[Trennmuster] Eigene Wortliste (1904) als Ergänzung zu pre1901

[Mojca Miklavec]

On Wed, Jun 11, 2014 at 10:21 PM, Werner LEMBERG wrote:
>
>> Weren't you the one saying that hopefully "ISO Latin 15 should do
>> for the foreseeable future" a few days ago?
>
> You either mean ISO-8859-15 or Latin-9 :-)

Yes, both of them, thank you. I'm sorry, I just wanted to write
ISO-8859-15 in a slightly shorter way and forgot that the numbers
don't match.

>> I'm not exactly sure, but there might be a valid reason to generate
>> separate patterns including ſ and separate patterns without ſ.
>> [...]
>
> As mentioned in another mail, it's straightforward to have unified
> patterns, provided a script generates the proper entries.

Despite the simplicity of generating the patterns, ſ in default
patterns might result in some unexpected minor problems. I would need
to patch the scripts that generate the 8-bit patterns, Hans would need
to patch ConTeXt etc. It's not a problem to fix hyph-utf8 to drop some
patterns for 8-bit engines and Hans wouldn't have a problem either. I
don't know what the derivative software packages do with the patterns.

There are certainly pros and contras to both approaches, either
fitting all patterns into the default file or making two separate
files (I don't know if that means 2 x 3 patterns then). I'm unable to
pick my favourite between the two.

Thank to you and Günter for explanations.

Mojca