[Pkg-ia32-libs-maintainers] ia32-libs update for lenny (2.7+lenny1)

[Goswin von Brederlow]

Moritz Muehlenhoff <jmm at inutil.org> writes:

> On Sat, Aug 14, 2010 at 02:40:52PM +0200, Thijs Kinkhorst wrote:
>> On tiisdei 20 July 2010, Thijs Kinkhorst wrote:
>> > Hi Goswin,
>> > 
>> > On moandei 28 Juny 2010, Thijs Kinkhorst wrote:
>> > > Besides from this thought the current updated packages look correct.
>> > > 
>> > > A number of them have had further updates since you prepared this package
>> > > version. If you wish you can update them to include those fixes aswell,
>> > > would  be nice.
>> > > 
>> > > As for the changes to ia32-libs packaging, the following three changes
>> > > are
>> > > 
>> > > not  acceptable in a stable-security upload:
>> > >   * Add misc depends for debhelper.
>> > >   * Add lots of lintian overrides where nothing can be done about them.
>> > >   * Bump debhelper compat to 5.
>> > > 
>> > > Perhaps they were made to address Lintian reports. It is not necessary
>> > > (or even desirable) to do that kind of cleanup in a stable upload.
>> > > Please revert  those.
>> > 
>> > Just curious how things are going with this?
>> 
>> Also, now that squeeze has been frozen, I think it's essential to also make an 
>> upload for ia32-libs for squeeze, perhaps in september or so when the biggest 
>> inflow of packages is over. Then we'll have a much better supportable version 
>> for the squeeze lifetime. Ideally, based on which packages will change during 
>> the freeze, another catch-up upload when the release is very near, and/or an 
>> upload in the first point release are also very much preferred.

Sorry for taking so long. ia32-libs isn't high on my priority list. I
promise to work on it tonight.

There are already a number of updates in the ia32-libs git but some more
polishing is needed for the next upload.

>>From what I heard during DebConf multiarch will still happen for
> Squeeze, so this shouldn't be needed?
>
> Cheers,
>         Moritz

No way. The critical parts are dpkg, apt, aptitude, synaptic, adept
support and none of them are anywhere done. The risk of breaking
esspecially dpkg by forcing a huge change past the freeze is just to
great and there isn't enough time to properly test any such change. At
least if the freeze isn't supposed to be longer than the last.

Also none of the relevant libs have changed to multiarch yet and that
would be some 50-100 freeze exceptions and library transitions right
there. Another showstopper.

MfG
        Goswin