[tex-hyphen] Special languages

[Mojca Miklavec]

On Thu, Apr 29, 2010 at 20:20, Stephan Hennig wrote:
> Am 29.04.2010 16:15, schrieb Manuel Pégourié-Gonnard:
>> Le 29/04/2010 14:27, Stephan Hennig a écrit :
>>
>>> How do the timestamped
>>> patterns break your scheme for all the other languages?
>>>
>> Short version: our choice is to use only plain text files for patterns
>> loaded at
>> runtime, similar to those found at [1], kindly provided by Mojca. If you
>> want to
>> provide such files in dehyph-exptl, then we'll happily support dynamic
>> loading
>> of these patterns.
>>
>> [1] http://www.tug.org/svn/texhyphen/trunk/plain/
>
> Thanks!  Any pointer to what the files with different extensions are? Here
> is what I guess
>
>  .pat      plain patterns
>  .chr      the set of characters used in the pattern file
>  .hyp      hyphenation exceptions
>  .lic      license
>
> Is that correct?

Yes

> What of those four files should be part of a contribution, e.g., is the .chr
> file automatically generated from the .pat file or is it in the
> responsibility of the contributor?

The chr file is auto-generated from patterns themselves.

http://tug.org/svn/texhyphen/branches/luatex/hyph-utf8/source/generic/hyph-utf8/generate-plain-patterns.rb

Ruby code:
    characters_indexes =
patterns.join('').gsub(/[.0-9]/,'').unpack('U*').sort.uniq
    characters_indexes.each do |c|
        ch = [c].pack('U')
        $file_let.puts ch + Unicode.upcase(ch)
    end

But it's there just because someone else outside of TeX community
might find that data handy (somebody has requested that information).

Actually, all the files in that folder are auto-generated from
original TeX files in order to make them easier to parse.

> Where should contributed files be committed?  (In case I can manage to
> recover my account data for the svn.)

My humble request: until we resolve the issue generally/globally, I
would like to avoid having to ship 10 times 6 different versions of
German patterns in hyph-utf8 (we could create a hyph-utf8-extras or
something) ... but that's something yet to be discussed.

Mojca