An attempt to import the latest version of hyph-utf8 (#558)

LuaTeX:
* add support for loading multiple pattern files in LuaTeX

TeX Live:
* remove copthyph.tex from texmf: it is now part of hyph-utf8
* update language.ptx and language.def for pTeX
* update some descriptions in tlpsrc files (not finished)
* change lefthyphenmin from 2 to 1 for Afrikaans (not sure when it was set to 2)

hyph-utf8:
* load patterns for Latin and Cyrillic scripts for Serbian simultaneously
* new version of German hyphenation patterns
* support both U+0027 and U+2019 as apostrophe in Unicode engines



git-svn-id: svn://tug.org/texlive/trunk@22909 c570f23f-e606-0410-a88d-b1316a301751

1 files changed, 11 insertions, 6 deletions

diff --git a/Master/texmf-dist/source/luatex/hyph-utf8/luatex-hyphen.dtx b/Master/texmf-dist/source/luatex/hyph-utf8/luatex-hyphen.dtx
index bca95daf0c7..c5a5af933a0 100644
--- a/Master/texmf-dist/source/luatex/hyph-utf8/luatex-hyphen.dtx
+++ b/Master/texmf-dist/source/luatex/hyph-utf8/luatex-hyphen.dtx
@@ -192,9 +192,11 @@ This work is under the CC0 license.
 %     \begin{itemize}
 %       \item For most languages:
 %         \begin{quote}
-%           \verb+patterns = <string> filename for patterns+\\
-%           \verb+hyphenation = <string> filename for exceptions+\\
+%           \verb+patterns = <string> filenames for patterns+\\
+%           \verb+hyphenation = <string> filenames for exceptions+\\
 %         \end{quote}
+%         Each string contains a coma-separated list of file names (whitespace
+%         before or after the coma is not accepted).
 %         The files given by \verb+patterns+ (resp. \verb+hypenation+) must be
 %         plain text files encoded in utf8, with only patterns (resp.
 %         exceptions) and not even comments: their content will be used
@@ -217,8 +219,8 @@ This work is under the CC0 license.
 %   \item Optional fields may be added. For example:
 %     \begin{quote}
 %       \verb+loader = <string> name of the TeX loader+\\
-%       \verb+lefthyphenmin = <number> value for \letfhyphenmin+\\
-%       \verb+righthyphenmin = <number> value for \letfhyphenmin+
+%       \verb+lefthyphenmin = <number> value for \lefthyphenmin+\\
+%       \verb+righthyphenmin = <number> value for \righthyphenmin+
 %     \end{quote}
 %     Those fields are present in \texttt{language.dat.lua} as generated bu
 %     \texttt{tlmgr}, for example, but they \emph{are not} used by the present
@@ -280,6 +282,7 @@ local dbfile = kpse.find_file(dbname)
 if not dbfile then
     err("file not found: "..dbname)
 else
+    wlog('using data file: %s', dbfile)
     language_dat = dofile(dbfile)
 end
 %    \end{macrocode}
@@ -342,13 +345,15 @@ function loadlanguage(lname, id)
 %    \begin{macrocode}
     wlog(msg, '', cname, id)
     for _, item in ipairs{'patterns', 'hyphenation'} do
-        local file = ldata[item]
-        if file ~= nil and file ~= '' then
+        local filelist = ldata[item]
+        if filelist ~= nil and filelist ~= '' then
+          for _, file in ipairs(filelist:explode(',')) do
             local file = kpse.find_file(file) or err("file not found: %s", file)
             local fh = io.open(file, 'r')
             local data = fh:read('*a') or err("file not readable: %s", f)
             fh:close()
             lang[item](lang.new(id), data)
+          end
         else
             if item == 'hyphenation' then item = item..' exceptions' end
             wlog("info: no %s for this language", item)