lang-hyp.mkiv / last modification: 2020-01-30 14:16
%D \module
%D   [       file=lang-ini,
%D        version=2014.08.10,
%D          title=\CONTEXT\ Language Macros,
%D       subtitle=Experimental Patterns,
%D         author=Hans Hagen,
%D           date=\currentdate,
%D      copyright={PRAGMA ADE \& \CONTEXT\ Development Team}]
%C
%C This module is part of the \CONTEXT\ macro||package and is
%C therefore copyrighted by \PRAGMA. See mreadme.pdf for
%C details.

%D This is an experimental module. We often have to deal with titles
%D that have conflicting demands:
%D
%D \startitemize
%D \startitem They go into a dedicated space (often a graphic). \stopitem
%D \startitem The words cannot be hyphenated. \stopitem
%D \startitem But as an escape they can get hyphenated. \stopitem
%D \startitem In that case we want at least an extra word on the last line. \stopitem
%D \stopitemize
%D
%D These and maybe more cases can be dealt with using dedicated hyphenation
%D mechanisms. At he same time we want to experiment with more extensive patterns
%D as discussed in {\em TUGboat, Volume 27 (2006), No. 2—Proceedings of EuroTEX2006}.

% lua: 5.341 5.354
% tex: 5.174 5.262

\writestatus{loading}{ConTeXt Language Macros / Initialization}

\registerctxluafile{lang-dis}{}
\registerctxluafile{lang-hyp}{}

\unprotect

\definesystemattribute[hyphenation][public]

%D After a decade of playing with these things in \LUATEX|/|\MKIV\ it's time to
%D finish the way we deal with discretionaries. Apart from the fact that they play a
%D role in hyphenation they also need to be dealt with in fonts. Flattening, cleanup
%D and such are now more or less default in \CONTEXT\ so we can simplify some of the
%D code. We also use the new penalty mechanism.

\newcount\compoundhyphenpenalty

\automatichyphenmode    \plusone
\hyphenpenaltymode      \plusfour

\hyphenpenalty            50 % hyphenator
\automatichyphenpenalty   50 % -
\explicithyphenpenalty    50 % \-
\compoundhyphenpenalty    50
\exceptionpenalty       1000

%D This command can change! At some point we will keep the setting with the
%D paragraph and then the \type {\par} can go.

% \unexpanded\def\atleastoneword#1%
%   {\begingroup
%    \enabledirectives[hyphenators.method=traditional]%
%    \enabledirectives[hyphenators.rightwordsmin=1]%
%    \lefthyphenmin \plusfour
%    \righthyphenmin\plusfour
%    #1\par
%    \disabledirectives[hyphenators.rightwordsmin]%
%    \enabledirectives[hyphenators.method]%
%    \endgroup}

% \exhyphenchar    \hyphenasciicode
% \preexhyphenchar \lessthanasciicode
% \postexhyphenchar\morethanasciicode

%D Here is the real way:

\installcorenamespace{hyphenation}
\installcorenamespace{hyphenationfeatures}

\installparameterhandler \??hyphenation {hyphenation}
\installsetuphandler     \??hyphenation {hyphenation}

\setuphyphenation
  [\c!method=\s!default,
   \s!righthyphenchar=0, % number tzt g: etc
   \s!lefthyphenchar=0]  % number

\appendtoks
    \clf_sethyphenationmethod{\hyphenationparameter\c!method}%
\to \everysetuphyphenation

%D These are mostly meant for manuals:

\unexpanded\def\starthyphenation[#1]%
  {\begingroup
   \clf_pushhyphenation{#1}}

\unexpanded\def\stophyphenation
  {\ifhmode\par\fi
   \clf_pophyphenation
   \endgroup}

% This is a global setting, so we need to disable it when needed. However, as
% we are (hopefully) compatible and attribute driven one can also just keep it
% enabled.
%
% \setuphyphenation
%   [\c!method=\s!traditional] % no translations

\unexpanded\def\definehyphenationfeatures
  {\dodoubleargument\lang_hyphenation_define_features}

\unexpanded\def\lang_hyphenation_define_features[#1][#2]%
  {\begingroup
   \letdummyparameter\c!characters\empty           % maybe \s!
   \letdummyparameter\c!hyphens\empty              % maybe \s!
   \letdummyparameter\c!joiners\empty              % maybe \s!
   \letdummyparameter\c!rightwords\zerocount       % maybe \s!
   \letdummyparameter\s!lefthyphenmin\zerocount
   \letdummyparameter\s!righthyphenmin\zerocount
   \letdummyparameter\s!hyphenmin\zerocount
   \letdummyparameter\s!lefthyphenchar\zerocount
   \letdummyparameter\s!righthyphenchar\zerocount
   \letdummyparameter\c!alternative\empty
   \letdummyparameter\c!rightedge\empty
   \letdummyparameter\c!rightchars\empty
   \getdummyparameters[#2]%
   \clf_definehyphenationfeatures
      {#1}%
      {
        characters   {\dummyparameter\c!characters}%
        hyphens      {\dummyparameter\c!hyphens}%
        joiners      {\dummyparameter\c!joiners}%
        rightwordmin \numexpr\dummyparameter\c!rightwords\relax
        rightchars   {\dummyparameter\c!rightchars}%
        charmin      \numexpr\dummyparameter\s!hyphenmin\relax
        leftcharmin  \numexpr\dummyparameter\s!lefthyphenmin\relax
        rightcharmin \numexpr\dummyparameter\s!righthyphenmin\relax
        leftchar     \numexpr\dummyparameter\s!lefthyphenchar\relax
        rightchar    \numexpr\dummyparameter\s!righthyphenchar\relax
        alternative  {\dummyparameter\c!alternative}%
rightedge    {\dummyparameter\c!rightedge}%
% autohyphen   {\dummyparameter\c!autohyphen}
% hyphenonly   {\dummyparameter\c!hyphenonly}
      }%
   \relax
   \endgroup}

\unexpanded\def\sethyphenationfeatures[#1]%
  {\clf_sethyphenationfeatures{#1}}

\unexpanded\def\resethyphenationfeatures
  {\hyphenationattribute\attributeunsetvalue}

\resethyphenationfeatures

% todo: \start ... \stop too

\unexpanded\def\registerhyphenationpattern
  {\dodoubleempty\lang_hyphenation_register_pattern}

\def\lang_hyphenation_register_pattern[#1][#2]%
  {\clf_registerhyphenationpattern\ifsecondargument{#1}{#2}\else{\currentlanguage}{#1}\fi\s!true\relax}

\unexpanded\def\unregisterhyphenationpattern
  {\dodoubleempty\lang_hyphenation_unregister_pattern}

\def\lang_hyphenation_unregister_pattern[#1][#2]%
  {\clf_registerhyphenationpattern\ifsecondargument{#1}{#2}\else{\currentlanguage}{#1}\fi\s!false\relax}

\unexpanded\def\registerhyphenationexception
  {\dodoubleempty\lang_hyphenation_register_exception}

\def\lang_hyphenation_register_exception[#1][#2]%
  {\clf_registerhyphenationexception\ifsecondargument{#1}{#2}\else{\currentlanguage}{#1}\fi\relax}

\unexpanded\def\showhyphenationtrace
  {\dodoubleempty\lang_hyphenation_show_trace}

\def\lang_hyphenation_show_trace[#1][#2]%
  {\begingroup
   \tt
   \clf_showhyphenationtrace\ifsecondargument{#1}{#2}\else{\currentlanguage}{#1}\fi\relax
   \endgroup}

% For old times sake:

\unexpanded\def\atleastoneword#1%
  {\begingroup
   \starthyphenation[traditional]% this might become default or a faster switch
   \sethyphenationfeatures[words]%
   #1\par
   \stophyphenation
   \endgroup}

%D For me:

\unexpanded\def\showdiscretionaries
  {\clf_showdiscretionaries}

%D These are (at least now) not cummulative:

\definehyphenationfeatures % just an example
  [fences]
  [\c!characters={[]()}]

\definehyphenationfeatures
  [words]
  [\c!rightwords=1,
   \s!lefthyphenmin=4,
   \s!righthyphenmin=4]

\definehyphenationfeatures
  [default]
  [%c!rightedge=\v!tex,
   \c!hyphens=\v!yes,
   \c!joiners=\v!yes]

\definehyphenationfeatures
  [strict]
  [\c!rightedge=\s!tex]

% \sethyphenationfeatures
%   [fences]

% \sethyphenationfeatures
%   [default,fences]

% \setuphyphenation % will be default
%   [method=expanded]

\protect \endinput

% \starttext
%
% \enabledirectives[hyphenators.method=traditional]
%
% % \dorecurse{1000}{\input tufte \par}
%
% \setupalign[verytolerant,flushleft]
% \setuplayout[width=140pt] \showframe
%
% longword longword long word longword longwordword \blank
%
% \enabledirectives[hyphenators.rightwordsmin=1]
%
% longword longword long word longword longwordword\blank
%
% \disabledirectives[hyphenators.rightwordsmin]
%
% longword longword long word longword longwordword\blank
%
% \atleastoneword{longword longword long word longword longwordword}
%
% \enabledirectives[hyphenators.method=traditional]
%
% \stoptext

% \startluacode
%     -- e1ë/e=e             reëel      re-eel
%     -- a1atje./a=t,1,3     omaatje    oma-tje
%     -- schif1f/ff=f,5,2    Schiffahrt Schiff-fahrt
%
%     languages.hyphenators.traditional.registerpattern("en","a1b",      { start = 1, length = 2, before = "CD", after = "EF"  } )
%     languages.hyphenators.traditional.registerpattern("en","e1ë",      { start = 1, length = 2, before = "e",  after = "e"  } )
%     languages.hyphenators.traditional.registerpattern("en","oo1ë",     { start = 2, length = 2, before = "o",  after = "e"  } )
%     languages.hyphenators.traditional.registerpattern("en","qqxc9xkqq",{ start = 3, length = 4, before = "ab", after = "cd" } ) -- replacement start length
%
%     --  print("reëel",       injecthyphens(dictionaries.nl,"reëel",       2,2))
%     --  print("reeëel",      injecthyphens(dictionaries.nl,"reeëel",      2,2))
%     --  print("rooëel",      injecthyphens(dictionaries.nl,"rooëel",      2,2))
%     --  print(   "QXcXkQ",   injecthyphens(dictionaries.de,   "QXcXkQ",   2,2))
%     --  print(  "QQXcXkQQ",  injecthyphens(dictionaries.de,  "QQXcXkQQ",  2,2))
%     --  print( "QQQXcXkQQQ", injecthyphens(dictionaries.de, "QQQXcXkQQQ", 2,2))
%     --  print("QQQQXcXkQQQQ",injecthyphens(dictionaries.de,"QQQQXcXkQQQQ",2,2))
%     --
%     --  print(  "QQXcXkQQ QQXcXkQQ",  injecthyphens(dictionaries.de,  "QQXcXkQQ QQXcXkQQ",  2,2))
% \stopluacode
%
% \starttext
%
% \blank
%
% xreëel rooëel \par xxabxx xxxabxxx \par
%
% \hsize1mm \lefthyphenmin2 \righthyphenmin2
%
% \blank Capacity \blank capacity \blank xyabxy \blank xreëel \blank rooëel \blank
%
% xy\discretionary{CD}{EF}{ab}xy % xxacceedxxx
%
% \stoptext