Re: [Pkg-ruby-extras-maintainers] Orphaning my packages

[Paul van Tilburg]

Hello David,

On Tue, Sep 12, 2006 at 04:21:18PM -0500, David Moreno Garza wrote:
> I'll be taking a long vacation of Debian and free software activity  
> for the next couple of months for personal reasons. [...]

I hope that the vacation will do you good and hope you'll come back to
us (if it feels right of course).

> If you think you could adopt some of the packages, feel free to do it  
> (filling an ITA would be nice). Just talk to the Debian Perl Group  
> first, if thinking on adopting some of the Perl modules; talk first  
> to the pkg-ruby-extras groups if thinking on adopting some of the  
> ruby modules, also.

I think the Debian/Ruby Extras team is interested in taking all
Ruby-related packages:  libform-validator-ruby, libgtkmozembed-ruby,
libgtk-trayicon-ruby, libsnmp-ruby, libtermios-ruby and revolution. 
This is mainly because I am already familiar with these packages and we,
as a team, can share the load.  
David, if I am overlooking a Ruby package, please let me know?

If someone is interested in maintaining these packages and doing this
within our team, feel free to contact us!

Regards,

Paul

-- 
Student @ Eindhoven | email: [EMAIL PROTECTED]
University of Technology, The Netherlands   | JID: [EMAIL PROTECTED]
>>> Using the Power of Debian GNU/Linux <<< | GnuPG key ID: 0x50064181


signature.asc
Description: Digital signature

Re: Orphaning my packages

[Francesco Pedrini]

On Wednesday 13 September 2006 08:58, Francesco Pedrini wrote:
>
> Best whishes for your life! :)
>
> I'll take care of hfsplus and tdfsb if anyone has already taken them
> :)

I can take care even of wyg :)

Thanks again.

Francesco
-- 
:wq


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

stale lock files

[Brian May]

Hello,

I have noticed GUI programs tend to be very inconsistent in behaviour
if I accidently kill them (e.g. by shutting down the X server) or
abort them (e.g. power failure).

For example:

gnucash (not tested newest version):

* displays a message box saying the file is locked, and asks if you
want to continue anyway.

* liferea: displays an error and terminates; the lock file must be
manually deleted.

* mozilla - occasionally will display a message that the profile is in
use, and won't let you reuse it until you delete the lock file. I am
not sure what triggers this condition.


In this era of making computers easier to use, forcing the user to
worry about cleaning up lock files like this seems very clumsy at
best, and can also be unreliable (e.g. I sometimes already do have a
copy of gnucash running, but I can't find it in the maze of windows,
and delete the lock file :-( ).


Is there any guidelines for checking for stale lock files in a sane
manner that doesn't involve the user?

Yes, I realize having a home directory on NFS or AFS or SMB might be
complications, but most users of these applications don't use these
protocols either.

Thanks.
-- 
Brian May <[EMAIL PROTECTED]>


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: transitioning config files between two packages

[sean finney]

hey james,

On Tue, 2006-09-12 at 22:27 -0400, James Vega wrote:
> It should just be a matter of removing the files from the old package
> and letting the new ones take their place (with a backup if there are
> any user changes).  A little grepping around in /var/lib/dpkg/info
> turned up this snippet for removing conffiles.

thanks for the suggestion, but if it's a choice between this and ucf, i
think ucf would be the lesser of two evils.  

as a general rule relying on /var/lib/dpkg/anything should be greatly
frowned upon.  what you suggest *would* work (though in the case that
the file is modified it could be kept there instead of made into a
backup copy), i can't in good conscience implement something like that.

i suppose i could store the md5sums of the conffiles in a seperate file
that is guaranteed to exist, but then i'm on my way to reimplementing
ucf and might as well just use ucf then.


thanks,
sean 



signature.asc
Description: This is a digitally signed message part

Re: transitioning config files between two packages

[sean finney]

hey steve,

On Tue, 2006-09-12 at 17:21 -0700, Steve Langasek wrote:
> I don't know what happens if there is a versioned Conflicts/Replaces
> instead, and the conflicted-with package remains installed in a newer
> version as a result.  It's certainly possible that doesn't work nearly as
> smoothly, a conclusion supported by your original post.

well if anyone is interested in reproducing this:

- use debootstrap/pbuilder to create a sarge.tgz
- chroot into the sarge.tgz, mounting /proc.
- apt-get install nagios-plugins
- change the sarge lines to etch lines in sources.list
- apt-get update
- apt-get dist-upgrade
- wait for the prompt

if anyone here has some dpkg-fu handy off the top of their heads that i
could use to further deduce what's going on i'd be happy to hear it.

> I do know also that beginning with the dpkg version in etch, the Conflicts:
> is no longer required when moving conffiles, it's possible to use Replaces:
> by itself.

okay.  the conflict is there for other reasons as well but i'll keep
that in mind for other packages.

> After an upgrade and answering all of the conffile prompts, does
> /var/lib/dpkg/info/nagios-plugins.conffiles still exist and reference these
> files?  Depending on what dpkg is really doing here, it may well be possible
> to handle the conffile transfer in maintainer scripts.  (And I thought
> dpkg.org once had recipes for exactly this, but unfortunately the site has
> been down for some time now. :/)

my guess is that they no longer exist in there (and are instead in
nagios-plugins-foo.conffilels), but i'll double check and report back on
that.

thanks,
sean


signature.asc
Description: This is a digitally signed message part

Re: stale lock files

[Frederic Peters]

Brian May wrote:

> * liferea: displays an error and terminates; the lock file must be
> manually deleted.

For the record, Lars Lindner (author of liferea) recently wrote:

  But for v1.1 I rewrote the code to use libbacon (which GUniqueApp also
  uses, there was a thread on this list some weeks ago). So the locking
  dialog is gone with v1.1.x.

So that may be an answer.


Regards,
Frederic


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: transitioning config files between two packages

[Steve Langasek]

On Wed, Sep 13, 2006 at 10:49:16AM +0200, sean finney wrote:

> On Tue, 2006-09-12 at 17:21 -0700, Steve Langasek wrote:
> > I don't know what happens if there is a versioned Conflicts/Replaces
> > instead, and the conflicted-with package remains installed in a newer
> > version as a result.  It's certainly possible that doesn't work nearly as
> > smoothly, a conclusion supported by your original post.

> well if anyone is interested in reproducing this:

> - use debootstrap/pbuilder to create a sarge.tgz
> - chroot into the sarge.tgz, mounting /proc.
> - apt-get install nagios-plugins
> - change the sarge lines to etch lines in sources.list
> - apt-get update
> - apt-get dist-upgrade
> - wait for the prompt

Done, and:

$ ls -l /var/lib/dpkg/info/nagios-plugins.conffiles
ls: /var/lib/dpkg/info/nagios-plugins.conffiles: No such file or directory
$

So everything works here except that dpkg doesn't recognize the files on
disk as unmodified conffiles after changing their ownership, I think.

So to fix this within your preinst, you could check whether each file's
md5sum matches the known md5sum from sarge, and if so remove the file.  If
the md5sum /doesn't/ match, the conffile prompt should happen as normal.

It's up to you whether you think ucf is a better way to handle this.

Cheers,
-- 
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 PROTECTED]   http://www.debian.org/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: transitioning config files between two packages

[Frank Küster]

James Vega <[EMAIL PROTECTED]> wrote:

> It should just be a matter of removing the files from the old package
> and letting the new ones take their place (with a backup if there are
> any user changes).  A little grepping around in /var/lib/dpkg/info
> turned up this snippet for removing conffiles.
>
> nagios-plugins.preinst:
>
> rm_conffile() {
> CONFFILE="$1"
>
> if [ -e "$CONFFILE" ]; then
> md5sum="`md5sum \"$CONFFILE\" | sed -e \"s/ .*//\"`"
> old_md5sum="`sed -n -e \"/^Conffiles:/,/^[^ ]/{' $CONFFILE'{s/.* 
> //;p}}\" /var/lib/dpkg/status`"
> if [ "$md5sum" != "$old_md5sum" ]; then
> echo "conffile $CONFFILE has been modified by you."
> echo "Saving as $CONFFILE.dpkg-bak ..."
> mv -f "$CONFFILE" "$CONFFILE.dpkg-bak"

This violates the spirit of Policy 10.7.3:

"local changes must be preserved during a package upgrade, and" 

and, I think, the word of the next sentence:

"configuration files must be preserved when the package is removed, and
only deleted when the package is purged."

Regards, Frank

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

Re: transitioning config files between two packages

[Frank Küster]

Steve Langasek <[EMAIL PROTECTED]> wrote:

> On Tue, Sep 12, 2006 at 01:28:34PM +0200, Frank Küster wrote:
>> sean finney <[EMAIL PROTECTED]> wrote:
>
>> > so the question is: what am i forgetting to do?  i'm guessing that the
>> > problem has something to do with the original package still being
>> > present (as a metapackage)?
>
>> No, it's a general problem: dpkg won't notice that a conffile has been
>> moved from one package to the other, no matter whether it declares
>> "Replaces" or whatever.  There's simply no solution within dpkg at the
>> moment.
>
> Where do you get this?  Conflicts:/Replaces: has been used quite
> successfully to transfer ownership of conffiles for, e.g., the Xorg
> packages, without spurious prompts.

Hm, I did get this from the fact that I got lots of spurious prompts,
both with my own and with others' packages.  However, in all cases I
remember there was either only a versioned Conflicts, or no Conflicts at
all.

I also don't see anything in the policy that indicates that Conflicts
has an effect on things that should be covered by Replaces; and I think
that it shouldn't.  If dpkg already has the means to cleanly take over
conffiles from an other package, why not just do this when the
taking-over package declares Replaces?

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

Re: transitioning config files between two packages

[Frank Küster]

sean finney <[EMAIL PROTECTED]> wrote:

> if anyone here has some dpkg-fu handy off the top of their heads that i
> could use to further deduce what's going on i'd be happy to hear it.

DPKg
{
options "--debug=221"
}

in a file in /etc/apt/conf.d/ should do (this is untested, please check
the details).

>> I do know also that beginning with the dpkg version in etch, the Conflicts:
>> is no longer required when moving conffiles, it's possible to use Replaces:
>> by itself.
>
> okay.  the conflict is there for other reasons as well but i'll keep
> that in mind for other packages.

But then you should make sure that dpkg is updated first:

apt-get update
apt-get install dpkg apt #and aptitude, if you like
apt-get dist-upgrade # or aptitude

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

Re: transitioning config files between two packages

[Frank Küster]

Steve Langasek <[EMAIL PROTECTED]> wrote:

> So to fix this within your preinst, you could check whether each file's
> md5sum matches the known md5sum from sarge, and if so remove the file.  If
> the md5sum /doesn't/ match, the conffile prompt should happen as normal.

The conffile present might also be yet a different one, from a version
earlier in etch's release cycle.  At least if the package splitting has
just been done in the last version in unstable, it's likely that there
are machines around with such versions installed.  Of course it's much
more important to not give dpkg prompts on dist-upgrades from stable to
new stable, but IMHO it should also be avoided during a development
cycle:  Many people use sid or testing for their workstations, although
they don't know much about internals and what triggers bogus prompts.

Ah, and I forgot:  Even oldstable deserves its md5sums to be recorded.
This has bitten us with tetex-extra:  People had tetex-extra installed
when the machine was woody, then removed it but did not purge - so the
conffiles were still there.  Only after switching to etch did they
reinstall the package, and...  The result can be "admired" in
/var/lib/dpkg/info/tetex-extra.preinst, look for ".*_md5sum_list".  We
didn't use ucf because these files all disappeared.

> It's up to you whether you think ucf is a better way to handle this.

At least it is able to record a set of md5sums for a given file.

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

Re: my CDBS gallery: real-world rules samples

[Stefano Zacchiroli]

On Tue, Sep 12, 2006 at 07:19:31PM +0200, Arnaud Fontaine wrote:
> Caio> Hmm maybe  I didn't get it, Arnaud. What  you meant?  Also, if
> Caio> you has found some blank  file in the gallery... please let me
> Caio> know.
> Indeed, often,  when you  are written a  debian/rules with cdbs  for the
> first time,  it isn't usually clean. I  think that it would  be great to
> remove the one that aren't really clean. It is just an idea though, tell
> me what you think about that...

I still don't get it, could you expand your concept of "clean"?

-- 
Stefano Zacchiroli -*- Computer Science PhD student @ Uny Bologna, Italy
[EMAIL PROTECTED],debian.org,bononia.it} -%- http://www.bononia.it/zack/
If there's any real truth it's that the entire multidimensional infinity
of the Universe is almost certainly being run by a bunch of maniacs. -!-


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#387273: ITP: senna -- Senna is an embeddable fulltext search engine, which you can use in conjunction with various scripting languages and databases. Senna is an inverted index based engine, and c

[Tasuku SUENAGA]

Package: wnpp
Severity: wishlist
Owner: Tasuku SUENAGA <[EMAIL PROTECTED]>


* Package name: senna
  Version : 0.8.1
  Upstream Author : Brazil <[EMAIL PROTECTED]>
* URL : http://qwik.jp/senna/
* License : LGPL
  Programming Lang: C
  Description : Senna is an embeddable fulltext search engine, which you 
can use in conjunction with various scripting languages and databases. Senna is 
an inverted index based engine, and combines the best of n-gram indexing and 
word indexing to achieve fast, precise searches. While senna codebase is rather 
compact it is scalable enough to handle large amounts of data and queries.

-- System Information:
Debian Release: testing/unstable
Architecture: i386 (i686)
Kernel: Linux 2.6.16-2-686-smp
Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968)


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Why not only support Sid and Testing?

[Thibaut Paumard]

Le mardi 12 septembre 2006 à 07:23 -0600, Joseph Smidt a écrit :
> At the risk of repeating myself, Desktop Users: For most non-server
> minded people who still want a stable OS, Debian takes too long to
> release.  On top of that,many of them want a "As easy aw Windows"
> distro. I don't know that Debian will provide that as well as Ubuntu,
> unless we want to alter many aspects of the project, like how often we
> release.

And I don't agree, even for Desktop Users. I think you're talking
"desktop tweakers". Two examples:
 - My wife at home likes Debian, she likes the entire FOSS concept etc.
   Yet, she wants hates it when tiny little subtle changes in her
   desktop or applications get in her way and require her to learn again
   that "this entry has moved to another menu" etc. Even if it happens
   every 3 months, it just gets on her nerves. When it does, she
   wants to kill me, and since I want to live, we run stable.
 - At work: the employees, including the sysadmin, should do something
   else with their time that getting back on tracks a system that was
   running fine the day before (even if it's only every three months or
   so). Sure, occasionally I need a more recent version of a software,
   then I can get a backport or compile it from source, takes half a day
   in the tricky cases. Of course I'm running an older version of SuSE
   at work, and it's not so great for backports. I'd much rather have a
   Debian stable, updated to the new stable each time it comes out.

Sure, once in a while I install a more experimental system at home, or
play enough with my stable system to get it to fail, or buy esoteric
hardware, and I like the thrill I get and the skills I develop. I'm just
a natural puzzle solver. Still, I do that out of mission-critical and
family range.

One side-note on the stress caused by freezes: I haven't been much
involved in Debian development so far, but from my viewpoint, this phase
is tremendously useful for keeping the entire distribution in shape,
with clearly defined goals, RC bug-squashing, removing obsolete and
unmaintained packages etc.

Regards, Thibaut.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Accepted lynx 2.8.5-2sarge2 (source i386)

[Thomas Dickey]

On Thu, Aug 31, 2006 at 08:20:14AM +0200, Martin Schulze wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
> 
> Format: 1.7
> Date: Sat, 13 May 2006 07:47:40 +0200
> Source: lynx
> Binary: lynx
> Architecture: source i386
> Version: 2.8.5-2sarge2
> Distribution: stable-security
> Urgency: high
> Maintainer: Martin Schulze <[EMAIL PROTECTED]>
> Changed-By: Martin Schulze <[EMAIL PROTECTED]>
> Description: 
>  lynx   - Text-mode WWW Browser
> Changes: 
>  lynx (2.8.5-2sarge2) stable-security; urgency=high
>  .
>* Non-maintainer upload by the Security Team
>* Added OpenBSD patch to fix infinete loop rendering broken HTML
>  [debian/patches/04_CVE-2004-1617.dpatch]

Martin Schulze has been told more than once that this was not an OpenBSD
patch.

After the second time, there is no plausible excuse.

Do you have an excuse?

-- 
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net


pgpMDsLV3pPby.pgp
Description: PGP signature

Re: Accepted lynx 2.8.5-2sarge1.2 (source i386 sparc amd64)

[Thomas Dickey]

In article <[EMAIL PROTECTED]> you wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1

> Format: 1.7
> Date: Wed, 13 Sep 2006 18:41:49 +1000
> Source: lynx
> Binary: lynx
> Architecture: source i386 sparc amd64
> Version: 2.8.5-2sarge1.2
> Distribution: unstable
> Urgency: high
> Maintainer: James Troup <[EMAIL PROTECTED]>
> Changed-By: Anibal Monsalve Salazar <[EMAIL PROTECTED]>
> Description: 
>  lynx   - Text-mode WWW Browser
> Closes: 238057 384725
> Changes: 
>  lynx (2.8.5-2sarge1.2) unstable; urgency=high
>  .
>* NMU.
>* Added OpenBSD patch to fix infinete loop rendering broken HTML
>  [debian/patches/04_CVE-2004-1617.dpatch]


This is the third time this year that patch has been incorrectly
ascribed to OpenBSD.

-- 
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Accepted lynx 2.8.5-2sarge2 (source i386)

[Thomas Viehmann]

Thomas Dickey wrote:
>> Date: Sat, 13 May 2006 07:47:40 +0200
[...]
> After the second time, there is no plausible excuse.
> Do you have an excuse?

Why do you ask if you know there isn't?
Hint: You could always look at the date of the actual update.

Maybe you just file a minor bug, that would help people noticing and
correcting the error.

Kind regards

T.
-- 
Thomas Viehmann, http://thomas.viehmann.net/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Accepted lynx 2.8.5-2sarge2 (source i386)

[Thomas Dickey]

On Wed, Sep 13, 2006 at 02:46:56PM +0200, Thomas Viehmann wrote:
> Thomas Dickey wrote:
> >> Date: Sat, 13 May 2006 07:47:40 +0200
> [...]
> > After the second time, there is no plausible excuse.
> > Do you have an excuse?
> 
> Why do you ask if you know there isn't?

Because Martin's actions are malicious.  He was told about this when _he_
put out an "OpenBSD" patch fix, then shortly afterwards wrote two more
announcements.

> Hint: You could always look at the date of the actual update.
> 
> Maybe you just file a minor bug, that would help people noticing and
> correcting the error.

It's already in the changelog.

-- 
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net


pgpj7TRm9xmjM.pgp
Description: PGP signature

Bug#72140: is it possible Margarette

[Mr. Sabine]

Want to see something shocking?

Everyone knows it's embarressing for people to talk with their physician about 
their or their partners private "issues". Well, you
don't need to be embarressed. We can offer the pill right to your door and at 
about one third of the cost.

We have hundreds of thousands of happy couples who changed their marriages. If 
you want to take a look, please check
up the site below. You'll be quite impressed:

http://mi.com>.qbqun5vbytwsus0baf5nap8ngannn.ysmicgor.cc







animal stiff


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Accepted lynx 2.8.5-2sarge2 (source i386)

[Thomas Dickey]

Thomas Viehmann <[EMAIL PROTECTED]> wrote:
> Thomas Dickey wrote:
>>> Date: Sat, 13 May 2006 07:47:40 +0200
> [...]
>> After the second time, there is no plausible excuse.
>> Do you have an excuse?

> Why do you ask if you know there isn't?
> Hint: You could always look at the date of the actual update.

You mean that re-issuing an incorrect announcement allows you to use
the same excuse(*) as before?

(*) an excuse which would go along the lines of "I found this patch on
someone else's ftp area, so I'll ignore your comments and changlog and
assume they wrote the code since I want to say nice things about them"

-- 
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Accepted lynx 2.8.5-2sarge2 (source i386)

[Otavio Salvador]

Thomas Dickey <[EMAIL PROTECTED]> writes:

> On Wed, Sep 13, 2006 at 02:46:56PM +0200, Thomas Viehmann wrote:
>> Thomas Dickey wrote:
>> >> Date: Sat, 13 May 2006 07:47:40 +0200
>> [...]
>> > After the second time, there is no plausible excuse.
>> > Do you have an excuse?
>> 
>> Why do you ask if you know there isn't?
>
> Because Martin's actions are malicious.  He was told about this when _he_
> put out an "OpenBSD" patch fix, then shortly afterwards wrote two more
> announcements.
>
>> Hint: You could always look at the date of the actual update.
>> 
>> Maybe you just file a minor bug, that would help people noticing and
>> correcting the error.
>
> It's already in the changelog.

He can fix a previous entry and cite it in next version.

-- 
O T A V I OS A L V A D O R
-
 E-mail: [EMAIL PROTECTED]  UIN: 5906116
 GNU/Linux User: 239058 GPG ID: 49A5F855
 Home Page: http://www.freedom.ind.br/otavio
-
"Microsoft gives you Windows ... Linux gives
 you the whole house."


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Bug#387286: Acknowledgement (postrm_hook is not run)

[Manoj Srivastava]

On Wed, 13 Sep 2006 15:11:11 +0200, Michael Biebl <[EMAIL PROTECTED]> said: 

> I have to add that I changed /etc/kernel-img.conf from
  ,[ /etc/kernel-img.conf ]
  | ...
  | postinst_hook = /sbin/update-grub
  | postrm_hook   = /sbin/update-grub
  `
> to
  ,[ /etc/kernel-img.conf ]
  | ...
  | postinst_hook = update-grub
  | postrm_hook   = update-grub
  `

> as promoted by NEWS.Debian, grub 0.97-14. If I use absolute paths
> again, the postrm hook is called, but the deinstallation fails:

The grub NEWS file is wrong. This is user error: the postinst
 of the kernel image being removed was created by a kernel image which
 was not designed for hooks which did not use the full path; so it
 fails. Nothing can be done about this.

This is hte reason I was promoting telling people to change
 from /sbin/update-grub => /usr/sbin/update-grub, so that older kernel
 images would not fall on their face. 

Feel free to file a bug report on grub for giving advice that 
 caused your kernel image removal to break. I am closing this report.

Sorry for the inconvenience.

manoj
-- 
The grand leap of the whale up the Fall of Niagara is esteemed, by all
who have seen it, as one of the finest spectacles in nature. --
Benjamin Franklin.
Manoj Srivastava   <[EMAIL PROTECTED]>  
1024D/BF24424C print 4966 F272 D093 B493 410B  924B 21BA DABB BF24 424C


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: my CDBS gallery: real-world rules samples

[Stefano Zacchiroli]

On Tue, Sep 12, 2006 at 01:57:08PM -0300, Caio Begotti wrote:
> > I would like to see is an (alternative) organization sliced  
> > according to which cdbs rules/classes are used.
> Indeed, that could be better. I even tried to find out some kind of  
> list of existing CDBS rules and variables to make hypertext links  
> with them (pointing to Duck's documentation) and that would allow me  
> to change the current organization of debian/rules and even have a  
> nice reference for those rules.

No idea for variables, but for rules and classes you can use the
following naive (but IMO working) heuristic.

  Look for all include lines in the debian/rules you're getting, if the
  included file is under /usr/share/cdbs then it is worth using it as a
  category to which the current debian/rules belongs to.

Cheers.

-- 
Stefano Zacchiroli -*- Computer Science PhD student @ Uny Bologna, Italy
[EMAIL PROTECTED],debian.org,bononia.it} -%- http://www.bononia.it/zack/
If there's any real truth it's that the entire multidimensional infinity
of the Universe is almost certainly being run by a bunch of maniacs. -!-


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Bug#387286: Acknowledgement (postrm_hook is not run)

[Manoj Srivastava]

On Wed, 13 Sep 2006 15:18:27 +0200, Michael Biebl <[EMAIL PROTECTED]> said: 

> And setting
> postinst_hook = /usr/sbin/update-grub
> postrm_hook   = /usr/sbin/update-grub

> finally fixes the problem, although the absolute path is not
> recommended by the grub maintainer.

The grub maintainer has not thought through all use cases,
 including yours -- changing /etc/kernel-img.conf to not contain
 absolute path names while there are kernel images on disk with the
 older postrm scripts will cause problems when those packages are
 removed.

Since there is no expectation that the location of update-grub
 is going to change in the near-to-mid term, there is no harm in using
 absolute path lengths right ow -- until your system cycles through
 all the older kernel images. Post-etch one can recommend putting in
 non-absolute path names in the configuration file, when there are few
 installations with kernels that would break. There is no hurry.

> Still, postrm should work if postrm_hook does not specify an
> absolute path (seems to work for postinst after all, which is a bit
> weird).

If you find me a time machine to retroactively fix the
 kernel-package used to generate your old kernel image, sure.

Any kernel images created with the current kernel-packagew
 will have a working postrm -- but that does not help with the postrem
 scripts already sitting in your /var/lib/dpkg/info/ directory.

manoj
-- 
You scratch my tape, and I'll scratch yours.
Manoj Srivastava   <[EMAIL PROTECTED]>  
1024D/BF24424C print 4966 F272 D093 B493 410B  924B 21BA DABB BF24 424C


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Accepted lynx 2.8.5-2sarge2 (source i386)

[Thomas Dickey]

Otavio Salvador <[EMAIL PROTECTED]> wrote:

> He can fix a previous entry and cite it in next version.

Perhaps 4 months is too short a time for him to correct it.

-- 
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Accepted lynx 2.8.5-2sarge2 (source i386)

[Steinar H. Gunderson]

On Wed, Sep 13, 2006 at 08:26:09AM -0400, Thomas Dickey wrote:
> Martin Schulze has been told more than once that this was not an OpenBSD
> patch.
> 
> After the second time, there is no plausible excuse.
> 
> Do you have an excuse?

Could you please tell me why this is such a huge deal?

/* Steinar */
-- 
Homepage: http://www.sesse.net/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Bug#387286: Acknowledgement (postrm_hook is not run)

[Frans Pop]

reassign 387286 grub
severity 387286 grave
retitle 387286 grub path transition is breaking kernel removals
thanks

On Wednesday 13 September 2006 16:28, Manoj Srivastava wrote:
> The grub maintainer has not thought through all use cases,
>  including yours -- changing /etc/kernel-img.conf to not contain
>  absolute path names while there are kernel images on disk with the
>  older postrm scripts will cause problems when those packages are
>  removed.

If the bug is caused by the grub transition, why don't you reassign it 
there with appropriate severity?

FWIW, I've seen the issue myself too.


pgppoOr4TDenE.pgp
Description: PGP signature

lilypond and thanks to Rob Browning

[Thomas Bushnell BSG]

Rob Browning, the guile maintainer, has been doing a lot of hard work
trying to get guile-1.8 into Debian.  Turns out there were some
critical timing bugs affecting the operation of fork in the guile
threading implementation, bugs which are too intractible to solve
immediately.

Rob has uploaded guile-1.8 with the threading option turned off, which
seems to work fine, and since lilypond doesn't rely on threads in
guile, should enable the long-awaited upload of lilypond 2.8.

It is my hope that #359855 will not exist in the new lilypond.
However, this is just a hope.  If ghostscript continues to have such a
bug, then solving it will become of critical priority for getting
lilypond into the release.

So right now, I'm waiting for ftpmaster to approve the new guile-1.8
packages, and then I can build and upload lilypond 2.8 without delay.

Thomas


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Bug#387286: Acknowledgement (postrm_hook is not run)

[Manoj Srivastava]

On Wed, 13 Sep 2006 18:25:54 +0200, Frans Pop <[EMAIL PROTECTED]> said: 

> reassign 387286 grub severity 387286 grave retitle 387286 grub path
> transition is breaking kernel removals thanks

Note that grub is not breaking, nor does grub transition break
 anything -- if users follow the advice in NEWS.Debian from grub, then
 they intriducean user error. So it is a documentation bug, which
 ought not to be grave, in my opinion.

> On Wednesday 13 September 2006 16:28, Manoj Srivastava wrote:
>> The grub maintainer has not thought through all use cases,
>> including yours -- changing /etc/kernel-img.conf to not contain
>> absolute path names while there are kernel images on disk with the
>> older postrm scripts will cause problems when those packages are
>> removed.

> If the bug is caused by the grub transition, why don't you reassign
> it there with appropriate severity?

Because I already filed a bug on grub with what I considered
 appropriate severity -- normal. See 

> FWIW, I've seen the issue myself too.

Yeah, it does not help there was a bug in kernel-package that
 I just fixed that mean that any images built so far would result in a
 postrm that does not work well with non-absolute paths.  I think I
 have fixed it (finally) in 10.056, in incoming as I write this.

manoj
-- 
"People don't form relationships, they take hostages." anon
Manoj Srivastava   <[EMAIL PROTECTED]>  
1024D/BF24424C print 4966 F272 D093 B493 410B  924B 21BA DABB BF24 424C


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Orphaning my packages

[Martín Ferrari]

On 9/12/06, David Moreno Garza <[EMAIL PROTECTED]> wrote:


If you think you could adopt some of the packages, feel free to do it
(filling an ITA would be nice). Just talk to the Debian Perl Group
first, if thinking on adopting some of the Perl modules; talk first
to the pkg-ruby-extras groups if thinking on adopting some of the
ruby modules, also.


I'd like to adopt some of the -perl packages, but I don't know what
has to be coordinated with the perl group. Can somebody comment?

Thanks.

--
Martín Ferrari

Re: Orphaning my packages

[Aurélien GÉRÔME]

Hi,

On Tue, Sep 12, 2006 at 04:21:18PM -0500, David Moreno Garza wrote:
> If you think you could adopt some of the packages, feel free to do it  
> (filling an ITA would be nice). Just talk to the Debian Perl Group  
> first, if thinking on adopting some of the Perl modules; talk first  
> to the pkg-ruby-extras groups if thinking on adopting some of the  
> ruby modules, also.

I filed 2 ITA against hfsplus (#387337) and hfsutils (#387338). Expect
uploads tomorrow or the day after. :)

Cheers,
-- 
 .''`.   Aurélien GÉRÔME
: :'  :
`. `'`   Free Software Developer
  `- Unix Sys & Net Admin


signature.asc
Description: Digital signature

Re: Orphaning my packages

[Gunnar Wolf]

Martín Ferrari dijo [Wed, Sep 13, 2006 at 03:26:43PM -0300]:
> On 9/12/06, David Moreno Garza <[EMAIL PROTECTED]> wrote:
> 
> >If you think you could adopt some of the packages, feel free to do it
> >(filling an ITA would be nice). Just talk to the Debian Perl Group
> >first, if thinking on adopting some of the Perl modules; talk first
> >to the pkg-ruby-extras groups if thinking on adopting some of the
> >ruby modules, also.
> 
> I'd like to adopt some of the -perl packages, but I don't know what
> has to be coordinated with the perl group. Can somebody comment?

Hi,

I was planning on starting the wide adopting process to the group, but
if you can help, much better. In my experience, the pkg-perl group has
helped me not appear like an irresponsable maintainer (which I am! :-P )
during my stress periods. So, Mart�n, if you are not currently part
of the group, I can add you - just give me your Alioth user name and
promise to do no intentional harm.

So, taking from Damog's developer page - We need to adopt:

libcontextual-return-perl
libcurses-widgets-perl
libend-perl
libio-prompt-perl
liblingua-es-numeros-perl
liblist-compare-perl
libmath-fibonacci-perl
libmath-nocarry-perl
libmath-randomorg-perl
libmath-vec-perl
libnumber-compare-perl
libopengl-perl
libterm-size-perl
libwww-freshmeat-perl
libwww-google-calculator-perl
libwww-myspace-perl

They look quite simple, the only three open bugs on them look trivial
to fix. We should check for updatedness, add watch files, and... Well,
simple stuff in the end. Perl group: Should we? Who starts?

David: Best luck. Hope to see you back here soon!

-- 
Gunnar Wolf - [EMAIL PROTECTED] - (+52-55)5623-0154 / 1451-2244
PGP key 1024D/8BB527AF 2001-10-23
Fingerprint: 0C79 D2D1 2C4E 9CE4 5973  F800 D80E F35A 8BB5 27AF


signature.asc
Description: Digital signature

Re: Orphaning my packages

[Martín Ferrari]

On 9/13/06, Gunnar Wolf <[EMAIL PROTECTED]> wrote:


I was planning on starting the wide adopting process to the group, but
if you can help, much better. In my experience, the pkg-perl group has
helped me not appear like an irresponsable maintainer (which I am! :-P )
during my stress periods. So, Mart�n, if you are not currently part
of the group, I can add you - just give me your Alioth user name and
promise to do no intentional harm.


it is tincho-guest.
I have no experience with maintaining within a group, but I will be
happy to learn :)

--
Martín Ferrari

Manpages in language-specific packages

[Gunnar Wolf]

Hi,

I'm adopting liblingua-es-numeros-perl, a Perl module for translating
numbers into their Spanish string representation. One of the first
things I noticed is that the manpage is completely (and only) written
in Spanish. So, besides sending any changes I do to the upstream
author (although the module looks unmaintained, with only version 0.01
existing since 2001), what would be wisest?

- Leaving it as it is
- Translating only the manpage description, perhaps first section -
  Don't think it would be the best, as it might have non-ES users
  (although the whole module API is in Spanish)
- Translating the whole manpage, and providing the Spanish translation
  as a separate file in /usr/share/doc (pointing at it from the
  manpage) 
- Going the inverse way, leaving the package as close as possible to
  its current status, and pointing to an English translation from
  /usr/share/doc 
- I18N in manpages? Sorry, haven't been there, haven't done that

Personally, I don't feel it natural having a Spanish-only
manpage. But then again, I'm among the nasty minority who does not
like his computer to use his mother tongue ;-)

Greetings,

-- 
Gunnar Wolf - [EMAIL PROTECTED] - (+52-55)5623-0154 / 1451-2244
PGP key 1024D/8BB527AF 2001-10-23
Fingerprint: 0C79 D2D1 2C4E 9CE4 5973  F800 D80E F35A 8BB5 27AF


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [Pkg-ruby-extras-maintainers] Orphaning my packages

[David Moreno Garza]

On Sep 13, 2006, at 2:13 AM, Paul van Tilburg wrote:

Hello David,

On Tue, Sep 12, 2006 at 04:21:18PM -0500, David Moreno Garza wrote:

I'll be taking a long vacation of Debian and free software activity
for the next couple of months for personal reasons. [...]


I hope that the vacation will do you good and hope you'll come back to
us (if it feels right of course).


Thank you.


If you think you could adopt some of the packages, feel free to do it
(filling an ITA would be nice). Just talk to the Debian Perl Group
first, if thinking on adopting some of the Perl modules; talk first
to the pkg-ruby-extras groups if thinking on adopting some of the
ruby modules, also.


I think the Debian/Ruby Extras team is interested in taking all
Ruby-related packages:  libform-validator-ruby, libgtkmozembed-ruby,
libgtk-trayicon-ruby, libsnmp-ruby, libtermios-ruby and revolution.
This is mainly because I am already familiar with these packages  
and we,

as a team, can share the load.
David, if I am overlooking a Ruby package, please let me know?


Only note that libgtk-mozembed-ruby was supposed to be merged with  
Ruby-GNOME2 (as it already was by upstream). This should be  
rescheduled with daf. Feel free to add the others to the repository,  
as I was supposed to do months ago.


Cheers,

David Moreno Garza <[EMAIL PROTECTED]>
 http://www.damog.net/ | Oh, my god! That's the funky shit!



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Paper on (debian and others) distribution managment at ASE'06

[Ralf Treinen]

Hi,

a paper on automatic tools to assist in distribution managment will
be presented at the 21st IEEE/ACM International Conference on
Automated Software Engineering (ASE 2006) next week in Tokyo [1].

  Managing the Complexity of Large Free and Open Source Package-Based
  Software Distributions

This paper presents a model of package dependencies, with the two
principal examples debian and RPM. This is the model used in the
debcheck and rpmcheck tools (which are packaged in debian as
edos-debcheck, resp. edos-rpmcheck).

The full paper (10 pages) in PDF is available here:

https://protactinium.pps.jussieu.fr:12345/svn/edos/reports/wp2/ASE/ase.pdf

-Ralf.

[1] http://www.ase-conference.jp/


signature.asc
Description: Digital signature

Re: Bug#386911: ITP: Claroline -- Course Management System for Online Learning

[Moritz Muehlenhoff]

Victor Manuel Mtz wrote:
> * Package name: Claroline
>   Version : 1.7.8
>   Upstream Author : Lederer Guillaume <[EMAIL PROTECTED]>
> * URL : http://www.claroline.net
> * License : GPL
>   Description : Course Management System for Online Learning
>
> Claroline is a free application based on PHP/MySQL allowing teachers or
> education organizations to create and administrate courses through the
> web.
>
> Developed from teachers to teachers, Claroline is built over sound
> pedagogical principles allowing a large variety of pedagogical setup
> including widening of traditional classroom and online collaborative
> learning.

However, it also seems to be built over unsound web programming principles
allowing a large variety of security exploits including widening of
SQL queries and online collaborative cross-site-scripting.

(CVE-2006-3257, CVE-2006-2868, CVE-2006-2284, CVE-2006-1596, CVE-2006-1595,
CVE-2006-1594, CVE-2006-0411, CVE-2005-1377, CVE-2005-1376, CVE-2005-1375,
CVE-2005-1374 and possibly more, I stopped digging deeper)

I don't think this should enter the archive.

Cheers,
Moritz


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Orphaning my packages

[Francesco Pedrini]

On Wednesday 13 September 2006 20:32, Aurélien GÉRÔME wrote:
>
> I filed 2 ITA against hfsplus (#387337) and hfsutils (#387338).
> Expect uploads tomorrow or the day after. :)

D'OH!

http://lists.debian.org/debian-devel/2006/09/msg00353.html 

:P 

if you need help with hfsplus please you just have to ask ;)

Regards,
Francesco
-- 
:wq

Re: Orphaning my packages

[Aurélien GÉRÔME]

On Wed, Sep 13, 2006 at 10:56:21PM +0200, Francesco Pedrini wrote:
> On Wednesday 13 September 2006 20:32, Aurélien GÉRÔME wrote:
> >
> > I filed 2 ITA against hfsplus (#387337) and hfsutils (#387338).
> > Expect uploads tomorrow or the day after. :)
> 
> D'OH!
> 
> http://lists.debian.org/debian-devel/2006/09/msg00353.html 
> 
> :P 
> 
> if you need help with hfsplus please you just have to ask ;)

Argh... Sorry for that. My habit of reading mailing-lists diagonally
is to blame. :)

I already started to work on it, should I stop and do you want to
take over?

Cheers,
-- 
 .''`.   Aurélien GÉRÔME
: :'  :
`. `'`   Free Software Developer
  `- Unix Sys & Net Admin


signature.asc
Description: Digital signature

Re: Orphaning my packages

[Francesco Pedrini]

On Wednesday 13 September 2006 23:03, Aurélien GÉRÔME wrote:
> On Wed, Sep 13, 2006 at 10:56:21PM +0200, Francesco Pedrini wrote:
> > On Wednesday 13 September 2006 20:32, Aurélien GÉRÔME wrote:
> > > I filed 2 ITA against hfsplus (#387337) and hfsutils (#387338).
> > > Expect uploads tomorrow or the day after. :)
> >
> > D'OH!
> >
> > http://lists.debian.org/debian-devel/2006/09/msg00353.html
> >
> > :P
> >
> > if you need help with hfsplus please you just have to ask ;)
>
> Argh... Sorry for that. My habit of reading mailing-lists diagonally
> is to blame. :)
>
> I already started to work on it, should I stop and do you want to
> take over?

No no, don't stop! :)
I'm quite busy at the moment, so go ahead ;)

It's all yours :)


Francesco
-- 
:wq

Re: Orphaning my packages

[Francesco Pedrini]

On Tuesday 12 September 2006 23:21, David Moreno Garza wrote:
> Hello,
>
> I'll be taking a long vacation of Debian and free software activity
> for the next couple of months for personal reasons. Because of that,
> I'm orphaning my non-comaintained packages. I really think those
> packages shouldn't make it to Etch with a non attending maintainer,
> just like I'm beginning to become (I already orphaned some of them in
> the last couple of months). Once I get more free time or motivation
> to work on my packages, I'll be coming back, but since that's not the
> case now, I'm stepping back for a while so I don't interfere with the
> project.

in short:
I've already filled ITA for wyg and tdfsb, i'm also intersed in cssed, 
and I'll fill an ITA for it tomorrow (where tomorrow is 14/09/2006) 
morning :)

cheers 
Francesco
-- 
:wq


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Orphaning my packages

[gregor herrmann]

On Wed, 13 Sep 2006 16:00:06 -0300, Martín Ferrari wrote:

> >I was planning on starting the wide adopting process to the group, 

Fine, I had the same idea.

> >but
> >if you can help, much better. 

Ack.

> I have no experience with maintaining within a group, but I will be
> happy to learn :)

Just take a look at http://pkg-perl.alioth.debian.org/ and start
working :-)
Feel free to ask any questions an [EMAIL PROTECTED]

Cheers,
gregor 
 
-- 
 .''`.   http://info.comodo.priv.at/ | gpg key ID: 0x00F3CFE4
 : :' :  debian: the universal operating system - http://www.debian.org/
 `. `'   member of https://www.vibe.at/ | how to reply: http://got.to/quote/
   `-NP: Don McLean: Dreidel


signature.asc
Description: Digital signature

Re: Accepted lynx 2.8.5-2sarge1.2 (source i386 sparc amd64)

[Aníbal Monsalve Salazar]

On Wed, Sep 13, 2006 at 08:19:02AM -0400, Thomas Dickey wrote:
>In article <[EMAIL PROTECTED]> you wrote:
>>-BEGIN PGP SIGNED MESSAGE-
>>Hash: SHA1
>
>>Format: 1.7
>>Date: Wed, 13 Sep 2006 18:41:49 +1000
>>Source: lynx
>>Binary: lynx
>>Architecture: source i386 sparc amd64
>>Version: 2.8.5-2sarge1.2
>>Distribution: unstable
>>Urgency: high
>>Maintainer: James Troup <[EMAIL PROTECTED]>
>>Changed-By: Anibal Monsalve Salazar <[EMAIL PROTECTED]>
>>Description: 
>> lynx   - Text-mode WWW Browser
>>Closes: 238057 384725
>>Changes: 
>> lynx (2.8.5-2sarge1.2) unstable; urgency=high
>> .
>>   * NMU.
>>   * Added OpenBSD patch to fix infinete loop rendering broken HTML
>> [debian/patches/04_CVE-2004-1617.dpatch]
>
>
>This is the third time this year that patch has been incorrectly
>ascribed to OpenBSD.

Sorry for the misundertanding. I'll fix the lynx changelog file.
I just used the patch from 2.8.5-2sarge2 in stable-security.

>-- 
>Thomas E. Dickey
>http://invisible-island.net
>ftp://invisible-island.net

Aníbal Monsalve Salazar
-- 
http://v7w.com/anibal


signature.asc
Description: Digital signature

Re: Manpages in language-specific packages

[Javier Fernández-Sanguino Peña]

On Wed, Sep 13, 2006 at 12:46:19PM -0500, Gunnar Wolf wrote:
> Hi,
> 
> I'm adopting liblingua-es-numeros-perl, a Perl module for translating
> numbers into their Spanish string representation. One of the first
> things I noticed is that the manpage is completely (and only) written
> in Spanish. So, besides sending any changes I do to the upstream
> author (although the module looks unmaintained, with only version 0.01
> existing since 2001), what would be wisest?

Translate the manpage to English, distribute that in /usr/share/man/man[1-9]
and put the Spanish manpage at /usr/share/man/es/man[1-9] (where it should
be). That way non-Spanish users installing the program will have a manual
page and Spanish users will be able to read it properly (if their locale is
configured properly)

> - I18N in manpages? Sorry, haven't been there, haven't done that

It's not that difficult, just put the (english) manpage in /usr/share/man/
and the i18n manpage in /usr/share/man/XX. Man takes care of the rest.

Regards

Javier


signature.asc
Description: Digital signature

Re: transitioning config files between two packages

[Javier Fernández-Sanguino Peña]

On Tue, Sep 12, 2006 at 05:21:50PM -0700, Steve Langasek wrote:
> After an upgrade and answering all of the conffile prompts, does
> /var/lib/dpkg/info/nagios-plugins.conffiles still exist and reference these
> files?  Depending on what dpkg is really doing here, it may well be possible
> to handle the conffile transfer in maintainer scripts.  (And I thought
> dpkg.org once had recipes for exactly this, but unfortunately the site has
> been down for some time now. :/)

Are you looking for http://wiki.debian.org/DpkgConffileHandling ? 
(just found it googling). I guess this information (if current) should be
moved over to the developer's reference. 

Regards

Javier


signature.asc
Description: Digital signature

Re: transitioning config files between two packages

[Steve Langasek]

On Thu, Sep 14, 2006 at 01:32:43AM +0200, Javier Fernández-Sanguino Peña wrote:
> On Tue, Sep 12, 2006 at 05:21:50PM -0700, Steve Langasek wrote:
> > After an upgrade and answering all of the conffile prompts, does
> > /var/lib/dpkg/info/nagios-plugins.conffiles still exist and reference these
> > files?  Depending on what dpkg is really doing here, it may well be possible
> > to handle the conffile transfer in maintainer scripts.  (And I thought
> > dpkg.org once had recipes for exactly this, but unfortunately the site has
> > been down for some time now. :/)

> Are you looking for http://wiki.debian.org/DpkgConffileHandling ? 
> (just found it googling). I guess this information (if current) should be
> moved over to the developer's reference. 

That seems to be the one.  It also doesn't seem to address changing the
owner of a conffile.. ohwell. :)

-- 
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 PROTECTED]   http://www.debian.org/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Accepted lynx 2.8.5-2sarge2 (source i386)

[Thomas Dickey]

Steinar H. Gunderson <[EMAIL PROTECTED]> wrote:
> On Wed, Sep 13, 2006 at 08:26:09AM -0400, Thomas Dickey wrote:
>> Martin Schulze has been told more than once that this was not an OpenBSD
>> patch.
>> 
>> After the second time, there is no plausible excuse.
>> 
>> Do you have an excuse?

> Could you please tell me why this is such a huge deal?

hmm.  I'm given to understand that you don't get annoyed when people
are (to be generous) careless with their descriptions of security-related
issues. 

In that case, I don't know how to explain it...

bye

-- 
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Accepted lynx 2.8.5-2sarge1.2 (source i386 sparc amd64)

[Thomas Dickey]

Aníbal Monsalve Salazar <[EMAIL PROTECTED]> wrote:

>>This is the third time this year that patch has been incorrectly
>>ascribed to OpenBSD.

> Sorry for the misundertanding. I'll fix the lynx changelog file.
> I just used the patch from 2.8.5-2sarge2 in stable-security.

ok.  Perhaps you should reference

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=296340

-- 
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Accepted lynx 2.8.5-2sarge2 (source i386)

[Steinar H. Gunderson]

On Thu, Sep 14, 2006 at 12:15:43AM -, Thomas Dickey wrote:
>>> Martin Schulze has been told more than once that this was not an OpenBSD
>>> patch.
>> Could you please tell me why this is such a huge deal?
> hmm.  I'm given to understand that you don't get annoyed when people
> are (to be generous) careless with their descriptions of security-related
> issues. 

I could understand a certain annoyance, but yelling at them and calling their
actions “malicious” on public mailing lists seems a bit... overblown for
misattributing a patch to lynx.

/* Steinar */
-- 
Homepage: http://www.sesse.net/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

how to treat upgrade bugs that only affect unstable->unstable?

[Peter Samuelson]

CC: debian-devel, as I'm asking about packaging best practices.

[Agustin Martin]
> >* python-subversion.{prerm,postinst}: use pyversions, fix stupid
> >  bug (Closes: #379278) in prerm.  Tighten python build-dep to
> >  ensure availability of pyversions.
> 
> Note that some upgrades might still be a problem,

> dpkg: warning - old pre-removal script returned error exit status 1
> dpkg - trying script from the new package instead ...

That raises a philosophical question:

If a bug was in a prerm script in unstable for 7 days, but never
appeared in stable or testing, should we include cruft in present and
future prerm versions to work around it?

Or, put another way: a prerm is designed to run with the package
version it is shipped with.  To what lengths should it go to do the
right thing when dpkg runs it with previous versions of the package?


signature.asc
Description: Digital signature

Re: Accepted lynx 2.8.5-2sarge1.2 (source i386 sparc amd64)

[Aníbal Monsalve Salazar]

On Thu, Sep 14, 2006 at 12:13:03AM -, Thomas Dickey wrote:
>Aníbal Monsalve Salazar <[EMAIL PROTECTED]> wrote:
>
>>>This is the third time this year that patch has been incorrectly
>>>ascribed to OpenBSD.
>
>>Sorry for the misundertanding. I'll fix the lynx changelog file.
>>I just used the patch from 2.8.5-2sarge2 in stable-security.
>
>ok.  Perhaps you should reference
>
>   http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=296340

Too late! 2.8.5-2sarge2.1 was uploaded before you posted this suggestion.

interdiff -z -p1 cache/lynx/lynx_2.8.5-2sarge1.2.diff.gz 
cache/lynx/lynx_2.8.5-2sarge2.1.diff.gz
diff -u lynx-2.8.5/debian/changelog lynx-2.8.5/debian/changelog
--- lynx-2.8.5/debian/changelog
+++ lynx-2.8.5/debian/changelog
@@ -1,10 +1,20 @@
+lynx (2.8.5-2sarge2.1) unstable; urgency=low
+
+  * NMU.
+  * Fixed changelog entry that closes #384725.
+  * Used a debian version greater than the one in stable-security.
+
+ -- Anibal Monsalve Salazar <[EMAIL PROTECTED]>  Thu, 14 Sep 2006 09:15:56 
+1000
+
 lynx (2.8.5-2sarge1.2) unstable; urgency=high
 
   * NMU.
-  * Added OpenBSD patch to fix infinete loop rendering broken HTML
+  * Fixed CVE-2004-1617 (see DSA-1076-1).
 [debian/patches/04_CVE-2004-1617.dpatch]
+Patch by Thomas E. Dickey <[EMAIL PROTECTED]>.
 Closes: #384725.
   * Fixed "FTBFS on GNU/Hurd (and GNU/k*BSD)".
+[debian/patches/05_FTBFS_on_GNUHurd_and_GNUkBSD]
 Patch by Robert Millan <[EMAIL PROTECTED]>.
 Closes: #238057.
 
>-- 
>Thomas E. Dickey
>http://invisible-island.net
>ftp://invisible-island.net

Aníbal Monsalve Salazar
-- 
http://v7w.com/anibal


signature.asc
Description: Digital signature

Bug#387385: ITP: shed -- Hex editor using ncurses, with a friendly pico-style interface

[Adam Cecile]

Package: wnpp
Severity: wishlist
Owner: Adam Cecile <[EMAIL PROTECTED]>


* Package name: shed
  Version : 1.13
  Upstream Author : Alex Sisson <[EMAIL PROTECTED]> 
* URL : http://shed.sourceforge.net/
* License : GPL
  Description : Hex editor using ncurses, with a friendly pico-style 
interface

 Shed (simple hex editor) is a ncurse based hex editor. 
 .
 It includes the following features :
   - Displays each byte as ascii, hex, decimal, octal and binary.
   - Allows changes to be input in all of the above.
   - Simple Pico-style interface.
   - Search.
   - Dump to file.
   - Small memory requirments because file is not loaded into memory.
   - Can handle files up to 2Gb.
 .
  Homepage: http://shed.sourceforge.net/

-- System Information:
Debian Release: testing/unstable
  APT prefers testing
  APT policy: (900, 'testing'), (400, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.17-2-686-aircrack
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Bug#387385: ITP: shed -- Hex editor using ncurses, with a friendly pico-style interface

[Ron Johnson]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 09/13/06 20:15, Adam Cecile wrote:
[snip]
>  It includes the following features :
[snip]
>- Can handle files up to 2Gb.

That's a feature?

- --
Ron Johnson, Jr.
Jefferson LA  USA

Is "common sense" really valid?
For example, it is "common sense" to white-power racists that
whites are superior to blacks, and that those with brown skins
are mud people.
However, that "common sense" is obviously wrong.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFCLzyS9HxQb37XmcRAm0mAJ4qRfgKkqoIwSrY2K2zwyXsjSvPfQCfUjZd
5Mb2w6ZA0/bWf5AbxHOVd78=
=DYoT
-END PGP SIGNATURE-


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Manpages in language-specific packages

[Felipe Augusto van de Wiel (faw)]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 09/13/2006 08:21 PM, Javier Fernández-Sanguino Peña wrote:
> On Wed, Sep 13, 2006 at 12:46:19PM -0500, Gunnar Wolf wrote:
> 
>>Hi,
>>
>>I'm adopting liblingua-es-numeros-perl, a Perl module for translating
>>numbers into their Spanish string representation. One of the first
>>things I noticed is that the manpage is completely (and only) written
>>in Spanish. So, besides sending any changes I do to the upstream
>>author (although the module looks unmaintained, with only version 0.01
>>existing since 2001), what would be wisest?
> 
> Translate the manpage to English, distribute that in /usr/share/man/man[1-9]
> and put the Spanish manpage at /usr/share/man/es/man[1-9] (where it should
> be). That way non-Spanish users installing the program will have a manual
> page and Spanish users will be able to read it properly (if their locale is
> configured properly)
> 
> 
>>- I18N in manpages? Sorry, haven't been there, haven't done that
> 
> It's not that difficult, just put the (english) manpage in /usr/share/man/
> and the i18n manpage in /usr/share/man/XX. Man takes care of the rest.

You could use po4a, it is an amazing tool to work with different
formats and convert them into PO files, than you could easily request
for translator on -i18n. ;)

http://po4a.alioth.debian.org/


> Regards
> Javier

Kind regards,

- --
Felipe Augusto van de Wiel (faw)
"Debian. Freedom to code. Code to freedom!"
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with Debian - http://enigmail.mozdev.org

iD8DBQFFCM0XCjAO0JDlykYRAh8dAJ0YTSRClqmuztq9kcX+JrOa5MS/6wCg0RN5
51RL76Gdu8xCOT+NGHk4NkY=
=hEBH
-END PGP SIGNATURE-


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: how to treat upgrade bugs that only affect unstable->unstable?

[Henrique de Moraes Holschuh]

On Wed, 13 Sep 2006, Peter Samuelson wrote:
> That raises a philosophical question:
> 
> If a bug was in a prerm script in unstable for 7 days, but never
> appeared in stable or testing, should we include cruft in present and
> future prerm versions to work around it?
> 
> Or, put another way: a prerm is designed to run with the package
> version it is shipped with.  To what lengths should it go to do the
> right thing when dpkg runs it with previous versions of the package?

It depends only on the ammount of damage the bug causes.

-- 
  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

A few problems in sight switching to Debian

[Gilles Pelletier]

I'm not a developer and I'd just like to point a few issues to check 
for the next release. I'm planning to switch to Debian when Etch 
comes out but, for now, my findings are from Knoppix.

My monitor, a Viewsonic VX922, is not set correctly. Il looks OK but, 
when a 1280x1024 image is downloaded it fits more than the screen. I 
now use a file made by PCLinuxOS whcih is correct. Here are the two 
sections that I copied in my XF86Config file:

Section "Device"
Identifier  "ATI Technologies, Inc. Rage 128 PF/PRO AGP 4x 
TMDS"
Driver  "ati"
BusID   "PCI:1:0:0"
EndSection

Section "Monitor"
Identifier  "VX922"
Option  "DPMS"
EndSection

Section "Screen"
Identifier  "Default Screen"
Device  "ATI Technologies, Inc. Rage 128 PF/PRO AGP 4x 
TMDS"
Monitor "VX922"
DefaultDepth24
SubSection "Display"
Depth   1
Modes   "1280x1024" "1152x864" "1024x768" 
"832x624" "800x600" "720x400" "640x480"
EndSubSection
SubSection "Display"
Depth   4
Modes   "1280x1024" "1152x864" "1024x768" 
"832x624" "800x600" "720x400" "640x480"
EndSubSection
SubSection "Display"
Depth   8
Modes   "1280x1024" "1152x864" "1024x768" 
"832x624" "800x600" "720x400" "640x480"
EndSubSection
SubSection "Display"
Depth   15
Modes   "1280x1024" "1152x864" "1024x768" 
"832x624" "800x600" "720x400" "640x480"
EndSubSection
SubSection "Display"
Depth   16
Modes   "1280x1024" "1152x864" "1024x768" 
"832x624" "800x600" "720x400" "640x480"
EndSubSection
SubSection "Display"
Depth   24
Modes   "1280x1024" "1152x864" "1024x768" 
"832x624" "800x600" "720x400" "640x480"
EndSubSection
EndSection

-

My printer, an HP Laserjet 1012, which used to work with Knoppix 3.9, 
now doesn't work with Knoppix 5.01. It seems the installation program 
doesn't find the driver. 

There are two keyboards used in Quebec, Canada. The oldest is called 
CF and a newer one is called CA, with which it's possible to write 
accented letters such as "à è ç" in just one stroke. I use the the CA 
keyboard and choosing it in KDE used to be very straightforward. 

In the latest version of KDE, though, the CA keyboard has disappered 
and the CF keyboard has been renamed CA. So all distros using KDE 
have the same problem and I can't find my keyboard anymore.

If this is not plain sabotage, tell me what it is.

Regards!

 

Re: Two versions of pan in etch?

[Lionel Elie Mamane]

On Thu, Jul 27, 2006 at 08:31:34PM +0200, Søren Boll Overgaard wrote:

> Pan[0] is currently undergoing a major rewrite, and being the
> maintainer, I am currently considering what version of pan to
> include in etch. This mail[1] from one of the pan mailing lists sums
> up the situation quite nicely.

I'm aware that it is a bit late now, but as a pan user whose pan got
upgraded under his feet: You consider the only showstopper is/was
unability to _delete_ servers? WTF? Pan 0.111 cannot _print_ articles!
I would have considered that a showstopper.

-- 
Lionel


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Orphaning my packages

[Martín Ferrari]

On 9/13/06, gregor herrmann <[EMAIL PROTECTED]> wrote:


Just take a look at http://pkg-perl.alioth.debian.org/ and start
working :-)
Feel free to ask any questions an [EMAIL PROTECTED]


Thanks!!

--
Martín Ferrari