=== charles_ is now known as charles
[18:18] <doug5> balloons, do you know when veebers is around more or less?
[18:42] <balloons> doug5, yes here in another hour he'll be about
[18:42] <doug5> ok cool
[18:43] <balloons> doug5, he wanted to talk to you about your mp
[18:45] <doug5> we have to fix the doc
[19:46] <doug5> veebers, hello!
[19:47] <veebers> hey doug5 hows it going?
[19:47] <doug5> veebers, not bad, thx...you?
[19:48] <veebers> doug5: a little tired but otherwise good :-)
[19:53] <doug5> veebers, I have updated the mp, but I'm unsure what to do with the docstrings
[19:54] <veebers> doug5: oh cool, I'm just in a meeting at the moment. I'll take a look after that and get back to you :-)
[19:54] <doug5> veebers, sure, np :)
[21:58] <thomi> barry: got a second?
[21:58] <barry> thomi: i got a third and fourth too
[21:58] <thomi> great - you might need them :D
[21:59]  * barry hides
[21:59] <thomi> barry: I made a quick and dirty package for selenium here: https://launchpad.net/~canonical-platform-qa/+archive/ubuntu/selenium/+packages
[21:59] <thomi> and the problem I'm having is that the distropatch doesn't apply
[21:59] <thomi> it's listed in d/patches/series
[21:59] <thomi> and the patch itself seems find
[21:59] <thomi> *fine
[21:59] <thomi> but in the binary package, the patch doesn't get applied
[21:59] <thomi> any ideas?
[22:00] <barry> thomi: in the package source, you can `quilt push -a` and `quilt pop -a` without error or warning?  sometimes (i forget in which instances) if a patch has fuzz, quilt will not be happy
[22:00] <thomi> hmm
[22:01] <thomi> no, I can't
[22:01] <thomi> ok, so the patch needs changing. how do I do that?
[22:01] <barry> thomi: here's how i do it:
[22:02] <barry> `quilt push -a` until it stops with an error
[22:02] <barry> `quilt push -f` to "force" the broken patch to apply
[22:02] <barry> fix the patch (e.g. <edit> <edit> <edit>)
[22:02] <barry> `quilt refresh`
[22:02] <barry> `quilt pop`
[22:02] <thomi> which file am I editing?
[22:02] <barry> `quilt push` should now have no errors, if it does, rinse/repeat
[22:03] <barry> thomi: the files in the quilt patch that fail to cleanly apply
[22:03] <barry> quilt should tell you which ones that is
[22:03] <barry> and since patch(1) is used,you'll have .rej files for those
[22:03] <thomi> hmm, I see a .rej file, but I'm not sure how to fix it
[22:03] <barry> thomi: that i can't help you with since it's specific to the patch.  it's like resolving a conflict in a $vcs
[22:04] <thomi> I assume it's because the offsets in the .patch file are wrong
[22:04] <barry> could be
[22:04] <barry> generally go through each hunk and manually apply them to the file
[22:04] <thomi> how do I know what the correct offsets should be?
[22:04] <thomi> ahh ok
[22:04] <thomi> then how do I get those changes out of the source tree and back into a patch?
[22:05] <barry> you don't have to.  once the source tree is happy, `quilt refresh` will fix the top quilt patch on the stack
[22:05] <thomi> ... how does it know how to do that? magic!
[22:05] <barry> yep
[22:05] <barry> `man quilt` talks about this i think
[22:09] <thomi> barry: awesome, thanks
[22:09] <barry> thomi: good luck!
[22:10] <thomi> thanks. last thing I need to do now is get it building python2 & 3 packages, which shouldn't be too hard
[22:11] <barry> thomi: esp. if it's setup.py based.  pybuild ftw!
[22:11] <thomi> yup
[22:25] <rpadovani> balloons, I don't want to be herald of bad news, but your fix for calculator app didn't work :S
[22:25] <rpadovani> https://code.launchpad.net/~rpadovani/ubuntu-calculator-app/bigNumber150122/+merge/248183
[22:26] <rpadovani> balloons, TBH I'm not sure the bug is in autopilot tests, but maybe (and I underline maybe because I'm pretty confused) in the packaging / files inclusion. Sometimes seems the external file isn't loaded
[22:27] <balloons> rpadovani, :-(
[22:28] <balloons> rpadovani, it definitely helped fix the tests locally for me.. as for bignumber, :-( I thought for sure i was running big number though
[22:29] <rpadovani> balloons, it works also for me, this is the strange, and this is why I'm thinking about package issues
[22:29] <rpadovani> balloons, look to this
[22:29] <rpadovani> http://91.189.93.70:8080/job/generic-mediumtests-vivid/921/testReport/junit/ubuntu_calculator_app.tests.test_main/MainTestCase/test_operators_precedence/
[22:29] <rpadovani> TypeError: Cannot call method 'format' of undefined
[22:30] <rpadovani> seems mathks isn't loaded
[22:30] <rpadovani> buttons don't count here
[22:34] <balloons> rpadovani, ohh that's not the same issue.. the app neve rlaunched
[22:35] <balloons> rpadovani, I re-ran, maybe it was a fluke
[22:36]  * rpadovani cross fingers
[22:37] <thomi> barry: are you able to take a peek at lp:~thomir/+junk/selenium-packaging and advise? It seems like pybuild doesn't honor the data_files kwarg in setup.py?
[22:37] <thomi> the build fails with 'error: can't copy 'py/selenium/webdriver/firefox/x86/x_ignore_nofocus.so': doesn't exist or not a regular file', but that file exists in the source tree, and is referenced in setup.py's data_files section
[22:37] <thomi> so I think it should work?
[23:16] <rpadovani> balloons, this time 4 fails instead of 5, but 3 out of 4 are different tests :D
[23:16] <rpadovani> https://code.launchpad.net/~rpadovani/ubuntu-calculator-app/bigNumber150122/+merge/248183
[23:16] <rpadovani> balloons, actually, it want to was a :S, not a :D
[23:16] <rpadovani> I'm quite confused