passivetex: destroys local configuration

Automatically imported from Debian bug report #337436 http://bugs.debian.org/337436

Message-ID: <email address hidden>
Date: Fri, 04 Nov 2005 13:02:59 +0100
From: =?iso-8859-1?q?Frank_K=FCster?= <email address hidden>
To: Debian Bug Tracking System <email address hidden>
Subject: passivetex: destroys local configuration

Package: passivetex
Severity: serious
Version: 1.25-2

The package has two serious bugs, and some normal ones:

1. The package has in its postinst the line:

/usr/bin/fmtutil --cnffile /usr/share/tetex-bin/fmtutil.cnf --all > /dev/nu=
ll 2>&1

   Beside the fact that it should use fmtutil-sys as soon as teTeX-3.0 is
   in testing, and that it should not prepend a path to the binary, it has
   a major flaw:
=20=20=20
   It statically uses a template configuration file in
   /usr/share/tetex-bin instead of the actual configuration file
   (/var/lib/texmf/web2c/fmtutil.cnf, a file generated from files in
   /etc/texmf/fmt.d). This does not make any sense at all, and it
   creates formats according to the tetex-bin package default instead of
   local customization.
=20=20=20
   Furthermore, since passivetex is not listed anywhere in
   /usr/share/tetex-bin/fmtutil.cnf, the line only recreates formats that
   are already available anyway (created by tetex-bin's postinst script).

2. The postinst snippet below has the effect that if the local admin
   deletes the configuration file /etc/texmf/texmf.d/96passivetex.cnf,
   it is resurrected from the template file, which is a violation of
   Policy 10.7.3:

        if [ -f $conffile ] ; then
           :
        elif [ -f $conffile.disable ]; then
            mv $conffile.disable $conffile
        else
            cp $TMPLDIR/`basename $conffile` $conffile
        fi

   Furthermore, there is no Policy or tradition that a configuration
   file with suffix disable is treated specially. Therefore it might
   well be that local admin renamed the configuration file manually to
   $conffile.disable, and this would also be destroyed by this snippet.

   As a sidenote, it seems as if this configuration file has not been
   updated for years - the main_memory setting for passivetex is obviously
   supposed to increase the available memory size for initex, but in fact
   it is only half the size as the setting for all other programs.

3. In the prerm script, the configuration file discussed above is moved
   to $conffile.disable. This is not necessary, since texmf variable
   settings for programs that won't be run don't do any harm.
   Furthermore, after this is done, the program update-fmtutil is
   called, but for the change to have an effect update-texmf would be
   needed. This is only of normal severity, but when it's fixed the
   postrm script has to be adapted.

I don't know how passivetex is actually used, but the Makefile in the
examples directory calls xsltproc from the package of the same name,
without Depending on it, or listing it in Suggests/Recommends.

Regards, Frank

-- System Information:
Debian Release: 3.1
Architecture: i386 (i686)
Kernel: Linux 2.6.8-2-386
Locale: LANG=3Dde_DE@euro, LC_CTYPE=3Dde_DE@euro (charmap=3DISO-8859-15)

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

Message-ID: <email address hidden>
Date: Thu, 22 Dec 2005 17:55:38 +0100
From: =?iso-8859-1?q?Frank_K=FCster?= <email address hidden>
To: Debian Bug Control Server <email address hidden>
Subject: Splitting this bug, separate issues

clone 337436 -1
retitle -1 Remove unnecessary, buggy fmtutil code
user <email address hidden>
usertags -1 + teTeX-3.0-misc
thanks

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

Message-ID: <email address hidden>
Date: Thu, 12 Jan 2006 18:41:47 +0100
From: =?iso-8859-1?q?Frank_K=FCster?= <email address hidden>
To: <email address hidden>
Cc: <email address hidden>, teTeX maintainers <email address hidden>
Subject: please remove passivetex from testing (was: passivetex: destroys
 local configuration)

Frank K=FCster <email address hidden> wrote:

> The package has two serious bugs, and some normal ones:

I was about to create a patch for the package, but I discovered more
issues. Most notably, the Makefile in the example directory fails
because it cannot find a file "tei.xsl", which does not exist in the
source package or anywhere in Debian (main). Thus it seems the package
is not even functional.=20=20

Given that

- The package has 2 RC bugs without maintainer reaction, one of them for
  more than 2 months

- and one normal without maintainer reaction (after it has been assigned
  to passivetex), and no action at all for 2 years

- passivetex seems to have functionality problems (see above, and the
  normal bug's log mentions that xmlto wants to switch away from
  passivetex)

- No package Depends on it, and only one (xmlto, see above) Suggests it,
  and one Recommends it (xmltex) it,

I suggest to remove the package from testing.

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

Message-ID: <email address hidden>
Date: Thu, 12 Jan 2006 13:19:01 -0800
From: Steve Langasek <email address hidden>
To: Frank =?iso-8859-1?Q?K=FCster?= <email address hidden>
Cc: <email address hidden>, <email address hidden>,
 teTeX maintainers <email address hidden>
Subject: Re: please remove passivetex from testing (was: passivetex: destroys local configuration)

--5mCyUwZo2JvN/JJP
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Thu, Jan 12, 2006 at 06:41:47PM +0100, Frank K=FCster wrote:
> Frank K=FCster <email address hidden> wrote:

> > The package has two serious bugs, and some normal ones:

> I was about to create a patch for the package, but I discovered more
> issues. Most notably, the Makefile in the example directory fails
> because it cannot find a file "tei.xsl", which does not exist in the
> source package or anywhere in Debian (main). Thus it seems the package
> is not even functional. =20

> Given that

> - The package has 2 RC bugs without maintainer reaction, one of them for
> more than 2 months

> - and one normal without maintainer reaction (after it has been assigned
> to passivetex), and no action at all for 2 years

> - passivetex seems to have functionality problems (see above, and the
> normal bug's log mentions that xmlto wants to switch away from
> passivetex)

> - No package Depends on it, and only one (xmlto, see above) Suggests it,
> and one Recommends it (xmltex) it,

However, gstreamer0.8 has a build-dependency on it, and I think we might
miss that. Do you have a solution for gstreamer0.8?

Thanks,
--=20
Steve Langasek Give me a lever long enough and a Free OS
Debian Developer to set it on, and I can move the world.
<email address hidden> http://www.debian.org/

--5mCyUwZo2JvN/JJP
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Digital signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQFDxsfEKN6ufymYLloRAtC5AJ0S/Ie5GD23EiAmwgEDMg7wmZvMTgCdHTRf
/O1Olmipci2gFdQtbTQ0GOY=
=YCjC
-----END PGP SIGNATURE-----

--5mCyUwZo2JvN/JJP--

Message-ID: <email address hidden>
Date: Fri, 13 Jan 2006 09:51:35 +0100
From: Norbert Preining <email address hidden>
To: Frank =?iso-8859-15?Q?K=FCster?= <email address hidden>,
 <email address hidden>, <email address hidden>,
 teTeX maintainers <email address hidden>
Subject: Re: please remove passivetex from testing (was: passivetex: destroys local configuration)

Hi Frank, hi Steve!

On Don, 12 Jan 2006, Frank K�wrote:
> source package or anywhere in Debian (main). Thus it seems the package
> is not even functional.

On Don, 12 Jan 2006, Steve Langasek wrote:
> However, gstreamer0.8 has a build-dependency on it, and I think we might
> miss that. Do you have a solution for gstreamer0.8?

I could do the following: Take the sources from texlive (where
passivetex was excluded due to its existence in debian) and try to fix
the passivetex packages. It shouldn't be too complicated, but probably
would be a complete rewrite of the scripts/rules.

Would this be ok for NMU?

Best wishes

Norbert

-------------------------------------------------------------------------------
Dr. Norbert Preining <preining AT logic DOT at> Universit�i Siena
gpg DSA: 0x09C5B094 fp: 14DF 2E6C 0307 BE6D AD76 A9C0 D2BF 4AA3 09C5 B094
-------------------------------------------------------------------------------
DITHERINGTON (n)
Sudden access to panic experienced by one who realises that he is
being drawn inexorably into a clabby (q.v.) conversion, i.e. one he
has no hope of enjoying, benefiting from or understanding.
   --- Douglas Adams, The Meaning of Liff

Message-ID: <email address hidden>
Date: Fri, 13 Jan 2006 10:34:43 +0100
From: =?iso-8859-1?q?Frank_K=FCster?= <email address hidden>
To: <email address hidden>
Cc: <email address hidden>, teTeX maintainers <email address hidden>
Subject: Re: please remove passivetex from testing

Steve Langasek <email address hidden> wrote:

> However, gstreamer0.8 has a build-dependency on it, and I think we might
> miss that. Do you have a solution for gstreamer0.8?

I'll look into it during the weekend.

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

Message-ID: <email address hidden>
Date: Fri, 13 Jan 2006 11:03:38 +0100
From: =?iso-8859-1?q?Frank_K=FCster?= <email address hidden>
To: Norbert Preining <email address hidden>
Cc: <email address hidden>, <email address hidden>,
 teTeX maintainers <email address hidden>
Subject: Re: please remove passivetex from testing

Norbert Preining <email address hidden> wrote:

> On Don, 12 Jan 2006, Steve Langasek wrote:
>> However, gstreamer0.8 has a build-dependency on it, and I think we might
>> miss that. Do you have a solution for gstreamer0.8?
>
> I could do the following: Take the sources from texlive (where
> passivetex was excluded due to its existence in debian) and try to fix
> the passivetex packages. It shouldn't be too complicated, but probably
> would be a complete rewrite of the scripts/rules.
>
> Would this be ok for NMU?

It would also be a new upstream release, I guess (at least current
upstream has a "tests" subdirectory, while the Debian package has
"examples" instead). I don't think this would be good for an NMU,
rather for a takeover. Which I wouldn't mind, if you want to take the
task.=20

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

Message-ID: <20060113110056.GA1432@preusse>
Date: Fri, 13 Jan 2006 12:00:56 +0100
From: Hilmar Preusse <email address hidden>
To: <email address hidden>,
 Debian-teTeX-maint <email address hidden>
Cc: Steve Langasek <email address hidden>
Subject: Re: please remove passivetex from testing (was: passivetex: destroys local configuration)

On 12.01.06 Steve Langasek (<email address hidden>) wrote:
> On Thu, Jan 12, 2006 at 06:41:47PM +0100, Frank K�wrote:

Hi *,

> > I was about to create a patch for the package, but I discovered
> > more issues. Most notably, the Makefile in the example directory
> > fails because it cannot find a file "tei.xsl", which does not
> > exist in the source package or anywhere in Debian (main). Thus
> > it seems the package is not even functional.
>
> However, gstreamer0.8 has a build-dependency on it, and I think we
> might miss that. Do you have a solution for gstreamer0.8?
>
I could build the docs of gstreamer0.8 without having passivetex
installed. I could not test the full build as it gave an FTBFS at
another place, which was definitely not caused by missing passivetex.
As Frank stated passivetex is not functional the generated doccs
should not look different than without passivetex.

H.
--
When your memory goes, forget it!
  http://www.hilmar-preusse.de.vu/

Message-ID: <email address hidden>
Date: Fri, 13 Jan 2006 12:04:34 +0100
From: Norbert Preining <email address hidden>
To: Frank =?iso-8859-15?Q?K=FCster?= <email address hidden>
Cc: <email address hidden>, <email address hidden>,
 teTeX maintainers <email address hidden>
Subject: Re: please remove passivetex from testing

On Fre, 13 Jan 2006, Frank K�wrote:
> > I could do the following: Take the sources from texlive (where
> > passivetex was excluded due to its existence in debian) and try to fix
> > the passivetex packages. It shouldn't be too complicated, but probably
> > would be a complete rewrite of the scripts/rules.
> >
> > Would this be ok for NMU?
>
> It would also be a new upstream release, I guess (at least current
> upstream has a "tests" subdirectory, while the Debian package has

What I meant is take the debian stuff as I would have used it for
passivetex, but leave the .orig.tar.gz.

> "examples" instead). I don't think this would be good for an NMU,
> rather for a takeover. Which I wouldn't mind, if you want to take the

Maybe I prepare a NMU over the weekend of next week, it will be a severe
restructuring of the debian part, but nothing else.

If someone else does it in the meantime, also good.

Best wishes

Norbert

-------------------------------------------------------------------------------
Dr. Norbert Preining <preining AT logic DOT at> Universit�i Siena
gpg DSA: 0x09C5B094 fp: 14DF 2E6C 0307 BE6D AD76 A9C0 D2BF 4AA3 09C5 B094
-------------------------------------------------------------------------------
YORK (vb.)
To shift the position of the shoulder straps on a heavy bag or
rucksack in a vain attempt to make it seem lighter. Hence : to laugh
falsely and heartily at an unfunny remark. 'Jasmine yorked politely,
loathing him to the depths of her being' - Virginia Woolf.
   --- Douglas Adams, The Meaning of Liff

Message-ID: <email address hidden>
Date: Fri, 13 Jan 2006 03:17:23 -0800
From: Steve Langasek <email address hidden>
To: Norbert Preining <email address hidden>, <email address hidden>,
 Debian-teTeX-maint <email address hidden>
Cc: Frank =?iso-8859-1?Q?K=FCster?= <email address hidden>,
 <email address hidden>
Subject: Re: please remove passivetex from testing (was: passivetex: destroys local configuration)

--k+w/mQv8wyuph6w0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Fri, Jan 13, 2006 at 09:51:35AM +0100, Norbert Preining wrote:
> Hi Frank, hi Steve!

> On Don, 12 Jan 2006, Frank K=FCster wrote:
> > source package or anywhere in Debian (main). Thus it seems the package
> > is not even functional. =20

> On Don, 12 Jan 2006, Steve Langasek wrote:
> > However, gstreamer0.8 has a build-dependency on it, and I think we might
> > miss that. Do you have a solution for gstreamer0.8?

> I could do the following: Take the sources from texlive (where
> passivetex was excluded due to its existence in debian) and try to fix
> the passivetex packages. It shouldn't be too complicated, but probably
> would be a complete rewrite of the scripts/rules.

> Would this be ok for NMU?

Not really; it sounds like a hijacking to me. Of course, the maintainer
field lists a mailing list, and there are no uploaders listed, so I'm not
sure anybody *cares*, but it's still not the sort of change to make unless
you're prepared to maintain the package long-term. Which also probably
means going through the orphaning process first.

On Fri, Jan 13, 2006 at 12:00:56PM +0100, Hilmar Preusse wrote:
> On 12.01.06 Steve Langasek (<email address hidden>) wrote:
> > On Thu, Jan 12, 2006 at 06:41:47PM +0100, Frank K=FCster wrote:

> Hi *,

> > > I was about to create a patch for the package, but I discovered
> > > more issues. Most notably, the Makefile in the example directory
> > > fails because it cannot find a file "tei.xsl", which does not
> > > exist in the source package or anywhere in Debian (main). Thus
> > > it seems the package is not even functional.

> > However, gstreamer0.8 has a build-dependency on it, and I think we
> > might miss that. Do you have a solution for gstreamer0.8?

> I could build the docs of gstreamer0.8 without having passivetex
> installed. I could not test the full build as it gave an FTBFS at
> another place, which was definitely not caused by missing passivetex.
> As Frank stated passivetex is not functional the generated doccs
> should not look different than without passivetex.

Then if Norbert opts not to adopt passivetex, it sounds like an RC bug
against gstreamer0.8 asking it not to depend on passivetex would be the
answer?

--=20
Steve Langasek Give me a lever long enough and a Free OS
Debian Developer to set it on, and I can move the world.
<email address hidden> http://www.debian.org/

--k+w/mQv8wyuph6w0
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Digital signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE---...

Message-ID: <email address hidden>
Date: Fri, 13 Jan 2006 12:25:41 +0100
From: =?iso-8859-1?q?Frank_K=FCster?= <email address hidden>
To: Steve Langasek <email address hidden>
Cc: <email address hidden>, Norbert Preining <email address hidden>,
 teTeX maintainers <email address hidden>, <email address hidden>
Subject: Re: please remove passivetex from testing

Steve Langasek <email address hidden> wrote:

> However, gstreamer0.8 has a build-dependency on it, and I think we might
> miss that. Do you have a solution for gstreamer0.8?

Yes: I tried to find out which problems would occur, and it turned out
that the build-dependency can simply be dropped: #347884.

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

Message-ID: <email address hidden>
Date: Fri, 13 Jan 2006 12:26:47 +0100
From: =?iso-8859-1?q?Frank_K=FCster?= <email address hidden>
To: <email address hidden>
Cc: Debian-teTeX-maint <email address hidden>, Steve Langasek <email address hidden>
Subject: Re: please remove passivetex from testing

Hilmar Preusse <email address hidden> wrote:

> I could build the docs of gstreamer0.8 without having passivetex
> installed. I could not test the full build as it gave an FTBFS at
> another place, which was definitely not caused by missing passivetex.

I didn't get a FTBFS in a sid chroot (i386).

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

Message-ID: <email address hidden>
Date: Fri, 13 Jan 2006 12:29:41 +0100
From: =?iso-8859-1?q?Frank_K=FCster?= <email address hidden>
To: <email address hidden>
Cc: <email address hidden>, Norbert Preining <email address hidden>,
 Debian-teTeX-maint <email address hidden>, <email address hidden>
Subject: Re: please remove passivetex from testing

severity 347884 serious
# passivetex will be removed from testing
thanks

Steve Langasek <email address hidden> wrote:

>> I could build the docs of gstreamer0.8 without having passivetex
>> installed. I could not test the full build as it gave an FTBFS at
>> another place, which was definitely not caused by missing passivetex.
>> As Frank stated passivetex is not functional the generated doccs
>> should not look different than without passivetex.
>
> Then if Norbert opts not to adopt passivetex, it sounds like an RC bug
> against gstreamer0.8 asking it not to depend on passivetex would be the
> answer?

Adjusted severity.

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

Confirming, since debian discussion shows that the problem is real.

Package has been removed from Debian. If a MOTU would please confirm the removal request I filed as bug 82342, it would be great. Thanks.

Closing this bug as passivetex has been removed from Ubuntu (bug 82342).