gregor herrmann <gre...@debian.org> writes:

> On Tue, 14 Aug 2012 20:05:50 +0200, Sébastien Villemot wrote:
>
>> >> The patch also includes fixes for packaging issues.
>> > This is not a good idea for an NMU for an RC bug during the freeze
>> The package is currently not cleaning properly (violation of policy
>> §4.1) and not building twice in a row. While the former is clearly an RC
>> issue, it looks like the latter is indeed not considered as such.
>> Anyways I don't mind if these changes are not incorporated.
>
> I see your point, and I appreciate the fixes in general, but I guess
> the release team would be a bit overwhelmed if we included fixes for
> all not-yet-reported policy violations :)
>
> BTW: When I try to build twice I get (after having added
> o2monitor/o2hbmonitor to debian/clean already):
>
>  dpkg-source -b ocfs2-tools-1.6.4
> dpkg-source: info: using source format `3.0 (quilt)'
> dpkg-source: info: building ocfs2-tools using existing 
> ./ocfs2-tools_1.6.4.orig.tar.gz
> dpkg-source: warning: ignoring deletion of file o2monitor/o2hbmonitor
> dpkg-source: warning: executable mode 0755 of 'vendor/common/o2cb.init' will 
> not be represented in diff
> dpkg-source: info: local changes detected, the modified files are:
>  ocfs2-tools-1.6.4/mkfs.ocfs2/mkfs.ocfs2.8
>  ocfs2-tools-1.6.4/o2image/o2image.8
>  ocfs2-tools-1.6.4/vendor/common/o2cb.init
>
> which doesn't exactly match the list of files you added to d/clean :)

The *.8 manpages are present in the original tarball, but just modified
by the build, so I put them in ignore in d/source/options rather than
d/clean. But they could indeed rather have been put in d/clean.

Anyways, let's just forget about that packaging part, I don't want to
argue.

>> >> +  * Remove popen2-ignore-depreciation-warn.patch: DeprecationWarnings 
>> >> are now
>> >> +    ignored by default in Python 2.7, and this patch messes things up
>> >> +    (Closes: #682517, LP: #923754)
>> > Hm, and what happens if ocfs2console is run with python2.6?
>> I did not consider that possibility, given that the app is run with a
>> #!/usr/bin/python shebang, and that the /usr/bin/python symlink is not
>> managed by an alternative. Am I missing something?
>
> No, it was me who was missing something (python details still confuse
> me), thanks for the enlightenment!
>
> And since the package has Depends: ...,  python (<< 2.8), python (>= 2.7), ...
> we also have a safety net here.

Then if you are convinced that this is the right fix don't hesitate to
go with your own NMU, I cannot upload mine anyways since I'm a DM.
Otherwise if it saves you from some work I am also willing to regenerate
my nmudiff.

Thanks,

-- 
 .''`.    Sébastien Villemot
: :' :    Debian Maintainer
`. `'     http://www.dynare.org/sebastien
  `-      GPG Key: 4096R/381A7594

Attachment: pgppwUHl2Z4eg.pgp
Description: PGP signature

Reply via email to