Re: language.dat: Should we just enable all the patterns?
Atsuhito Kohda <kohda@pm.tokushima-u.ac.jp> schrieb:
> From: frank@kuesterei.ch (Frank Küster)
> Subject: language.dat: Should we just enable all the patterns?
> Date: Mon, 01 Mar 2004 12:57:29 +0100
>
>> Hi all,
>
> Hi Frank, please take it easy.
Oh, I'm taking it easy. It's just that Hilmar and I had this discussion
in de.comp.text.tex with Walter Schmidt (author of VTeX/free[1] and the
PS-font-guru in the group). He strongly argued for enabling all patterns
and couldn't imagine any reason why not to do this.
So I decided to think that over, and even asked Hilmar. He said I should
contact you to stop the upload if I intended to kick this out. But I
answered, essentially, that we are _not_ in a hurry, because for the
user the changes are quite the same if we do it before or after applying
my patch.
So I had decided to discuss this in the context of tetex-3. But now,
since there are bugs, I changed opinion and rose the question at
once. That's all.
>> What about kicking out all my nice code for handling language.dat, and
>> all debconf related language.dat stuff, and simply enabling _all_
>> patterns we ship?
>
> Well, I suspect that if we enabled all patterns then some
> fmt files could fail to be generated with TeX capacity exceeded
> error, if I remembered correctly.
Couldn't we just put the limits high enough? I'm not familiar with the
internals, but I guess on a standard, 3-year old PC this isn't a problem
at all. And if it is, for some old boxes of a not-so-fast developping
architecture, we can document this in README.Debian and the local admin
can edit language.dat and texmf.cnf.
> Perhaps ptex and/or jtex might be the case and there could be
> some more so I think it is not practical solution to enable
> all patterns.
I just installed ptex-{bin,base} and jtex-ptex on sid, and ptex
generated its format with only one language. It simply doesn't use
language.dat - whereas babel's hyphen.cfg reads it (\openin1 =
language.dat), ptex's doesn't even try to. Unfortunately, I wasn't able
to read the documentation - xdvi segfaults after some font errors, and
dvips doesn't generate output. Brr.
>> I guess it will save us bug reports, and work to be done (actually I
>> found on error in my code this morning, though it doesn't seem to be the
>> reason for the bug).
>
> At present I think candidates are;
>
> - improve your nice code (aggressive solution)
> - kick out debconf-handling and leave language.dat to user's manual
> editing (defensive solution)
with a) enabling nothing but english and nohyphenation by default
and b) by default enabling all available patterns
> Anyway please don't be so haste.
Oh, I'm not hastily. I'm just thinking loudly.
Regards, Frank
[1] free as in free beer
--
Frank Küster, Biozentrum der Univ. Basel
Abt. Biophysikalische Chemie
Reply to: