[00:17] spm: here? [00:28] lifeless: am now [00:28] nvm [00:28] sorry [00:29] np [01:38] * mwhudson lunching === barry` is now known as barry [04:35] * mwhudson has a nasty feeling that it's possible to have two concurrent processes trying to pull the same branch now [04:35] that doesn't seem like it's gong to work well [04:35] oh man [04:35] I have a dirty mind [04:38] two processes one pid? [06:36] Good morning! [08:15] mwhudson_: still here? [08:18] thumper: not really [08:19] mwhudson_: ok, np [08:29] good morning [08:38] moin adeuring [08:38] hi al-maisan! [08:52] Morning === henninge_ is now known as henninge [10:28] jtv: Hi! [10:28] henninge: hi [10:28] again [10:28] :) [10:28] jtv: could you help David? [10:28] ;) [10:29] henninge: david who? david where? [10:29] jtv: I am looking at the RemoveTranslations script. [10:29] jtv: glad to see you are on top of things [10:29] :-) [10:29] jtv: it has an option "origin". What is that? [10:30] jtv: I mean, what is the "origin code" ? [10:30] henninge: that's that field saying "this came from the web UI" or "this came in from bzr" [10:30] ah! [10:30] RosettaTranslationOrigin [10:30] jtv: there seems to be no way of restricting to a certain Project/Distro ? [10:30] in lp.translations.interfaces.translationmessage [10:31] jtv: yes, I remember now [10:31] of restricting deletions to a project or distro? [10:31] no, the script wasn't particularly made for real mass deletions [10:31] but if you have a particular need, feel free to extend it. [10:31] The options processing is still broken, which is a shame because it'd be really nice to have elsewhere too. [10:32] jtv: https://answers.edge.launchpad.net/rosetta/+question/83585 [10:32] It really needs better ways to select templates. [10:32] henninge: that's not currently something that the remove-translations script can do. :( [10:33] jtv: well, I'd have to look up all those chinese translators and do a run for each, restricting to ace. [10:33] henninge: this doesn't look urgent; it may be better in the long run to extend the script. [10:34] jtv: ok, I'll file a bug and make mention of it in the question. [10:34] henninge: ok [11:17] rockstar: got some more code import questions... can you handle them? [11:18] Well, one code import question and one problem with bzr send. [12:20] henninge: are you free for a pre-imp? [12:21] jtv: I have 10 minutes .. [12:21] henninge: should be enough! [12:21] henninge: skype doesn't show you as online though [12:22] henninge: ah, sound problem on your end [12:22] jtv: should work now === matsubara-afk is now known as matsubara === danilos is now known as Guest95679 === danilo__ is now known as danilos [13:23] cprov: still fighting the "what is the latest translations upload for this package" issue... is it normal that I'm not finding any upload sources in distroseriesqueue-translations.txt? [13:25] jtv: probably, if you need consistent data you better create it, paste me some code, let me help you. [13:26] cprov: very kind, thanks. Maybe the easiest is if I point you to my branch: lp:~jtv/launchpad/redo-upload [13:26] adeuring, hi [13:26] jtv, hi, have you seen the bzr export problem? henninge, can you please look at it? (and thanks for requesting the CP) [13:26] danilos: yes, I mentioned it in the call yesterday. [13:26] jtv: k [13:27] danilos: some branches still locked from when things were failing. [13:27] jtv, "some"? 71 out of 78? :) [13:27] danilos: yes, some. :-P [13:27] jtv, what we need to do to fix it? [13:27] danilos: break the locks. [13:28] jtv, ok, so who has the privileges to do that, and who'll do it? :) [13:28] danilos: some users seem to have done it for themselves. I wanted to ask the losas about it. [13:28] henninge, can you help with this while jtv is busy figuring out the re-uploading stuff? [13:29] danilos: sure, I'll see what I can do! [13:29] jtv, is henninge around at all (I didn't see anyone online a few hours ago, and I have trouble being online at all times) [13:29] (and be careful only to break locks that are several days old) [13:29] henninge, excellent, thanks a lot [13:29] danilos: no, he's not around, you're talking to a bot [13:29] jtv: is your branch pushed ? [13:29] cprov: think so... hang on [13:29] jtv, go away :) [13:29] * jtv goes away [13:30] * henninge too === jtv is now known as jtv-afk === mup_ is now known as mup === mrevell changed the topic of #launchpad-dev to: Code hosting offline 10.00-10.30 UTC 1st Oct -- http://blog.launchpad.net/general/code-hosting-offline-2009-10-01 | This is Launchpad Development Channel | Week 00 of 3.1.10 | PQM is OPEN | https://dev.launchpad.net/ | Get the code: https://dev.launchpad.net/Getting | On-call review in #launchpad-reviews | Use http://paste.ubuntu.com/ for pastes | This channel is logged: http://irclogs.ubuntu.com/ === mrevell changed the topic of #launchpad-dev to: Code hosting offline 10.00-10.30 UTC 1st Oct -- http://is.gd/3MyHY | This is Launchpad Development Channel | Week 00 of 3.1.10 | PQM is OPEN | https://dev.launchpad.net/ | Get the code: https://dev.launchpad.net/Getting | On-call review in #launchpad-reviews | Use http://paste.ubuntu.com/ for pastes | This channel is logged: http://irclogs.ubuntu.com/ === jtv-afk is now known as jtv [13:49] cprov: that branch should be up there... look at what I did in model/sourcepackage.py, with julian's guidance [13:53] jtv:k [13:54] cprov: ohhhh, I'm assuming main_archive... is that a problem? [13:57] jtv: not necessarily [14:05] jtv: in getLatestTranslationsUploads() implementation 'histories' is wrong, we can have multiple publications of the same SPR in the same distroseries in different pockets. [14:05] cprov: hang on, in standup! [14:05] sorry! [14:06] jtv: the assertion is too strict, in fact [14:08] jtv: once you have a published SPPH you can user pub.sourcepackagerelease.package_upload.customfiles directly, no need to user the PUSet lookup [14:13] mrjazzcat: https://edge.launchpad.net/~chromium-daily/+archive/ppa === mrevell changed the topic of #launchpad-dev to: Launchpad's slow right now -- we're looking into it | Code hosting offline 10.00-10.30 UTC 1st Oct -- http://is.gd/3MyHY | This is Launchpad Development Channel | Week 00 of 3.1.10 | PQM is OPEN | https://dev.launchpad.net/ | Get the code: https://dev.launchpad.net/Getting | On-call review in #launchpad-reviews | Use http://paste.ubuntu.com/ for pastes | This channel is logged: http://irclogs.ubuntu.com/ [14:23] mpt, how do you feel about dropping the "overview" tab? [14:23] re: bug 185030 [14:23] I'm 50/50 on it... [14:24] cprov: sorry to be away at that crucial time. Back now. [14:24] Bug #185030: "Overview" is an awkward name for the first tab [14:24] jtv: no worries, did you follow my comments ? [14:24] cprov: yes, though somehow these long identifiers always make my head swim :) [14:24] Matching them up to code now. [14:25] beuno, I don't think the project icon alone would be obviously mutually exclusive with the tabs. [14:26] beuno, and it would worsen the problem of navigation in distribution packages vs. distributions. [14:26] hm [14:26] yeah [14:26] beuno, because (for example) you'd be clicking on an Ubuntu icon to go to a source package page. [14:26] (Or to go to a distribution series page.) [14:27] agreed [14:27] I'll keep chewing on it then [14:28] * mpt can load other Bugs pages, but can't load that bug report [14:28] yeah [14:28] LP is slow again [14:28] something off with apache I think [14:29] beuno, I think renaming it to "Profile", "Package", "Project", etc would work pretty well [14:29] That's what I was working on (but didn't finish) when I left the LP team. :-) [14:29] henninge: about that pre-imp... is 30 minutes from now ok for you? [14:29] mpt, that approach itches for a reason. I should mock it up and see how it looks [14:30] jtv: so at 15:59? OK. [14:30] ;) [14:30] henninge: roughly :) [14:31] jtv: about the locked branches - where do I see them? [14:31] henninge: error-reports. Hang on, I'll paste you the latest. [14:35] henninge, sanitized list: https://pastebin.canonical.com/22715/ [14:36] cprov: looks like SPPH.sourcepackagerelease.package_upload can be None..? [14:37] jtv: in the sampledata, yes [14:38] cprov: ah. Still getting the exact same symptoms in that test... Do you think splicing into distroeseriesqueue-translations.txt is the wrong approach for exercising this? [14:39] jtv: I dunno, let me look at the test. [14:41] jtv: how do I get information about the locks, like the lock age? [14:42] henninge: normally, if they were your branches, you'd use bzr break-lock. As it is, I don't know. This one needs help from codehosting folks, I think [14:42] jtv: ah, ok. So there is not an easy way. [14:43] henninge: they may have a trick, like just deleting a bunch of lock files. [14:44] abentley: ping [14:49] henninge: pong [14:50] abentley: Hi! [14:50] abentley: We have a number of branches that were left locked by the translations export script. [14:51] abentley: we are looking for an easy way to break them while making sure, the locks are really stale. [14:51] abentley: do you have a suggestion? [14:52] henninge: I can have a look at bzrlib. [14:53] thumper, reported bug 438711 [14:53] Bug #438711: Can launchpad-login but not launchpad-logout [14:54] abentley: I mainly thought of the permissions issue. I cannot break other people's locks, can I? [14:54] henninge: You can break a lock on any branch you have write access to. [14:55] henninge: If we get a losa to run a script on crowberry, then you can break any lock. [14:55] abentley: Ok, that is what we'll need, I guess. [14:58] abentley: Does bzr break-lock detect if a lock is really dead or not? [14:58] henninge: I'm investigating that. [14:58] abentley: cool, thank you [15:05] henninge: There doesn't appear to be any check that the holder is dead. [15:06] henninge: In the sftp case, the holder will be a process on a remote machine. [15:06] henninge: I am not certain about the bzr+ssh case. [15:10] abentley: is there a way to read the age of a lock? [15:11] henninge: Yes. [15:12] abentley: oh cool, that should help. How? [15:12] henninge: For a bzr+ssh lock, the process id is the server-side process. [15:12] abentley: oh cool, too, I think. [15:12] abentley: these are all designated as lp-hosted. What protocol is that? [15:13] henninge: calling "peek" on the lock will tell you pid and start_time. [15:14] henninge: For the case where they were created by a script, they will be accessed as local files. [15:15] henninge: So your script didn't use a protocol, but users can access them via bzr+ssh or sftp. [15:16] I think if you ensure the PID isn't active and the lock is old, the risk of breaking an active lock is low, but not 0. [15:17] abentley: yes, I guess I should just get a listing of the locks, first. [15:21] cprov: code's a lot simpler and cleaner now, but still not finding anything in that test. [15:22] cprov: I've pushed an update. [15:27] jtv: will update in one sec [15:43] intellectronica: ping :) [15:44] sinzui: ping [15:44] Hi EdwinGrubbs [15:45] jtatum: hi. i got some test failures for your branch, but it looks unrelated to your change. let me forward them to you [15:45] sinzui: I had some difficulties running close_bugs_from_commits.py, and matsubara informed me that he can run it for all the launchpad projects, but that some of the team leaders had opted out. Can we opt in? [15:46] EdwinGrubbs: I do not want to [15:47] EdwinGrubbs: The script does not give you credit for your work. Barry was put out by it. [15:48] EdwinGrubbs: Did you get the script to work for yourself? [15:48] EdwinGrubbs: I ran it (or a derivation of it) Last Wednesday and Friday. [15:49] sinzui: I got it to work, but it only limits changes to given projects not people, so I actually Fix Committed some other people's bugs. [15:49] sinzui: when you ran it, it didn't close any of my Blueprint bugs. [15:49] oh [15:49] EdwinGrubbs: I'll send you my script [15:50] sinzui: when you say you don't get credit for closing the bug, do you mean that you don't get karma? [15:51] yes, + we loose tracking of that...we want to use karmaevents to report person's activities, even if there are no point. That script (or our API) wrongly reports who did the work === mrevell changed the topic of #launchpad-dev to: Code hosting offline 10.00-10.30 UTC 1st Oct -- http://is.gd/3MyHY | This is Launchpad Development Channel | Week 00 of 3.1.10 | PQM is OPEN | https://dev.launchpad.net/ | Get the code: https://dev.launchpad.net/Getting | On-call review in #launchpad-reviews | Use http://paste.ubuntu.com/ for pastes | This channel is logged: http://irclogs.ubuntu.com/ === matsubara is now known as matsubara-lunch === sidnei_ is now known as sidnei [17:28] cprov: had a chance to look at why that test has no package uploads? === EdwinGrubbs is now known as Edwin-lunch [18:37] * rockstar lunches [18:49] hi abentley [18:49] abentley, do you know if it's a known issue that lp doesn't support http branch urls for git? [18:49] Ursinha: hi [18:50] issue or limitation [18:50] Ursinha: I think I've seen a bug for that... [18:50] abentley, I just searched and couldn't find one [18:51] maybe I'm just doing it wrong :) [18:52] Ursinha: It may have been a question. Still looking. [18:55] Ursinha: I can't find anything. I don't know whether it's a known issue. [18:56] cprov: not having much luck deciphering the situation in distroseriesqueue-translations.txt... Does the SPR get created in that test? [18:56] abentley, well, I'll file a bug then [18:56] thanks abentley [18:56] Ursinha: np [18:56] jtv: uhm, forgot about it. [18:57] I'm really too tired to work on this, but it needs to get done. :/ [18:58] jtv: duderino, translations come from binaries, not sources (!) [18:58] hi, I'm trying to import a git repository inside launchpad, but it's just available in http (webdav) and currently launchpad doesn't support it [18:58] cprov: so this test doesn't upload a source package? [18:59] jtv: no, it creates a base source, so we can upload a binary with a translation tarball [18:59] I'd like to know why exaclty this is not supported, and if possible, where is the code that actually implements the git import from launchpad [18:59] jtv: your lookup on SP is wrong. [18:59] so I could take a look and maybe hack it to make the support [18:59] cprov: argh [18:59] don't know if it's a bzr problem or launchpad itself [18:59] cprov: I should be finding the binary packages for the source package? [18:59] abentley, ^ [18:59] jtv: looks for custom files uploaded with binaries built by the context SPR [19:00] rsalveti, I'm on my way to file a bug about it, abentley, with whom should rsalveti talk to get to know what to do? [19:01] I'm quite new to both launchpad and bzr, but would like to know what could I do to have this support :-) [19:02] jtv: yes, history.getBuilds() -> build.package_upload.customfiles [19:02] rsalveti, abentley and rockstar are the people that would be available in a reasonable timezone :P [19:02] * jtv cries quietly into his keyboard [19:02] :-) [19:02] rsalveti: launchpad uses bzr-git to do git imports. The launchpad code is will be in launchpad/stable/lib/lp/code/ [19:02] sorry, "lib/lp/code". [19:03] abentley: cool, will take a look at it [19:03] going to a meeting now =\, in one hour will be back, thanks [19:03] rsalveti: There will be more in lib/lp/codehosting/codeimport [19:03] cprov: will getBuilds give me any superseded builds? [19:04] jtv: all builds for the context source are relevant, if they got superseded their 'package_upload' property is None. [19:05] cprov: so at least that's a mercy [19:05] julian said this was easy, damn his salad tongs [19:09] cprov: progress! I'm seeing the upload! [19:10] jtv: cool [19:17] cprov: ooooh, I have passing tests! thank you thank you thank you [19:17] jtv: you know my address, send cookies ;) [19:18] cprov: how do I explain this... I know your _email_ address. [19:18] That's not useful for this type of cookie. [19:18] jtv: :) [19:21] jtv: you know iterating over builds and getting their corresponding package_upload is somehow expensive, right ? [19:21] cprov: no I don't know, and I'm not sure I care much. :) This is for a one-off, not production use. [19:21] jtv: will your script dealing with multiple sources ? [19:21] cprov: a few dozen source packages. [19:21] cool, no worries then === Edwin-lunch is now known as EdwinGrubbs [19:53] morning === mwhudson_ is now known as mwhudson [20:26] rsalveti, I've filed bug 438929 for your import problem [20:26] Bug #438929: Launchpad doesn't accept git imports with http [20:26] Ursinha: cool, thanks, going to take a look [20:27] rsalveti, np [20:56] Ursinha, I didn't see your PM earlier. What's up? [21:01] rockstar, abentley: morning [21:01] rockstar, abentley: standup now? [21:01] mwhudson, yo [21:01] ish [21:01] mwhudson, sure. [21:01] mwhudson: Okay. [21:32] * mwhudson away for a speel [21:32] speel [21:32] spell [22:18] mwhudson: hi [22:18] rockstar: hi [22:19] thumper, hi [22:22] thumper, I think mwhudson is off to pick up a suit. [22:31] the hotel internet is flirting with me [22:31] first it is there [22:35] thumper, is it there now? [22:35] not really [22:35] it lets me get one irc message in [22:35] and then dies [22:35] OMG three in a row! [22:36] you haven't appeared to disconnect yet [22:54] thumper: good evening === matsubara is now known as matsubara-afk [23:10] err [23:10] from https://answers.edge.launchpad.net/launchpad-code/+question/84184, how can i see the current status of the question? [23:10] oh, there it is [23:30] omg this internet connection sucks [23:31] thumper: hi