Bug#657393: RFS: skstream/0.3.6-1 [ITA] -- IOStream C++ socket Library
On 01/26/2012 06:49 AM, Jakub Wilk wrote:
>> . * new maintainer (closes: #653977)
> This is a bit misleading. It would normally interpret such item as
> "I set myself as Maintainer". But this is not what happened here:
> you set Debian Games Team as maintainer, and added yourself to
> Uploaders. I think this should be written explicitly in the
Fair enough. Modified.
>> * renamed binary packages due to SONAME change
> But here are reverse-dependencies of the old binary package. Which
> means that uploading this to unstable starts a transition. What
> this discussed with the release team? It probably should, even
> though the number of involved packages is small.
> That said, the best moment to talk to the release team would be
> after the package has been thoroughly reviewed (thus: not yet).
The old and new library packages are parallel-installable. I consider
this a feature, since the library is a part of an MMORPG stack, and I
anticipate a newer client app revision getting in to Debian long
before a new server app, so the coexistence of both old and new
SONAMEs will be required, at least for a little while.
But, as you say, this will need to be discussed with the release team
after this package (and other upgraded packages in the stack) has been
>> * moved to debhelper 8
> What does this mean?
It means converted to use the dh sequencer from a hand-rolled
debian/rules script. I have adjusted the debian/changelog wording to
> I see that you rewrote debian/rules from scratch, apparently
> introducing regressions... Is that a part of "moved to debhelper
> Does you new d/rules support DEB_BUILD_OPTIONS=noopt like the old
> one did? Are you sure that there are no other regressions?
I have the greatest confidence that the dh sequencer support Debian
policy much better than the previous hand-rolled debian/rules script.
I have confirmed that the new debian/rules does indeed support
DEB_BUILD_OPTIONS=noopt and DEB_BUILD_OPTIONS=nostrip. It is true
that if you build with the noopt option you will get warnings about
the ABI changing. That is a feature, since a changed ABI can cause
hard-to-find runtime misbehaviour and at least a build-time warning
should be issued. Most packages handle this by silently ignoring the
ABI change. I can modify the debian/rules to not issue the warning if
that's the preferred Debian way.
I would appreciate an explicit list of any apparent regressions, since
they aren't apparent to me from the build logs or runtime testing of
>> * added debian/symbols file
> This looks a bit suspicious. Symbols that exist only on amd64? I
> seriously doubt it...
Right, I loosened the symbols file so it should not fail on arches I
have not tested on. Since the symbol differences were all the result
of arch-specific std::size_t, this should not in fact cause unintended
side effects. Fixed.
>> * debian/copyright: convert to DEP-5 format
> I see no such changes to debian/copyright in my debdiff.
Oooh, workflow problems. I grabbed a newer changelog than the one on
mentors.d.n to attach to the RFS. Well, taken care of now by
including the debian/copyright change.
> You converted the package to source format 3.0 (quilt), but this is
> not documented in the changelog.
Gah, don't know how I missed that one. Fixed.
> Why is the patch name 0001-gcc-4.4.patch if the description is
> "fixes compilation errors with GCC **3.3**" (emphasis mine).
Typo. The "3" and "4" keys are too close together on the keyboard.
I have made the above documented changes and uploaded a new source
package to mentors.debian.net. It can be obtained from here:
Here is the new changelog.
skstream (0.3.8-1) unstable; urgency=low
* new maintainer: Debian games team (closes: #653977)
- added myself as uploader
* new upstream release
* renamed binary packages due to SONAME change
* converted packaging to use dh sequencer
* converted packaging to 3.0 (quilt) format
* updated Standards-Version to 3.9.2 (no changes required)
* added debian/symbols file
* added VCS- fields to debian/control
* debian/copyright: convert to DEP-5 format
Stephen M. Webb <email@example.com>