diff options
Diffstat (limited to 'Master/texmf-dist/source/latex/l3experimental/l3str/l3str-expl.dtx')
-rw-r--r-- | Master/texmf-dist/source/latex/l3experimental/l3str/l3str-expl.dtx | 1046 |
1 files changed, 1046 insertions, 0 deletions
diff --git a/Master/texmf-dist/source/latex/l3experimental/l3str/l3str-expl.dtx b/Master/texmf-dist/source/latex/l3experimental/l3str/l3str-expl.dtx new file mode 100644 index 00000000000..9305a8fe07c --- /dev/null +++ b/Master/texmf-dist/source/latex/l3experimental/l3str/l3str-expl.dtx @@ -0,0 +1,1046 @@ +% \iffalse meta-comment +% +%% File: l3str.dtx Copyright (C) 2011-2014 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 +%% license or (at your option) any later version. The latest version +%% of this license is in the file +%% +%% http://www.latex-project.org/lppl.txt +%% +%% This file is part of the "l3experimental bundle" (The Work in LPPL) +%% and all files in that bundle must be distributed together. +%% +%% The released version of this bundle is available from CTAN. +%% +%% ----------------------------------------------------------------------- +%% +%% The development version of the bundle can be found at +%% +%% http://www.latex-project.org/svnroot/experimental/trunk/ +%% +%% for those people who are interested. +%% +%%%%%%%%%%% +%% NOTE: %% +%%%%%%%%%%% +%% +%% Snapshots taken from the repository represent work in progress and may +%% not work or may contain conflicting material! We therefore ask +%% people _not_ to put them into distributions, archives, etc. without +%% prior consultation with the LaTeX3 Project. +%% +%% ----------------------------------------------------------------------- +% +%<*driver|package> +\RequirePackage{expl3} +\GetIdInfo$Id: l3str-expl.dtx 5226 2014-07-18 18:08:01Z joseph $ + {L3 Experimental strings} +\def\ExplSyntaxName{l3str} +%</driver|package> +%<*driver> +\documentclass[full]{l3doc} +\usepackage{amsmath} +\begin{document} + \DocInput{\jobname.dtx} +\end{document} +%</driver> +% \fi +% +% +% \title{^^A +% The \textsf{l3str} package: manipulating strings of characters^^A +% \thanks{This file describes v\ExplFileVersion, +% last revised \ExplFileDate.}^^A +% } +% +% \author{^^A +% The \LaTeX3 Project\thanks +% {^^A +% E-mail: +% \href{mailto:latex-team@latex-project.org} +% {latex-team@latex-project.org}^^A +% }^^A +% } +% +% \date{Released \ExplFileDate} +% +% \maketitle +% +% \tableofcontents +% +% \begin{documentation} +% +% \LaTeX3 provides a set of functions to manipulate token lists +% as strings of characters, ignoring the category codes of those +% characters. +% +% String variables are simply specialised token lists, but by convention +% should be named with the suffix \texttt{\ldots{}str}. Such variables +% should contain characters with category code $12$ (other), except +% spaces, which have category code $10$ (blank space). All the +% functions in this module which accept a token list argument first +% convert it to a string using \cs{tl_to_str:n} for internal processing, +% and will not treat a token list or the corresponding string +% representation differently. +% +% Most expandable functions in this module come in three flavours: +% \begin{itemize} +% \item \cs{str_...:N}, which expect a token list or string +% variable as their argument; +% \item \cs{str_...:n}, taking any token list (or string) as an +% argument; +% \item \cs{str_..._ignore_spaces:n}, which ignores any space +% encountered during the operation: these functions are typically +% faster than those which take care of escaping spaces +% appropriately. +% \end{itemize} +% +% \section{Building strings} +% +% \begin{variable} +% { +% \c_backslash_str, +% \c_left_brace_str, +% \c_right_brace_str, +% \c_hash_str, +% \c_tilde_str, +% \c_percent_str +% } +% Constant strings, containing a single character token, with category +% code $12$. Any character can be accessed as \cs{iow_char:N} +% |\|\meta{character}. +% \end{variable} +% +% \begin{function}{\str_new:N, \str_new:c} +% \begin{syntax} +% \cs{str_new:N} \meta{str~var} +% \end{syntax} +% Creates a new \meta{str~var} or raises an error if the name is +% already taken. The declaration is global. The \meta{str~var} will +% initially be empty. +% \end{function} +% +% \begin{function}{\str_const:Nn, \str_const:Nx, \str_const:cn, \str_const:cx} +% \begin{syntax} +% \cs{str_const:Nn} \meta{str~var} \Arg{token list} +% \end{syntax} +% Creates a new constant \meta{str~var} or raises an error if the name +% is already taken. The value of the \meta{str~var} will be set +% globally to the \meta{token list}, converted to a string. +% \end{function} +% +% \begin{function} +% { +% \str_set:Nn, \str_set:Nx, \str_set:cn, \str_set:cx, +% \str_gset:Nn, \str_gset:Nx, \str_gset:cn, \str_gset:cx +% } +% \begin{syntax} +% \cs{str_set:Nn} \meta{str var} \Arg{token list} +% \end{syntax} +% Converts the \meta{token list} to a \meta{string}, and stores the +% result in \meta{str var}. +% \end{function} +% +% \begin{function} +% { +% \str_put_left:Nn, \str_put_left:Nx, +% \str_put_left:cn, \str_put_left:cx, +% \str_gput_left:Nn, \str_gput_left:Nx, +% \str_gput_left:cn, \str_gput_left:cx +% } +% \begin{syntax} +% \cs{str_put_left:Nn} \meta{str var} \Arg{token list} +% \end{syntax} +% Converts the \meta{token list} to a \meta{string}, and prepends the +% result to \meta{str var}. The current contents of the \meta{str +% var} are not automatically converted to a string. +% \end{function} +% +% \begin{function} +% { +% \str_put_right:Nn, \str_put_right:Nx, +% \str_put_right:cn, \str_put_right:cx, +% \str_gput_right:Nn, \str_gput_right:Nx, +% \str_gput_right:cn, \str_gput_right:cx +% } +% \begin{syntax} +% \cs{str_put_right:Nn} \meta{str var} \Arg{token list} +% \end{syntax} +% Converts the \meta{token list} to a \meta{string}, and appends the +% result to \meta{str var}. The current contents of the \meta{str +% var} are not automatically converted to a string. +% \end{function} +% +% \section{Accessing the contents of a string} +% +% \begin{function}[EXP] +% {\str_count:N, \str_count:n, \str_count_ignore_spaces:n} +% \begin{syntax} +% \cs{str_count:n} \Arg{token list} +% \end{syntax} +% Leaves in the input stream the number of characters in the string +% representation of \meta{token list}, as an integer denotation. The +% functions differ 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} function +% leaves the number of non-space characters in the input stream. +% \end{function} +% +% \begin{function}[EXP]{\str_count_spaces:N, \str_count_spaces:n} +% \begin{syntax} +% \cs{str_count_spaces:n} \Arg{token list} +% \end{syntax} +% Leaves in the input stream the number of space characters in the +% string representation of \meta{token list}, as an integer +% denotation. Of course, this function has no \texttt{_ignore_spaces} +% variant. +% \end{function} +% +% \begin{function}[EXP]{\str_head:N, \str_head:n, \str_head_ignore_spaces:n} +% \begin{syntax} +% \cs{str_head:n} \Arg{token list} +% \end{syntax} +% Converts the \meta{token list} into a \meta{string}. The first +% character in the \meta{string} is then left in the input stream, +% with category code \enquote{other}. The functions differ if the +% first character is a space: \cs{str_head:N} and \cs{str_head:n} +% return a space token with category code~$10$ (blank space), while +% the \cs{str_head_ignore_spaces:n} function ignores this space +% character and leaves the first non-space character in the input +% stream. If the \meta{string} is empty (or only contains spaces in +% the case of the \texttt{_ignore_spaces} function), then nothing is +% left on the input stream. +% \end{function} +% +% \begin{function}[EXP]{\str_tail:N, \str_tail:n, \str_tail_ignore_spaces:n} +% \begin{syntax} +% \cs{str_tail:n} \Arg{token list} +% \end{syntax} +% Converts the \meta{token list} to a \meta{string}, removes the first +% character, and leaves the remaining characters (if any) in the input +% stream, with category codes $12$ and $10$ (for spaces). The +% functions differ in the case where the first character is a space: +% \cs{str_tail:N} and \cs{str_tail:n} will trim only that space, while +% \cs{str_tail_ignore_spaces:n} removes the first non-space character +% and any space before it. If the \meta{token list} is empty (or +% blank in the case of the \texttt{_ignore_spaces} variant), then +% nothing is left on the input stream. +% \end{function} +% +% \begin{function}[EXP]{\str_item:Nn, \str_item:nn, \str_item_ignore_spaces:nn} +% \begin{syntax} +% \cs{str_item:nn} \Arg{token list} \Arg{integer expression} +% \end{syntax} +% Converts the \meta{token list} to a \meta{string}, and leaves in the +% input stream the character in position \meta{integer expression} of +% the \meta{string}, starting at $1$ for the first (left-most) +% character. In the case of \cs{str_item:Nn} and \cs{str_item:nn}, +% all characters including spaces are taken into account. The +% \cs{str_item_ignore_spaces:nn} function skips spaces when counting +% characters. 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.} +% \end{function} +% +% \begin{function}[EXP] +% {\str_range:Nnn, \str_range:nnn, \str_range_ignore_spaces:nnn} +% \begin{syntax} +% \cs{str_range:nnn} \Arg{token list} \Arg{start index} \Arg{end index} +% \end{syntax} +% Converts the \meta{token list} to a \meta{string}, and leaves in the +% input stream the characters from the \meta{start index} to the +% \meta{end index} inclusive. Positive \meta{indices} are counted +% from the start of the string, $1$~being the first character, and +% negative \meta{indices} are counted from the end of the string, +% $-1$~being the last character. If either of \meta{start index} or +% \meta{end index} is~$0$, the result is empty. For instance, +% \begin{verbatim} +% \iow_term:x { \str_range:nnn { abcdef } { 2 } { 5 } } +% \iow_term:x { \str_range:nnn { abcdef } { -4 } { -1 } } +% \iow_term:x { \str_range:nnn { abcdef } { -2 } { -1 } } +% \iow_term:x { \str_range:nnn { abcdef } { 0 } { -1 } } +% \end{verbatim} +% will print \texttt{bcd}, \texttt{cdef}, \texttt{ef}, and an empty +% line to the terminal. +% \end{function} +% +% \section{String conditionals} +% +% \begin{function}[EXP,pTF] +% { +% \str_if_eq:nn, \str_if_eq:Vn, \str_if_eq:on, \str_if_eq:no, +% \str_if_eq:nV, \str_if_eq:VV +% } +% \begin{syntax} +% \cs{str_if_eq_p:nn} \Arg{tl_1} \Arg{tl_2} +% \cs{str_if_eq:nnTF} \Arg{tl_1} \Arg{tl_2} \Arg{true code} \Arg{false code} +% \end{syntax} +% Compares the string representations of the 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_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}. +% \end{function} +% +% \begin{function}[added = 2013-07-24, EXP, TF]{\str_case:nn, \str_case:on} +% \begin{syntax} +% \cs{str_case:nnTF} \Arg{test string} \\ +% ~~|{| \\ +% ~~~~\Arg{string case_1} \Arg{code case_1} \\ +% ~~~~\Arg{string case_2} \Arg{code case_2} \\ +% ~~~~\ldots \\ +% ~~~~\Arg{string case_n} \Arg{code case_n} \\ +% ~~|}| \\ +% ~~\Arg{true code} +% ~~\Arg{false code} +% \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 any of the +% cases are matched, the \meta{true code} is also inserted into the +% input stream (after the code for the appropriate case), while if none +% match then the \meta{false code} is inserted. The function +% \cs{str_case:nn}, which does nothing if there is no match, is also +% available. +% \end{function} +% +% \begin{function}[added = 2013-07-24, EXP, TF]{\str_case_x:nn} +% \begin{syntax} +% \cs{str_case_x:nnn} \Arg{test string} \\ +% ~~|{| \\ +% ~~~~\Arg{string case_1} \Arg{code case_1} \\ +% ~~~~\Arg{string case_2} \Arg{code case_2} \\ +% ~~~~\ldots \\ +% ~~~~\Arg{string case_n} \Arg{code case_n} \\ +% ~~|}| \\ +% ~~\Arg{true code} +% ~~\Arg{false code} +% \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 any of the +% cases are matched, the \meta{true code} is also inserted into the +% input stream (after the code for the appropriate case), while if none +% match then the \meta{false code} is inserted. The function +% \cs{str_case_x:nn}, which does nothing if there is no match, is also +% available. +% 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} +% +% \begin{function}{\str_show:N, \str_show:c, \str_show:n} +% \begin{syntax} +% \cs{str_show:N} \meta{tl~var} +% \end{syntax} +% 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{Internal \pkg{l3str} functions} +% +% \begin{function}[EXP]{\__str_to_other:n} +% \begin{syntax} +% \cs{__str_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}. 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. It takes a +% time quadratic in the character count of the string, but there exist +% non-expandable ways to reach linear time. +% \end{function} +% +% \begin{function}[EXP]{\__str_count_unsafe:n} +% \begin{syntax} +% \cs{__str_count_unsafe:n} \Arg{other string} +% \end{syntax} +% This function expects an argument that is entirely made of +% characters with category \enquote{other}, as produced by +% \cs{__str_to_other:n}. It leaves in the input stream the number of +% character tokens in the \meta{other string}, faster than the +% analogous \cs{str_count:n} function. +% \end{function} +% +% \begin{function}[EXP]{\__str_range_unsafe:nnn} +% \begin{syntax} +% \cs{__str_range_unsafe:nnn} \Arg{other string} \Arg{start index} \Arg{end index} +% \end{syntax} +% Identical to \cs{str_range:nnn} except that the first argument is +% expected to be entirely made of characters with category +% \enquote{other}, as produced by \cs{__str_to_other:n}, and the +% result is also an \meta{other string}. +% \end{function} +% +% \section{Possible additions to \pkg{l3str}} +% +% Semantically correct copies of some \texttt{tl} functions. +% \begin{itemize} +% \item \cs{c_space_str} +% \item \cs{str_clear:N}, \cs{str_gclear:N}, \cs{str_clear_new:N}, +% \cs{str_gclear_new:N}. +% \item \cs{str_concat:NNN}, \cs{str_gconcat:NNN} +% \item \cs{str_set_eq:NN}, \cs{str_gset_eq:NN} +% \item \cs{str_if_empty:NTF}, \cs{str_if_empty_p:N} +% \item \cs{str_if_exist:NTF}, \cs{str_if_exist_p:N} +% \item \cs{str_use:N} +% \end{itemize} +% +% Some functions that are not copies of \texttt{tl} functions. +% \begin{itemize} +% \item \cs{str_if_blank:NTF}, \cs{str_if_blank_p:N}. +% \item \cs{str_map_inline:Nn}, \cs{str_map_function:NN}, +% \cs{str_map_variable:NNn}, and \texttt{:n} analogs. +% \item Expandable \cs{str_if_in:nnTF}? +% \item \cs{str_if_head_eq:nNTF}, \cs{str_if_head_eq_p:nN} +% \item \cs{str_if_numeric/decimal/integer:n}, perhaps in \pkg{l3fp}? +% \end{itemize} +% +% \end{documentation} +% +% \begin{implementation} +% +% \section{\pkg{l3str} implementation} +% +% \begin{macrocode} +%<*initex|package> +% \end{macrocode} +% +% \begin{macrocode} +%<@@=str> +% \end{macrocode} +% +% \begin{macrocode} +\ProvidesExplPackage + {\ExplFileName}{\ExplFileDate}{\ExplFileVersion}{\ExplFileDescription} +% \end{macrocode} +% +% The following string-related functions are currently defined in +% \pkg{l3kernel}. +% \begin{itemize} +% \item \cs{str_if_eq:nn}[pTF] and variants, +% \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: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{String assignments} +% +% \begin{macro}{\str_new:N, \str_new:c} +% A string is simply a token list. +% \begin{macrocode} +\cs_new_eq:NN \str_new:N \tl_new:N +\cs_generate_variant:Nn \str_new:N { c } +% \end{macrocode} +% \end{macro} +% +% \begin{macro} +% { +% \str_set:Nn, \str_set:Nx, +% \str_set:cn, \str_set:cx, +% \str_gset:Nn, \str_gset:Nx, +% \str_gset:cn, \str_gset:cx, +% \str_const:Nn, \str_const:Nx, +% \str_const:cn, \str_const:cx, +% \str_put_left:Nn, \str_put_left:Nx, +% \str_put_left:cn, \str_put_left:cx, +% \str_gput_left:Nn, \str_gput_left:Nx, +% \str_gput_left:cn, \str_gput_left:cx, +% \str_put_right:Nn, \str_put_right:Nx, +% \str_put_right:cn, \str_put_right:cx, +% \str_gput_right:Nn, \str_gput_right:Nx, +% \str_gput_right:cn, \str_gput_right:cx, +% } +% Simply convert the token list inputs to \meta{strings}. +% \begin{macrocode} +\tl_map_inline:nn + { + { set } + { gset } + { const } + { put_left } + { gput_left } + { put_right } + { gput_right } + } + { + \cs_new_protected:cpx { str_ #1 :Nn } ##1##2 + { \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 } + } +% \end{macrocode} +% \end{macro} +% +% \subsection{String variables and constants} +% +% \begin{variable} +% { +% \c_backslash_str, +% \c_left_brace_str, +% \c_right_brace_str, +% \c_hash_str, +% \c_tilde_str, +% \c_percent_str +% } +% For all of those strings, use \cs{cs_to_str:N} to get characters with +% the correct category code. +% \begin{macrocode} +\str_const:Nx \c_backslash_str { \cs_to_str:N \\ } +\str_const:Nx \c_left_brace_str { \cs_to_str:N \{ } +\str_const:Nx \c_right_brace_str { \cs_to_str:N \} } +\str_const:Nx \c_hash_str { \cs_to_str:N \# } +\str_const:Nx \c_tilde_str { \cs_to_str:N \~ } +\str_const:Nx \c_percent_str { \cs_to_str:N \% } +% \end{macrocode} +% \end{variable} +% +% \begin{variable}{\l_tmpa_str, \l_tmpb_str, \g_tmpa_str, \g_tmpb_str} +% Scratch strings. +% \begin{macrocode} +\str_new:N \l_tmpa_str +\str_new:N \l_tmpb_str +\str_new:N \g_tmpa_str +\str_new:N \g_tmpb_str +% \end{macrocode} +% \end{variable} +% +% \subsection{Counting characters} +% +% \begin{macro}[EXP]{\str_count_spaces:N, \str_count_spaces:n} +% \begin{macro}[EXP, aux]{\@@_count_spaces_loop:wwwwwwwww} +% To speed up this function, we grab and discard $9$ space-delimited +% arguments in each iteration of the loop. 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 $9$ that precedes, to adjust +% the result. +% \begin{macrocode} +\cs_new_nopar:Npn \str_count_spaces:N + { \exp_args:No \str_count_spaces:n } +\cs_new:Npn \str_count_spaces:n #1 + { + \int_eval:n + { + \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 \@@_count_spaces_loop:wwwwwwwww #1~#2~#3~#4~#5~#6~#7~#8~#9~ + { + \if_meaning:w X #9 + \use_i_delimit_by_q_stop:nw + \fi: + \c_nine + \@@_count_spaces_loop:wwwwwwwww + } +% \end{macrocode} +% \end{macro} +% \end{macro} +% +% \begin{macro}[EXP]{\str_count:N, \str_count:n, \str_count_ignore_spaces:n} +% \begin{macro}[EXP, int]{\@@_count_unsafe:n} +% \begin{macro}[EXP, aux]{\@@_count:n, \@@_count_loop:NNNNNNNNN} +% To count characters in a string we could first escape all spaces +% using \cs{@@_to_other:n}, then pass the result to \cs{tl_count:n}. +% However, the escaping step would be quadratic in the number of +% characters in the string, and we can do better. Namely, sum the +% number of spaces (\cs{str_count_spaces:n}) and the result of +% \cs{tl_count:n}, which ignores spaces. Since strings tend to be +% longer than token lists, we use specialized functions to count +% characters ignoring spaces. Namely, loop, grabbing $9$ non-space +% characters at each step, and end as soon as we reach one of the $9$ +% trailing items. The \texttt{_unsafe} variant expects a token list +% already converted to category code $12$ characters, and is used by +% \cs{str_item:nn} and \cs{str_range:nnn}. +% \begin{macrocode} +\cs_new_nopar:Npn \str_count:N { \exp_args:No \str_count:n } +\cs_new:Npn \str_count:n #1 + { + \@@_count:n + { + \str_count_spaces:n {#1} + + \exp_after:wN \@@_count_loop:NNNNNNNNN \tl_to_str:n {#1} + } + } +\cs_new:Npn \@@_count_unsafe:n #1 + { + \@@_count:n + { \@@_count_loop:NNNNNNNNN #1 } + } +\cs_new:Npn \str_count_ignore_spaces:n #1 + { + \@@_count:n + { \exp_after:wN \@@_count_loop:NNNNNNNNN \tl_to_str:n {#1} } + } +\cs_new:Npn \@@_count:n #1 + { + \int_eval:n + { + #1 + { X \c_eight } { X \c_seven } { X \c_six } + { X \c_five } { X \c_four } { X \c_three } + { X \c_two } { X \c_one } { X \c_zero } + \q_stop + } + } +\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 + \@@_count_loop:NNNNNNNNN + } +% \end{macrocode} +% \end{macro} +% \end{macro} +% \end{macro} +% +% \subsection{Head and tail of a string} +% +% \begin{macro}[EXP]{\str_head:N, \str_head:n, \str_head_ignore_spaces:n} +% \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{@@_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{@@_head:w} takes an empty argument, and the +% single (initially 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 +% \cs{use_i_delimit_by_q_stop:nw}. +% \begin{macrocode} +\cs_new_nopar:Npn \str_head:N { \exp_args:No \str_head:n } +\cs_set:Npn \str_head:n #1 + { + \exp_after:wN \@@_head:w + \tl_to_str:n {#1} + { { } } ~ \q_stop + } +\cs_set:Npn \@@_head:w #1 ~ % + { \use_i_delimit_by_q_stop:nw #1 { ~ } } +\cs_new:Npn \str_head_ignore_spaces:n #1 + { + \exp_after:wN \use_i_delimit_by_q_stop:nw + \tl_to_str:n {#1} { } \q_stop + } +% \end{macrocode} +% \end{macro} +% \end{macro} +% +% \begin{macro}[EXP]{\str_tail:N, \str_tail:n, \str_tail_ignore_spaces:n} +% \begin{macro}[EXP, aux]{\@@_tail_auxi:w, \@@_tail_auxii:w} +% Getting the tail is a little bit more convoluted than the head of a +% string. We hit the front of the string with \cs{reverse_if:N} +% \cs{if_charcode:w} \cs{scan_stop:}. This removes the first +% character, and necessarily makes the test true, since the character +% cannot match \cs{scan_stop:}. The auxiliary function then inserts +% the required \cs{fi:} to close the conditional, and leaves the tail +% of the string in the input stream. The details are such that an +% empty string has an empty tail (this requires in particular that the +% end-marker |X| be unexpandable and not a control sequence). The +% \texttt{_ignore_spaces} is rather simpler: after converting the +% input to a string, \cs{@@_tail_auxii:w} removes one undelimited +% argument and leaves everything else until an end-marker \cs{q_mark}. +% One can check that an empty (or blank) string yields an empty +% tail. +% \begin{macrocode} +\cs_new_nopar:Npn \str_tail:N { \exp_args:No \str_tail:n } +\cs_set:Npn \str_tail:n #1 + { + \exp_after:wN \@@_tail_auxi:w + \reverse_if:N \if_charcode:w + \scan_stop: \tl_to_str:n {#1} X X \q_stop + } +\cs_set:Npn \@@_tail_auxi:w #1 X #2 \q_stop { \fi: #1 } +\cs_new:Npn \str_tail_ignore_spaces:n #1 + { + \exp_after:wN \@@_tail_auxii:w + \tl_to_str:n {#1} \q_mark \q_mark \q_stop + } +\cs_new:Npn \@@_tail_auxii:w #1 #2 \q_mark #3 \q_stop { #2 } +% \end{macrocode} +% \end{macro} +% \end{macro} +% +% \subsection{Accessing specific characters in a string} +% +% \begin{macro}[EXP, int]{\@@_to_other:n} +% \begin{macro}[EXP, aux]{\@@_to_other_loop:w, \@@_to_other_end:w} +% 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. The end +% is detected when \cs{@@_to_other_loop:w} finds one of the trailing +% |A|, distinguished from any contents of the initial token list by +% their category. Then \cs{@@_to_other_end:w} is called, and finds +% the result between \cs{q_mark} and the first |A| (well, there is +% also the need to remove a space). +% \begin{macrocode} +\group_begin: +\char_set_lccode:nn { `\* } { `\ } +\char_set_lccode:nn { `\A } { `\A } +\tl_to_lowercase:n + { + \group_end: + \cs_new:Npn \@@_to_other: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 \@@_to_other_loop:w + #1 ~ #2 ~ #3 ~ #4 ~ #5 ~ #6 ~ #7 ~ #8 ~ #9 \q_stop + { + \if_meaning:w A #8 + \@@_to_other_end:w + \fi: + \@@_to_other_loop:w + #9 #1 * #2 * #3 * #4 * #5 * #6 * #7 * #8 * \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}[EXP, aux]{\@@_skip_c_zero:w} +% \begin{macro}[EXP, aux] +% {\@@_skip_loop:wNNNNNNNN, \@@_skip_end:w, \@@_skip_end: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{@@_skip_end:NNNNNNNN}. If the number of characters to remove +% is $6$, say, then there are two \cs{or:} left, and the $8$ arguments +% of \cs{@@_skip_end: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{@@_skip_c_zero:w} is called). +% \begin{macrocode} +\cs_new:Npn \@@_skip_c_zero:w #1; + { + \if_int_compare:w #1 > \c_eight + \exp_after:wN \@@_skip_loop:wNNNNNNNN + \else: + \exp_after:wN \@@_skip_end:w + \int_use:N \__int_eval:w + \fi: + #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 \@@_skip_end:NNNNNNNN + \if_case:w #1 \exp_stop_f: \or: \or: \or: \or: \or: \or: \or: \or: + } +\cs_new:Npn \@@_skip_end:NNNNNNNN #1#2#3#4#5#6#7#8 { \fi: \c_zero } +% \end{macrocode} +% \end{macro} +% \end{macro} +% +% \begin{macro}[EXP, aux]{\@@_collect_delimit_by_q_stop:w} +% \begin{macro}[EXP, aux] +% { +% \@@_collect_loop:wn, \@@_collect_loop:wnNNNNNNN, +% \@@_collect_end:wn, \@@_collect_end:nnnnnnnnw +% } +% Collects |max(#1,0)| characters, and removes everything else until +% \cs{q_stop}. This is somewhat similar to \cs{@@_skip_c_zero:w}, but +% accepts integer expression arguments. 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{@@_collect_end: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 \@@_collect_delimit_by_q_stop:w #1; + { \@@_collect_loop:wn #1 ; { } } +\cs_new:Npn \@@_collect_loop:wn #1 ; + { + \if_int_compare:w #1 > \c_seven + \exp_after:wN \@@_collect_loop:wnNNNNNNN + \else: + \exp_after:wN \@@_collect_end:wn + \fi: + #1 ; + } +\cs_new:Npn \@@_collect_loop:wnNNNNNNN #1; #2 #3#4#5#6#7#8#9 + { + \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 \@@_collect_end:wn #1 ; + { + \exp_after:wN \@@_collect_end: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 \@@_collect_end:nnnnnnnnw #1#2#3#4#5#6#7#8 #9 \q_stop + { #1#2#3#4#5#6#7#8 } +% \end{macrocode} +% \end{macro} +% \end{macro} +% +% \begin{macro}[EXP]{\str_item:Nn, \str_item:nn, \str_item_ignore_spaces:nn} +% \begin{macro}[EXP, aux]{\@@_item_unsafe:nn, \@@_item:ww} +% The \cs{str_item:nn} hands its argument with spaces escaped to +% \cs{@@_item_unsafe:nn}, and makes sure to turn the result back into +% a proper string (with category code~$10$ spaces) eventually. The +% \cs{str_item_ignore_spaces:nn} function cheats a little bit in that +% it doesn't hand to \cs{@@_item_unsafe:nn} an \enquote{other string}. +% This is safe, as everything else is done with undelimited arguments. +% Then evaluate the \meta{index} argument~|#2| and count characters in +% the string, passing those two numbers to \cs{@@_item:ww} for further +% analysis. If the \meta{index} is negative, shift by |#2|, and +% remove that number of characters before returning the next item in +% the input stream (and if |#1| is smaller than $-|#2|$, nothing is +% returned). If the \meta{index} is positive, ignore that number +% (minus one) of characters before returning the next one. The shift +% by one is obtained by inserting an empty brace group before the +% string in that case: that brace group also covers the case where the +% \meta{index} is zero. +% \begin{macrocode} +\cs_new_nopar:Npn \str_item:Nn { \exp_args:No \str_item:nn } +\cs_new:Npn \str_item:nn #1#2 + { + \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 \@@_item_unsafe:nn { \tl_to_str:n {#1} } } +\cs_new:Npn \@@_item_unsafe:nn #1#2 + { + \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 \@@_item:ww #1; #2; + { + \int_compare:nNnTF {#1} < \c_zero + { + \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 \exp_after:wN \@@_skip_c_zero:w + \int_use:N \__int_eval:w #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 \@@_skip_c_zero:w #1 ; { } + } + } + } +% \end{macrocode} +% \end{macro} +% \end{macro} +% +% \begin{macro}[EXP, aux]{\@@_range_normalize:nn} +% This function converts an \meta{index} argument into an explicit +% position in the string (a result of $0$ denoting \enquote{out of +% bounds}). Expects two explicit integer arguments: the +% \meta{index} |#1| and the string count~|#2|. If |#1| is negative, +% replace it by $|#1| + |#2| + 1$, then limit to the range $[0, +% |#2|]$. +% \begin{macrocode} +\cs_new:Npn \@@_range_normalize:nn #1#2 + { + \int_eval:n + { + \if_int_compare:w #1 < \c_zero + \if_int_compare:w #1 < -#2 \exp_stop_f: + \c_zero + \else: + #1 + #2 + \c_one + \fi: + \else: + \if_int_compare:w #1 < #2 \exp_stop_f: + #1 + \else: + #2 + \fi: + \fi: + } + } +% \end{macrocode} +% \end{macro} +% +% \begin{macro}[EXP] +% {\str_range:Nnn, \str_range:nnn, \str_range_ignore_spaces:nnn} +% \begin{macro}[EXP, int]{\@@_range_unsafe:nnn} +% \begin{macro}[EXP, aux]{\@@_range:www, \@@_range:nnw} +% Sanitize the string. Then evaluate the arguments. At this stage we +% also decrement the \meta{start index}, since our goal is to know how +% many characters should be removed. Then limit the range to be +% non-negative and at most the length of the string (this avoids +% needing to check for the end of the string when grabbing +% characters), shifting negative numbers by the appropriate amount. +% Afterwards, skip characters, then keep some more, and finally drop +% the end of the string. +% \begin{macrocode} +\cs_new_nopar:Npn \str_range:Nnn { \exp_args:No \str_range:nnn } +\cs_new:Npn \str_range:nnn #1#2#3 + { + \exp_args:Nf \tl_to_str:n + { + \exp_args:Nf \@@_range_unsafe:nnn + { \@@_to_other:n {#1} } {#2} {#3} + } + } +\cs_new:Npn \str_range_ignore_spaces:nnn #1 + { \exp_args:No \@@_range_unsafe:nnn { \tl_to_str:n {#1} } } +\cs_new:Npn \@@_range_unsafe:nnn #1#2#3 + { + \exp_after:wN \@@_range:www + \__int_value:w \@@_count_unsafe:n {#1} \exp_after:wN ; + \int_use:N \__int_eval:w #2 - \c_one \exp_after:wN ; + \int_use:N \__int_eval:w #3 ; + #1 \q_stop + } +\cs_new:Npn \@@_range:www #1; #2; #3; + { + \exp_args:Nf \@@_range:nnw + { \@@_range_normalize:nn {#2} {#1} } + { \@@_range_normalize:nn {#3} {#1} } + } +\cs_new:Npn \@@_range:nnw #1#2 + { + \exp_after:wN \@@_collect_delimit_by_q_stop:w + \int_use:N \__int_eval:w #2 - #1 \exp_after:wN ; + \tex_romannumeral:D \@@_skip_c_zero:w #1 ; + } +% \end{macrocode} +% \end{macro} +% \end{macro} +% \end{macro} +% +% \subsection{String conditionals} +% +% \begin{macro}[EXP, pTF]{\str_if_eq:NN} +% \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 } + { + \if_int_compare:w \__str_if_eq_x:nn { \tl_to_str:N #1 } { \tl_to_str:N #2 } + = \c_zero \prg_return_true: \else: \prg_return_false: \fi: + } +% \end{macrocode} +% \end{macro} +% \end{macro} +% +% \begin{macro}[EXP, TF]{\str_case:nn, \str_case:on, \str_case_x:nn} +% Defined in \pkg{l3basics} at present. +% \end{macro} +% +% \subsection{Viewing strings} +% +% \begin{macro}{\str_show:n, \str_show:N, \str_show:c} +% Displays a string on the terminal. +% \begin{macrocode} +\cs_new_eq:NN \str_show:n \tl_show:n +\cs_new_eq:NN \str_show:N \tl_show:N +\cs_generate_variant:Nn \str_show:N { c } +% \end{macrocode} +% \end{macro} +% +% \subsection{Deprecated string functions} +% +% Deprecated 2013-01-20 for removal by 2013-04-30 +% \begin{macro}[EXP] +% {\str_substr:Nnn, \str_substr:nnn, \str_substr_ignore_spaces:nnn} +% \begin{macro}[EXP, aux]{\@@_substr:nnn} +% These functions used to allow for an empty argument to denote the +% start/end of the string. We reimplement them here by first checking +% for an empty argument, then only calling the appropriate version of +% the \cs{str_range:nnn} function. +% \begin{macrocode} +\cs_new:Npn \str_substr:Nnn #1 { \@@_substr:nnn { \str_range:Nnn #1 } } +\cs_new:Npn \str_substr:nnn #1 { \@@_substr:nnn { \str_range:nnn {#1} } } +\cs_new:Npn \str_substr_ignore_spaces:nnn #1 + { \@@_substr:nnn { \str_range_ignore_spaces:nnn {#1} } } +\cs_new:Npn \@@_substr:nnn #1#2#3 + { + \tl_if_empty:nTF {#2} + { \tl_if_empty:nTF {#3} { #1 { 1} { -1 } } { #1 { 1} {#3} } } + { \tl_if_empty:nTF {#3} { #1 {#2} { -1 } } { #1 {#2} {#3} } } + } +% \end{macrocode} +% \end{macro} +% \end{macro} +% +% Deprecated 2013-01-20 for removal by 2013-04-30 +% \begin{variable}{\c_lbrace_str, \c_rbrace_str} +% \begin{macrocode} +\cs_new_eq:NN \c_lbrace_str \c_left_brace_str +\cs_new_eq:NN \c_rbrace_str \c_right_brace_str +% \end{macrocode} +% \end{variable} +% +% \begin{macrocode} +%</initex|package> +% \end{macrocode} +% +% \end{implementation} +% +% \PrintIndex |