[07:43] <dholbach> good morning
[08:20] <geser> good morning
[10:24] <obounaim> Hello
[10:24] <obounaim> what are the disadvantages of using udd?
[10:25] <geser> there aren't really any (if the packaging branch is up-to-date), it's more a preference if one uses it or not
[10:28] <obounaim> I heard that udd have some problems with quilt patchs is that right?
[11:55] <tumbleweed> geser: well, sure there are. Checkout time vs a local mirror can be quite high (but once you have a local checkout, it's cheaper next time...)
[11:56] <tumbleweed> obounaim: yes, tehre are some issues with quilt patches. I thought that had improved in the last year or so, but can't remember the details
[11:56] <tumbleweed> play with it yourself, and see how it works for you
[12:31] <obounaim> How changes made in package branch (UDD) find its way to the Ubuntu archive.
[12:32] <Laney> someone uploads them as normal
[12:32] <Laney> by building a source package and uploading it to the correct location
[12:36] <obounaim> I mean when somebody uses (UDD) to make changes in a package. How the source package will be made available in the Ubuntu archive.
[12:36] <obounaim> source and binary packages
[12:37] <maxb> manual uploading same as any other upload
[12:37] <Laney> that's what I answered
[12:38] <maxb> yes. obounaim just didn't seem to understand :-)
[12:38] <Laney> heh
[12:44] <obounaim> No, I mean how changes made in a package branch using the UDD will be available to everybody out there, is the UDD importer responsible for this?
[12:44] <obounaim> When we make changes in a package branch we don't upload a source package.
[12:44] <Laney> a developer has to take the branch and turn it into a source package and upload it
[12:45] <obounaim> So it is not done automatically?
[12:45] <maxb> correct
[12:47] <Laney> you always want some deliberate action to trigger a build - it's not true that any arbitrary revision of a branch should be given out to users
[12:48] <Laney> currently that's uploading a source package, but in the future it could be pushing a signed tag or issuing a special Launchpad command
[12:48] <Laney> still will be an explicit request though
[12:48] <jtaylor> is this an armhf libc issue? https://launchpadlibrarian.net/123081185/buildlog_ubuntu-raring-armhf.gluegen2_2.0-rc11-1~exp1_FAILEDTOBUILD.txt.gz
[12:49] <jtaylor> scroll right down
[12:49] <jtaylor>  /usr/include/stdc-predef.h:30:26: fatal error: bits/predefs.h: No such file or directory
[12:50] <obounaim> Laney, maxb. Thank you very much
[12:54] <Laney> jtaylor: still exists according to Contents-armhf.gz
[12:58] <jtaylor> strange
[13:07] <jtaylor> micahg: anything more you want tested for bug 1074765
[14:56] <micahg> jtaylor: anything that's a Reverse-Depends
[14:56] <jtaylor> there are only reverse recommends and suggests
[14:57] <micahg> python-surfer, python-polybori?
[15:04] <jtaylor> those should not be depends
[15:04] <jtaylor> but I'll check them
[15:06] <micahg> jtaylor: if there's a bug in the script, let's get it fixed
[15:06] <jtaylor> no its a bug in the packages
[15:06] <jtaylor> the rdeps are all just embedded shells
[15:06] <jtaylor> all but accerciser just fall back to regular python if ipython is not there
[15:07] <jtaylor> fyi 0.13 is backported to precise since it existed in a ppa (~60 users)
[15:14] <jtaylor> micahg: checked both, surfer is quantal only (and broken) the backport is a nop there
[15:15] <jtaylor> polybori is already broken in precise and the functionality is completely optional
[15:16] <micahg> yade, nipype, python-pytango, connectomeviewer, statsmodels, rabbitvcs-core, those also appears as Depends, can you please comment in the bug as to the state of all of them
[15:16] <jtaylor> build depends
[15:16] <jtaylor> do those have to work too?=
[15:18] <jtaylor> because yade build is broken by it
[15:18] <jtaylor> but nobody wants to build that anyway it needs like 8gb of ram ._.
[15:19] <micahg> they're supposed to all work, if the version in precise is already broken, then a backport is fine
[15:33] <jtaylor> got to file some bugs, far to strict dependencies on ipython everywhere
[15:33] <jtaylor> rabbitvcs does not even expose the option unless you hack the code but its a depend ...