[08:07] <dholbach> good morning
[08:09] <micahg> achiang: this was a little off topic for #launchpad, so I moved here, anyways, you can get sponsored uploads here: http://ubuntu-dev.alioth.debian.org/cgi-bin/ubuntu-sponsorships.cgi
[08:11] <achiang> micahg: whizbang. thanks
[08:11] <achiang> and... bed time. :)
[08:11] <micahg> for me as well
[15:13] <alo21> hi all... I am merging gnome-photo-printer, and as in the changelog, the bug #662978 has been closed. But when I downloaded the package, I can still see that bug
[15:14] <alo21> sorry the bug is: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=662978
[15:14] <alo21> can I fix the bug while the merging?
[15:17] <TheLordOfTime> alo21, actually, the bug's expired.
[15:17] <TheLordOfTime> oh, the debian bug is open.
[15:18]  * TheLordOfTime leaves this be and waits for MOTUs
[15:19] <TheLordOfTime> speaking of MOTUs, got an unrelated question for them, but it can wait :P
[15:20] <alo21> hmm.. ok. thanks for your support
[15:21] <Adri2000> alo21: yes you can. then send your fixes to the debian maintainer / bug report
[15:23] <Adri2000> from what I can see, the only .desktop thing not fixed in -1.2 is that the Icon line should not contain the file extension
[15:25] <alo21> Adri2000, yes... so I can merge the package with my edited .desktop file, and then send the patch in debian.
[15:26] <alo21> Adri2000, why the bug is in the changelog?
[15:29] <Adri2000> the changelog says "adresses: 662978" which probably explains why the bug wasn't automatically closed. maybe the uploader did that because he didn't fix the bug completely. so the real question is why he didn't fix it completely :)
[15:31] <alo21> Adri2000, have I put in the changelog that I 'really fixed the bug'?
[15:32] <alo21> Close: #662978
[15:32] <TheLordOfTime> I've seen it written as "Closes: #number"
[15:32] <TheLordOfTime> not sure how strict their BTS is on looking for "Closes:"
[15:33] <Adri2000> I'm not sure it works without the 's' indeed. anyway that is only for debian so it won't have any impact for an ubuntu upload
[15:34] <TheLordOfTime> mhm.
[15:34] <Adri2000> the syntax for closing a bug in ubuntu is "LP: #nnnnnn"
[15:34] <TheLordOfTime> indeed.
[15:35] <alo21> Adri2000, a part 's', which I will include. Have I to say the fix explicitly ?
[15:37] <Adri2000> alo21: for your ubuntu upload you can write "fixed this and this in the desktop file" and you may indeed mention the debian bug report (using the Closes: syntax or not won't change anything here).
[15:38] <Adri2000> then when you forward your fixes to debian, if you provide a complete debdiff for a debian upload (instead of just the patch corresping to your fixes), you will have to use the correct syntax to close the bug
[15:38] <Adri2000> corresponding*
[15:39] <alo21> ok
[15:52] <psusi> someone tell me I'm not losing my mind... dpkg --compare-versions '2 > 1' complains that --compare-versiosn needs 3 arguments... umm... I'm giving it 3...
[15:53] <psusi> ohh... nevermind.. damnit... it's *also* interpreting the quotes...
[15:58] <micahg> psusi: that's one argument
[15:58] <psusi> yea, had to escape rather than quote to get through the shell
[16:15] <psusi> did I see somewhere that this last dev cycle incoming packages started automatically being sent to -proposed?
[16:16] <psusi> does that mean they are built against other packages in -proposed? or are they still built against what's in the release pocket?
[19:28] <jtaylor> if I backport a new soversion library and then a rdep of it
[19:29] <jtaylor> will the rdep backport build against the library backport ?
[19:31] <jtaylor> psusi: they are built against proposed
[19:32] <psusi> won't that lead to breakage if the app is moved to the release pocket but not the lib it was built against?
[19:32] <jtaylor> if so then there should be versioned depends
[19:33] <psusi> well yea,  but that's what I'm talking about... it will depend on the newer version of the lib and thus, won't be installable
[19:33] <Laney> jtaylor: you'll get the release versions of stuff
[19:33] <jtaylor> then it won't migrate
[19:33] <Laney> unless you version the BD in which case you'll get uninstallable BDs
[19:33] <Laney> you can't win :>
[19:33] <jtaylor> so don't do it?
[19:33] <psusi> ohh, the command to migrate checks the deps and bails if they aren't already moved?
[19:33] <jtaylor> psusi: it checks installability
[19:34] <jtaylor> just like in debian
[19:34] <psusi> got ya
[19:34] <jtaylor> but without rc bugs as blockers
[19:34] <jtaylor> at least currently
[19:34] <Laney> (hassle ∞ to get that fixed)
[19:35] <micahg> wait, why wouldn't versioned BDs work?
[19:35] <jtaylor> the backports can't depend on backports bug?
[19:35] <Laney> same reason they always dont
[19:36] <micahg> that's only alternate BDs
[19:36] <Laney> nope
[19:38] <micahg> hrmm....are you referring just to backports, or in general?
[19:39] <jtaylor> just backports
[19:39] <Laney> just the usual backports bug
[19:39] <micahg> right, because sbuild skips the versions...
[19:40] <micahg> or rather LPs implementation
[19:42] <Laney> I wonder at which level you'd fix this
[19:42] <Laney> inject a preferences file to ignore NotAutomatic when the chroot is created?
[19:44] <ajmitch> is a separate chroot used for the backports pocket?
[19:44] <Laney> I dunno how it works
[19:44] <Laney> but if not it still shouldn't be harmful to have a preferences file for backports all the time, should it?
[19:45] <ajmitch> no, I think that was suggested in the bug
[19:45] <ajmitch> slightly easier than un-forking sbuild
[19:45] <Laney> that's a larger project, yeah
[19:46] <Laney> so find out where the script to make chroots is... it's not in lp-buildd
[19:47] <ajmitch> maybe scripts/ftpmaster-tools/manage-chroot.py in lp
[19:47] <ajmitch> which is a wrapper around soyuz chrootmanager
[19:47] <Laney> don't have lp checked out here
[19:48] <Laney> can you see if that's where it mangles sources.list when building for different pockets?
[19:49] <ajmitch> I can try, it may drive me mad in the process
[19:49] <Laney> should call 'override-sources-list'
[19:49] <Laney> bah, I'll start a branch of lp:launchpad :P
[19:49]  * ajmitch greps the whole lp tree for that
[19:50] <Laney> need to cook dinner anyway
[19:50] <ajmitch> no mention of override-sources-list in lp:launchpad
[19:51] <Laney> must be called another way
[19:52] <Laney> hmm
[19:53] <Laney> I see
[19:54] <ajmitch> found something useful?
[19:54] <Laney> that script is called from within launchpad-buildd
[20:00] <micahg> laney: infinity was going to do something like that (inject a preference to ignore notautomatic for backports)
[20:00] <Laney> yeah
[20:01] <Laney> I am trying to figure out enough of launchpad-buildd to see where you'd do it