[02:54] wgrant: My only concern with that branch was: [02:54] 291 - u'There is no project in Launchpad named "N\xf6 Such Product&<...] [02:54] 294 + There is no project in Launchpad named "N... Such Product&<... [02:55] I couldn't get print_feedback_messages() to show me \x..., it always printed the codepoint [02:55] That's correct. [02:55] Because it prints the string, not its repr. [02:55] Right [06:10] wgrant: Hm, I seem to have tripped over https://code.launchpad.net/~stevenk/launchpad/drop-kick-getBugTasks/+merge/74941 ; getBugTasks isn't in the monthly PPR [06:12] StevenK: What do you mean? [06:14] wgrant: I had forgotten all about that MP -- looking at the latest monthly PPR, I can't see getBugTasks at all, but it's still exported [06:14] Ah, didn't notice it was old. [06:14] If it's not in the PPR that shows the other methods, then it wasn't called at all. [06:18] wgrant: As long as pageids is the right page to be searching [06:18] StevenK: I believe so. Do the other methods appear there? [06:21] IDistribution.getBranchTips shows up as Distribution:EntryResource:getBranchTips, so looks like. (DSP doesn't have any other exported methods, but has other exported properties) [06:21] Right [06:23] It was called 38 times in March, that is the only time I've seen it [06:23] But other monthly PPRs have been eaten [06:51] Not sure anymore where to ask, but is there any way we can get a user unsubscribed from bugs on the heat project? Their out of office mail is spamming every change we make. [07:04] wgrant: StevenK: ^ halp. [07:05] Have you adequately slapped the user in the face for having a moronic autoresponder? [07:05] Which user? [07:06] SpamapS: [07:08] wgrant: https://launchpad.net/~faramir [07:08] * SpamapS is quite certain this, as most problems, is Lotus Notes' fault. [07:09] SpamapS, lifeless: lp-shell production devel; lp.load('/heat/+subscription/faramir').lp_delete() [07:09] The project owner should be able to do that. [07:10] wgrant: maintainer? as in, ~heat-drivers? [07:11] Yes [07:12] * SpamapS tries it [07:13] wgrant: lovely, thank you. subscription removed. [07:13] Excellent. [07:13] Do stab the user and/or their email admins as well, though. :) [07:14] wgrant: So, do you think that MP is landable? [07:14] StevenK: I don't see why not. [07:14] The only thing I can come up with is it drops the method from 1.0 [07:14] But we probably don't care [07:15] If it's not called frequently then we do not care. [07:15] wgrant: right, slapping subscribed. :) === vila is now known as vila-afk-biab === vila-afk-biab is now known as vila === Ursinha is now known as Ursinha-afk [15:38] Ursinha-afk: Just having a go at your QA now in order to unblock myself === Ursinha-afk is now known as Ursinha [15:41] Copied https://dogfood.launchpad.net/%7Eubuntu-unity/+archive/daily-build/+sourcepub/3476923/+listing-archive-extra to the primary archive on DF [15:51] wgrant: I restored the dogfood-publish config and its associated paraphernalia on labbu [15:53] Ursinha: This looks like a failure, I'm afraid - notes in the bug [15:53] cjwatson, will look [15:53] I wonder if this is DF-specific, since I thought you mostly just moved that code around ... [15:54] (sorry the delay in the answers, my ISP is not helping today) [15:54] Ah, but I guess that code never had to deal with copies from a PPA before [15:55] Odd, though, since _publishCustom uses self.packageupload too and it's clearly never None in those cases [15:56] So the PackageUploadCustom mustn't be being constructed in the right way, but why ... [15:56] cjwatson, yeah, I recall talking to wgrant about such cases and that never should be none [15:56] when I was transforming that code in a job [15:57] And for that matter how [15:59] I don't see anything other than PackageUpload.addCustom that constructs new PackageUploadCustom objects, and that sets packageupload=self [16:00] But the whole publisher looks kind of unhappy [16:01] cjwatson, have you tried to copy any other kind of package other than from a PPA? [16:01] No [16:02] That's the only case that matters though :) [16:03] cjwatson, yeah :/ [16:05] In fact this makes zero sense because the PUC is surely looked up from the PU in the first place [16:06] Oh, they have a PU but no SPR, I misread [16:10] Ursinha: OK, so PackageUploads that refer to copies (PackageCopyJobs) don't have a sourcepackagerelease - only direct uploads do [16:10] Ursinha: I think you need to do the same kind of things that PackageUpload.package_name and friends do, and look at either bits of packageupload.package_copy_job or packageupload.sourcepackagerelease [16:11] Ursinha: And of course have a test that exercises this in the context of copies [16:12] cjwatson, yeah.. I should have tested the ppa copies, I felt that was missing after submitted to pqm [16:12] I'm into it [16:13] Cool - I think you can use package_copy_job.target_distroseries (I think) or package_copy_job.component_name [16:13] For the various cases [16:14] But then you have to do something with the PackageTranslationsUploadJob creation ... [16:14] cjwatson, right, thanks for the huge help [16:14] yeah, I'll have to adapt it all [16:14] But I guess you know what to do there :) [16:14] :) [16:15] Yikes, attachTranslationFiles lives on SPR [16:16] I guess maybe there needs to be a similar thing for PCJ? Not sure [16:17] cjwatson, hm. that sounds ugly but might be the way to go [16:19] as pu doesn't have a spr it doesn't make sense to use the one it's there for such cases... but I'll check if there any other (less intrusive) way [16:20] I think you have to go through PCJ - I remember dealing with this for the queue API exposure [16:23] Ursinha: (see also my comments in https://bugs.launchpad.net/launchpad/+bug/851562, tangentially) [16:23] <_mup_> Bug #851562: Diffs not available for syncs on +queue page like for regular uploads

[16:24] cjwatson, yep, I was reading that right now [16:25] cjwatson, I couldn't understand well in the call, internet is crappy, are you working on fixing that? [16:25] Ursinha: Not actively [16:26] Ursinha: I don't think it's a blocker for this, just context [16:26] You don't really *need* the SPR here, it's just that that's where the code currently lives and it uses some attributes from it [16:28] cjwatson, that's okay, I ask because I could look at that next, as the context is similar [16:29] Awesome === BradCrittenden is now known as bac