Bug#403026: TeX Policy, TEXMFSYSCONFIG and TeXlive
Norbert Preining <preining@logic.at> wrote:
> On Don, 14 Dez 2006, Frank Küster wrote:
>> Consequently, the files that are now below /etc/texmf/texlive could be
>> at their "ordinary" places instead,
>>
>> /etc/texmf/texlive/dvips/config.ps -> /etc/texmf/dvips/config
>
> [...]
>
>> Maybe shared configuration files are not a bad idea, I should look up
>> the Debian Policy about that.
>
> Sound bad:
> ---
> 10.7.4. Sharing configuration files
> -----------------------------------
>
[...]
> If two or more packages use the same configuration file and it is
> reasonable for both to be installed at the same time, one of these
> packages must be defined as _owner_ of the configuration file, i.e.,
> it will be the package which handles that file as a configuration
> file. Other packages that use the configuration file must depend on
> the owning package if they require the configuration file to operate.
> If the other package will use the configuration file if present, but
> is capable of operating without it, no dependency need be declared.
I don't know which programs will not work at all with their
configuration files. Some are needed for sure (pdftexconfig.tex), other
might be mandatory (config.ps?)
> If it is desirable for two or more related packages to share a
> configuration file _and_ for all of the related packages to be able to
> modify that configuration file, then the following should be done:
This is probably not needed except in special cases. For the libpaper
thing, we might have to provide a script that does nothing but call the
libpaper hook.
> So we have three options:
> - Move all the stuff to tex-common
> - eliminate tetex and move all the stuff from /etc/texmf/texlive/ to
> /etc/texmf
(and assume we will never get miktex)
> - leave it as it is
I do not understand which particular problem with Debian Policy we have
here, it's just the usual "configuration files require thorough
consideration" thing. And I don't have energy for thorough
consideration ATM.
But we should probably not make any changes wrt libpaper before we have
sorted this out (and not for etch, anyway).
Regards, Frank
--
Dr. Frank Küster
Single Molecule Spectroscopy, Protein Folding @ Inst. f. Biochemie, Univ. Zürich
Debian Developer (teTeX/TeXLive)
Reply to: