typgloss.sty

% typgloss
% A LaTeX package to typeset linguistic abbreviations and syntactic annotations in interlinear glossed examples
% Repository: https://github.com/tpellard/typgloss
% E-mail: thomas.pellard@cnrs.fr

\NeedsTeXFormat{LaTeX2e}[1996/10/24]%
\ProvidesPackage{typgloss}%
[2021-06-16 v.0.0.3 typgloss package for linguistic abbreviations and syntactic annotations]%

% The package should be loaded *after* leipzig and gb4e.
% It converts all upper case grammatical abbreviations to small caps
% within interlinear glossed texts. If leipzig is loaded, it passes
% them in lower case form to \acrshort{}. It also works with person
% number labels (1, 2, 3). It also correctly formats bracketed annotations.

% New boolean checking whether the leipzig package is loaded or not
\newif\ifleipzig
\@ifpackageloaded{leipzig}{\leipzigtrue}{\leipzigfalse}
  
% Gloss formatting command
% The definition depends on whether is loaded or not.
% If leipzig is loaded, pass the lower case abbreviation to \acrshort{}.
\ifleipzig \newcommand{\grlblformat}[1]{\lowercase{\acrshort{#1}}}%
% If not, put the lower case abreviation in small caps.
\else \newcommand{\grlblformat}[1]{\textsc{\lowercase{#1}}}%
\fi

% Formatting of syntactic annotations
% brackets are always upright
\newcommand{\syntbformat}[1]{\textup{#1}}
% alphanumeric annotations are always upright and subscript
\newcommand{\syntannformat}[1]{\textsubscript{\textup{#1}}}
% negative space to correct the alignment in interlinar glossed examples
\newlength{\wspace}
\setlength{\wspace}{\fontdimen2\font plus\fontdimen3\font minus\fontdimen4\font}
\newcommand{\syntnegspace}{\hskip -\wspace}

% Regular expression replacements
\RequirePackage{expl3}
\ExplSyntaxOn
% wrap strings of upper case letters, single upper case letters that
% are not at the beginning of word, and the numbers 1, 2, and 3 inside \grlblformat
\newcommand{\grlbl}[1]{{
%   \XeTeXinterchartokenstate=0
  \tl_set:Nn \l_temp_tl {#1}
  
  % strings of 2 or more upper case letters
  \regex_replace_all:nnN {[A-Z][A-Z]+} {\c{grlblformat}\cB\{\0\cE\}} \l_temp_tl

  % one-letter abbreviations A, S, and P, as well as person numbers require special attention with leipzig
  \ifleipzig
  \regex_replace_all:nnN {([^a-zA-Z])([ASP])\Z} {\1\c{grlblformat}\cB\{\2arg\cE\}} \l_temp_tl
  \regex_replace_all:nnN {([^a-zA-Z])([ASP])([^a-zA-Z])} {\1\c{grlblformat}\cB\{\2arg\cE\}\3} \l_temp_tl
  \regex_replace_all:nnN {1} {\c{grlblformat}\cB\{first\cE\}} \l_temp_tl
  \regex_replace_all:nnN {2} {\c{grlblformat}\cB\{second\cE\}} \l_temp_tl
  \regex_replace_all:nnN {3} {\c{grlblformat}\cB\{third\cE\}} \l_temp_tl
  \else\fi

  % other one-letter abbreviations
  \regex_replace_all:nnN {([^a-zA-Z])([A-Z])([^a-zA-Z])} {\1\c{grlblformat}\cB\{\2\cE\}\3} \l_temp_tl
  \regex_replace_all:nnN {([^a-zA-Z])([A-Z])\Z} {\1\c{grlblformat}\cB\{\2\cE\}} \l_temp_tl

  \tl_use:N \l_temp_tl
}}

% wrap brackets and following annotations inside \syntbformat{} and \syntannformat{}
% add \syntnegspace in case of opening brackets
\newcommand{\syntbann}[1]{{
%   \XeTeXinterchartokenstate=0
  \tl_set:Nn \l_temp_tl {#1}
  
  \regex_replace_all:nnN {(\[+)(\S*)} {\c{syntbformat}\cB\{\1\cE\}\c{syntannformat}\cB\{\2\cE\}\c{syntnegspace}} \l_temp_tl
  \regex_replace_all:nnN {(\]+)(\S*)} {\c{syntbformat}\cB\{\1\cE\}\c{syntannformat}\cB\{\2\cE\}} \l_temp_tl

  \tl_use:N \l_temp_tl
}}
\ExplSyntaxOff

% Each word in the first line of gets wrapped inside \syntbann{}
% Font formatting should be done by redefining \eachwordonestyle
\newcommand{\eachwordonestyle}{}
\renewcommand{\eachwordone}{\eachwordonestyle\syntbann}

% Each word in the second line of gets wrapped inside \grlbl{}
% Font formatting should be done by redefining \eachwordtwostyle
\newcommand{\eachwordtwostyle}{}
\renewcommand{\eachwordtwo}{\eachwordtwostyle\grlbl}

% langsci-gb4e requires a patch since it uses an older code
\RequirePackage{xpatch}
\xpatchcmd{\getwords}%
  {\strut#3{}}%
  {{\strut#3}}%
  {}{}%

% patch for linguex
\@ifpackageloaded{linguex}{%
\def\getwords(#1,#2)#3 #4\\% #1=linebox, #2=\each, #3=1st word, #4=remainder
   {\setbox#1=\vbox{\hbox{#2{\strut\ifindex\expandafter\checkforbr\fi#3} }%
%    {\setbox#1=\vbox{\hbox{#2{\strut#3} }% adds space
                    \unvbox#1}%
    \def\more{#4}%
    \ifx\more\empty\let\more=\donewords
    \else\let\more=\getwords 
    \fi
    \more(#1,#2)#4\\%
   }%
}{}