promote to main

[Martin Pitt]

Binary package hint: elisa

Scott asked me to take a look at elisa for main promotion.

[Philippe Normand (philn)]

Great!

Feel free to ask any question and/or come over #elisa on IRC (freenode)

[Matthew McGowan (mmcg069)]

It would be good also so we ensure that python-pigment and libpigment are up do date. Seems to me people would quite like to have these up to date for development of their own flashy apps.

[Matthias Klose (doko)]

- can't find the MIR
- package dependencies: please avoid python-ctypes and python-pysqlite2 found in python2.5 (same for the -elementtree recommends)
- mir's for python-pigment and libpigment are missing (didn't look for mir's for the recommends).

[Martin Pitt (pitti)]

- current version does not work at all on hardy; promotion pending a fix (which is already in upstream SVN according to Loic; he will update soon)
- dependency problems reported to Debian (http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=46672)

[Philippe Normand (philn)]

Hi!

I worked on some unofficial packages for Hardy. They are against recent (the upcoming 0.3.4 release) SVN snapshots of both Pigment and Elisa.
The comments above (about python 2.5 dependencies) have been taken in account.

https://edge.launchpad.net/%7Ephilippe-normand/+archive

The gusty packages there need an update too, please don't bother about them, I think the most interesting one for you are the Hardy ones :)

[Philippe Normand (philn)]

Some heads up, the Pigment and Elisa Debian packages have been uploaded to experimental by Loïc.. We plan an elisa 0.3.5 bugfix release next monday.

[Alessandro Decina (alessandro.decina)]

Elisa is now in hardy. In the meantime we released a new bugfix version, 0.3.5, which is going to be packaged by lool soon.
Is there anything else that we can do? Maybe help writing the MIR?

[Martin Pitt (pitti)]

Helping with MIR writing is always appreciated, of course.

Elisa works now, so it's finally eligible for MIR.

[Martin Pitt (pitti)]

https://wiki.ubuntu.com/MainInclusionReports/Elisa

Looks good to me, pending approval of pigment-python.

[Martin Pitt (pitti)]

https://wiki.ubuntu.com/MainInclusionReports/PigmentPython

Looks good to me, approved. Promotion pending MIR of pigment.

[Martin Pitt (pitti)]

plugins-bad and plugins-good are covered in https://wiki.ubuntu.com/MainInclusionReports/Elisa.

[Martin Pitt (pitti)]

To whoever added the -ugly task: Do we really need this in main, too? The package name doesn't suggest good things...

[Martin Pitt (pitti)]

https://wiki.ubuntu.com/MainInclusionReports/Pigment

pigment approved. I promoted the lot.

[Martin Pitt (pitti)]

I also put elisa into the 'supported' seed.

[Alessandro Decina (alessandro.decina)]

Thanks for looking at this.

Regarding elisa-plugins-ugly, we don't strictly depend on it. Our policy for -ugly is the same as for gstreamer0.10-plugins-ugly and gstreamer0.10-plugins-ugly is in universe so I guess elisa-plugins-ugly can stay in universe.

[Martin Pitt (pitti)]

Ok, thanks for the followup. Let's leave it like it is for now, then.

[Martin Pitt (pitti)]

plugins-good needs pylirc.

[Martin Pitt (pitti)]

pylirc is a tiny and harmless python binding to liblircclient, which we have in main for ages. Approved and promoted.

[Martin Pitt (pitti)]

Reopening -plugins-bad: this needs the following additional packages: python-bluez python-coherence python-daap. They need MIR review, or we move the GUI to plugins-good and drop -bad to universe again.

[Loïc Minier (lool)]

Upstream would rather not split the package to not confuse users; they prefer we demote the Depends: on python-bluez, -coherence, and -daap to Recommends.

-bluez is not too big an issue, -daap and -coherence are a bit more of a concern.

[Alessandro Decina (alessandro.decina)]

It's ok for us to move -bluez and -daap to Recommends.

Regarding coherence, although we could live with it being in universe I think that it should be considered for inclusion in main. Rhythmbox has a powerful upnp plugin that depends on coherence too. The library is well written and relatively small. Upstream is very active and I'm sure that they would be happy to help with the MIR if needed.

[Martin Pitt (pitti)]

Hi,

Alessandro Decina [2008-03-19 13:35 -0000]:
> Regarding coherence, although we could live with it being in universe I
> think that it should be considered for inclusion in main. Rhythmbox has
> a powerful upnp plugin that depends on coherence too. The library is
> well written and relatively small. Upstream is very active and I'm sure
> that they would be happy to help with the MIR if needed.

Maybe, but we have to consider that we are deep in feature freeze
mode, and shouldn't do intrusive changes in Hardy any more.

Another idea that came up was to split the UI into its own binary
package and leave -plugins-bad with the other plugins in universe.
Would that work for you?

[Alessandro Decina (alessandro.decina)]

On Wed, Mar 19, 2008 at 3:24 PM, Martin Pitt <email address hidden> wrote:
> Maybe, but we have to consider that we are deep in feature freeze
> mode, and shouldn't do intrusive changes in Hardy any more.
Right.

> Another idea that came up was to split the UI into its own binary
> package and leave -plugins-bad with the other plugins in universe.
> Would that work for you?
I would really like to avoid this as it would confuse our users. I'd
much rather demote the missing deps
to recommends for now if it's ok for you.

[Martin Pitt (pitti)]

Hi,

Alessandro Decina [2008-03-19 16:22 -0000]:
> I would really like to avoid this as it would confuse our users. I'd
> much rather demote the missing deps
> to recommends for now if it's ok for you.

Yes, it is. python-bluez should get a MIR, that sounds harmless
enough. I haven't heard about the others so far. If they are truly
harmless (i. e. bindings to libraries in main), we can promote them,
otherwise we shuold find a way to keep them in universe without
changing too much.

Thank you!

Martin

--
Martin Pitt | http://www.piware.de
Ubuntu Developer (www.ubuntu.com) | Debian Developer (www.debian.org)

My 5 today: #200832, #197777 (jockey), #195706 (apport), #198341
(jockey), #198433
Do 5 a day - every day! https://wiki.ubuntu.com/5-A-Day

[Alessandro Decina (alessandro.decina)]

On Wed, Mar 19, 2008 at 5:44 PM, Martin Pitt <email address hidden> wrote:
> Yes, it is. python-bluez should get a MIR, that sounds harmless
> enough. I haven't heard about the others so far. If they are truly
> harmless (i. e. bindings to libraries in main), we can promote them,
> otherwise we shuold find a way to keep them in universe without
> changing too much.
Coherence is a pure python library not bindings. It depends on
python-louie and python-configobj that are in universe.
Rhythmbox recommends python-coherence for upnp support. FWIW, we know
the library pretty well and can guarantee on its quality.

python-daap is a small (around 500 lines of code or so) py file plus a
C python extension to compute a "modified" md5 hash used by iTunes.
The same C code is in rhythmbox as well afaik.

I'm fairly sure that they are safe to be included in main. On the
other hand I understand that we are really close to the release so if
it's too late we can live without them for this cycle.

[Martin Pitt (pitti)]

Alessandro Decina [2008-03-19 17:24 -0000]:
> Coherence is a pure python library not bindings. It depends on
> python-louie and python-configobj that are in universe.
> Rhythmbox recommends python-coherence for upnp support. FWIW, we know
> the library pretty well and can guarantee on its quality.

Hm, this sounds too much of a new feature to me TBH.

> python-daap is a small (around 500 lines of code or so) py file plus a
> C python extension to compute a "modified" md5 hash used by iTunes.
> The same C code is in rhythmbox as well afaik.

That sounds safe for main.

Thank you for the comments!

[Martin Pitt (pitti)]

pybluez: good maintenance in Debian, no bugs, relatively harmless Python binding for the BlueZ stack (which we have in main). Approved and promoted.

[Martin Pitt (pitti)]

I just synced a new python-daap from Debian to get some packaging improvements.

Package looks fine, and we have similar code in Rhythmbox already. Approved for promotion to main. I'll actually promote it once the new version built and published.

[Martin Pitt (pitti)]

That leaves coherence, which should be demoted to a Suggests: for Hardy (or the package needs to be split).

[Loïc Minier (lool)]

@Martin: I've uploaded elisa-plugins-bad 0.3.5-1ubuntu2 with the suggests.

[Martin Pitt (pitti)]

python-daap promoted.