Comment 8 for bug 28104

Message-ID: <email address hidden>
Date: Wed, 11 Jan 2006 10:31:12 +0100
From: =?iso-8859-1?q?Frank_K=FCster?= <email address hidden>
To: =?iso-8859-1?q?J=F6rg_Sommer?= <email address hidden>
Cc: <email address hidden>
Subject: Re: Bug#346326: tetex-bin: postinst fails -- dvips35.map not found

J=F6rg Sommer <email address hidden> wrote:

> Hello Frank,
>
> Frank K=FCster schrieb am Tue 10. Jan, 10:39 (+0100):
>> J=F6rg Sommer <email address hidden> wrote:
>>=20
>> > Did you ever discuss this issue in debian-devel?
>>=20
>> Yes, and it's clear that it's not a bug to fail if the local admin
>> refuses to update the configuration
>
> It's legal that the package installation fails?=20

You've been asked, you've been given the opportunity to view the diff,
and you chose. You can even fix the problem manually instead of
reporting a bug.

> Why you don't refuse to
> install, e.g. the preinstall script should check and fail or bring a
> warning

The preinst script is run before you decide about the conffiles.

> and the postinst should not run the hazardous parts and the admin
> must do it by hand (i.e. call dpkg-reconfigure tex-common or update-XY).

No, in future versions the postinst script (of tex-common) will check
for essential entries and fail with a user-friendly message if some are
missing or wrong. That's our service to users (and to us...). But it's
impossible to change the postinst script to not run the problematic
parts and *not* fail, since a succeeding postinst script tells dpkg that
the package is ready to use, which it wouldn't be.

Regards, Frank
--=20
Frank K=FCster
Inst. f. Biochemie der Univ. Z=FCrich
Debian Developer