w - { - \l_tl_analysis_type_int = - \if_meaning:w \l_tl_analysis_token \c_space_token + \int_zero:N \l__tl_analysis_normal_int + \int_zero:N \l__tl_analysis_index_int + \int_zero:N \l__tl_analysis_nesting_int + \if_false: { \fi: \__tl_analysis_i_loop:w #1 } + \int_decr:N \l__tl_analysis_index_int + } +\cs_new_protected_nopar:Npn \__tl_analysis_i_loop:w + { \tex_futurelet:D \l__tl_analysis_token \__tl_analysis_i_type:w } +\cs_new_protected_nopar:Npn \__tl_analysis_i_type:w + { + \l__tl_analysis_type_int = + \if_meaning:w \l__tl_analysis_token \c_space_token \c_zero \else: - \if_catcode:w \exp_not:N \l_tl_analysis_token \c_group_begin_token + \if_catcode:w \exp_not:N \l__tl_analysis_token \c_group_begin_token \c_one \else: - \if_catcode:w \exp_not:N \l_tl_analysis_token \c_group_end_token + \if_catcode:w \exp_not:N \l__tl_analysis_token \c_group_end_token \c_minus_one \else: \c_two \fi: \fi: \fi: - \if_case:w \l_tl_analysis_type_int - \exp_after:wN \tl_analysis_i_space:w - \or: \exp_after:wN \tl_analysis_i_bgroup:w - \or: \exp_after:wN \tl_analysis_i_safe:N - \else: \exp_after:wN \tl_analysis_i_egroup:w + \if_case:w \l__tl_analysis_type_int + \exp_after:wN \__tl_analysis_i_space:w + \or: \exp_after:wN \__tl_analysis_i_bgroup:w + \or: \exp_after:wN \__tl_analysis_i_safe:N + \else: \exp_after:wN \__tl_analysis_i_egroup:w \fi: } -\cs_new_protected_nopar:Npn \tl_analysis_i_space:w +\cs_new_protected_nopar:Npn \__tl_analysis_i_space:w { - \tex_afterassignment:D \tl_analysis_i_space_test:w + \tex_afterassignment:D \__tl_analysis_i_space_test:w \exp_after:wN \cs_set_eq:NN - \exp_after:wN \l_tl_analysis_char_token + \exp_after:wN \l__tl_analysis_char_token \token_to_str:N } -\cs_new_protected_nopar:Npn \tl_analysis_i_space_test:w +\cs_new_protected_nopar:Npn \__tl_analysis_i_space_test:w { - \if_meaning:w \l_tl_analysis_char_token \c_space_token - \tex_toks:D \l_tl_analysis_index_int { \exp_not:n { ~ } } - \tl_analysis_i_store: + \if_meaning:w \l__tl_analysis_char_token \c_space_token + \tex_toks:D \l__tl_analysis_index_int { \exp_not:n { ~ } } + \__tl_analysis_i_store: \else: - \int_incr:N \l_tl_analysis_normal_int + \int_incr:N \l__tl_analysis_normal_int \fi: - \tl_analysis_i_loop:w + \__tl_analysis_i_loop:w } \group_begin: \char_set_catcode_group_begin:N \^^@ \char_set_catcode_group_end:N \^^E - \cs_new_protected_nopar:Npn \tl_analysis_i_bgroup:w - { \tl_analysis_i_group:nw { \exp_after:wN ^^@ \if_false: ^^E \fi: } } + \cs_new_protected_nopar:Npn \__tl_analysis_i_bgroup:w + { \__tl_analysis_i_group:nw { \exp_after:wN ^^@ \if_false: ^^E \fi: } } \char_set_catcode_group_begin:N \^^B \char_set_catcode_group_end:N \^^@ - \cs_new_protected_nopar:Npn \tl_analysis_i_egroup:w - { \tl_analysis_i_group:nw { \if_false: ^^B \fi: ^^@ } } + \cs_new_protected_nopar:Npn \__tl_analysis_i_egroup:w + { \__tl_analysis_i_group:nw { \if_false: ^^B \fi: ^^@ } } \group_end: -\cs_new_protected:Npn \tl_analysis_i_group:nw #1 +\cs_new_protected:Npn \__tl_analysis_i_group:nw #1 { - \tex_lccode:D \c_zero = \tl_analysis_extract_charcode: \scan_stop: - \tl_to_lowercase:n { \tex_toks:D \l_tl_analysis_index_int {#1} } - \if_num:w \tex_lccode:D \c_zero = \tex_escapechar:D + \tex_lccode:D \c_zero = \__tl_analysis_extract_charcode: \scan_stop: + \tl_to_lowercase:n { \tex_toks:D \l__tl_analysis_index_int {#1} } + \if_int_compare:w \tex_lccode:D \c_zero = \tex_escapechar:D \int_set:Nn \tex_escapechar:D { 139 - \tex_escapechar:D } \fi: - \tex_afterassignment:D \tl_analysis_i_group_test:w + \tex_afterassignment:D \__tl_analysis_i_group_test:w \exp_after:wN \cs_set_eq:NN - \exp_after:wN \l_tl_analysis_char_token + \exp_after:wN \l__tl_analysis_char_token \token_to_str:N } -\cs_new_protected_nopar:Npn \tl_analysis_i_group_test:w +\cs_new_protected_nopar:Npn \__tl_analysis_i_group_test:w { - \if_charcode:w \l_tl_analysis_token \l_tl_analysis_char_token - \tl_analysis_i_store: + \if_charcode:w \l__tl_analysis_token \l__tl_analysis_char_token + \__tl_analysis_i_store: \else: - \int_incr:N \l_tl_analysis_normal_int + \int_incr:N \l__tl_analysis_normal_int \fi: - \tl_analysis_i_loop:w + \__tl_analysis_i_loop:w } -\cs_new_protected_nopar:Npn \tl_analysis_i_store: +\cs_new_protected_nopar:Npn \__tl_analysis_i_store: { - \tex_advance:D \l_tl_analysis_nesting_int \l_tl_analysis_type_int - \if_num:w \tex_lccode:D \c_zero = \c_thirty_two - \tex_multiply:D \l_tl_analysis_type_int \c_two + \tex_advance:D \l__tl_analysis_nesting_int \l__tl_analysis_type_int + \if_int_compare:w \tex_lccode:D \c_zero = \c_thirty_two + \tex_multiply:D \l__tl_analysis_type_int \c_two \fi: - \tex_skip:D \l_tl_analysis_index_int - = \l_tl_analysis_normal_int sp plus \l_tl_analysis_type_int sp \scan_stop: - \int_incr:N \l_tl_analysis_index_int - \int_zero:N \l_tl_analysis_normal_int - \if_num:w \l_tl_analysis_nesting_int = \c_minus_one - \cs_set_eq:NN \tl_analysis_i_loop:w \scan_stop: + \tex_skip:D \l__tl_analysis_index_int + = \l__tl_analysis_normal_int sp plus \l__tl_analysis_type_int sp \scan_stop: + \int_incr:N \l__tl_analysis_index_int + \int_zero:N \l__tl_analysis_normal_int + \if_int_compare:w \l__tl_analysis_nesting_int = \c_minus_one + \cs_set_eq:NN \__tl_analysis_i_loop:w \scan_stop: \fi: } -\cs_new_protected:Npn \tl_analysis_i_safe:N #1 +\cs_new_protected:Npn \__tl_analysis_i_safe:N #1 { \if_charcode:w \scan_stop: \exp_after:wN \use_none:n \token_to_str:N #1 \prg_do_nothing: \scan_stop: - \int_incr:N \l_tl_analysis_normal_int + \int_incr:N \l__tl_analysis_normal_int \else: - \tl_analysis_cs_space_count:NN \tl_analysis_i_cs:ww #1 + \__tl_analysis_cs_space_count:NN \__tl_analysis_i_cs:ww #1 \fi: - \tl_analysis_i_loop:w + \__tl_analysis_i_loop:w } -\cs_new_protected:Npn \tl_analysis_i_cs:ww #1; #2; +\cs_new_protected:Npn \__tl_analysis_i_cs:ww #1; #2; { - \if_num:w #1 > \c_zero - \tex_skip:D \l_tl_analysis_index_int - = \int_eval:w \l_tl_analysis_normal_int + \c_one sp \scan_stop: - \tex_advance:D \l_tl_analysis_index_int #1 \exp_stop_f: - \l_tl_analysis_normal_int #2 \exp_stop_f: + \if_int_compare:w #1 > \c_zero + \tex_skip:D \l__tl_analysis_index_int + = \__int_eval:w \l__tl_analysis_normal_int + \c_one sp \scan_stop: + \tex_advance:D \l__tl_analysis_index_int #1 \exp_stop_f: + \l__tl_analysis_normal_int #2 \exp_stop_f: \else: - \tex_advance:D \l_tl_analysis_normal_int #2 \exp_stop_f: + \tex_advance:D \l__tl_analysis_normal_int #2 \exp_stop_f: \fi: } -\cs_new_protected:Npn \tl_analysis_ii:n #1 +\cs_new_protected:Npn \__tl_analysis_ii:n #1 { - \tl_gset:Nx \g_tl_analysis_result_tl + \tl_gset:Nx \g__tl_analysis_result_tl { - \tl_analysis_ii_loop:w 0; #1 - \prg_break_point:n { } + \__tl_analysis_ii_loop:w 0; #1 + \__prg_break_point: } } -\cs_new:Npn \tl_analysis_ii_loop:w #1; +\cs_new:Npn \__tl_analysis_ii_loop:w #1; { - \exp_after:wN \tl_analysis_ii_normals:ww - \int_value:w \tex_skip:D #1 ; #1 ; + \exp_after:wN \__tl_analysis_ii_normals:ww + \__int_value:w \tex_skip:D #1 ; #1 ; } -\cs_new:Npn \tl_analysis_ii_normals:ww #1; +\cs_new:Npn \__tl_analysis_ii_normals:ww #1; { - \if_num:w #1 = \c_zero - \tl_analysis_ii_special:w + \if_int_compare:w #1 = \c_zero + \__tl_analysis_ii_special:w \fi: - \tl_analysis_ii_normal:wwN #1; + \__tl_analysis_ii_normal:wwN #1; } -\cs_new:Npn \tl_analysis_ii_normal:wwN #1; #2; #3 +\cs_new:Npn \__tl_analysis_ii_normal:wwN #1; #2; #3 { - \exp_not:n { \exp_not:n { #3 } } \s_tl + \exp_not:n { \exp_not:n { #3 } } \s__tl \if_charcode:w \scan_stop: \exp_after:wN \use_none:n \token_to_str:N #3 \prg_do_nothing: \scan_stop: - \exp_after:wN \tl_analysis_ii_char:Nww + \exp_after:wN \__tl_analysis_ii_char:Nww \else: - \exp_after:wN \tl_analysis_ii_cs:Nww + \exp_after:wN \__tl_analysis_ii_cs:Nww \fi: #3 #1; #2; } @@ -267,7 +267,7 @@ \char_set_uccode:nn { `? } { `D } \tl_to_uppercase:n { - \cs_new:Npn \tl_analysis_ii_char:Nww #1 + \cs_new:Npn \__tl_analysis_ii_char:Nww #1 { \if_meaning:w #1 \c_undefined:D ? \else: \if_catcode:w #1 \c_catcode_other_token C \else: @@ -279,40 +279,40 @@ \if_catcode:w #1 \c_space_token A \else: 6 \fi: \fi: \fi: \fi: \fi: \fi: \fi: \fi: - \int_value:w `#1 \s_tl - \exp_after:wN \tl_analysis_ii_normals:ww - \int_use:N \int_eval:w \c_minus_one + + \__int_value:w `#1 \s__tl + \exp_after:wN \__tl_analysis_ii_normals:ww + \int_use:N \__int_eval:w \c_minus_one + } } \group_end: -\cs_new:Npn \tl_analysis_ii_cs:Nww #1 +\cs_new:Npn \__tl_analysis_ii_cs:Nww #1 { - 0 -1 \s_tl - \tl_analysis_cs_space_count:NN \tl_analysis_ii_cs_test:ww #1 + 0 -1 \s__tl + \__tl_analysis_cs_space_count:NN \__tl_analysis_ii_cs_test:ww #1 } -\cs_new:Npn \tl_analysis_ii_cs_test:ww #1 ; #2 ; #3 ; #4 ; +\cs_new:Npn \__tl_analysis_ii_cs_test:ww #1 ; #2 ; #3 ; #4 ; { - \exp_after:wN \tl_analysis_ii_normals:ww - \int_use:N \int_eval:w - \if_num:w #1 = \c_zero + \exp_after:wN \__tl_analysis_ii_normals:ww + \int_use:N \__int_eval:w + \if_int_compare:w #1 = \c_zero #3 \else: - \tex_skip:D \int_eval:w #4 + #1 \int_eval_end: + \tex_skip:D \__int_eval:w #4 + #1 \__int_eval_end: \fi: - #2 \exp_after:wN ; - \int_use:N \int_eval:w #4 + #1 ; + \int_use:N \__int_eval:w #4 + #1 ; } \group_begin: \char_set_catcode_other:N A - \cs_new:Npn \tl_analysis_ii_special:w - \fi: \tl_analysis_ii_normal:wwN 0 ; #1 ; + \cs_new:Npn \__tl_analysis_ii_special:w + \fi: \__tl_analysis_ii_normal:wwN 0 ; #1 ; { \fi: - \if_num:w #1 = \l_tl_analysis_index_int - \exp_after:wN \prg_map_break: + \if_int_compare:w #1 = \l__tl_analysis_index_int + \exp_after:wN \__prg_break: \fi: - \tex_the:D \tex_toks:D #1 \s_tl + \tex_the:D \tex_toks:D #1 \s__tl \if_case:w \etex_gluestretch:D \tex_skip:D #1 \exp_stop_f: A \or: 1 @@ -320,131 +320,131 @@ \else: 2 \fi: \if_int_odd:w \etex_gluestretch:D \tex_skip:D #1 \exp_stop_f: - \exp_after:wN \tl_analysis_ii_special_char:wN \int_use:N + \exp_after:wN \__tl_analysis_ii_special_char:wN \int_use:N \else: - \exp_after:wN \tl_analysis_ii_special_space:w \int_use:N + \exp_after:wN \__tl_analysis_ii_special_space:w \int_use:N \fi: - \int_eval:w \c_one + #1 \exp_after:wN ; + \__int_eval:w \c_one + #1 \exp_after:wN ; \token_to_str:N } \group_end: -\cs_new:Npn \tl_analysis_ii_special_char:wN #1 ; #2 +\cs_new:Npn \__tl_analysis_ii_special_char:wN #1 ; #2 { - \int_value:w `#2 \s_tl - \tl_analysis_ii_loop:w #1 ; + \__int_value:w `#2 \s__tl + \__tl_analysis_ii_loop:w #1 ; } -\cs_new:Npn \tl_analysis_ii_special_space:w #1 ; ~ +\cs_new:Npn \__tl_analysis_ii_special_space:w #1 ; ~ { - 32 \s_tl - \tl_analysis_ii_loop:w #1 ; + 32 \s__tl + \__tl_analysis_ii_loop:w #1 ; } -\cs_new_protected:Npn \tl_analysis_map_inline:nn #1 +\cs_new_protected:Npn \__tl_analysis_map_inline:nn #1 { - \tl_analysis:n {#1} - \int_gincr:N \g_prg_map_int - \exp_args:Nc \tl_analysis_map_inline_aux:Nn - { tl_analysis_map_inline_ \int_use:N \g_prg_map_int :wNw } + \__tl_analysis:n {#1} + \int_gincr:N \g__prg_map_int + \exp_args:Nc \__tl_analysis_map_inline_aux:Nn + { __tl_analysis_map_inline_ \int_use:N \g__prg_map_int :wNw } } -\cs_new_protected:Npn \tl_analysis_map_inline_aux:Nn #1#2 +\cs_new_protected:Npn \__tl_analysis_map_inline_aux:Nn #1#2 { - \cs_gset_protected:Npn #1 ##1 \s_tl ##2 ##3 \s_tl + \cs_gset_protected:Npn #1 ##1 \s__tl ##2 ##3 \s__tl { \use_none:n ##2 #2 #1 } \exp_after:wN #1 - \g_tl_analysis_result_tl - \s_tl { ? \prg_map_break: } \s_tl - \prg_break_point:n { \int_gdecr:N \g_prg_map_int } + \g__tl_analysis_result_tl + \s__tl { ? \tl_map_break: } \s__tl + \__prg_break_point:Nn \tl_map_break: { \int_gdecr:N \g__prg_map_int } } \cs_new_protected:Npn \tl_show_analysis:N #1 { - \exp_args:No \tl_analysis:n {#1} - \msg_aux_show:Nnx #1 + \exp_args:No \__tl_analysis:n {#1} + \__msg_show_variable:Nnn #1 { tl-analysis } { - \exp_after:wN \tl_show_analysis_loop:wNw \g_tl_analysis_result_tl - \s_tl { ? \prg_map_break: } \s_tl - \prg_break_point:n { } + \exp_after:wN \__tl_analysis_show_loop:wNw \g__tl_analysis_result_tl + \s__tl { ? \__prg_break: } \s__tl + \__prg_break_point: } } \cs_new_protected:Npn \tl_show_analysis:n #1 { - \tl_set:Nn \l_tl_analysis_internal_tl {#1} - \tl_show_analysis:N \l_tl_analysis_internal_tl + \tl_set:Nn \l__tl_analysis_internal_tl {#1} + \tl_show_analysis:N \l__tl_analysis_internal_tl } -\cs_new:Npn \tl_show_analysis_loop:wNw #1 \s_tl #2 #3 \s_tl +\cs_new:Npn \__tl_analysis_show_loop:wNw #1 \s__tl #2 #3 \s__tl { \use_none:n #2 \iow_newline: > \c_space_tl \c_space_tl - \if_num:w "#2 = \c_zero - \exp_after:wN \tl_show_analysis_cs:n + \if_int_compare:w "#2 = \c_zero + \exp_after:wN \__tl_analysis_show_cs:n \else: - \if_num:w "#2 = \c_thirteen + \if_int_compare:w "#2 = \c_thirteen \exp_after:wN \exp_after:wN - \exp_after:wN \tl_show_analysis_active:n + \exp_after:wN \__tl_analysis_show_active:n \else: \exp_after:wN \exp_after:wN - \exp_after:wN \tl_show_analysis_normal:n + \exp_after:wN \__tl_analysis_show_normal:n \fi: \fi: {#1} - \tl_show_analysis_loop:wNw + \__tl_analysis_show_loop:wNw } -\cs_new:Npn \tl_show_analysis_normal:n #1 +\cs_new:Npn \__tl_analysis_show_normal:n #1 { \exp_after:wN \token_to_str:N #1 ~ ( \exp_after:wN \token_to_meaning:N #1 ) } -\cs_new:Npn \tl_show_analysis_value:N #1 +\cs_new:Npn \__tl_analysis_show_value:N #1 { \token_if_expandable:NF #1 { - \token_if_chardef:NTF #1 \prg_map_break: { } - \token_if_mathchardef:NTF #1 \prg_map_break: { } - \token_if_dim_register:NTF #1 \prg_map_break: { } - \token_if_int_register:NTF #1 \prg_map_break: { } - \token_if_skip_register:NTF #1 \prg_map_break: { } - \token_if_toks_register:NTF #1 \prg_map_break: { } - \use_none:nnnn - \prg_break_point:n { } + \token_if_chardef:NTF #1 \__prg_break: { } + \token_if_mathchardef:NTF #1 \__prg_break: { } + \token_if_dim_register:NTF #1 \__prg_break: { } + \token_if_int_register:NTF #1 \__prg_break: { } + \token_if_skip_register:NTF #1 \__prg_break: { } + \token_if_toks_register:NTF #1 \__prg_break: { } + \use_none:nnn + \__prg_break_point: \use:n { = \tex_the:D #1 } } } -\cs_new:Npn \tl_show_analysis_cs:n #1 - { \exp_args:No \tl_show_analysis_long:nn {#1} { control~sequence= } } -\cs_new:Npn \tl_show_analysis_active:n #1 - { \exp_args:No \tl_show_analysis_long:nn {#1} { active~character= } } -\cs_new:Npn \tl_show_analysis_long:nn #1 +\cs_new:Npn \__tl_analysis_show_cs:n #1 + { \exp_args:No \__tl_analysis_show_long:nn {#1} { control~sequence= } } +\cs_new:Npn \__tl_analysis_show_active:n #1 + { \exp_args:No \__tl_analysis_show_long:nn {#1} { active~character= } } +\cs_new:Npn \__tl_analysis_show_long:nn #1 { - \tl_show_analysis_long_aux:oofn + \__tl_analysis_show_long_aux:oofn { \token_to_str:N #1 } { \token_to_meaning:N #1 } - { \tl_show_analysis_value:N #1 } + { \__tl_analysis_show_value:N #1 } } -\cs_new:Npn \tl_show_analysis_long_aux:nnnn #1#2#3#4 +\cs_new:Npn \__tl_analysis_show_long_aux:nnnn #1#2#3#4 { \int_compare:nNnTF - { \str_length:n { #1 ~ ( #4 #2 #3 ) } } - > { \l_iow_line_length_int - \c_three } + { \str_count:n { #1 ~ ( #4 #2 #3 ) } } + > { \l_iow_line_count_int - \c_three } { - \str_substr:nnn { #1 ~ ( #4 #2 #3 ) } \c_zero + \str_substr:nnn { #1 ~ ( #4 #2 #3 ) } \c_one { - \l_iow_line_length_int - \c_three - - \str_length:N \c_tl_show_analysis_etc_str + \l_iow_line_count_int - \c_three + - \str_count:N \c__tl_analysis_show_etc_str } - \c_tl_show_analysis_etc_str + \c__tl_analysis_show_etc_str } { #1 ~ ( #4 #2 #3 ) } } -\cs_generate_variant:Nn \tl_show_analysis_long_aux:nnnn { oof } -\tl_const:Nx \c_tl_show_analysis_etc_str % ( +\cs_generate_variant:Nn \__tl_analysis_show_long_aux:nnnn { oof } +\tl_const:Nx \c__tl_analysis_show_etc_str % ( { \token_to_str:N \ETC.) } -\msg_kernel_new:nnn { tl-analysis } { show } +\__msg_kernel_new:nnn { kernel } { show-tl-analysis } { The~token~list~ - \str_if_eq:nnF {#1} { \l_tl_analysis_internal_tl } + \str_if_eq:nnF {#1} { \l__tl_analysis_internal_tl } { \token_to_str:N #1 ~ } \tl_if_empty:NTF #1 { is~empty } diff --git a/Master/texmf-dist/tex/latex/l3experimental/l3str/l3tl-build.sty b/Master/texmf-dist/tex/latex/l3experimental/l3str/l3tl-build.sty index 415a89f4897..1ad40197bab 100644 --- a/Master/texmf-dist/tex/latex/l3experimental/l3str/l3tl-build.sty +++ b/Master/texmf-dist/tex/latex/l3experimental/l3str/l3tl-build.sty @@ -47,74 +47,74 @@ %% ----------------------------------------------------------------------- \RequirePackage{expl3} \GetIdInfo$Id: l3tl-build.dtx 3039 2011-12-08 09:22:35Z bruno $ - {L3 Experimental Build of Token lists} + {L3 Experimental token list construction} \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} -\int_new:N \l_tl_build_start_index_int -\int_new:N \l_tl_build_index_int -\tl_new:N \l_tl_build_result_tl -\cs_new_protected_nopar:Npn \tl_build_unpack: +\int_new:N \l__tl_build_start_index_int +\int_new:N \l__tl_build_index_int +\tl_new:N \l__tl_build_result_tl +\cs_new_protected_nopar:Npn \__tl_build_unpack: { - \tl_put_right:Nx \l_tl_build_result_tl + \tl_put_right:Nx \l__tl_build_result_tl { - \exp_after:wN \tl_build_unpack_loop:w - \int_use:N \l_tl_build_start_index_int ; - \prg_break_point:n { } + \exp_after:wN \__tl_build_unpack_loop:w + \int_use:N \l__tl_build_start_index_int ; + \__prg_break_point: } } -\cs_new:Npn \tl_build_unpack_loop:w #1 ; +\cs_new:Npn \__tl_build_unpack_loop:w #1 ; { - \if_num:w #1 = \l_tl_build_index_int - \exp_after:wN \prg_map_break: + \if_int_compare:w #1 = \l__tl_build_index_int + \exp_after:wN \__prg_break: \fi: \tex_the:D \tex_toks:D #1 \exp_stop_f: - \exp_after:wN \tl_build_unpack_loop:w - \int_use:N \int_eval:w #1 + \c_one ; + \exp_after:wN \__tl_build_unpack_loop:w + \int_use:N \__int_eval:w #1 + \c_one ; } -\cs_new_protected_nopar:Npn \tl_set_build:Nw - { \tl_set_build_aux:NNw \tl_set:Nn } -\cs_new_protected_nopar:Npn \tl_set_build_x:Nw - { \tl_set_build_aux:NNw \tl_set:Nx } -\cs_new_protected_nopar:Npn \tl_gset_build:Nw - { \tl_set_build_aux:NNw \tl_gset:Nn } -\cs_new_protected_nopar:Npn \tl_gset_build_x:Nw - { \tl_set_build_aux:NNw \tl_gset:Nx } -\cs_new_protected:Npn \tl_set_build_aux:NNw #1#2 +\cs_new_protected_nopar:Npn \__tl_build:Nw + { \__tl_build_aux:NNw \tl_set:Nn } +\cs_new_protected_nopar:Npn \__tl_build_x:Nw + { \__tl_build_aux:NNw \tl_set:Nx } +\cs_new_protected_nopar:Npn \__tl_gbuild:Nw + { \__tl_build_aux:NNw \tl_gset:Nn } +\cs_new_protected_nopar:Npn \__tl_gbuild_x:Nw + { \__tl_build_aux:NNw \tl_gset:Nx } +\cs_new_protected:Npn \__tl_build_aux:NNw #1#2 { \group_begin: - \cs_set_nopar:Npn \tl_build_end_assignment:n + \cs_set_nopar:Npn \__tl_build_end_assignment:n { \group_end: #1 #2 } - \int_zero:N \l_tl_build_start_index_int - \int_zero:N \l_tl_build_index_int - \tl_clear:N \l_tl_build_result_tl + \int_zero:N \l__tl_build_start_index_int + \int_zero:N \l__tl_build_index_int + \tl_clear:N \l__tl_build_result_tl } -\cs_new_protected_nopar:Npn \tl_build_end: +\cs_new_protected_nopar:Npn \__tl_build_end: { - \tl_build_unpack: + \__tl_build_unpack: \exp_args:No - \tl_build_end_assignment:n \l_tl_build_result_tl + \__tl_build_end_assignment:n \l__tl_build_result_tl } -\cs_new_eq:NN \tl_build_end_assignment:n \use_none:n -\cs_new_protected:Npn \tl_build_one:n #1 +\cs_new_eq:NN \__tl_build_end_assignment:n \use_none:n +\cs_new_protected:Npn \__tl_build_one:n #1 { - \tex_toks:D \l_tl_build_index_int {#1} - \tex_advance:D \l_tl_build_index_int \c_one - \if_num:w \l_tl_build_index_int > \c_max_register_int - \tl_build_unpack: - \l_tl_build_index_int \l_tl_build_start_index_int + \tex_toks:D \l__tl_build_index_int {#1} + \tex_advance:D \l__tl_build_index_int \c_one + \if_int_compare:w \l__tl_build_index_int > \c_max_register_int + \__tl_build_unpack: + \l__tl_build_index_int \l__tl_build_start_index_int \fi: } -\cs_new_protected:Npn \tl_build_one:o #1 +\cs_new_protected:Npn \__tl_build_one:o #1 { - \tex_toks:D \l_tl_build_index_int \exp_after:wN {#1} - \tex_advance:D \l_tl_build_index_int \c_one - \if_num:w \l_tl_build_index_int > \c_max_register_int - \tl_build_unpack: - \l_tl_build_index_int \l_tl_build_start_index_int + \tex_toks:D \l__tl_build_index_int \exp_after:wN {#1} + \tex_advance:D \l__tl_build_index_int \c_one + \if_int_compare:w \l__tl_build_index_int > \c_max_register_int + \__tl_build_unpack: + \l__tl_build_index_int \l__tl_build_start_index_int \fi: } -\cs_new_protected:Npn \tl_build_one:x #1 - { \use:x { \tl_build_one:n {#1} } } +\cs_new_protected:Npn \__tl_build_one:x #1 + { \use:x { \__tl_build_one:n {#1} } } %% %% %% End of file `l3tl-build.sty'. diff --git a/Master/texmf-dist/tex/latex/l3experimental/xcoffins/xcoffins.sty b/Master/texmf-dist/tex/latex/l3experimental/xcoffins/xcoffins.sty index 0c88f04de5e..2eeee54d1ab 100644 --- a/Master/texmf-dist/tex/latex/l3experimental/xcoffins/xcoffins.sty +++ b/Master/texmf-dist/tex/latex/l3experimental/xcoffins/xcoffins.sty @@ -47,25 +47,25 @@ %% ----------------------------------------------------------------------- %% \RequirePackage{xparse} -\GetIdInfo$Id: xcoffins.dtx 3483 2012-03-03 18:56:29Z bruno $ +\GetIdInfo$Id: xcoffins.dtx 3876 2012-06-30 15:35:39Z joseph $ {L3 Experimental design-level coffins} \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} \keys_define:nn { coffin } { - coffin1-hpole .tl_set:N = \l_coffin_A_hpole_tl , + coffin1-hpole .tl_set:N = \l__coffin_A_hpole_tl , coffin1-hpole .value_required: , - coffin1-vpole .tl_set:N = \l_coffin_A_vpole_tl , + coffin1-vpole .tl_set:N = \l__coffin_A_vpole_tl , coffin1-vpole .value_required: , - coffin2-hpole .tl_set:N = \l_coffin_B_hpole_tl , + coffin2-hpole .tl_set:N = \l__coffin_B_hpole_tl , coffin2-hpole .value_required: , - coffin2-vpole .tl_set:N = \l_coffin_B_vpole_tl , + coffin2-vpole .tl_set:N = \l__coffin_B_vpole_tl , coffin2-vpole .value_required: , - grow-bounding-box .bool_set:N = \l_coffin_bound_box_grow_bool , + grow-bounding-box .bool_set:N = \l__coffin_bound_box_grow_bool , grow-bounding-box .default:n = true , - hoffset .dim_set:N = \l_coffin_hoffset_dim , + hoffset .dim_set:N = \l__coffin_hoffset_dim , hoffset .value_required: , - voffset .dim_set:N = \l_coffin_voffset_dim , + voffset .dim_set:N = \l__coffin_voffset_dim , voffset .value_required: } \keys_set:nn { coffin } @@ -82,10 +82,10 @@ { \coffin_new:N #1 } \NewDocumentCommand \BoxToCoffin { m } { - \prop_clear_new:c { l_coffin_corners_ \int_value:w #1 _prop } - \prop_clear_new:c { l_coffin_poles_ \int_value:w #1 _prop } - \coffin_update_corners:N #1 - \coffin_update_poles:N #1 + \prop_clear_new:c { l__coffin_corners_ \__int_value:w #1 _prop } + \prop_clear_new:c { l__coffin_poles_ \__int_value:w #1 _prop } + \__coffin_update_corners:N #1 + \__coffin_update_poles:N #1 } \NewDocumentCommand \SetHorizontalCoffin { m +m } { \hcoffin_set:Nn #1 {#2} } @@ -115,22 +115,22 @@ { \group_begin: \keys_set:nn { coffin } {#1} - \tl_set:Nx \l_coffin_tmp_tl + \tl_set:Nx \l__coffin_tmp_tl { \group_end: - \bool_if:NTF \l_coffin_bound_box_grow_bool + \bool_if:NTF \l__coffin_bound_box_grow_bool { \coffin_join:NnnNnnnn } { \coffin_attach:NnnNnnnn } \exp_not:N #3 - { \exp_not:o { \l_coffin_A_hpole_tl } } - { \exp_not:o { \l_coffin_A_vpole_tl } } + { \exp_not:o { \l__coffin_A_hpole_tl } } + { \exp_not:o { \l__coffin_A_vpole_tl } } \exp_not:N #5 - { \exp_not:o { \l_coffin_B_hpole_tl } } - { \exp_not:o { \l_coffin_B_vpole_tl } } - { \dim_use:N \l_coffin_hoffset_dim } - { \dim_use:N \l_coffin_voffset_dim } + { \exp_not:o { \l__coffin_B_hpole_tl } } + { \exp_not:o { \l__coffin_B_vpole_tl } } + { \dim_use:N \l__coffin_hoffset_dim } + { \dim_use:N \l__coffin_voffset_dim } } - \l_coffin_tmp_tl + \l__coffin_tmp_tl } } \NewDocumentCommand \TypesetCoffin @@ -151,7 +151,7 @@ \NewDocumentCommand \CoffinHeight { m } { \coffin_ht:N #1 } \NewDocumentCommand \CoffinTotalHeight { m } - { \dim_eval:w \coffin_ht:N #1 + \coffin_dp:N #1 \dim_eval_end: } + { \__dim_eval:w \coffin_ht:N #1 + \coffin_dp:N #1 \__dim_eval_end: } \NewDocumentCommand \CoffinWidth { m } { \coffin_wd:N #1 } \NewDocumentCommand \DisplayCoffinHandles { m m } diff --git a/Master/texmf-dist/tex/latex/l3experimental/xgalley/l3galley.sty b/Master/texmf-dist/tex/latex/l3experimental/xgalley/l3galley.sty index cb77ed14d8f..dcf3c5871fc 100644 --- a/Master/texmf-dist/tex/latex/l3experimental/xgalley/l3galley.sty +++ b/Master/texmf-dist/tex/latex/l3experimental/xgalley/l3galley.sty @@ -48,18 +48,15 @@ %% ----------------------------------------------------------------------- %% \RequirePackage{xparse} -\GetIdInfo$Id: l3galley.dtx 3325 2012-02-06 22:07:00Z joseph $ +\GetIdInfo$Id: l3galley.dtx 3918 2012-07-05 00:32:37Z bruno $ {L3 Experimental galley code} \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} -\package_check_loaded_expl: -\cs_new_protected_nopar:Npn \galley_leave_vmode: - { \hbox_unpack:N \c_empty_box } -\clist_new:N \l_galley_parshape_left_indent_clist -\clist_new:N \l_galley_parshape_right_indent_clist -\bool_new:N \l_galley_parshape_multipar_bool -\bool_new:N \l_galley_parshape_resume_std_bool -\bool_new:N \l_galley_parshape_fixed_lines_bool +\clist_new:N \l__galley_parshape_left_indent_clist +\clist_new:N \l__galley_parshape_right_indent_clist +\bool_new:N \l__galley_parshape_multipar_bool +\bool_new:N \l__galley_parshape_resume_std_bool +\bool_new:N \l__galley_parshape_fixed_lines_bool \int_new:N \l_galley_parshape_std_lines_int \cs_new_eq:NN \l_galley_text_width_dim \linewidth \cs_new_eq:NN \l_galley_total_left_margin_dim \@totalleftmargin @@ -67,154 +64,154 @@ \int_new:N \l_galley_interpar_penalty_int \skip_new:N \l_galley_interpar_vspace_skip \cs_new_eq:NN \l_galley_width_dim \tex_hsize:D -\bool_new:N \g_galley_begin_level_bool -\bool_new:N \l_galley_begin_level_bool +\bool_new:N \g__galley_begin_level_bool +\bool_new:N \l__galley_begin_level_bool \bool_new:N \g_galley_omit_next_indent_bool -\bool_new:N \l_galley_omit_next_indent_bool -\bool_new:N \g_galley_parshape_set_bool -\bool_new:N \l_galley_parshape_set_bool -\bool_new:N \g_galley_nobreak_next_bool -\bool_new:N \l_galley_nobreak_next_bool +\bool_new:N \l__galley_omit_next_indent_bool +\bool_new:N \g__galley_parshape_set_bool +\bool_new:N \l__galley_parshape_set_bool +\bool_new:N \g_galley_no_break_next_bool +\bool_new:N \l__galley_no_break_next_bool \tl_new:N \g_galley_par_begin_hook_tl -\tl_new:N \l_galley_par_begin_hook_tl +\tl_new:N \l__galley_galley_par_begin_hook_tl \tl_new:N \g_galley_par_end_hook_tl -\tl_new:N \l_galley_par_end_hook_tl +\tl_new:N \l__galley_galley_par_end_hook_tl \tl_new:N \g_galley_par_after_hook_tl -\tl_new:N \l_galley_par_after_hook_tl +\tl_new:N \l__galley_par_after_hook_tl \int_new:N \g_galley_previous_par_lines_int -\int_new:N \l_galley_previous_par_lines_int +\int_new:N \l__galley_previous_par_lines_int \tl_new:N \g_galley_restore_running_tl -\tl_new:N \l_galley_restore_running_tl +\tl_new:N \l__galley_restore_running_tl \tl_new:N \g_galley_whatsit_next_tl -\tl_new:N \l_galley_whatsit_next_tl +\tl_new:N \l__galley_whatsit_next_tl \tl_new:N \g_galley_whatsit_previous_tl -\tl_new:N \l_galley_whatsit_previous_tl -\tl_new:N \g_galley_interpar_penalty_user_tl -\tl_new:N \l_galley_interpar_penalty_user_tl -\tl_new:N \g_galley_interpar_vspace_user_tl -\tl_new:N \l_galley_interpar_vspace_user_tl -\cs_new_protected_nopar:Npn \galley_initialise_variables: - { - \bool_gset_true:N \g_galley_begin_level_bool - \tl_gclear:N \g_galley_interpar_penalty_user_tl - \tl_gclear:N \g_galley_interpar_vspace_user_tl +\tl_new:N \l__galley_whatsit_previous_tl +\tl_new:N \g__galley_interpar_penalty_user_tl +\tl_new:N \l__galley_interpar_penalty_user_tl +\tl_new:N \g__galley_interpar_vspace_user_tl +\tl_new:N \l__galley_interpar_vspace_user_tl +\cs_new_protected_nopar:Npn \__galley_initialise_variables: + { + \bool_gset_true:N \g__galley_begin_level_bool + \tl_gclear:N \g__galley_interpar_penalty_user_tl + \tl_gclear:N \g__galley_interpar_vspace_user_tl \bool_gset_true:N \g_galley_omit_next_indent_bool - \bool_gset_false:N \g_galley_nobreak_next_bool + \bool_gset_false:N \g_galley_no_break_next_bool \tl_gclear:N \g_galley_par_begin_hook_tl \tl_gclear:N \g_galley_par_end_hook_tl \tl_gclear:N \g_galley_par_after_hook_tl - \bool_gset_false:N \g_galley_parshape_set_bool + \bool_gset_false:N \g__galley_parshape_set_bool \int_gzero:N \g_galley_previous_par_lines_int \tl_gclear:N \g_galley_restore_running_tl \tl_gclear:N \g_galley_whatsit_previous_tl \tl_gclear:N \g_galley_whatsit_next_tl } -\galley_initialise_variables: -\cs_new_protected_nopar:Npn \galley_initialise_settings: +\__galley_initialise_variables: +\cs_new_protected_nopar:Npn \__galley_initialise_settings: { \dim_set_eq:NN \l_galley_text_width_dim \l_galley_width_dim \dim_zero:N \l_galley_total_left_margin_dim \dim_zero:N \l_galley_total_right_margin_dim } -\cs_new_protected_nopar:Npn \galley_save_parameters: +\cs_new_protected_nopar:Npn \__galley_save_parameters: { - \bool_set_eq:NN \l_galley_begin_level_bool - \g_galley_begin_level_bool - \tl_set_eq:NN \l_galley_interpar_penalty_user_tl - \g_galley_interpar_penalty_user_tl - \tl_set_eq:NN \l_galley_interpar_vspace_user_tl - \g_galley_interpar_vspace_user_tl - \bool_set_eq:NN \l_galley_omit_next_indent_bool + \bool_set_eq:NN \l__galley_begin_level_bool + \g__galley_begin_level_bool + \tl_set_eq:NN \l__galley_interpar_penalty_user_tl + \g__galley_interpar_penalty_user_tl + \tl_set_eq:NN \l__galley_interpar_vspace_user_tl + \g__galley_interpar_vspace_user_tl + \bool_set_eq:NN \l__galley_omit_next_indent_bool \g_galley_omit_next_indent_bool - \bool_set_eq:NN \l_galley_nobreak_next_bool - \g_galley_nobreak_next_bool - \tl_set_eq:NN \l_galley_par_begin_hook_tl + \bool_set_eq:NN \l__galley_no_break_next_bool + \g_galley_no_break_next_bool + \tl_set_eq:NN \l__galley_galley_par_begin_hook_tl \g_galley_par_begin_hook_tl - \tl_set_eq:NN \l_galley_par_end_hook_tl + \tl_set_eq:NN \l__galley_galley_par_end_hook_tl \g_galley_par_end_hook_tl - \tl_set_eq:NN \l_galley_par_after_hook_tl + \tl_set_eq:NN \l__galley_par_after_hook_tl \g_galley_par_after_hook_tl - \bool_set_eq:NN \l_galley_parshape_set_bool - \g_galley_parshape_set_bool - \int_set_eq:NN \l_galley_previous_par_lines_int + \bool_set_eq:NN \l__galley_parshape_set_bool + \g__galley_parshape_set_bool + \int_set_eq:NN \l__galley_previous_par_lines_int \g_galley_previous_par_lines_int - \tl_set_eq:NN \l_galley_restore_running_tl + \tl_set_eq:NN \l__galley_restore_running_tl \g_galley_restore_running_tl - \tl_set_eq:NN \l_galley_whatsit_previous_tl + \tl_set_eq:NN \l__galley_whatsit_previous_tl \g_galley_whatsit_previous_tl - \tl_set_eq:NN \l_galley_whatsit_next_tl + \tl_set_eq:NN \l__galley_whatsit_next_tl \g_galley_whatsit_next_tl } -\cs_new_protected_nopar:Npn \galley_restore_parameters: +\cs_new_protected_nopar:Npn \__galley_restore_parameters: { - \bool_gset_eq:NN \g_galley_begin_level_bool - \l_galley_begin_level_bool - \tl_gset_eq:NN \g_galley_interpar_penalty_user_tl - \l_galley_interpar_penalty_user_tl - \tl_gset_eq:NN \g_galley_interpar_vspace_user_tl - \l_galley_interpar_vspace_user_tl + \bool_gset_eq:NN \g__galley_begin_level_bool + \l__galley_begin_level_bool + \tl_gset_eq:NN \g__galley_interpar_penalty_user_tl + \l__galley_interpar_penalty_user_tl + \tl_gset_eq:NN \g__galley_interpar_vspace_user_tl + \l__galley_interpar_vspace_user_tl \bool_gset_eq:NN \g_galley_omit_next_indent_bool - \l_galley_omit_next_indent_bool - \bool_gset_eq:NN \g_galley_nobreak_next_bool - \l_galley_nobreak_next_bool + \l__galley_omit_next_indent_bool + \bool_gset_eq:NN \g_galley_no_break_next_bool + \l__galley_no_break_next_bool \tl_gset_eq:NN \g_galley_par_begin_hook_tl - \l_galley_par_begin_hook_tl + \l__galley_galley_par_begin_hook_tl \tl_gset_eq:NN \g_galley_par_end_hook_tl - \l_galley_par_end_hook_tl + \l__galley_galley_par_end_hook_tl \tl_gset_eq:NN \g_galley_par_after_hook_tl - \l_galley_par_after_hook_tl - \bool_gset_eq:NN \g_galley_parshape_set_bool - \l_galley_parshape_set_bool + \l__galley_par_after_hook_tl + \bool_gset_eq:NN \g__galley_parshape_set_bool + \l__galley_parshape_set_bool \int_gset_eq:NN \g_galley_previous_par_lines_int - \l_galley_previous_par_lines_int + \l__galley_previous_par_lines_int \tl_gset_eq:NN \g_galley_restore_running_tl - \l_galley_restore_running_tl + \l__galley_restore_running_tl \tl_gset_eq:NN \g_galley_whatsit_previous_tl - \l_galley_whatsit_previous_tl + \l__galley_whatsit_previous_tl \tl_gset_eq:NN \g_galley_whatsit_next_tl - \l_galley_whatsit_next_tl + \l__galley_whatsit_next_tl } \cs_new_protected_nopar:Npn \galley_level: { - \galley_save_parameters: + \__galley_save_parameters: \group_begin: - \galley_initialise_variables: - \galley_initialise_settings: - \group_insert_after:N \galley_level_end: + \__galley_initialise_variables: + \__galley_initialise_settings: + \group_insert_after:N \__galley_level_end: } -\cs_new_protected_nopar:Npn \galley_level_end: +\cs_new_protected_nopar:Npn \__galley_level_end: { \par - \galley_restore_parameters: + \__galley_restore_parameters: \group_end: } -\scan_new:N \s_par_omit -\cs_new_protected_nopar:Npn \galley_std_par: +\__scan_new:N \s__par_omit +\cs_new_protected_nopar:Npn \__galley_std_par: { - \s_par_omit - \exp_after:wN \galley_std_par_aux_i: \tex_romannumeral:D - `0 + \s__par_omit + \exp_after:wN \__galley_std_par_aux_i: \tex_romannumeral:D - `0 } -\cs_new_protected:Npn \galley_std_par_aux_i: +\cs_new_protected:Npn \__galley_std_par_aux_i: { - \peek_meaning:NTF \s_par_omit - { \galley_std_par_aux:N } - { \galley_std_par_aux_ii: } + \peek_meaning:NTF \s__par_omit + { \__galley_std_par_aux:N } + { \__galley_std_par_aux_ii: } } -\cs_new_protected:Npn \galley_std_par_aux:N #1 +\cs_new_protected:Npn \__galley_std_par_aux:N #1 { - \str_if_eq:xxF {#1} { \s_par_omit } + \str_if_eq_x:nnF {#1} { \s__par_omit } { - \galley_std_par_aux_ii: + \__galley_std_par_aux_ii: #1 } } -\cs_new_protected_nopar:Npn \galley_std_par_aux_ii: +\cs_new_protected_nopar:Npn \__galley_std_par_aux_ii: { \mode_if_vertical:TF { \tex_par:D } { \g_galley_par_end_hook_tl - \galley_set_measure_and_parshape: + \__galley_set_measure_and_parshape: \group_begin: \tex_par:D \group_end: @@ -224,33 +221,33 @@ \tl_gclear:N \g_galley_par_after_hook_tl \tex_penalty:D \c_ten_thousand } -\cs_new_protected:Npn \galley_end_par:n #1 +\cs_new_protected:Npn \__galley_end_par:n #1 { - \s_par_omit - \bool_if:nF \g_galley_begin_level_bool + \s__par_omit + \bool_if:nF \g__galley_begin_level_bool { #1 - \galley_std_par: + \__galley_std_par: } } -\cs_set_protected_nopar:Npn \par { \galley_std_par: } -\tl_set:Nn \@par { \galley_std_par: } +\cs_set_protected_nopar:Npn \par { \__galley_std_par: } +\tl_set:Nn \@par { \__galley_std_par: } \cs_new_protected_nopar:Npn \galley_display_begin: { \group_begin: - \galley_save_parameters: + \__galley_save_parameters: \mode_if_vertical:TF { - \galley_display_penalty:N \l_galley_display_begin_par_penalty_tl - \galley_display_vspace:N \l_galley_display_begin_par_vspace_tl + \__galley_display_penalty:N \l_galley_display_begin_par_penalty_tl + \__galley_display_vspace:N \l_galley_display_begin_par_vspace_tl } { - \galley_display_penalty:N \l_galley_display_begin_penalty_tl - \galley_display_vspace:N \l_galley_display_begin_vspace_tl + \__galley_display_penalty:N \l_galley_display_begin_penalty_tl + \__galley_display_vspace:N \l_galley_display_begin_vspace_tl } \par } -\cs_new_protected:Npn \galley_display_penalty:N #1 +\cs_new_protected:Npn \__galley_display_penalty:N #1 { \tl_if_empty:NF #1 { @@ -262,7 +259,7 @@ \int_gset:Nn \g_galley_penalty_int {#1} } } -\cs_new_protected:Npn \galley_display_vspace:N #1 +\cs_new_protected:Npn \__galley_display_vspace:N #1 { \tl_if_empty:NF #1 { @@ -277,34 +274,34 @@ \cs_new_protected_nopar:Npn \galley_display_end: { \par - \galley_restore_parameters: + \__galley_restore_parameters: \group_end: \group_insert_after:N \group_insert_after:N - \group_insert_after:N \galley_display_par_setup: + \group_insert_after:N \__galley_display_par_setup: } -\cs_new_protected_nopar:Npn \galley_display_par_setup: +\cs_new_protected_nopar:Npn \__galley_display_par_setup: { \bool_gset_false:N \g_galley_omit_next_indent_bool - \cs_set_eq:NN \par \galley_display_par: + \cs_set_eq:NN \par \__galley_display_par: } -\cs_new_protected_nopar:Npn \galley_display_par: +\cs_new_protected_nopar:Npn \__galley_display_par: { - \cs_set_eq:NN \par \galley_std_par: + \cs_set_eq:NN \par \__galley_std_par: \mode_if_vertical:TF { \par \bool_gset_false:N \g_galley_omit_next_indent_bool - \galley_display_penalty:N \l_galley_display_end_par_penalty_tl - \galley_display_vspace:N \l_galley_display_end_par_vspace_tl + \__galley_display_penalty:N \l_galley_display_end_par_penalty_tl + \__galley_display_vspace:N \l_galley_display_end_par_vspace_tl } { - \galley_set_measure_and_parshape: + \__galley_set_measure_and_parshape: \group_begin: \tex_par:D \group_end: \int_gadd:Nn \g_galley_previous_par_lines_int \tex_prevgraf:D - \galley_display_penalty:N \l_galley_display_end_penalty_tl - \galley_display_vspace:N \l_galley_display_end_vspace_tl + \__galley_display_penalty:N \l_galley_display_end_penalty_tl + \__galley_display_vspace:N \l_galley_display_end_vspace_tl } } \cs_undefine:N \everypar @@ -313,84 +310,84 @@ { \tex_everypar:D { - \bool_if:NTF \g_galley_begin_level_bool - { \galley_start_paragraph_first: } - { \galley_start_paragraph_std: } + \bool_if:NTF \g__galley_begin_level_bool + { \__galley_start_paragraph_first: } + { \__galley_start_paragraph_std: } \tex_the:D \everypar } } -\box_new:N \g_galley_last_box -\cs_new_protected_nopar:Npn \galley_start_paragraph_std: +\box_new:N \g__galley_last_box +\cs_new_protected_nopar:Npn \__galley_start_paragraph_std: { \group_begin: - \box_gset_to_last:N \g_galley_last_box + \box_gset_to_last:N \g__galley_last_box \tex_par:D - \galley_insert_vertical_items: + \__galley_insert_vertical_items: \tex_everypar:D { } \tex_noindent:D \group_end: \int_gzero:N \g_galley_previous_par_lines_int - \galley_insert_horizontal_items: - \galley_restore_running_parameters: + \__galley_insert_horizontal_items: + \__galley_restore_running_parameters: } -\cs_new_protected_nopar:Npn \galley_start_paragraph_first: +\cs_new_protected_nopar:Npn \__galley_start_paragraph_first: { - \bool_gset_false:N \g_galley_begin_level_bool + \bool_gset_false:N \g__galley_begin_level_bool \mode_if_horizontal:TF { \group_begin: - \box_gset_to_last:N \g_galley_last_box + \box_gset_to_last:N \g__galley_last_box \tex_par:D - \galley_insert_vspace: + \__galley_insert_vspace: \tex_everypar:D { } \tex_noindent:D \group_end: } - { \galley_insert_vspace: } - \galley_insert_horizontal_items: - \galley_restore_running_parameters: + { \__galley_insert_vspace: } + \__galley_insert_horizontal_items: + \__galley_restore_running_parameters: } -\cs_new_protected_nopar:Npn \galley_insert_vertical_items: +\cs_new_protected_nopar:Npn \__galley_insert_vertical_items: { \g_galley_whatsit_previous_tl \tl_gclear:N \g_galley_whatsit_previous_tl - \tl_if_empty:NTF \g_galley_interpar_penalty_user_tl + \tl_if_empty:NTF \g__galley_interpar_penalty_user_tl { - \bool_if:NTF \g_galley_nobreak_next_bool + \bool_if:NTF \g_galley_no_break_next_bool { \tex_penalty:D \c_ten_thousand } { \tex_penalty:D \l_galley_interpar_penalty_int } } { \tex_penalty:D - \int_eval:w \g_galley_interpar_penalty_user_tl \int_eval_end: - \tl_gclear:N \g_galley_interpar_penalty_user_tl + \__int_eval:w \g__galley_interpar_penalty_user_tl \__int_eval_end: + \tl_gclear:N \g__galley_interpar_penalty_user_tl } - \bool_gset_false:N \g_galley_nobreak_next_bool - \galley_insert_vspace: + \bool_gset_false:N \g_galley_no_break_next_bool + \__galley_insert_vspace: } -\cs_new_protected_nopar:Npn \galley_insert_vspace: +\cs_new_protected_nopar:Npn \__galley_insert_vspace: { - \tl_if_empty:NTF \g_galley_interpar_vspace_user_tl + \tl_if_empty:NTF \g__galley_interpar_vspace_user_tl { \skip_vertical:N \l_galley_interpar_vspace_skip } { - \skip_vertical:n { \g_galley_interpar_vspace_user_tl } - \tl_gclear:N \g_galley_interpar_vspace_user_tl + \skip_vertical:n { \g__galley_interpar_vspace_user_tl } + \tl_gclear:N \g__galley_interpar_vspace_user_tl } } -\cs_new_protected_nopar:Npn \galley_insert_horizontal_items: +\cs_new_protected_nopar:Npn \__galley_insert_horizontal_items: { \g_galley_whatsit_next_tl \tl_gclear:N \g_galley_whatsit_next_tl \bool_if:NF \g_galley_omit_next_indent_bool { - \box_if_empty:NF \g_galley_last_box + \box_if_empty:NF \g__galley_last_box { \hbox_to_wd:nn \l_galley_par_indent_dim { } } } \skip_horizontal:N \l_galley_par_begin_skip \g_galley_par_begin_hook_tl \bool_gset_false:N \g_galley_omit_next_indent_bool } -\cs_new_protected_nopar:Npn \galley_restore_running_parameters: +\cs_new_protected_nopar:Npn \__galley_restore_running_parameters: { \g_galley_restore_running_tl \tl_gclear:N \g_galley_restore_running_tl @@ -419,50 +416,50 @@ } \cs_new_protected:Npn \galley_parshape_fixed_lines:nnn #1#2#3 { - \bool_gset_true:N \g_galley_parshape_set_bool - \bool_set_true:N \l_galley_parshape_fixed_lines_bool + \bool_gset_true:N \g__galley_parshape_set_bool + \bool_set_true:N \l__galley_parshape_fixed_lines_bool \int_set:Nn \l_galley_parshape_std_lines_int {#1} - \clist_set:Nn \l_galley_parshape_left_indent_clist {#2} - \clist_set:Nn \l_galley_parshape_right_indent_clist {#3} - \bool_set_true:N \l_galley_parshape_resume_std_bool + \clist_set:Nn \l__galley_parshape_left_indent_clist {#2} + \clist_set:Nn \l__galley_parshape_right_indent_clist {#3} + \bool_set_true:N \l__galley_parshape_resume_std_bool } \cs_new_protected:Npn \galley_parshape_multi_par:nnnN #1#2#3#4 { - \bool_gset_true:N \g_galley_parshape_set_bool - \bool_set_true:N \l_galley_parshape_multipar_bool - \bool_set_false:N \l_galley_parshape_fixed_lines_bool + \bool_gset_true:N \g__galley_parshape_set_bool + \bool_set_true:N \l__galley_parshape_multipar_bool + \bool_set_false:N \l__galley_parshape_fixed_lines_bool \int_set:Nn \l_galley_parshape_std_lines_int {#1} - \clist_set:Nn \l_galley_parshape_left_indent_clist {#2} - \clist_set:Nn \l_galley_parshape_right_indent_clist {#3} - \bool_set_eq:NN \l_galley_parshape_resume_std_bool #4 + \clist_set:Nn \l__galley_parshape_left_indent_clist {#2} + \clist_set:Nn \l__galley_parshape_right_indent_clist {#3} + \bool_set_eq:NN \l__galley_parshape_resume_std_bool #4 } \cs_new_protected:Npn \galley_parshape_single_par:nnnN #1#2#3#4 { - \bool_gset_true:N \g_galley_parshape_set_bool - \bool_set_false:N \l_galley_parshape_multipar_bool - \bool_set_false:N \l_galley_parshape_fixed_lines_bool + \bool_gset_true:N \g__galley_parshape_set_bool + \bool_set_false:N \l__galley_parshape_multipar_bool + \bool_set_false:N \l__galley_parshape_fixed_lines_bool \int_set:Nn \l_galley_parshape_std_lines_int {#1} - \clist_set:Nn \l_galley_parshape_left_indent_clist {#2} - \clist_set:Nn \l_galley_parshape_right_indent_clist {#3} - \bool_set_eq:NN \l_galley_parshape_resume_std_bool #4 + \clist_set:Nn \l__galley_parshape_left_indent_clist {#2} + \clist_set:Nn \l__galley_parshape_right_indent_clist {#3} + \bool_set_eq:NN \l__galley_parshape_resume_std_bool #4 } \cs_generate_variant:Nn \galley_parshape_fixed_lines:nnn { nVV } \cs_generate_variant:Nn \galley_parshape_multi_par:nnnN { nVV } \cs_generate_variant:Nn \galley_parshape_single_par:nnnN { nVV } -\cs_new_protected_nopar:Npn \galley_set_measure_and_parshape: +\cs_new_protected_nopar:Npn \__galley_set_measure_and_parshape: { - \bool_if:NTF \g_galley_parshape_set_bool + \bool_if:NTF \g__galley_parshape_set_bool { - \bool_if:NTF \l_galley_parshape_fixed_lines_bool + \bool_if:NTF \l__galley_parshape_fixed_lines_bool { \int_compare:nNnTF \g_galley_previous_par_lines_int > \c_zero - { \galley_generate_parshape_lines: } - { \galley_generate_parshape: } + { \__galley_generate_parshape_lines: } + { \__galley_generate_parshape: } } { - \bool_gset_eq:NN \g_galley_parshape_set_bool - \l_galley_parshape_multipar_bool - \galley_generate_parshape: + \bool_gset_eq:NN \g__galley_parshape_set_bool + \l__galley_parshape_multipar_bool + \__galley_generate_parshape: } } { @@ -473,16 +470,16 @@ \dim_use:N \l_galley_text_width_dim } } -\cs_new_protected_nopar:Npn \galley_generate_parshape: +\cs_new_protected_nopar:Npn \__galley_generate_parshape: { \tex_global:D \tex_parshape:D - \int_eval:w + \__int_eval:w \l_galley_parshape_std_lines_int + \int_min:nn - { \clist_length:N \l_galley_parshape_left_indent_clist } - { \clist_length:N \l_galley_parshape_right_indent_clist } - \bool_if:NT \l_galley_parshape_resume_std_bool { + 1 } - \int_eval_end: + { \clist_count:N \l__galley_parshape_left_indent_clist } + { \clist_count:N \l__galley_parshape_right_indent_clist } + \bool_if:NT \l__galley_parshape_resume_std_bool { + 1 } + \__int_eval_end: \prg_replicate:nn \l_galley_parshape_std_lines_int { \dim_use:N \l_galley_total_left_margin_dim @@ -490,10 +487,10 @@ \dim_use:N \l_galley_text_width_dim \c_space_tl } - \galley_set_parshape_map:oo - \l_galley_parshape_left_indent_clist - \l_galley_parshape_right_indent_clist - \bool_if:NT \l_galley_parshape_resume_std_bool + \__galley_set_parshape_map:oo + \l__galley_parshape_left_indent_clist + \l__galley_parshape_right_indent_clist + \bool_if:NT \l__galley_parshape_resume_std_bool { \c_space_tl \dim_use:N \l_galley_total_left_margin_dim @@ -501,10 +498,10 @@ \dim_use:N \l_galley_text_width_dim } } -\cs_new:Npn \galley_set_parshape_map:nn #1#2 - { \galley_set_parshape_map_aux:nw { } #1 , \q_mark #2 , \q_stop } -\cs_generate_variant:Nn \galley_set_parshape_map:nn { oo } -\cs_new:Npn \galley_set_parshape_map_aux:nw #1#2 , #3 \q_mark #4 , #5 \q_stop +\cs_new:Npn \__galley_set_parshape_map:nn #1#2 + { \__galley_set_parshape_map_aux:nw { } #1 , \q_mark #2 , \q_stop } +\cs_generate_variant:Nn \__galley_set_parshape_map:nn { oo } +\cs_new:Npn \__galley_set_parshape_map_aux:nw #1#2 , #3 \q_mark #4 , #5 \q_stop { \bool_if:nTF { \tl_if_empty_p:n {#3} || \tl_if_empty_p:n {#5} } { @@ -514,7 +511,7 @@ \dim_eval:n { \l_galley_text_width_dim - ( ( #2 ) + ( #4 ) ) } } { - \galley_set_parshape_map_aux:nw + \__galley_set_parshape_map_aux:nw { #1 \dim_eval:n { \l_galley_total_left_margin_dim + ( #2 ) } @@ -525,19 +522,19 @@ #3 \q_mark #5 \q_stop } } -\cs_new_protected_nopar:Npn \galley_generate_parshape_lines: +\cs_new_protected_nopar:Npn \__galley_generate_parshape_lines: { \int_compare:nNnTF \tex_parshape:D > \g_galley_previous_par_lines_int { \tex_global:D \tex_parshape:D - \int_eval:w \tex_parshape:D - \g_galley_previous_par_lines_int - \int_eval_end: - \prg_stepwise_function:nnnN + \__int_eval:w \tex_parshape:D - \g_galley_previous_par_lines_int + \__int_eval_end: + \int_step_function:nnnN { \g_galley_previous_par_lines_int + \c_one } - \c_one \tex_parshape:D \galley_generate_parshape_lines_aux:n + \c_one \tex_parshape:D \__galley_generate_parshape_lines:n } { - \bool_gset_false:N \g_galley_parshape_set_bool + \bool_gset_false:N \g__galley_parshape_set_bool \tex_global:D \tex_parshape:D \c_one \dim_use:N \l_galley_total_left_margin_dim @@ -545,16 +542,16 @@ \dim_use:N \l_galley_text_width_dim } } -\cs_new:Npn \galley_generate_parshape_lines_aux:n #1 +\cs_new:Npn \__galley_generate_parshape_lines:n #1 { \etex_parshapeindent:D #1 ~ \etex_parshapelength:D #1 } \cs_new_protected:Npn \galley_set_user_penalty:n #1 - { \tl_gset:Nx \g_galley_interpar_penalty_user_tl { \int_eval:n {#1} } } + { \tl_gset:Nx \g__galley_interpar_penalty_user_tl { \int_eval:n {#1} } } \cs_new_protected:Npn \galley_set_user_vspace:n #1 - { \tl_gset:Nx \g_galley_interpar_vspace_user_tl { \skip_eval:n {#1} } } + { \tl_gset:Nx \g__galley_interpar_vspace_user_tl { \skip_eval:n {#1} } } \dim_set:Nn \parskip \c_zero_dim \cs_undefine:N \parskip \skip_new:N \parskip @@ -591,7 +588,7 @@ \cs_new_protected:Npn \galley_break_line:Nn #1#2 { \mode_if_vertical:TF - { \msg_kernel_error:nn { galley } { no-line-to-end } } + { \__msg_kernel_error:nn { galley } { no-line-to-end } } { \tex_unskip:D \bool_if:NF #1 @@ -614,7 +611,7 @@ \cs_new_protected:Npn \galley_set_display_widow_penalties:n #1 { \etex_displaywidowpenalties:D - \int_eval:w \clist_length:n {#1} + \c_one \int_eval_end: + \__int_eval:w \clist_count:n {#1} + \c_one \__int_eval_end: \clist_map_function:nN {#1} \galley_set_aux:n \c_zero } @@ -622,7 +619,7 @@ \cs_new_protected:Npn \galley_set_widow_penalties:n #1 { \etex_widowpenalties:D - \int_eval:w \clist_length:n {#1} + \c_one \int_eval_end: + \__int_eval:w \clist_count:n {#1} + \c_one \__int_eval_end: \clist_map_function:nN {#1} \galley_set_aux:n \c_zero } @@ -631,107 +628,105 @@ \cs_new_protected:Npn \galley_set_club_penalties:n #1 { \clist_set:Nn \l_galley_club_penalties_clist {#1} - \galley_calc_interline_penalties: + \__galley_calc_interline_penalties: } \cs_generate_variant:Nn \galley_set_club_penalties:n { V , v } \cs_new_protected:Npn \galley_set_interline_penalties:n #1 { \clist_set:Nn \l_galley_line_penalties_clist {#1} - \galley_calc_interline_penalties: + \__galley_calc_interline_penalties: } \cs_generate_variant:Nn \galley_set_interline_penalties:n { V , v } \cs_new_protected:Npn \galley_set_display_club_penalties:n #1 { \etex_clubpenalties:D - \int_eval:w \clist_length:n {#1} + \c_one \int_eval_end: + \__int_eval:w \clist_count:n {#1} + \c_one \__int_eval_end: \clist_map_function:nN {#1} \galley_set_aux:n \c_zero - \galley_calc_interline_penalties: + \__galley_calc_interline_penalties: } \cs_generate_variant:Nn \galley_set_display_club_penalties:n { V , v } \cs_new_protected:Npn \galley_set_interline_penalty:n #1 { \int_compare:nNnTF { \etex_interlinepenalties:D \c_zero } = \c_zero { - \etex_interlinepenalties:D \c_one \int_eval:w #1 \int_eval_end: - \galley_calc_interline_penalties: + \etex_interlinepenalties:D \c_one \__int_eval:w #1 \__int_eval_end: + \__galley_calc_interline_penalties: } { - \cs_set:Npn \galley_set_interline_penalty_aux_ii:n ##1 + \cs_set:Npn \__galley_set_interline_penalty_ii:n ##1 { - \int_eval:w + \__int_eval:w \etex_interlinepenalties:D ##1 - \etex_interlinepenalties:D \etex_interlinepenalties:D \c_zero + #1 - \int_eval_end: + \__int_eval_end: } - \exp_args:Nf \galley_set_interline_penalty_aux:nn - { \clist_length:N \l_galley_line_penalties_clist } {#1} + \exp_args:Nf \__galley_set_interline_penalty:nn + { \clist_count:N \l_galley_line_penalties_clist } {#1} } } -\cs_new_protected:Npn \galley_set_interline_penalty_aux:nn #1#2 +\cs_new_protected:Npn \__galley_set_interline_penalty:nn #1#2 { \etex_interlinepenalties:D \etex_interlinepenalties:D \c_zero - \prg_stepwise_function:nnnN \c_one \c_one {#1} - \galley_set_interline_penalty_aux_i:n - \prg_stepwise_function:nnnN { #1 + \c_one } \c_one + \int_step_function:nnnN \c_one \c_one {#1} + \__galley_set_interline_penalty_i:n + \int_step_function:nnnN { #1 + \c_one } \c_one { \etex_interlinepenalties:D \c_zero - \c_one } - \galley_set_interline_penalty_aux_ii:n - \int_eval:w #2 \int_eval_end: + \__galley_set_interline_penalty_ii:n + \__int_eval:w #2 \__int_eval_end: } -\cs_new:Npn \galley_set_interline_penalty_aux_i:n #1 - { \etex_interlinepenalties:D \int_eval:w #1 \int_eval_end: } -\cs_new:Npn \galley_set_interline_penalty_aux_ii:n #1 { } -\cs_new_protected_nopar:Npn \galley_calc_interline_penalties: +\cs_new:Npn \__galley_set_interline_penalty_i:n #1 + { \etex_interlinepenalties:D \__int_eval:w #1 \__int_eval_end: } +\cs_new:Npn \__galley_set_interline_penalty_ii:n #1 { } +\cs_new_protected_nopar:Npn \__galley_calc_interline_penalties: { - \exp_args:Nff \galley_calc_interline_penalties_aux:nn + \exp_args:Nff \__galley_calc_interline_penalties:nn { \int_eval:n { \int_max:nn { - \clist_length:N \l_galley_club_penalties_clist + \clist_count:N \l_galley_club_penalties_clist + \c_one } { \int_max:nn { - \clist_length:N \l_galley_line_penalties_clist + \clist_count:N \l_galley_line_penalties_clist + \c_one } { \etex_clubpenalties:D \c_zero } } } } - { \clist_length:N \l_galley_line_penalties_clist } + { \clist_count:N \l_galley_line_penalties_clist } } -\cs_new_protected:Npn \galley_calc_interline_penalties_aux:nn #1#2 +\cs_new_protected:Npn \__galley_calc_interline_penalties:nn #1#2 { \etex_interlinepenalties:D #1 ~ - \prg_stepwise_function:nnnN \c_one \c_one {#2} - \galley_calc_interline_penalties_aux_i:n - \prg_stepwise_function:nnnN { #2 + \c_one } \c_one { #1 - \c_one } - \galley_calc_interline_penalties_aux_ii:n + \int_step_function:nnnN \c_one \c_one {#2} + \__galley_calc_interline_penalties_i:n + \int_step_function:nnnN { #2 + \c_one } \c_one { #1 - \c_one } + \__galley_calc_interline_penalties_ii:n \etex_interlinepenalties:D \etex_interlinepenalties:D \c_zero } -\cs_new:Npn \galley_calc_interline_penalties_aux_i:n #1 +\cs_new:Npn \__galley_calc_interline_penalties_i:n #1 { - \int_eval:w - \clist_item:Nn \l_galley_line_penalties_clist { #1 - \c_one } - + 0 \clist_item:Nn \l_galley_club_penalties_clist - { #1 - \c_one } + \__int_eval:w + \clist_item:Nn \l_galley_line_penalties_clist {#1} + + 0 \clist_item:Nn \l_galley_club_penalties_clist {#1} - \etex_clubpenalties:D #1 ~ - \int_eval_end: + \__int_eval_end: } -\cs_new:Npn \galley_calc_interline_penalties_aux_ii:n #1 +\cs_new:Npn \__galley_calc_interline_penalties_ii:n #1 { - \int_eval:w + \__int_eval:w \etex_interlinepenalties:D \etex_interlinepenalties:D \c_zero - + 0 \clist_item:Nn \l_galley_club_penalties_clist - { #1 - \c_one } + + 0 \clist_item:Nn \l_galley_club_penalties_clist {#1} - \etex_clubpenalties:D #1 ~ - \int_eval_end: + \__int_eval_end: } \cs_new_protected:Npn \galley_save_club_penalties:N #1 { \clist_set_eq:NN #1 \l_galley_club_penalties_clist } @@ -741,38 +736,38 @@ { \clist_set:Nx #1 { - \prg_stepwise_function:nnnN \c_one \c_one + \int_step_function:nnnN \c_one \c_one { \etex_clubpenalties:D \c_zero - \c_one } \galley_save_display_club_penalties:_aux:n } } \cs_new:Npn \galley_save_display_club_penalties:_aux:n #1 - { \int_use:N \etex_clubpenaltes:D \int_eval:w #1 \int_eval_end: , } + { \int_use:N \etex_clubpenaltes:D \__int_eval:w #1 \__int_eval_end: , } \cs_new_protected:Npn \galley_save_display_widow_penalties:N #1 { \clist_set:Nx #1 { - \prg_stepwise_function:nnnN \c_one \c_one + \int_step_function:nnnN \c_one \c_one { \etex_displaywidowpenalties:D \c_zero - \c_one } \galley_save_display_widow_penalties:_aux:n } } \cs_new:Npn \galley_save_display_widow_penalties:_aux:n #1 - { \int_use:N \etex_displaywidowpenalties:D \int_eval:w #1 \int_eval_end: , } + { \int_use:N \etex_displaywidowpenalties:D \__int_eval:w #1 \__int_eval_end: , } \cs_new_protected:Npn \galley_save_widow_penalties:N #1 { \clist_set:Nx #1 { - \prg_stepwise_function:nnnN \c_one \c_one + \int_step_function:nnnN \c_one \c_one { \etex_widowpenalties:D \c_zero - \c_one } \galley_save_widow_penalties:_aux:n } } \cs_new:Npn \galley_save_widow_penalties:_aux:n #1 - { \int_use:N \etex_widowpenalties:D \int_eval:w #1 \int_eval_end: , } + { \int_use:N \etex_widowpenalties:D \__int_eval:w #1 \__int_eval_end: , } \cs_new_protected_nopar:Npn \galley_interline_penalty: { \int_use:N \etex_interlinepenalties:D \etex_interlinepenalties:D \c_zero } -\msg_kernel_new:nnn { galley } { no-line-to-end } +\__msg_kernel_new:nnn { galley } { no-line-to-end } { There's~no~line~here~to~end. } \RenewDocumentCommand \clearpage { } { @@ -807,7 +802,7 @@ { \galley_break_line:Nn #1 {#2} } \RenewDocumentCommand \newline { } { \galley_break_line:Nn \c_true_bool { 0 pt } } -\cs_set_eq:NN \@@par \galley_std_par: +\cs_set_eq:NN \__galleypar \__galley_std_par: \cs_set_protected_nopar:Npn \@afterheading { \bool_gset_true:N \g_galley_no_break_next_bool diff --git a/Master/texmf-dist/tex/latex/l3experimental/xgalley/xgalley.sty b/Master/texmf-dist/tex/latex/l3experimental/xgalley/xgalley.sty index 3e3d251791a..31fe465cef3 100644 --- a/Master/texmf-dist/tex/latex/l3experimental/xgalley/xgalley.sty +++ b/Master/texmf-dist/tex/latex/l3experimental/xgalley/xgalley.sty @@ -48,13 +48,13 @@ %% ----------------------------------------------------------------------- %% \RequirePackage{l3names} -\GetIdInfo$Id: xgalley.dtx 3253 2012-01-28 14:59:55Z joseph $ +\GetIdInfo$Id: xgalley.dtx 3879 2012-07-01 09:17:47Z joseph $ {L3 Experimental galley} \ProvidesExplPackage {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} \RequirePackage{xparse,xtemplate,l3galley} -\clist_new:N \l_galley_tmpa_clist -\clist_new:N \l_galley_tmpb_clist +\clist_new:N \l__galley_tmpa_clist +\clist_new:N \l__galley_tmpb_clist \DeclareObjectType { measure } { 0 } \DeclareTemplateInterface { measure } { absolute } { 0 } { @@ -66,27 +66,27 @@ left-margin : length = 0 pt , right-margin : length = 0 pt } -\cs_new_eq:NN \l_galley_left_margin_dim \leftmargin -\cs_new_eq:NN \l_galley_right_margin_dim \rightmargin +\cs_new_eq:NN \l__galley_left_margin_dim \leftmargin +\cs_new_eq:NN \l__galley_right_margin_dim \rightmargin \DeclareTemplateCode { measure } { absolute } { 0 } { - left-margin = \l_galley_left_margin_dim , - right-margin = \l_galley_right_margin_dim + left-margin = \l__galley_left_margin_dim , + right-margin = \l__galley_right_margin_dim } { \AssignTemplateKeys - \galley_margins_set_absolute:nn \l_galley_left_margin_dim - \l_galley_right_margin_dim + \galley_margins_set_absolute:nn \l__galley_left_margin_dim + \l__galley_right_margin_dim } \DeclareTemplateCode { measure } { relative } { 0 } { - left-margin = \l_galley_left_margin_dim , - right-margin = \l_galley_right_margin_dim + left-margin = \l__galley_left_margin_dim , + right-margin = \l__galley_right_margin_dim } { \AssignTemplateKeys - \galley_margins_set_relative:nn \l_galley_left_margin_dim - \l_galley_right_margin_dim + \galley_margins_set_relative:nn \l__galley_left_margin_dim + \l__galley_right_margin_dim } \DeclareObjectType { parshape } { 0 } \DeclareTemplateInterface { parshape } { hang } { 0 } @@ -103,58 +103,58 @@ } \DeclareTemplateCode { parshape } { hang } { 0 } { - indent = \l_galley_parshape_indent_dim , - on-left-side = \l_galley_parshape_on_left_bool , - lines = \l_galley_parshape_lines_int + indent = \l__galley_parshape_indent_dim , + on-left-side = \l__galley_parshape_on_left_bool , + lines = \l__galley_parshape_lines_int } { \AssignTemplateKeys - \bool_if:NTF \l_galley_parshape_on_left_bool + \bool_if:NTF \l__galley_parshape_on_left_bool { \galley_parshape_single_par:nVVN - \l_galley_parshape_lines_int - \l_galley_parshape_indent_dim + \l__galley_parshape_lines_int + \l__galley_parshape_indent_dim \c_zero_dim \c_false_bool } { \galley_parshape_single_par:nVVN - \l_galley_parshape_lines_int + \l__galley_parshape_lines_int \c_zero_dim - \l_galley_parshape_indent_dim + \l__galley_parshape_indent_dim \c_false_bool } } \DeclareTemplateCode { parshape } { initial } { 0 } { - indent = \l_galley_parshape_indent_dim , - on-left-side = \l_galley_parshape_on_left_bool , - lines = \l_galley_parshape_lines_int + indent = \l__galley_parshape_indent_dim , + on-left-side = \l__galley_parshape_on_left_bool , + lines = \l__galley_parshape_lines_int } { \AssignTemplateKeys - \clist_clear:N \l_galley_tmpa_clist - \clist_clear:N \l_galley_tmpb_clist - \prg_replicate:nn { \l_galley_parshape_lines_int } + \clist_clear:N \l__galley_tmpa_clist + \clist_clear:N \l__galley_tmpb_clist + \prg_replicate:nn { \l__galley_parshape_lines_int } { - \clist_put_right:Nn \l_galley_tmpa_clist - { \l_galley_parshape_indent_dim } - \clist_put_right:Nn \l_galley_tmpb_clist + \clist_put_right:Nn \l__galley_tmpa_clist + { \l__galley_parshape_indent_dim } + \clist_put_right:Nn \l__galley_tmpb_clist { \c_zero_dim } } - \bool_if:NTF \l_galley_parshape_on_left_bool + \bool_if:NTF \l__galley_parshape_on_left_bool { \galley_parshape_single_par:nVVN \c_zero - \l_galley_tmpa_clist - \l_galley_tmpb_clist + \l__galley_tmpa_clist + \l__galley_tmpb_clist \c_true_bool } { \galley_parshape_single_par:nVVN \c_zero - \l_galley_tmpb_clist - \l_galley_tmpa_clist + \l__galley_tmpb_clist + \l__galley_tmpa_clist \c_true_bool } } @@ -207,7 +207,7 @@ } \UseTemplate { hyphenation } { std } { } \tex_defaulthyphenchar:D 45 \scan_stop: -\tl_new:N \l_galley_justification_other_tl +\tl_new:N \l__galley_justification_other_tl \DeclareObjectType { justification } { 0 } \DeclareTemplateInterface { justification } { std } { 0 } { @@ -241,7 +241,7 @@ } { \AssignTemplateKeys - \tl_clear:N \l_galley_justification_other_tl + \tl_clear:N \l__galley_justification_other_tl \galley_set_interword_spacing:N \l_galley_fixed_spacing_bool \bool_if:NTF \l_galley_par_stretch_last_bool { \int_set_eq:NN \l_galley_last_line_fit_int \c_one_thousand } @@ -259,7 +259,7 @@ stretch-last-line = \l_galley_par_stretch_last_bool } { - \tl_put_left:Nx \l_galley_justification_other_tl + \tl_put_left:Nx \l__galley_justification_other_tl { \skip_set:Nn \exp_not:N \l_galley_par_end_skip { \skip_use:N \l_galley_par_end_skip } @@ -282,7 +282,7 @@ { \skip_use:N \l_galley_line_right_skip } } \tl_gput_right:Nn \g_galley_restore_running_tl - { \l_galley_justification_other_tl } + { \l__galley_justification_other_tl } \AssignTemplateKeys \galley_set_interword_spacing:N \l_galley_fixed_spacing_bool \bool_if:NTF \l_galley_par_stretch_last_bool @@ -337,16 +337,16 @@ } \DeclareTemplateCode { justification } { compound } { 0 } { - first-paragraph = \galley_justification_first: , - other-paragraphs = \galley_justification_other: + first-paragraph = \__galley_justification_first: , + other-paragraphs = \__galley_justification_other: } { \AssignTemplateKeys - \galley_justification_first: - \tl_set:Nn \l_galley_justification_other_tl - { \galley_justification_other: } + \__galley_justification_first: + \tl_set:Nn \l__galley_justification_other_tl + { \__galley_justification_other: } \tl_gput_right:Nn \g_galley_restore_running_tl - { \l_galley_justification_other_tl } + { \l__galley_justification_other_tl } } \DeclareObjectType { line-breaking } { 0 } \DeclareTemplateInterface { line-breaking } { std } { 0 } @@ -366,15 +366,15 @@ \DeclareTemplateCode{ line-breaking } { std } { 0 } { badness = \l_galley_linebreak_badness_int , - binop-penalty = \l_galley_binop_penalty_int , + binop-penalty = \l__galley_binop_penalty_int , double-hyphen-demerits = \l_galley_double_hyphen_demerits_int , emergency-stretch = \l_galley_emergency_stretch_skip , final-hyphen-demerits = \l_galley_final_hyphen_demerits_int , fuzz = \l_galley_linebreak_fuzz_dim , - line-penalty = \l_galley_linebreak_penalty_int , + line-penalty = \l__galley_linebreak_penalty_int , mismatch-demerits = \l_galley_mismatch_demerits_int , pretolerance = \l_galley_linebreak_pretolerance_int , - relation-penalty = \l_galley_relation_penalty_int , + relation-penalty = \l__galley_relation_penalty_int , tolerance = \l_galley_linebreak_tolerance_int } { \AssignTemplateKeys } @@ -397,23 +397,23 @@ \DeclareTemplateCode { paragraph-breaking } { std } { 0 } { badness = \l_galley_parbreak_badness_int , - broken-penalty = \l_galley_broken_penalty_int , - club-penalty = \l_galley_club_penalty_int , - display-club-penalty = \l_galley_display_club_penalty_int , - display-widow-penalty = \l_galley_display_widow_penalty_int , + broken-penalty = \l__galley_broken_penalty_int , + club-penalty = \l__galley_club_penalty_int , + display-club-penalty = \l__galley_display_club_penalty_int , + display-widow-penalty = \l__galley_display_widow_penalty_int , fuzz = \l_galley_parbreak_fuzz_dim , - interline-penalty = \l_galley_interline_penalty_int , - post-display-penalty = \l_galley_post_display_penalty_int , - pre-display-penalty = \l_galley_pre_display_penalty_int , - widow-penalty = \l_galley_widow_penalty_int + interline-penalty = \l__galley_interline_penalty_int , + post-display-penalty = \l__galley_post_display_penalty_int , + pre-display-penalty = \l__galley_pre_display_penalty_int , + widow-penalty = \l__galley_widow_penalty_int } { \AssignTemplateKeys - \galley_set_club_penalties:V \l_galley_club_penalty_int - \galley_set_display_club_penalties:V \l_galley_display_club_penalty_int - \galley_set_display_widow_penalties:V \l_galley_display_widow_penalty_int - \galley_set_interline_penalty:n \l_galley_interline_penalty_int - \galley_set_widow_penalties:V \l_galley_widow_penalty_int + \galley_set_club_penalties:V \l__galley_club_penalty_int + \galley_set_display_club_penalties:V \l__galley_display_club_penalty_int + \galley_set_display_widow_penalties:V \l__galley_display_widow_penalty_int + \galley_set_interline_penalty:n \l__galley_interline_penalty_int + \galley_set_widow_penalties:V \l__galley_widow_penalty_int } \DeclareInstance { paragraph-breaking } { std } { std } { } \UseInstance { paragraph-breaking } { std } @@ -444,39 +444,39 @@ } \DeclareTemplateCode { paragraph-breaking } { single } { 0 } { - badness = \l_galley_parbreak_badness_tl , - broken-penalty = \l_galley_broken_penalty_tl , - club-penalty = \l_galley_club_penalties_tl , - display-club-penalty = \l_galley_display_club_penalties_tl , - display-widow-penalty = \l_galley_display_widow_penalties_tl , - fuzz = \l_galley_parbreak_fuzz_tl , - interline-penalty = \l_galley_interline_penalty_tl , - post-display-penalty = \l_galley_post_display_penalty_tl , - pre-display-penalty = \l_galley_pre_display_penalty_tl , - widow-penalty = \l_galley_widow_penalties_tl + badness = \l__galley_parbreak_badness_tl , + broken-penalty = \l__galley_broken_penalty_tl , + club-penalty = \l__galley_club_penalties_tl , + display-club-penalty = \l__galley_display_club_penalties_tl , + display-widow-penalty = \l__galley_display_widow_penalties_tl , + fuzz = \l__galley_parbreak_fuzz_tl , + interline-penalty = \l__galley_interline_penalty_tl , + post-display-penalty = \l__galley_post_display_penalty_tl , + pre-display-penalty = \l__galley_pre_display_penalty_tl , + widow-penalty = \l__galley_widow_penalties_tl } { \AssignTemplateKeys - \tl_if_empty:NF \l_galley_interline_penalty_tl + \tl_if_empty:NF \l__galley_interline_penalty_tl { \tl_gput_right:Nx \g_galley_par_after_hook_tl { - \int_set:Nn \exp_not:N \l_galley_interline_penalty_int + \int_set:Nn \exp_not:N \l__galley_interline_penalty_int { \galley_interline_penalty: } } - \int_set:Nn \l_galley_interline_penalty_int - { \l_galley_interline_penalty_tl } + \int_set:Nn \l__galley_interline_penalty_int + { \l__galley_interline_penalty_tl } } - \tl_if_empty:NF \l_galley_parbreak_fuzz_tl + \tl_if_empty:NF \l__galley_parbreak_fuzz_tl { \tl_gput_right:Nx \g_galley_par_after_hook_tl { \dim_set:Nn \exp_not:N \l_galley_parbreak_fuzz_dim { \dim_use:N \l_galley_parbreak_fuzz_dim } } - \dim_set:Nn \l_galley_parbreak_fuzz_dim { \l_galley_parbreak_fuzz_tl } + \dim_set:Nn \l_galley_parbreak_fuzz_dim { \l__galley_parbreak_fuzz_tl } } - \seq_map_inline:Nn \c_galley_parbreak_single_seq + \seq_map_inline:Nn \c__galley_parbreak_single_seq { \tl_if_empty:cF { l_galley_ ##1 _tl } { @@ -489,30 +489,30 @@ { \tl_use:c { l_galley_ ##1 _tl } } } } - \seq_map_inline:Nn \c_galley_parbreak_multi_seq + \seq_map_inline:Nn \c__galley_parbreak_multi_seq { \tl_if_empty:cF { l_galley_ ##1 _tl } { - \use:c { galley_save_ ##1 :N } \l_galley_tmpa_clist + \use:c { galley_save_ ##1 :N } \l__galley_tmpa_clist \tl_gput_right:Nx \g_galley_par_after_hook_tl { \exp_not:c { galley_set_ ##1 :n } - { \exp_not:o \l_galley_tmpa_clist } + { \exp_not:o \l__galley_tmpa_clist } } \use:c { galley_set_ ##1 :v } { l_galley_ ##1 _tl } } } } -\seq_new:N \c_galley_parbreak_multi_seq -\seq_gput_right:Nn \c_galley_parbreak_multi_seq { club_penalties } -\seq_gput_right:Nn \c_galley_parbreak_multi_seq { display_club_penalties } -\seq_gput_right:Nn \c_galley_parbreak_multi_seq { display_widow_penalties } -\seq_gput_right:Nn \c_galley_parbreak_multi_seq { widow_penalties } -\seq_new:N \c_galley_parbreak_single_seq -\seq_gput_right:Nn \c_galley_parbreak_single_seq { parbreak_badness } -\seq_gput_right:Nn \c_galley_parbreak_single_seq { broken_penalty } -\seq_gput_right:Nn \c_galley_parbreak_single_seq { post_display_penalty } -\seq_gput_right:Nn \c_galley_parbreak_single_seq { pre_display_penalty } +\seq_new:N \c__galley_parbreak_multi_seq +\seq_gput_right:Nn \c__galley_parbreak_multi_seq { club_penalties } +\seq_gput_right:Nn \c__galley_parbreak_multi_seq { display_club_penalties } +\seq_gput_right:Nn \c__galley_parbreak_multi_seq { display_widow_penalties } +\seq_gput_right:Nn \c__galley_parbreak_multi_seq { widow_penalties } +\seq_new:N \c__galley_parbreak_single_seq +\seq_gput_right:Nn \c__galley_parbreak_single_seq { parbreak_badness } +\seq_gput_right:Nn \c__galley_parbreak_single_seq { broken_penalty } +\seq_gput_right:Nn \c__galley_parbreak_single_seq { post_display_penalty } +\seq_gput_right:Nn \c__galley_parbreak_single_seq { pre_display_penalty } \DeclareInstance { paragraph-breaking } { single-std } { single } { } \DeclareInstance { paragraph-breaking } { single-nobreak } { single } { -- cgit v1.2.3