=== fuzzy is now known as Fuzai === Fuzai is now known as Ponyo [04:49] Is anyone available to review my trivial PR to the names package? https://github.com/juju/names/pull/28 [05:05] Also, http://reviews.vapour.ws/r/158 which will depend on the names package PR too === uru_ is now known as urulama === benji is now known as Guest37515 === uru_ is now known as urulama [13:04] thanks mgz [13:05] jcw4: no probs [13:07] anastasiamac: congrats on your first blocker bug fix yesterday :) [13:10] jcw4: it's not quite done yet sadly :) [13:10] anastasiamac: poke poke :) [13:10] it was not me fixing it tho :-) I was just a PA that typed it [13:10] anastasiamac: lol [13:10] mgz: it's still not working? [13:11] mgz: was that your ordering bugfix? [13:12] jcw4: no, that's amusingly an unrelated regression that was masked by the fact that other bug was already causing the test job to fail [13:12] mgz: I see. [13:13] once both land the ppc64el job might be happy again [13:13] anastasiamac: my first big PR was step by step laid out for my by thumper - I was pretty much a PA too. [13:13] mgz: hopefully [13:17] TheMue: I responded to your comments on https://github.com/juju/names/pull/28 [13:33] mgz: wrt. to the un-exercised error case you mentioned in the review... the only error that would get returned is if the mgo iterator returns an error on closing in a defer block (state/state.go:1620) [13:33] mgz: I'm not quite sure how to cause that error [13:34] mgz: suggestions? or should I make a fake ActionReceiver and force a bogus error return there just to exercise that line? [13:52] jcw4: you can fake, but if it's that edgy it's likely not worth bothering with [13:56] okay; thanks mgz - shall I drop that issue? [13:59] mgz: dropped - as soon as I get my names PR landed I'll update this one with the dependencies.tsv change. Should I take "looks great overall" as an LGTM or wait for an official one? :) [14:07] jcw4: I may bug someone else as well for the shipit [14:13] mgz: ta [14:19] axw, anastasiamac: https://code.google.com/p/go/issues/detail?id=8654 [14:23] mgz: yep [15:30] 346022 [15:30] 250720 [15:30] 962491 [15:30] 279768 [15:30] 376365 [15:30] Erm [15:30] huwshimi: ok [15:53] axw: bug 1379380 - can you go/gccgo compile and compare goyaml.Marshal(map[string]string{{"pew": "pew\npew\n"}}) - something very odd is happening... [15:53] Bug #1379380: Test failure on gccgo in RelationGetSuite [15:53] mgz: axw fwiw, I was seeing the same error with go 1.3.3 recently [15:57] looking [15:57] anastasia had this issue before, but I couldn't repro [15:58] with gccgo? maybe it's some other thing, like wrong goyaml version [15:58] yes, with gccgo [15:59] testing again now anyway [16:00] gccgo 4.9.1 vs. gc 1.3 makes no diff to me [16:00] worth asking anastasiamac what goyaml she hath [16:00] jcw4: if you can repo, what goyaml? [16:00] yes, I was going to ask to do godeps, but then got sidetracked [16:01] mgz: I'd have to reboot my mac into ubuntu since it was on an experimental dual boot instance [16:01] mgz: but I'm pretty sure it was a stock godeps -u dependencies.tsv version of goyaml [16:01] jcw4: 's not urgent === robbiew is now known as robbiew-afk [16:38] mgz, axw my goyaml was at a different revision - confirming the goyaml suspicion