[09:44] gmb, btw, do you need any help with unused potmsgset purging, or did you not get to that state yet? :) [09:44] danilos: I'm looking at in now actually. [09:45] cool, just ping me if you need any help [09:45] danilos: How can I find out if a POTMsgSet is not participating in any translations? [09:46] gmb, "all instances of TranslationTemplateItem's that reference that potmsgset have a sequence of 0" [09:47] gmb, we could add an "or they reference potemplate that is inactive" [09:48] gmb, if you look at TranslationTemplateItem (\d translationtemplateitem from within "psql launchpad_dev" is useful), it should be clearer what I meant above [09:48] danilos: Okay, I think I'm with you. [09:49] danilos: I was thinking that we could use Garbo.daily for this (I know you said weekly, but I don't thin we have a facility for that unless we add a new cronscript). [09:49] We could do an initial run manually before rolling this out, maybe. [09:50] gmb, right, I feel better about separate scripts then munging it all together into one big behemoth, but it's ultimately your call :) [09:50] danilos: Well, it's more that last time I tried this I got a stern "Use Garbo!" from Stub and Lifeless. [09:50] (there is an argument to be made about a gazillion scripts as well, so I am partial to both solutions) [09:50] gmb, ack [09:51] gmb, looking at the docstring of cronscripts/garbo-daily.py, it is indeed the right place to add it :) [09:52] :) [09:52] Righto, I'll crack on then. [09:52] danilos: I'll ping you if I need any more help. So far it seems clear enough. [09:53] gmb, cool, note that you'll have to first drop TranslationMessages for the potmsgsets you are removing as well [09:53] danilos: Ah, right. Noted, thanks. [09:56] gmb, also, note that there might be potmsgsets which have no TranslationTemplateItem rows at all (which means they are not in any of the templates), and you can get that more easily with a separate query [09:57] Ok. [09:57] Thanks. [12:12] hello === bac` is now known as bac [12:14] CHR is in a very bad shape :/ [12:15] deryck squad bad! [12:16] * bac thinks squad colors should be somehow tied to team lead name so i can remember them. [12:16] :-( [12:16] that's no fun [12:16] and hi [12:16] we should be the green team [12:16] heh, gren gary? [12:16] geen [12:16] eh blah :-) [12:17] j & d are hard [12:17] * gary_poster tries to think of colors beginning with J and D. [12:17] yeah [12:19] danilos, is it bad enough for me to raise the issue with deryck? If you think so, I'll investigate to prepare for the conversation [12:20] gary_poster, well, there's like 15 open questions from last week [12:22] danilos, gotcha. there were a lot of questions from us when deryck's week started. It would have been nice if they had made more of a headway, but I suspect that we're still paying the price of the sprint. [12:22] For myself, questions are typically near the end of my hour, so I don't make as much progress as I might (and they are often the hardest to deal with). [12:23] gary_poster, right, makes sense then [12:24] gary_poster, fwiw, I think we should start on them earlier to make sure we clear the back log (the two I hit were from people inside canonical, and some re-raised them on IRC now) [12:25] danilos, makes sense [12:29] benji bac danilos gmb, call RSN [12:29] gary_poster, also, it might be that the way we split work means that items at the bottom of the list never get done [12:29] ack [12:50] gary_poster: did you get bunch of rain yesterday? we had thunder, power flickering, and about 5 minutes of rain. i think it passed us by and went to wake county. [12:50] bac, no not really. It kept looking like it might rain, but nothing :-/ [12:51] oh, i heard parts of wake got 5" [13:23] benji, would you be willing to help me qa bug 812335? If so, I'd like you to help me figure ourt how to do so. :-) [13:23] ("getBranchTips") [13:25] gary_poster: now that I have my coffee I'm up for anything [13:26] benji, :-) [13:27] so, did you figure out how to create a branch that would be returned from getBranchTips when you qa'd this before? [13:27] I think we need to do that, and then run getBranchTips, [13:27] and then ask LOSAs to make the branch private, and then see that we don't see the branch any more [13:28] I have a reasonable idea of how to do all of those except the first :-) [13:28] do you know? [13:29] danilos: do you have time before EOD to review https://code.launchpad.net/~bac/launchpad/bug-788685/+merge/69082 ? [13:30] benji ^^^ [13:32] bac, sure [13:32] thanks [13:33] gary_poster: I started to do the QA with "ubuntu", it has lots of qualifying branches (IIRC), but then realized that wgrant had QAed it [13:34] so, I believe that "ubuntu" should work, but am not completely certain [13:38] bac, r=me, small suggestions :) [13:38] danilos: great! [13:39] bac, I love the version comparison, very good to learn about things like that :) [13:39] yes! so much more sane and cheaper than that awful thing i'd cooked up on friday [13:41] benji, flacoste has instructions for me I think. Now will try them :-) [13:41] bac, oh, ignore the comment re IDistroSeriesSet import, I missed that you just reformatted it [13:41] yep [13:42] k [13:44] benji, shouldn't this work? [13:44] >>> lp.distributions('principia').getBranchTips() [13:44] Traceback... [13:44] AttributeError: 'Entry' object has no attribute 'getBranchTips' [13:45] gary_poster: it should. Be sure you're using the right version of the API. [13:46] benji, pretty sure I am ("./bin/py /usr/bin/lp-shell qastaging devel") [13:46] "Connected to LP service "https://api.qastaging.launchpad.net/" with API version "devel"" [13:46] gary_poster: I'm not certain how the caching works, you may want to use a different cache dir -- or nuke your current one -- so you aren't using an old WADL [13:47] benji ack. (It works on production, just tested) [13:57] bac, btw, qaing that change is probably going to be possible only on dogfood (not sure if staging is up to the par yet, but I doubt it is) [13:57] danilos: yeah, i was going to ask you for advice on that [14:00] bac, basically, we usually asked soyuz people about re-uploading (re-publishing or something) existing ubuntu packages on dogfood so the same code paths would be triggered [14:01] bac, after that, it's easy enough to go to the translations.lp.net/ubuntu//+source//+imports to check what was put in the import queue [14:01] danilos: sounds good. [15:07] CHR awaits! [15:07] Did we ever get gmb around? [15:08] No [15:08] I'll start the counts, then [15:20] danilos, could you take a look at https://answers.launchpad.net/launchpad/+question/165699 [15:43] Hello? Is this thing on? [15:44] gary_poster: Hi. Apologies for the long absence. My plan to move to a wifi hotspot was thwarted by a telecom engineer using me as a test subject. [15:44] Upside: All fixed now. [15:45] gary_poster: I'll make up the time this evening, FTR. [15:45] bac benji danilos gmb, (1) I set up numbers on https://dev.launchpad.net/MaintenanceRotationSchedule ; please review and suggest/make changes. [15:45] (2) deryck apologized for number of questions. he offered to help. I said I'd contact him Thursday if we still had problems. [15:45] abentley should be back then, which will help [15:45] gmb, glad you have internet again. :-) thanks for heads up [16:02] my CHR hour was less productuve than usual [16:02] but I need a break [16:02] * gary_poster not feeling 100% [16:03] lunch [16:32] * gmb tries to un-break make schema. [17:23] A pox on tests where I have to switchDBUser() to make them work. [19:17] * gmb -> calling it a night. [19:35] * gary_poster slept for 2.5 hours (after 10 hours last night) [19:35] * gary_poster going to take a sick day [19:35] hopefully I'll be ok tomorrow [19:35] ttyl [20:56] thunder and lightning, very very frightening [20:56] yeah, we've got it here too. [20:57] grass can't live on thunder alone [20:58] but i haven't had to mow since before going to dublin