[19:06] <thomi> good morning.
[19:06] <thomi> wgrant: or cjwatson, I wonder if I could bug one of you to take a peek at https://code.launchpad.net/~thomir/launchpad/devel-add-bug-header/+merge/245921 please?
[19:50] <wgrant> thomi: Morning.
=== Peng__ is now known as Peng
[21:18] <thomi> hey wgrant
[21:19] <wgrant> Hi
[22:18] <thomi> wgrant: I ran those tests, and I'm getting a lot of test failures, but most of them seem unrelated to my change: http://pastebin.ubuntu.com/9905467/
[22:18] <thomi> the first failure is clearly related (and fixed now)
[22:19] <thomi> wgrant: but "ProgrammingError: column sprint.is_physical does not exist" seems like some sort of sql schema mismatch?
[22:19] <thomi> do I have to do something to keep the schema up to date?
[22:19] <wgrant> thomi: make schema
[22:20] <thomi> hmnmm, ok
[22:20] <thomi> wgrant: there's no CI set up for launchpad branches?
[22:20] <wgrant> thomi: Not until landing.
[22:20] <thomi> ahh ok
[22:20] <wgrant> The test suite is very expensive to run.
[22:20] <wgrant> Doing it automatically over everything would require a lot of hardware.
[22:21] <thomi> hmmm
[22:21] <thomi> hardware is cheap :D
[22:24] <wgrant> Tell that to my 7 year old DB servers :)
[22:24] <blr> wgrant: any more news om the SSDs?
[22:24] <blr> *on
[22:25] <wgrant> The ticket for the new servers is in the queue.
[22:26] <blr> is that analogous to "the cheque's in the mail"? :)
[22:27] <wgrant> Pretty much, I think.
=== Peng__ is now known as Peng
[23:28] <thomi> wgrant: I'm still having issues getting a clean test run... the rocketfuel setup should have installed  all the build-deps, right? How do I make sure that's kept up to date in my lpdev lxc container?
[23:28] <thomi> For example, I seem to be getting errors about Gtk...
[23:30] <thomi> hmmm "RuntimeError: Gtk couldn't be initialized" - probably because it's running in an lxc container?
[23:37] <wgrant> thomi: Ah yes, the JS tests require DISPLAY to be set. Try it in xvfb-run
[23:38] <wgrant> But you probably haven't broken those :)
[23:38] <thomi> ok
[23:39] <thomi> down to one doctest failure
[23:39] <wgrant> Excellent.
[23:39] <thomi> hmmm, which may have been caused by my editor stripping whitespace
[23:39] <thomi> *sigh*
[23:40] <thomi> have I mentioned how much I dislike doctests?
[23:40] <wgrant> We all loathe them :)
[23:40] <wgrant> There is a universal authorisation to shoot them on sight.
[23:41] <thomi> yup... looks like the doctest fails due to a whitespace change...
[23:41] <wgrant> What sort of whitespace?
[23:42] <thomi> can I get 'bzr diff' to print whitespace as tokens?
[23:42] <wgrant> Not sure the commandline diff can do that.
[23:43] <wgrant> Pipe it through tr or use a graphical one?
[23:43] <thomi> wgrant: looks like a trailing space
[23:44] <wgrant> Sounds easy enough.
[23:44]  * wgrant lunches.