[11:06] morning [11:06] Morning bac. [11:06] gary_poster, Am I right in thinking that today's meeting is going to be at 12:10 UTC? (i.e. just over an hour) [11:13] * gmb assumes "yes", goes for lunch [11:58] gmb, yes :-) [11:58] benji, bac, call in 12 minutes, just to clarify because of the time zone changes lately [11:59] 11 minutes :-P [11:59] yep, thanks for the hint [11:59] cool welcome [11:59] ack [11:59] * gary_poster sees if there is anything worth updating... [12:00] * gary_poster wonders if printing works now... [12:08] bac benji frankban gmb call in 2 [12:12] Waitimg for a useable browser [12:14] gary_poster, Can you invite non-canonical me to the hangout? [12:14] It seems something is preventing the Google Talk plugin from launching [12:14] my phone might be a better bet. [12:14] gmb, address completion no longer works for me :-/ [12:14] what's your non-canonical address? [12:14] gary_poster, graham.binns@gmail.com [12:15] gmb invited [12:33] Right [12:33] Out tomorrow, yes. [12:34] Also out Wednesday morning for an eye exam (though hopefully I'll just get the yellow dye, not the pupil dilation stuff). [12:34] frankban, I'll be happy to pair with you, but I need to reboot and possibly roll back the latest updates. [12:35] gmb: np, ping me when you are ready [12:35] Will do. [13:02] Argh. [13:02] * gmb -> rolling back to latest snapshot. BBIAB [13:11] frankban, Okay, I'm ready to go. I'm broadcasting the terminal here: (export OTHER=graham REMOTE=chinstrap.canonical.com; trap reset 0 1 2 3 15; stty raw -echo isig && ssh chinstrap.canonical.com "cat /tmp/$OTHER-term") [13:11] I'll start a hangout [13:14] frankban, https://plus.google.com/hangouts/extras/canonical.com/mostly--zopeless# [13:50] frankban, could you please verify that you now have the task to do your self evaluation on allhands? If you don't I should ask dragnob about it [13:50] gary_poster: I have that task [13:50] thanks frankban [14:06] gah, the "next" button on the allhands form doesn't do a damned thing. stuck. [14:37] gary_poster, frankban and I have a working solution! (Because pair programming rocks!). Wanna see it? [14:37] gmb, yeah! :-) [14:37] yay frankban and gmb [14:38] :-) [14:39] gary_poster, http://pastebin.ubuntu.com/900548/ [14:39] Just noticed some extraneous doctest in there which can be removed. [14:45] gary_poster, updated diff without all the cruft from earlier revisions: http://pastebin.ubuntu.com/900560/ [14:49] oh, duh, I was telling it to run the wrong test, eh? "-ssample3" [14:50] gmb, frankban, looks nice and elegant. I *think* we need brackets or something around the traceback though? [14:50] * gary_poster goes to look at bug again [14:51] bac, that usually means JS is giving you an error. scroll up [14:51] gary_poster: yeah, it's neat how on error they color everything red making it harder to find what is bugging them [14:51] lol, yes [14:52] frankban, gmb, https://bugs.launchpad.net/launchpad/+bug/609986/comments/2 [14:52] <_mup_> Bug #609986: layer setup failures don't output a failure message (they spew straight to console) < https://launchpad.net/bugs/609986 > [14:54] frankban, gmb, do you see what I mean? "[" and so on? [14:58] gary_poster, looking... [14:58] gary_poster: yes, and afaict we have those brackets [14:58] Hmm. [14:59] frankban, I don't think we do. We did, and then we cribbed from how setUp() failures are handled. [14:59] Now we don't. [14:59] right, I was looking at http://pastebin.ubuntu.com/900560/ [14:59] line 97 [15:00] and line 109 [15:02] gary_poster: hum... [15:02] gmb: https://plus.google.com/hangouts/extras/canonical.com/mostly--zopeless [15:02] Oh, this is encouraging... [15:02] 798 def error(self, message): [15:02] 799 """Report an error.""" [15:02] 800 # XXX: Mostly used for user errors, sometimes used for errors in the [15:02] 801 # test framework, sometimes used to record layer setUp failure (!!!). [15:02] 802 self._stream.write('%s\n' % (message,)) [15:02] heh [15:02] The !!! is a particularly soothing touch. [15:03] yeah [15:03] Also "sometimes" [15:03] Bodes well. [15:03] :-) [15:03] so, the good news is that your tear down is now handled as well as set up. The bad news is that it turns out set up isn't quite right either, so it tuens out you have to fix it too. Right? [15:04] gary_poster, Looks that way. [15:04] frankban, Okay, so I guess we need to get back to it; I'll restart the hangout in a minute; let me just grab a drink first. [15:04] gmb: sure [15:04] bah. :-) well, good progres nonetheless. Go go pair programming! [15:04] progess, even [15:05] oh heck, never mind on the spelling :-P [15:06] gary_poster: so maybe it's summer time fault if I've seen square brackets around... [15:06] lol [15:06] summer comes, you get all relaxed, you start seeing brackets everywhere... :-) [15:06] lol [15:30] * gary_poster tries again [15:42] frankban: I'm going to move your bug 963461 card to landing, since it has been reviewed [15:43] <_mup_> Bug #963461: lib/lp/services/webservice/stories/xx-hostedfile.txt fails intermittently in parallel tests < https://launchpad.net/bugs/963461 > [15:43] (to free up space for another card) [15:43] benji: ok thank you [15:44] parenthetically: it seems a little odd to have the review count connected to the coding count; I can understand the desire to keep the number in review down, but a seperate WIP limit for reviews would seem to accomplish that [16:18] gary_poster, New version, now with subunit formatter magic! http://pastebin.ubuntu.com/900690/ [18:07] gmb, a thing of beauty. benji, the reason they are connected is because someone who might start coding but is blocked by a review would then be encouraged to go and perform the review. If the "review" has turned into post-review coding, then in fact it is still coding and the WIP is still apt. [18:09] gary_poster: yeah, I figured something like that. I can't tell which is a better situation (connected or separate).