Bug#402994: use new libpaper hook to track system paper size

[Frank Küster <frank@kuesterei.ch>]

Stephen Gildea <stepheng+debian-bts@gildea.com> wrote:

>>     As we found out, libpaper support didn't work in the old
>>   setup, anyway.  So we are not dropping any functionality if we just add
>>   the hook and the file in /etc/default, setting the variable to "no
>>   action".  We are just adding the possibility to have it done with
>>   libpaper.  Why would that warrant a debconf question?  Or am I missing
>>   something? 
>
> That would indeed be not dropping any functionality, but neither would
> it be fixing this bug.  If the default is "no action", then we still have
> the same problem we have now where tetex isn't configured with the system
> paper size.
>
> Tetex should work out of the box; it should not be necessary for a
> sysadmin to know to change a /etc/default file.  

teTeX does work out of the box, it produces correct paper for those who
know how to write their documents.  Only the lazy folks who never have
changed their habits since the old days need to configure the papersize
site-wide.  I never had the wish to do anything of that sort.

In my opinion, all this specifying of the papersize after document
processing is legacy.  I hope it will become completely obsolete with a
new LaTeX version (although I must admit that it might take years), and
I'm sure it is already not needed with ConTeXt.  It simply doesn't make
sense to print on something else than what you typeset for, but what you
typeset for needs to be specified in the document source.

> Asking a debconf
> question, "do you want tetex's paper size managed by libpaper", at
> least alerts sysadmins to the possible problem, and at best allows a
> "yes" answer to be pre-seeded.

Hm, I'm still unsure this setup will work.  No matter whether we set the
default to no or yes, won't that mean that the postinst script
programmatically changes the file (by calling the libpaper hook, and
libpaper will do the same), which is forbidden for conffiles?

Regards, Frank

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