summaryrefslogtreecommitdiff
path: root/Master/texmf-dist/source/latex/phfparen
diff options
context:
space:
mode:
authorKarl Berry <karl@freefriends.org>2016-08-15 20:38:41 +0000
committerKarl Berry <karl@freefriends.org>2016-08-15 20:38:41 +0000
commitd939f20185ca3eeb907009d125b3b6f8a69203c4 (patch)
tree2e96d66c663f04e40ae06ea889bd7f05e691b3f7 /Master/texmf-dist/source/latex/phfparen
parent6cf79a03641bce0310d2a380e33a2684780c8c19 (diff)
phfparen (16aug16)
git-svn-id: svn://tug.org/texlive/trunk@41859 c570f23f-e606-0410-a88d-b1316a301751
Diffstat (limited to 'Master/texmf-dist/source/latex/phfparen')
-rw-r--r--Master/texmf-dist/source/latex/phfparen/phfparen.dtx781
-rw-r--r--Master/texmf-dist/source/latex/phfparen/phfparen.ins55
2 files changed, 836 insertions, 0 deletions
diff --git a/Master/texmf-dist/source/latex/phfparen/phfparen.dtx b/Master/texmf-dist/source/latex/phfparen/phfparen.dtx
new file mode 100644
index 00000000000..b746f6c51f7
--- /dev/null
+++ b/Master/texmf-dist/source/latex/phfparen/phfparen.dtx
@@ -0,0 +1,781 @@
+% \iffalse meta-comment
+%
+% Copyright (C) 2016 by Philippe Faist <philippe.faist@bluewin.ch>
+% -------------------------------------------------------
+%
+% This file may be distributed and/or modified under the
+% conditions of the LaTeX Project Public License, either version 1.3
+% of this license or (at your option) any later version.
+% The latest version of this license is in:
+%
+% http://www.latex-project.org/lppl.txt
+%
+% and version 1.3 or later is part of all distributions of LaTeX
+% version 2005/12/01 or later.
+%
+% \fi
+%
+% \iffalse
+%<*driver>
+\ProvidesFile{phfparen.dtx}
+%</driver>
+%<package>\NeedsTeXFormat{LaTeX2e}[2005/12/01]
+%<package>\ProvidesPackage{phfparen}
+%<*package>
+ [2016/08/15 v1.0 phfparen package]
+%</package>
+%
+%<*driver>
+\documentclass{ltxdoc}
+\usepackage{xcolor}
+\usepackage{phfqit}
+\usepackage{amsmath}
+\usepackage[backtickon]{phfparen}
+\makeatletter
+\providecommand\phfnote@pkgdoc@setupmainfont{
+ \renewcommand{\rmdefault}{futs}% only rm font, not math
+}\makeatother
+\usepackage[preset=xpkgdoc]{phfnote}
+\EnableCrossrefs
+\CodelineIndex
+\RecordChanges
+
+\verbdef\phfverbpren+\paren+
+\robustify\phfverbpren
+
+\begin{document}
+ \DocInput{phfparen.dtx}
+\end{document}
+%</driver>
+% \fi
+%
+% \CheckSum{0}
+%
+% \CharacterTable
+% {Upper-case \A\B\C\D\E\F\G\H\I\J\K\L\M\N\O\P\Q\R\S\T\U\V\W\X\Y\Z
+% Lower-case \a\b\c\d\e\f\g\h\i\j\k\l\m\n\o\p\q\r\s\t\u\v\w\x\y\z
+% Digits \0\1\2\3\4\5\6\7\8\9
+% Exclamation \! Double quote \" Hash (number) \#
+% Dollar \$ Percent \% Ampersand \&
+% Acute accent \' Left paren \( Right paren \)
+% Asterisk \* Plus \+ Comma \,
+% Minus \- Point \. Solidus \/
+% Colon \: Semicolon \; Less than \<
+% Equals \= Greater than \> Question mark \?
+% Commercial at \@ Left bracket \[ Backslash \\
+% Right bracket \] Circumflex \^ Underscore \_
+% Grave accent \` Left brace \{ Vertical bar \|
+% Right brace \} Tilde \~}
+%
+%
+% \changes{v1.0}{2016/04/20}{Initial version}
+%
+% \GetFileInfo{phfparen.dtx}
+%
+% \DoNotIndex{\newcommand,\newenvironment,\def,\gdef,\edef,\xdef,\if,\else,\fi,\ifx}
+%
+% \title{\phfqitltxPkgTitle{phfparen}}
+% \author{Philippe Faist\quad\email{philippe.faist@bluewin.ch}}
+% \date{\pkgfmtdate\filedate}
+% \maketitle
+%
+% \begin{abstract}
+% \pkgname{phfparen}---Parenthetic math expressions made simpler and less
+% redundant.
+% \end{abstract}
+%
+% \inlinetoc
+%
+% \section{Introduction}
+%
+% The \pkgname{phfparen} package provides an easier way to handle
+% parenthetic expressions in math formulae, by reducing the amount of
+% redundancy in the \LaTeX{} code.
+%
+% The basic command is |\paren|\hspace{0.25ex}\meta{optional size
+% arg}\hspace{0.25ex}\meta{opening delimiter} \ldots{} \meta{closing
+% delimiter}. It puts the given contents (``\ldots'') within the given
+% delimiters (such as ``[...]'', ``(...)'' or ``\{...\}''), and where the
+% delimiters are sized according to the default size (no argument), the
+% explicitly given size argument (``|\big|''), or the automatic sizing
+% (``|*|''). As a shorthand, you may use a backtick character (``|`|'')
+% instead of |\paren|.
+%
+% For example, you can use |`\Big(1+\hat{f}`\big(x))| or
+% |\paren\Big(1+\hat{f}\paren\big(x))| instead of
+% |\Bigl(1+\hat{f}\bigl(x\bigr)\Bigr)|, to display
+% $`\Big(1+\hat{f}`\big(x))$. In this way, the writing is more condensed.
+% Moreover, you don't have to repeat the size for the second delimiter,
+% making it easier to change the delimiter sizes if you later change your
+% mind. See further examples in \autoref{sec:examples}.
+%
+% \iffalse META-COMMENT :: TOO LONG
+% \begin{verbatim}
+% \begin{align}
+% P\,`\bigg[\Phi_{XA\to X'A}`*(\sigma_{XR}\otimes
+% \frac{P_A\Gamma_A P_A}{\tr P_A\Gamma_A})
+% \,,\; \rho_{X'R}\otimes
+% \frac{P'_{A'}\Gamma_{A'}P'_{A'}}{\tr P'_{A'}\Gamma_{A'}}
+% ] \leqslant \epsilon\ .
+% \end{align}
+% \end{verbatim}
+% for:
+% \begin{align}
+% P\,`\bigg[\Phi_{XA\to X'A}`*(\sigma_{XR}\otimes
+% \frac{P_A\Gamma_A P_A}{\tr P_A\Gamma_A})
+% \,,\; \rho_{X'R}\otimes
+% \frac{P'_{A'}\Gamma_{A'}P'_{A'}}{\tr P'_{A'}\Gamma_{A'}}
+% ] \leqslant \epsilon\ .
+% \end{align}
+% \fi
+%
+%
+% \section{The \phfverbpren{} command and \phfverb` backtick shorthand}
+%
+% \DescribeMacro{\paren} The basic macro of this package is |\paren|. It
+% handles a parenthetic expression including parsing matching delimiters
+% and adjusting their sizing.
+%
+% \leavevmode\marginpar{\raggedleft\phfverb`\texttt\ldots} The backtick
+% char provides a shorthand for specifying |\paren|. It is useful to
+% de-clutter long formulas.
+%
+% The syntax for |\paren| and |`| is:
+% \begin{center}
+% |\paren|\meta{optional size argument}\meta{open-delimiter} \ldots \meta{close-delimiter}
+% \par or\par
+% |`|\meta{optional size argument}\meta{open-delimiter} \ldots \meta{close-delimiter}
+% \end{center}
+%
+% By default, the following pairs of delimiters are recognized:
+% |\paren(expression in parenthesis)|,\\
+% |\paren[expression in brackets]|,\\
+% |\paren<expression in angle brackets>|,\\
+% |\paren\{expression in curly braces\}|, and\\
+% |\paren{expression in curly braces}| (alternative syntax for curly
+% braces). Of course, the same are available with the backtick shorthand:
+% |`(expression)|, |`[expression]|, |`<expression>|, and so on.
+%
+% Delimiters are balanced, meaning that %
+% |\paren[g[x] + f[x]]| is parsed as you would expect (the expression is %
+% ``|g[x] + f[x]|'') and not as \TeX/\LaTeX{} usually parses default
+% arguments (in which case the argument would be ``|g[x|''). \iffalse]\fi
+%
+%
+% The optional size argument must be a single token. It may be one of |*|,
+% |\big|, |\Big|, |\bigg|, |\Bigg|, or another standard sizing command. If
+% the optional size argument is |*|, then the delimiters are sized with
+% |\left(| \ldots{} |\right)|, that is, they are sized
+% automatically.
+%
+% \subsection{Examples}
+% \label{sec:examples}
+%
+% \makeatletter
+% \def\showcaseexample{\begingroup\catcode`\\=12\catcode`\_=12\catcode`\^=12\catcode`\{=12\catcode`\}=12\relax\showcaseexample@}
+% \def\showcaseexample@!!#1!!{\endgroup\showcaseexample@@{#1}{\scantokens{#1}}}
+% \def\showcaseexample@@#1#2{\par\noindent\begin{minipage}{0.5\textwidth}\centering\ttfamily #1\end{minipage}\hspace{0.05\textwidth}\begin{minipage}{0.45\textwidth}\centering $\displaystyle #2$\end{minipage}\par\vskip 0.5em\relax}
+% \makeatother
+%
+% \showcaseexample!! \log\paren\big( a + \sin(x) - y) !!
+% \showcaseexample!! \log`\Big( a + \sin(x) - y) !!
+% \showcaseexample!! \log`*( \sum_k`\big[ a_k^\dagger - a_k ] ) !!
+% \showcaseexample!! F = `\big< f(x) >_x = `*< \sum g(`<p>) > !!
+% \showcaseexample!! `\Big[\sum_j `[x]^2 - \sum [y]^2 + \sum f`*(x_3^\dagger)^2] !!
+% \showcaseexample!! `(z + \backtick x) !!
+% \showcaseexample!! `\{\vec z\} !!
+% \showcaseexample!! `{\vec z} !!
+% \showcaseexample!! `*{ \vec z : \sum_i z_i = 1 } !!
+% \showcaseexample!! `\big{ \vec z : \sum_i z_i = 1 } !!
+%
+%
+%
+% \subsection{Controlling the behavior of the \phfverb` backtick shorthand}
+%
+% The behavior of the backtick character is only altered in math mode. The
+% character works as normal in text mode. (Its catcode is in fact not
+% changed, only its mathcode.)
+%
+% \DescribeMacro{\backtick} If you need a literal backtick, you may in any
+% case use the macro |\backtick|.
+%
+% You may use the |backtick| package option to enable or disable the |`|
+% backtick shorthand.
+%
+% \begin{pkgoptions}
+% \item[backtick=true] Enable the |`| backtick shorthand. This is the
+% default.
+% \item[backtick=false] Disable the |`| backtick shorthand. A backtick
+% character in math mode will now simply display a backtick character.
+% \end{pkgoptions}
+%
+% \DescribeMacro{\parenMakeBacktickActiveParen} You may use the command
+% |\parenMakeBacktickActiveParen| to enable the backtick shorthand at any
+% time in the document. Similarly, the command
+% \DescribeMacro{\parenMakeNormalBacktick} |\parenMakeNormalBacktick|
+% disables the backtick shorthand. These changes are local to \LaTeX{}
+% groups, which means that you can temporarily disable the backtick
+% shorthand with
+% \begin{verbatim}
+% {\parenMakeNormalBacktick Normal backtick in math mode: $`$.}
+% \end{verbatim}
+%
+%
+% \section{Declaring matching delimiters and corresponding representations}
+%
+% \DescribeMacro{\parenRegister} Register a new delimiter type with
+% |\parenRegister|. This macro should only be called in the preamble.
+%
+% Syntax is: |\parenRegister|\marg{internal name}\marg{open parse
+% delimiter}\marg{close parse delimiter}\marg{open display
+% delimiter}\marg{close display delimiter}.
+%
+% The \meta{internal name} is just an internal identifier: use plain ascii
+% chars please (e.g. ``|angbrackets|'').
+%
+% The \meta{open parse delimiter} and \meta{close parse delimiter} are
+% those characters which will be used to parse the delimited expression.
+% They may differ from the actual delimiters used to display the
+% expression, given by \marg{open display delimiter} and \marg{close
+% display delimiter}.
+%
+% For example, you may use
+% \begin{verbatim}
+% \parenRegister{angbrackets}{<}{>}{\langle}{\rangle}
+% \end{verbatim}
+% to instruct |\paren| (and |`|) to do the following: whenever the ``|<|''
+% open delimiter is encountered, then the expression is to be parsed within
+% the delimiters %
+% ``|< ... >|,'' and the final expression is to be delimited as %
+% ``|\langle ... \rangle|,'' where the |\langle| and |\rangle| are possibly
+% sized according to an optional size argument.
+%
+% You may repeat calls to |\parenRegister|, with different internal
+% identifiers, to register different kinds of delimiters. It is always the
+% opening delimiter which determines which registered info to use.
+%
+% \begin{pkgnote}
+% With older versions of \pkgname{xparse} (older than
+% \texttt{2015/11/04}), only single-character tokens or single-character
+% escape sequences (such as |\{|) could be used as delimiters for parsing
+% (second and third arguments to |\parenRegister|).
+%
+% Otherwise, with versions of \pkgname{xparse} newer than
+% \texttt{2015/11/04}, you may use full macros to delimit content, for
+% example |\parenRegister{testang}{\start}{\stop}{\langle}{\rangle}|
+% allows you to write e.g.\@ \fbox{\phfverb{`\start content...\stop}},
+% and the content will be displayed in angle brackets.
+% \end{pkgnote}
+%
+% \DescribeMacro{\parenRegsiterSimpleBraces} The case where the delimiters
+% are a normal \LaTeX{} group, that is ``|`{...}|,'' is treated specially.
+% In this case, you need to specify which internal identifier to fall back
+% to. For example, if you defined
+% |\parenRegister{braces}{\{}{\}}{\{}{\}}|, then you may use
+% |\parenRegisterSimpleBraces{braces}| to instruct |\paren| (and |`|) to
+% use curly braces whenever both the forms |`\{...\}| and |`{...}| are
+% used.
+%
+% \DescribeMacro{\parenRegisterDefaults} Register the default set of
+% delimiters: |`(...)|, |`[...]|, |`<...>|, |`\{...\}|, and with |`{...}|
+% the same as |`\{...\}|.
+%
+% There is a package option |registerdefaults| to instruct to load or not
+% to load the default set of delimiters:
+% \begin{pkgoptions}
+% \item[registerdefaults=true] Set up the default set of delimiters
+% provided by |\parenRegisterDefaults| when loading the package. This is
+% the default.
+% \item[registerdefaults=false] Do not set up any delimiters. It is your
+% job to call |\parenRegister| as necessary to register all your
+% delimiters.
+% \end{pkgoptions}
+%
+%
+%
+%
+%
+%
+%
+%
+%
+%
+%
+%
+%
+%
+%
+% \StopEventually{\PrintChangesAndIndex}
+%
+% \section{Implementation}
+%
+% Load some utility packages.
+%
+% \begin{macrocode}
+\RequirePackage{etoolbox}
+\RequirePackage{kvoptions}
+\RequirePackage{xparse}
+\RequirePackage{amsmath}
+\RequirePackage{mathtools}
+% \end{macrocode}
+%
+%
+% \subsection{Internal \phfverbpren{} API for different delimiters}
+% \label{sec:internal-API-for-delimiters}
+%
+% The idea is to combine the power of the delimiter-parsing mechanism of
+% \pkgname{xparse} with the display power of paired delimiters provided by
+% \pkgname{mathtools}. With \pkgname{xparse}, you can define commands
+% which parse an argument given by arbitrary (but fixed) delimiters. The
+% \pkgname{mathtools} package allows you to define commands such as |\abs|
+% with |\DeclarePairedDelimiter|, such that you can use them as |\abs{x}|
+% for default delimiter sizes, |\abs*{x}| for automatic delimiter sizing,
+% and |\abs[\big]{x}| to use an explicitly given size.
+%
+% The idea for |\paren| is the following: detect what the open delimiter
+% is, and relay the call to a macro which knows how to parse a balanced
+% argument with those delimiters. That macro then relays the call to the
+% paired-delimiter-display macro with appropriate star, optional argument
+% and mandatory argument.
+%
+% When parsing, a delimiter set is recognized by its opening delimiter
+% token. Furthermore, to each delimiter set corresponds an internal
+% identifier name.
+%
+% A delimiter set is declared by |\parenRegister| via three macros. First,
+% the macro |\paren@registered@delims@|\meta{open delimiter token (as
+% \phfverb\string)} is defined to expand to the internal identifier name
+% for this delimiter pair.
+%
+% Then, the macro |\paren@impl@|\meta{internal identifier name} is defined
+% to accept one argument, the possible modifiers passed to |\paren| before
+% the opening delimiter.
+%
+% Finally, the macro |\paren@impl@|\meta{internal identifier name}|@go| is
+% a |\DeclarePairedDelimiter|-declared macro which can display the
+% delimited expression.
+%
+%
+% \subsection{Implementation of \phfverbpren}
+%
+% \begin{macro}{\paren}
+% Define the main |\paren| macro. The syntax is:
+% |\paren|\meta{modifier-token}\meta{delimited-argument}, where
+% modifier-token is optional and can be |*| or a sizing command such as
+% |\big|.
+%
+% Because of the way \LaTeX{} parses arguments, we need to treat the case
+% |\paren{...}| separately form, say, |\paren\{...\}|. Recall that
+% |\@ifnextchar\bgroup| checks for an opening brace. Then, delegate call
+% to appropriate macro for further processing.
+% \begin{macrocode}
+\def\paren{%
+ \@ifnextchar\bgroup\paren@impl@bgroup\paren@impl@nobgroup%
+}
+% \end{macrocode}
+%
+% Treat the special case in which we have a single \LaTeX{} group as
+% argument, no modifiers. In this case, directly call the display macro
+% corresponding to the delimiter set we should use in this case. (We know
+% that the main argument follows in the input token chain.)
+% \begin{macrocode}
+\def\paren@impl@bgroup{%
+ \csname paren@impl@\paren@registered@default @go\endcsname%
+}
+% \end{macrocode}
+%
+%
+% Parse the following token. See if |#1| is an opening delimiter. Do that
+% by seeing if |#1| is a known and registered delimiter by checking if
+% |\paren@registered@delims@|\meta{open delimiter token as \phfverb\string}
+% is defined. If it is not a recognized open delimiter, assume that it is
+% a modifier token.
+% \begin{macrocode}
+\def\paren@impl@nobgroup#1{%
+ \ifcsdef{paren@registered@delims@\string#1}{%
+ \paren@impl@nomod{#1}%
+ }{%
+ \paren@impl@mod{#1}%
+ }%
+}
+% \end{macrocode}
+%
+% We have just parsed and recognized an open delimiter token, repeated here
+% as argument |#1|. So get the internal delimiter identifier via
+% |\paren@registered@delims@\string|\meta{open delim}, and relay the call
+% to |\paren@impl@|\meta{internal delim identifier}. Don't forget an empty
+% argument as we saw no modifier tokens.
+% \begin{macrocode}
+\def\paren@impl@nomod#1{%
+ \letcs\paren@tmp@delimname{paren@registered@delims@\string#1}%
+ \csname paren@impl@\paren@tmp@delimname\endcsname{}#1%
+ %
+}
+% \end{macrocode}
+%
+% We have just seen a token which is not recognized as an open delimiter,
+% so we've assumed it's a modifier token (repeated here as |#1|). Store it
+% in |\paren@tmp@modtoken|, and parse further, keeping in mind that again
+% we may have a \LaTeX{} braced group instead of a single token.
+% \begin{macrocode}
+\def\paren@impl@mod#1{%
+ \def\paren@tmp@modtoken{#1}%
+ \@ifnextchar\bgroup\paren@impl@modBgroup\paren@impl@modNobgroup%
+}
+% \end{macrocode}
+%
+% We have seen a modifier token, then a \LaTeX{} braced group. First, use
+% |\paren@util@parseModifs| to parse the modifier token and to get the
+% actual argument to the |\DeclarePairedDelimiter|-defined macro. Then,
+% see which delimiter set to fall back to, and relay the call to that
+% including the modifier flags.
+% \begin{macrocode}
+\def\paren@impl@modBgroup{%
+ \paren@util@parseModifs\paren@tmp@modifsForMathtools\paren@tmp@modtoken%
+ \letcs\paren@tmp@gocmd{paren@impl@\paren@registered@default @go}%
+ \expandafter\paren@tmp@gocmd\paren@tmp@modifsForMathtools%
+}
+% \end{macrocode}
+%
+% We have seen a modifier token, stored in |\paren@tmp@modtoken|, and now
+% there is a next token coming in (|#1|) which has to be the open delimiter
+% (there cannot be several modifier tokens). First, check that the given
+% delimiter is known and registered.
+% |\paren@registered@delims@|\meta{delimiter token as \phfverb\string}
+% should be defined to be the internal name of the registered delimiter.
+% Then, call the implementation for that delimiter with the modifier-token
+% as argument, and let that do the actual parsing.
+% \begin{macrocode}
+\def\paren@impl@modNobgroup#1{%
+ \ifcsdef{paren@registered@delims@\string#1}{%
+ \letcs\paren@tmp@delimname{paren@registered@delims@\string#1}%
+ \letcs\paren@tmp@delimcmd{paren@impl@\paren@tmp@delimname}%
+ \edef\paren@tmp@modtokenarg{{\expandonce\paren@tmp@modtoken}}%
+ \expandafter\paren@tmp@delimcmd\paren@tmp@modtokenarg#1%
+ }{%
+% \end{macrocode}
+%
+% In the event the open delimiter is not recognized, display an error
+% message, and attempt to recover by keeping all arguments as text tokens.
+% \begin{macrocode}
+ \PackageError{phfqit}{Unknown delimiter: (or can't parse args?)
+ {\expandafter\string\paren@tmp@modtoken}{\string#1}}{Your call
+ to \string\paren couldn't be parsed, presumably because I didn't
+ recognize your delimiter, or because there's a bug in this
+ package.}%
+ \paren@tmp@modtoken#1%
+ }%
+}
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Some Utilities}
+%
+% \begin{macro}{\paren@util@parseModifs}
+% Parse a modifier token, and generates the appropriate size argument
+% tokens (star or optional argument) for a
+% |\DeclarePairedDelimiter|-declared command.
+%
+% \par |#1| = new macro name to be defined.
+% \par |#2| = macro containing the modifier token. (Will be expanded
+% once.)
+%
+% A call to |\paren@util@parseModifs| defines |#1| to contain
+% mathtools-compatible size args. |#2| is expanded once, so it is
+% expected to be a macro which contains the relevant modifiers.
+%
+% \begin{macrocode}
+\def\paren@util@parseModifs#1#2{%
+% \end{macrocode}
+%
+% If the modifier token is a star |*|, then the argument should be a simple star.
+% \begin{macrocode}
+ \expandafter\ifstrequal\expandafter{#2}{*}{%
+ \def#1{*}%
+ }%
+% \end{macrocode}
+% Otherwise, if there is something and |#2| is not blank, put it in an
+% optional argument (square braces).
+% \begin{macrocode}
+ {% (else:)
+ \expandafter\ifblank\expandafter{#2}{%
+ \def#1{}%
+ }{%
+ \edef#1{[\expandonce{#2}]}%
+ }%
+ }%
+}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\paren@xparsefix@storeparenarg}
+% The second utility is a fix for a bug (or feature?) in
+% \pkgname{xparse}. It seems that when specifying a delimited argument
+% with |r|\meta{open}\meta{close}, if the \meta{open} delimiter is not a
+% single character, then the argument is not properly prepared: for
+% example, with some versions of \pkgname{xparse}, for delimiters
+% |r\{\}|, we get for the macro call |\mymacro\{hi there\}| as parsed
+% argument |{hi there|. This bug was not present in older versions of
+% \pkgname{xparse}, where it was impossible to use full macros as
+% delimiters, but it seemed to have been introduced in more recent
+% versions (tracked to starting from version \texttt{2015/11/04}).
+%
+% So we wrap the fix into a macro
+% |\paren@xparsefix@storeparenarg|\marg{main argument which may need
+% fix}. This defines the macro |\paren@tmp@contents| which is the actual
+% argument, possibly fixed.
+%
+% The caller must ensure that the macro |\paren@impl@tmp@delimtoken| is
+% defined to expand to the opening delimiter token.
+%
+% The default implementation assumes that \pkgname{xparse} doesn't have
+% the bug.
+% \begin{macrocode}
+\def\paren@xparsefix@storeparenarg#1{%
+ \def\paren@tmp@contents{#1}%
+}
+% \end{macrocode}
+%
+% Now, redefine this macro to the correct fix depending on the
+% \pkgname{xparse} package version. If the version is more recent than
+% \texttt{2015/11/04}, then we need to strip macro name length (excluding
+% backslash char) from beginning of argument.
+% \begin{macrocode}
+\@ifpackagelater{xparse}{2015/11/04}{
+ \RequirePackage{xstring}%
+ \def\paren@xparsefix@storeparenarg#1{%
+ \fullexpandarg\StrLen{\expandafter\string\paren@impl@tmp@delimtoken}%
+ [\paren@tmp@delimstrlen]%
+ \ifnum\paren@tmp@delimstrlen=1\relax%
+ \def\paren@tmp@contents{#1}%
+ \else%
+ \def\paren@tmp@delimstrlen@{\numexpr\paren@tmp@delimstrlen-1\relax}%
+ \noexpandarg\StrGobbleLeft{#1}{\the\paren@tmp@delimstrlen@}%
+ [\paren@tmp@contents]%
+ \fi%
+ }%
+}{%
+}
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Registering Delimiters}
+%
+% Helper to register a delimiter set for use with |\paren|.
+%
+% \begin{macro}{\parenRegister}
+% The macro |\parenRegister| defines the necessary macros documented in
+% \autoref{sec:internal-API-for-delimiters}. Namely,
+% |\paren@registered@delims@|\meta{open delimiter token, \phfverb\string
+% ified} expands to the internal registered \meta{identifier} for this
+% delimiter. Then, defines |\paren@impl@<name>| which is a parser for
+% this delimiter and calls the mathtools goodies
+% etc. |\paren@impl@<name>| takes one mandatory argument, which is all
+% the stuff to insert before the open delimiter and to pass on to
+% mathtools (just a |*| or a delimiter size), and then reads the
+% delimited (balanced) content.
+%
+% The syntax is |\parenRegister|\marg{identifier}\marg{open delimiter
+% token}\marg{close delimiter token}\marg{display open
+% delimiter}\marg{display close delimiter}.
+%
+% This macro should only be called in the preamble.
+% \begin{macrocode}
+\newcommand\parenRegister[5]{%
+% \end{macrocode}
+% \iffalse meta-comment %}{ -- for matching braces in my editor
+% \fi
+%
+% First, define the actual |\DeclarePairedDelimiter|-type command
+% |\paren@impl@#1@go| which displays the final delimited expression.
+% Recall that |#1| = the internal identifier, while |#4| and |#5| are the
+% delimiters we should use to display the expression.
+%
+% Use |\DeclarePairedDelimiterX| so that we put the main argument inside a
+% \LaTeX{} group, just to be sure. (For example, we want to make sure that
+% in arguments such as |-1|, the minus sign is a unary minus and not seen
+% as a binary operator.)
+% \begin{macrocode}
+ \expandafter\DeclarePairedDelimiterX\csname paren@impl@#1@go\endcsname[1]{#4}{#5}{{##1}}
+% \end{macrocode}
+%
+% The second macro is the one which is capable of parsing the balanced,
+% delimited expression, while first taking a mandatory argument containing
+% the possible modifier token.
+%
+% Here, we just read the modifier token and store them in a temporary
+% macro, while relaying the delimiter parsing to a helper macro (not sure
+% why this is necessary).
+% \begin{macrocode}
+ \csdef{paren@impl@#1}##1{%
+ \gdef\paren@impl@tmp@modifs{##1}%
+ \csname paren@impl@#1@parsedelim\endcsname%
+ }
+ \expandafter\DeclareDocumentCommand\csname paren@impl@#1@parsedelim\endcsname{r#2#3}{%
+ \letcs\paren@impl@tmp@gocmd{paren@impl@#1@go}%
+%%\message{********MESSAGE*********}%
+%%\message{**********\detokenize\expandafter{\paren@impl@tmp@modifs}**********}%
+%%\message{**********\detokenize{##1}***********}%
+ \paren@util@parseModifs\paren@impl@tmp@modifsForMathtools\paren@impl@tmp@modifs%
+%%\show\paren@impl@tmp@modifsForMathtools%
+ \def\paren@impl@tmp@delimtoken{#2}%
+ \paren@xparsefix@storeparenarg{##1}%
+%%\show\paren@impl@tmp@modifsForMathtools
+%%\show\paren@tmp@contents
+ \expandafter\paren@impl@tmp@gocmd\paren@impl@tmp@modifsForMathtools{\paren@tmp@contents}%
+}
+% \end{macrocode}
+%
+% Finally, register this open delimiter to be detected by the |\paren|
+% macro. With this, |\paren| also knows the internal identifier associated
+% to this open delimiter token.
+% \begin{macrocode}
+ \csdef{paren@registered@delims@\string#2}{#1}
+}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\parenRegisterSimpleBraces}
+% Specify the internal name for the delimiter which should be used in the
+% syntax |`{...}| or |\paren{...}| (with or without modifier token).
+%
+% If this macro is called multiple times, the last specified name is
+% used.
+% \begin{macrocode}
+\newcommand\parenRegisterSimpleBraces[1]{%
+ \def\paren@registered@default{#1}%
+}
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \begin{macro}{\parenRegisterDefaults}
+% A macro which registers the default set of delimiters.
+% \begin{macrocode}
+\newcommand\parenRegisterDefaults{
+% \end{macrocode}
+% \iffalse meta-comment %}{ -- for matching braces in my editor
+% \fi
+%
+% The default set of delimiters:
+% \begin{macrocode}
+ \parenRegister{parens}{(}{)}{(}{)}
+ \parenRegister{brackets}{[}{]}{[}{]}
+ \parenRegister{angbrackets}{<}{>}{\langle}{\rangle}
+ \parenRegister{braces}{\{}{\}}{\{}{\}}
+% \end{macrocode}
+%
+% And the registered paren type to use when given the syntax |`{...}|:
+% curly braces.
+% \begin{macrocode}
+ \parenRegisterSimpleBraces{braces}
+}
+% \end{macrocode}
+% \end{macro}
+%
+%
+% \subsection{Handle backtick shorthand}
+%
+% \begin{macro}{\parenMakeBacktickActiveParen}
+% Change the behavior of the backtick character in math mode to
+% (|\parenMakeBacktickActiveParen|) be an alias for |\paren|.
+% \begin{macrocode}
+\def\parenMakeBacktickActiveParen{%
+% \end{macrocode}
+%
+% Set the backtick character to be active, but only in math mode. It still
+% has its normal |\catcode|, only its |\mathcode| is altered.
+% \begin{macrocode}
+\mathcode`\`="8000\relax%
+% \end{macrocode}
+%
+% Use the usual |\lccode|-trick to do ``|\def`{...}|'' but for an active |`| char
+% \begin{macrocode}
+ \begingroup%
+ \lccode`\~=`\`%
+ \lowercase{\endgroup\def~}{\paren}%
+}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\parenMakeNormalBacktick}
+% The command |\parenMakeNormalBacktick| changes the behavior of the
+% backtick character in math mode to be a normal backtick char
+% (``$\backtick$'').
+% \begin{macrocode}
+\def\parenMakeNormalBacktick{\mathcode`\`="0060\relax}
+% \end{macrocode}
+% \end{macro}
+%
+% \begin{macro}{\backtick}
+% A literal backtick. Simply done by temporarily disabling our backtick
+% shorthand within a \LaTeX{} group.
+% \begin{macrocode}
+\def\backtick{\begingroup\parenMakeNormalBacktick`\endgroup}
+% \end{macrocode}
+% \end{macro}
+%
+%
+%
+%
+% \subsection{Parse package options}
+%
+%
+% Set up \pkgname{kvoptions} option parsing.
+% \begin{macrocode}
+\SetupKeyvalOptions{
+ family=phfparen,
+ prefix=phfparen@opt@
+}
+% \end{macrocode}
+%
+%
+% Option \pkgoptionfmt{backtick}, and complementary option \pkgoptionfmt{nobacktick}.
+% \begin{macrocode}
+\DeclareBoolOption[true]{backtick}
+\DeclareComplementaryOption{nobacktick}{backtick}
+% \end{macrocode}
+%
+% For backwards compatibility with my previous versions of
+% \pkgname{phfparen}: provide the |backtickon| package option.
+% \begin{macrocode}
+\DeclareVoidOption{backtickon}{%
+ \phfparen@opt@backticktrue
+ \PackageWarning{phfparen}{Option 'backtickon' is deprecated. It still works,
+ but please consider using 'backtick=true|false' instead.}%
+}
+% \end{macrocode}
+%
+%
+%
+% Option \pkgoptionfmt{registerdefaults} to specify whether to register the
+% default set of delimiters or not.
+% \begin{macrocode}
+\DeclareBoolOption[true]{registerdefaults}
+\DeclareComplementaryOption{noregisterdefaults}{registerdefaults}
+% \end{macrocode}
+%
+%
+% Finally, process the options.
+% \begin{macrocode}
+\DeclareDefaultOption{%
+ \@unknownoptionerror
+}
+\ProcessKeyvalOptions*
+% \end{macrocode}
+%
+%
+% Now, execute the settings dictated by the user options.
+% \begin{macrocode}
+\ifphfparen@opt@backtick
+ \parenMakeBacktickActiveParen
+\fi
+\ifphfparen@opt@registerdefaults
+ \parenRegisterDefaults
+\fi
+% \end{macrocode}
+%
+%\Finale
+\endinput
diff --git a/Master/texmf-dist/source/latex/phfparen/phfparen.ins b/Master/texmf-dist/source/latex/phfparen/phfparen.ins
new file mode 100644
index 00000000000..2462474b482
--- /dev/null
+++ b/Master/texmf-dist/source/latex/phfparen/phfparen.ins
@@ -0,0 +1,55 @@
+%%
+%% Copyright (C) 2016 by Philippe Faist <philippe.faist@bluewin.ch>
+%%
+%% This file may be distributed and/or modified under the conditions of
+%% the LaTeX Project Public License, either version 1.3 of this license
+%% or (at your option) any later version. The latest version of this
+%% license is in:
+%%
+%% http://www.latex-project.org/lppl.txt
+%%
+%% and version 1.3 or later is part of all distributions of LaTeX version
+%% 2005/12/01 or later.
+%%
+
+\input docstrip.tex
+\keepsilent
+
+\usedir{tex/latex/phfqitltx}
+
+\preamble
+
+This is a generated file.
+
+Copyright (C) 2016 by Philippe Faist <philippe.faist@bluewin.ch>
+
+This file may be distributed and/or modified under the conditions of
+the LaTeX Project Public License, either version 1.3 of this license
+or (at your option) any later version. The latest version of this
+license is in:
+
+ http://www.latex-project.org/lppl.txt
+
+and version 1.3 or later is part of all distributions of LaTeX version
+2005/12/01 or later.
+
+\endpreamble
+
+\generate{\file{phfparen.sty}{\from{phfparen.dtx}{package}}}
+
+\obeyspaces
+\Msg{*************************************************************}
+\Msg{* *}
+\Msg{* To finish the installation you have to move the following *}
+\Msg{* file into a directory searched by TeX: *}
+\Msg{* *}
+\Msg{* phfparen.sty *}
+\Msg{* *}
+\Msg{* To produce the documentation run the .dtx file through *}
+\Msg{* (PdF-)LaTeX. *}
+\Msg{* *}
+\Msg{* Happy TeXing! *}
+\Msg{* *}
+\Msg{*************************************************************}
+
+\endbatchfile