Bug#390129: Rebuilding only provided formats - will that work? (was: Bug#390129: tetex-base: Does not properly call update-* scripts)

[Frank Küster <frank@debian.org>]

Hi Ralf, hi Norbert,

Frank Küster <frank@debian.org> wrote:

> Frank Küster <frank@kuesterei.ch> wrote:
>
>> Ralf Stubner <ralf.stubner@physik.uni-erlangen.de> wrote:
>>
>>> A general point: Recently I was thinking whether it would make sense to
>>> change the behaviour of 'format providing packages' such as tetex-base
>>> or jadetex: Right now they regerenate all formats, even though this is
>>> not necessary. Why not have them regenerate only those formats they
>>> actually provide? A simple way to achieve this would be to call
>>>
>>> fmtutil-sys --all --cnffile /etc/texmf/fmt.d/foo.cnf
>>>
>>> That way these packages would be more self contained and could not get
>>> as easily broken by other packages. Of course, the 'binary providing
>>> packages' such as tetex-bin would still have to regenerate all formats.
>>
>> I think that's a very good idea.
[...]
> Should we just go on and let tetex-base's and tetex-extra's postinst
> call 
>
> create_tetex_formats --all --cnffile /etc/texmf/fmt.d/01tetex.cnf
>
> I think that would help, too, at least with jadetex and friends.

There's an other potential problem here.  If a new tetex-base contains a
change that requires the LaTeX formats to be regenerated, it might well
be that other formats that preload LaTeX during format generation also
need to be rebuilt.  The same then is also true for a TeXlive based
system. 

But somehow this seems to be a more general problem.  Even a format that
does not preload LaTeX might become unusable, and it might even require
configuration file changes before it works again.  Just imagine that it
turns out that some hyphenation patterns are non-free and need to be
removed.  A package that keeps its own language.dat won't work any
more.  

So should we simply ignore such potential problems, and go on with using
--cnffile during format renewal?  Or rather try to recreate all of them?
We should also consider what the effect would have been for jadetex or
xmltex users on the previous frequent bugs if we had already only
regenerated our own formats.  Hm, I guess they mostly were related to
changes in tetex-bin, weren't they?

Regards, Frank
-- 
Frank Küster
Single Molecule Spectroscopy, Protein Folding @ Inst. f. Biochemie, Univ. Zürich
Debian Developer (teTeX/TeXLive)