Bug#791173: libstxxl: library transition may be needed when GCC 5 is the default

Sebastiaan Couwenberg sebastic at xs4all.nl
Sun Aug 23 17:34:19 UTC 2015


On 23-08-15 19:15, Simon McVittie wrote:
> On 23/08/15 16:10, Sebastiaan Couwenberg wrote:
>> On 23-08-15 16:59, Simon McVittie wrote:
>>> The "SONAME bump" option was only really meant to be taken if the library
>>> had an upstream SONAME bump pending anyway (for instance icu and boost
>>> went this route). If there is not a SONAME change already in the pipeline,
>>> you should do the "v5" rename instead. My NMUs of gtkmm2.4, gtkmm3.0,
>>> atlas-cpp, bullet etc. should make a reasonable template for how this works.
>>
>> I have the packaging for the libstxll v5 rename mostly ready in my local
>> git, I can push this to Alioth and/or NMU it to DELAYED/2 if you want.
> 
> Regardless of whether you NMU, please compare what you have done with
> the Ubuntu patch at
> <http://patches.ubuntu.com/libs/libstxxl/libstxxl_1.4.1-1ubuntu1.patch>
> (you'll probably find it ended up identical), and send a diff against
> current unstable to this bug.

Anton has pushed his changes already, but so far I've not seen the
upload yet. That's a good thing because the Breaks/Replaces needs to use
a version constraint:

 Breaks: libstxxl1 (<< 1.4.1-2~)
 Replaces: libstxxl1 (<< 1.4.1-2~)

Ubuntu uses Conflicts/Replaces without a version constraint. And
bothered to rename the -dbg package too, which neither I nor Anton have
done.

My proposed changes are attached because cannot push to debian-science,
I think that should be used instead of Conflicts/Replaces as Ubuntu has
done.

> If you've done the work already, and if all the library build-deps
> either don't need a transition or have already started theirs, then I
> would personally say you might as well NMU to an appropriate DELAYED
> queue. This overall transition has broken unstable for 3 weeks already,
> during which lots of packages are either uninstallable or
> non-functional, and basically no C++ can migrate to testing; the sooner
> we can get through it and have our distribution back, the better.

It doesn't look like any of the libstxxl build dependencies are a
blocker, so I do think we need to go ahead with an upload to unstable soon.

Kind Regards,

Bas

-- 
 GPG Key ID: 4096R/6750F10AE88D4AF1
Fingerprint: 8182 DE41 7056 408D 6146  50D1 6750 F10A E88D 4AF1
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Use-versioned-Breaks-Replaces.patch
Type: application/pgp-encrypted
Size: 1502 bytes
Desc: not available
URL: <http://lists.alioth.debian.org/pipermail/debian-science-maintainers/attachments/20150823/d0e379b3/attachment-0001.pgp>


More information about the debian-science-maintainers mailing list