[00:05] beuno: i am [00:07] barry, want to have that heading call? [00:08] beuno: sure! [00:08] beuno: give me 2m [00:10] barry, give me 10 :) [00:10] beuno: okay, but if jane gets back i'll have to make dinner :) i'm on skype now, ring me when you're ready [00:18] hey rockstar, do you have half a minute? [00:19] Ursinha, sure, what's up? [00:21] barry, now? [00:21] beuno: yep [00:24] barry, https://code.edge.launchpad.net/~sinzui/launchpad/official-portlets/+merge/10946/+review?claim=launchpad&review_type [00:26] barry, https://edge.launchpad.net/bzr [00:31] barry, https://wiki.canonical.com/Launchpad/UI/Navigation [00:36] barry, https://bugs.edge.launchpad.net/bzr/+bug/165293 [00:36] Bug #165293: collisions through uploading same-named .pack files not handled correctly [00:38] beuno: ? [00:38] lifeless, it's my example bug :) [00:38] oh kk [00:42] beuno: https://launchpad.dev/people/+newteam [00:44] beuno: http://people.canonical.com/~barry/screen.png [00:49] * barry -> dinner === flacoste_afk is now known as flacoste [01:03] thumper: actually, we should be pretty close [01:03] thumper: zope is upgraded as is all of our lazr deps [01:03] thumper: gary wants to tackle the chameleon branch before finishing the 2.5 upgrade [01:04] thumper: but you might try it, it might actually just work [01:04] thumper: python2.5 bootstrap.py && ./bin/buildout [01:07] flacoste: I'm happy that we're close [01:08] yeah, gary wanted to tackle the chameleon integration before it bit rots more [01:08] and didn't want to risk having it break more because of other deps upgrade [01:08] besides, a nice performance boost on rendering would be nice for 3.0 [01:10] flacoste, any udeas why I'd get this in karmic? https://pastebin.canonical.com/21664/ [01:12] salgado-afk: i submitted a patch for the typo poolie reopened as bug 156919 [01:12] Bug #156919: More information on where to download files needed [01:14] beuno: no python-psycopg2 installed or installed against the wrong python? [01:15] bac, it's ssntalled [01:15] installed [01:15] how do I know against which version of python? [01:15] ah, I think karmic's version is newer, so it's probably taking over the PPA one [01:15] beuno: you can dpkg -L python-psycopg2 to see [01:17] beuno: i've got 2.0.8-0ubuntu3~launchpad1 [01:17] bac, you are tight [01:17] 2.5 and 2.6 [01:17] DAMN [01:17] you are *right* [01:18] i *am* cheap, if that's what you meant [01:18] heh [01:18] beuno: this early adoption thing -- is it serving you well? didn't you go through this about 6 months ago? [01:19] bac, it is not. But I'm 50% massochist. [01:32] beuno: maxb should have this fixed soon, he's really on top of the dependencies [01:32] thanks to him, we can expect the karmic updates to be a lot much smoother than Jaunty [01:33] bac: what are you doing on karmic? [01:33] already, i mean? [01:33] flacoste: dude, i'd be on feisty if i could [01:34] i'm *not* on karmic -- i'm still all jaunty [01:34] ah, i see [01:34] you were speculating [01:34] i was just poking fun at beuno [01:34] good guess! [01:42] flacoste, it's fixed [01:42] but something else is broken now [02:00] * jml frowns [02:01] so here's the thing [02:01] pocket-based upload permissions have nothing to do with the uploader. [02:03] who do they have to do with [02:03] also, you're @ café? [02:03] I am. [02:04] no one. they are a function of the archive and the distroseries [02:21] * jml grumps at soyuz and timezones [02:24] jml: why is this? surely fine grained permissions will give individual users privilege via team-or-something inheritance? [02:25] lifeless, I have no idea what you mean, I'm sorry. [02:25] jml: ok [02:25] well I'll be there about 10 past 12 [02:25] heh [02:26] if you want to discuss in more detail before ring me [02:28] thanks [02:29] jml: Is canUploadToPocket so hard to use? [02:29] wgrant, it's not hard to use [02:29] wgrant, the question is whether it's relevant. [02:31] jml: Not for normal permission checking. I think it's only relevant for whether the branch should be completely immutable. [02:31] time for nommage [02:31] Which goes beyond 'does this user have additional privileges', as it affects even the branch owner. [02:32] wgrant, yes, this is what I'm thinking. [02:32] abentley: Can you forward me the failure email? [02:33] wgrant, also, while james_w is still uploading branches, I can see a pocket check getting in the way [02:33] jml: Crap, you are right. [02:33] How awkward. [02:34] wgrant, yes. [02:34] Sounds like you need a new distroseries or even distribution column. [02:35] well, at the least, I think I want to talk to james_w [02:36] jml: Actually, maybe you don't need to. [02:36] jml: Because the branches are probably only made official at the end. [02:37] So it might not matter that they become immutable. [02:37] good point. [02:37] but even that means changing my plans a little [02:38] Why? [02:38] since I was going to change the permissions for "make official" from what they are now to "own the branch & can upload" [02:39] 'own the branch' is probably wrong. [02:39] now, it would mean "(own the branch & can upload) OR are local deity" [02:39] wgrant, why? [02:40] jml: Mmm, I guess it's a good idea to require that if it means that privileges will be revoked. [02:40] wgrant, well, the thing is that it's unacceptable to allow random uploaders to expand branch permissions [02:41] But ISTM that launchpad.Edit in general shouldn't be revoked based on IDS.canUploadToPocket. You probably want to still allow editing of the description and all that. [02:41] wgrant: There wasn't a failure email. I ran it on my own box before submitting it. [02:41] And I don't think the officialness setting should be forbidden based on that either. [02:41] abentley: Ah. Can you then advise me which tests failed? [02:41] Or shall I run it myself? [02:42] wgrant: I don't remember which tests failed. [02:42] abentley: OK. [02:49] wgrant, in an ideal world, I think officialness should be set through a handshake [02:50] jml: Probably. But that's overcomplex. [02:50] I agree :) [02:50] jml: (owner & uploader) | ~ubuntu-branches is probably the best you can do. [02:50] wgrant, yeah, that's what I'm leaning toward. [02:50] (where ubuntu-branch == local deity) [02:50] jml: But it would have been reallly nice to kill off ~ubuntu-branches. [02:51] yes, it would have been nice [02:51] maybe we still can. [02:51] but one step at a time. [02:53] and I think the best next step is to change the bug nomination logic to use my newly extracted permissions function [02:53] since neither cares about pockets [02:54] hey jml! [02:54] flacoste, hello [02:54] sorry I missed your privmsg [02:54] xchat hates me [02:58] sinzui: ping [02:58] Hi EdwinGrubbs [03:00] jml: Oh! Did anything come of my branch? [03:00] sinzui: it looks like all the story-ui-3 registry bugs are triaged. Do I just steal one and hope that they haven't started on it? I was going to work on it some tonight so I can be ready for reviews in the morning. [03:01] EdwinGrubbs: Triaged means it is acknowledged an waiting for someone like your self to take and mark in progress. [03:03] EdwinGrubbs: All the Update...UI 3.0 that are triaged are available: https://edge.launchpad.net/launchpad-registry/+milestone/3.0 [03:14] jml: Was that the instance during which the network died? [03:15] wgrant, yes. it was [03:15] wgrant, sorry, I'm on the phone right now [03:15] wgrant, I've fired off another ec2test instance for merging [03:16] jml: Isn't that just going to die silently again? [03:16] wgrant, who knows! [03:16] We shall see. Thanks. [04:45] Can somebody please ec2test lp:~wgrant/launchpad/refactor-librarian-log-parser? Apparently there are failures, but I can't run the whole test suite sufficiently reliably here at the moment. [05:15] wgrant, hey [05:15] wgrant, I can do that. [05:16] jml: Thanks. [05:21] wgrant, have you got any email about the structural subscription branch? [05:22] jml: No :( [05:22] ok [05:22] the instance isn't running === jtv1 is now known as jtv [05:23] this is obviously a critical bug in the test suite [05:23] ec2test, rather [05:24] Well, ec2test is meant to terminate the instance in situations where the test suite dies. [05:25] So it's doing exactly what its meant to, however useless that may be. [05:27] it's also meant to tell you what happened before that [05:27] Hm, so it is. [05:27] Hadn't looked at ec2test-remote in detail. [05:27] So it is being bad after all. [05:34] wgrant, any situation where you can run tests and not get results reliably is critical, imo [05:34] wgrant, otherwise, wtf does 'test' mean? [05:35] jml: 'no results' == 'uberfail' [05:38] jml: Do you have any idea how far through it died? [05:38] wgrant, no, none at all. [05:39] wgrant, basically, I threw it over the wall, and nothing came back. [05:39] jml: Right. Damn. [05:41] jml: There's nothing that can really be done besides running it headless and hoping the connection survives, is there? [05:43] wgrant, running headful, itym [05:43] jml: That one. [05:46] For anyone who is interested: https://dev.launchpad.net/Code/Review/Plans [05:49] thumper, thanks! [05:51] mwhudson, how much pain would be involved in setting up a buildbot that ran with --coverage [05:51] Wouldn't that take like forever to run? [05:52] wgrant, yes. [05:53] * jml files some bugs [05:54] i think a rule of thumb is that --coverage takes 5x as long as normal [05:54] jml: setup, not much pain [05:55] Fortunately, every test that uses subprocesses will seem 5x as fast with --coverage ;) [05:55] spiv has a point [05:56] you could probably get around that by running the tests with a mangled ./bin/py [05:56] Or just change site.py in the ec2 image... [05:59] well, I'm happy to have an incomplete coverage report to start with [06:00] the report is still interesting even if it ignores coverage via subprocess execution [06:00] Yes. [06:00] Plus, it adds motivation to fix tests that use subprocesses :) [06:01] or delete code only used in subprocesses :) [06:25] * mwhudson is more or less done for the day [06:46] * jml is off [07:03] so coverage times are distorted [07:03] measure time and coverage separately. KThanks. [07:04] mwhudson: you can get some rad coverage using lspro [07:04] f [07:29] mwhudson, you've probably EODd, but https://bugs.edge.launchpad.net/launchpad-foundations/+bug/5814 is interesting [07:29] Bug #5814: want to know breakdown of test run time by area of development

[07:34] jml: four figure bug! [07:34] * mwhudson is not here [07:34] mwhudson, yes! [08:28] Hallo noodles775! [08:29] Moin al-maisan! [08:29] Hi henninge and al-maisan :) [08:29] good morning noodles775 and henninge :) [08:29] noodles775: I am for the first time trying to use a sub-view. I see two different approaches. [08:30] noodles775: Register it as a page in zcml, including the template, and call that page from the enclosing template (@@,,,) [08:31] noodles775: or not register in zcml and tell the sub-view in-line about its template using ViewTemplateFile (I think it is called) and then just calling the view from TAL (structure ...). [08:31] noodles775: are you with me? ;_) [08:31] ;) [08:31] yep [08:32] noodles775: is there any preference on one of the approaches? [08:33] So, afaik, at least in soyuz code we usually go for the former (ie. keeping the configuration in the zcml - not that we like zcml, but for consistency ;) ). [08:33] hmm .. zcml .. my love ;) [08:33] I'm not sure of other advantages/disadvantages, wgrant, what are your thoughts? [08:35] I was wondering if using the @@ approach engages more zope machinery than necessary and there might be a performance penalty. [08:35] I am using this in a repeat for a list of translatable strings. [08:35] yeah, that would be very interesting to know. [08:35] one tic [08:36] henninge: why isn't a simple tal:repeat with a template-snippet/macro enough in that case? [08:36] henninge: I think keeping it in ZCML is better, but I'm not thoroughly opinionated on that yet. [08:36] Er, noodles775 ^^ [08:36] Why do you need a separate view for each repetition? [08:36] wgrant: thanks. [08:37] * noodles775 looks for similar example for henninge [08:37] noodles775: Bug comments, maybe. [08:37] noodles775: well, the display is a bit more complex than just a string. [08:37] * wgrant looks. [08:38] henninge: A tal:repeat can contain anything, though. [08:38] true [08:38] henninge: sure, I was thinking more of printing a table row for each repetition... [08:39] henninge: but if you need some complicated manipulation of the repeated item's attributes, then yeah, that's a pain in the template (and is usually where we'd use a view). [08:40] noodles775: yes, there is more to it. [08:40] good morning [08:40] Hi adeuring! [08:40] henninge: so in templates/archive-index.pt [08:40] hi henninge! [08:40] hi adeuring :) [08:40] hi noodles775! [08:40] noodles775: I am not sure I want to put this all into the enclosing view [08:40] * henninge looks at that [08:41] henninge: you'll see at the bottom [08:41] See also bugs/templates/bugtask-index.pt just after line 300. [08:41] That might be more like what you want. [08:42] yeah, that's a good example, so the original view is providing an enum of structured information that can be easily repeated. [08:43] wgrant: yes, thanks [08:43] Great. [08:44] s/enum/iterable [08:47] noodles775: thanks, too [08:47] np! /me loves learning too :) [08:48] ;) [08:59] henninge: jfyi, the reason we use a macro on that archive template is because we re-use the source-package-list in other places too. [09:00] I'm not sure how expensive the context/@@+macros/macro-name is though. [09:01] It's not creating a new view though - it just uses the current view (similar to the new navigation menus, afaics) [09:02] noodles775: yes, I was just trying to figure out how that works (from bugtask-index.pt, though). [09:02] noodles775: but I'll go with an extra page and view for now. [09:02] yep. [09:02] sub-page and sub-view [09:03] * henninge needs to get results ... ;) [09:03] Hmm... I should have said dis-similar to the new menus, they do use their own sub-views too (NavigationMenuTabs). [09:04] jml: ec2test won't have submitted that branch, will it? [09:04] jml: (it succeeded -- abentley must have caught devel at a bad time) [09:05] wgrant, it doesn't say it submitted [09:05] hmm. [09:05] I must have forgot the -s [09:05] jml: Well, I didn't ask you to submit. [09:06] jml: Because I believed it would fail. [09:06] oh right. [09:07] wgrant, do you want me to submit? [09:07] jml: So, can you submit it please? "[r=abentley][ui=none] (wgrant) Refactor the librarian apache log parser so pieces can be reused for PPA download counts." [09:07] wgrant, can do [09:08] jml: Thanks. [09:12] Why does our CSS make labels non-wrappable? It wreaks total havoc with our radio buttons! [09:16] jtv: I assume it makes sense in the general case - it's been in the style.css for a long time (r6771). But you can always provide a more specific rule for your exception? [09:17] noodles775: I'm not sure how I'd do that cleanly though, since these are LaunchpadForm-generated. [09:17] wgrant, it's in PQM now [09:17] statik, are you actually around? [09:18] jml: Thanks. [09:20] jtv: in your template, if the call to the form macro is within a div with a class of your choice, you can then style it easily. If you're using generic-edit etc., then I'm not sure :/ [09:21] Hrm, but even in the first case, it would apply to all labels on that form. [09:21] noodles775: it's in there, but CSS says "label { white-space: ... }" [09:22] I wouldn't even mind if I could override it for all labels on that form, except that the problem happens elsewhere as well. [09:22] bigjools, I don't have time this evening, but I'd really appreciate the chance to chat with you or cprov about package permissions. [09:22] jml: of course [09:23] maybe tomorrow, I can get up early [09:23] bigjools, that'd be great [09:23] noodles775: maybe the solution is to give these labels an extra class that doesn't have this setting. Right now it's just "label," and all labels have wrapping prohibited. [09:23] bigjools, thanks. [09:23] jtv: I'm not sure I'm understanding... if your form is rendered within a

... then you can add a style rule div.foo label {white-space: normal} and it will override the other one (as it's more specific). [09:25] noodles775: oh, hadn't tried it that way... I just tried setting it directly on the div, but that would be overridden by the label style of course. [09:36] noodles775: nope, that doesn't work either. :( I can override label locally, but that's probably not the way Mommy Would Want. [09:36] jtv: push your branch and I'll try it. [09:37] noodles775: Thanks. Pushing to lp:~jtv/launchpad/translations-person-3.0-mechanical [09:38] noodles775: will advise when it's pushed in all the way. [09:39] jtv: great, thanks! And the demo url you're using. [09:39] noodles775: the home page for whomever you're logged in as; Translations tab. [09:39] Perfect, thanks. [09:39] There, go to "Translations licensing." [09:40] Then make your window narrow enough that at least the text for the bottom radio button want to wrap. [09:40] *wants [09:40] noodles775: pushed! [09:47] jtv: http://pastebin.ubuntu.com/262973/ [09:48] That works, but I'd not really recommend doing it... [09:48] *facepalm* [09:48] noodles775: I agree, but what's the alternative? [09:50] jtv: just doing it with the current style and creating a bug explaining the problem so it can be solved generally - there might be other pages that would benefit too. [09:50] noodles775: can't argue with you there; I want this solved for other templates. [09:51] beuno: I'm a little confused by your problem earlier, since python-psycopg2 has the same version in jaunty and karmic. (And I can't see your paste) [09:51] Great. The less work-arounds we have, the easier it will be for us to fix/maintain etc. [10:12] Whoa, new /builders page. [10:13] Overall, quite shiny, though I don't like the way all the builders are now squished into one table, instead of being visually demarcated by architecture and official/ppa status [10:13] The summary side portlet is neat, though [10:13] maxb: you can click on the table headers to sort by arch/status etc. [10:15] that's simultaneously neat yet kinda pointless, as it disrupts the default sort that separates official/ppa, and I can't imagine any use case where you want the two mixed [10:16] maxb: bug 225738 and bug 245459 [10:16] Bug #225738: new build farm page UI is a regression for buildd admins [10:16] Bug #245459: Sort the list of builders by architecture then status, not by hostname [10:16] maxb: the default ordering has all the official first and then all the ppa builders (see the first column). It was a hard problem to solve - the input that cprov got was that it *needed* to be sortable. [10:19] hmm, why does it suddenly need to be sortable when the 2.x page is not? [10:20] maxb: I think the 1.0 was, and the 2.0 was considered a regression, as per the bug above (I wasn't around, but am just guessing). [10:22] Hmm. I think a good start at fixing the 3.0 layout would be to put ppa and official in separate tables [10:25] maxb: you can read a bit about the purpose and target audiences that cprov used for the current 3-0 design at https://dev.launchpad.net/SoyuzBuildersIndexPage [10:25] and chat with him later when he's around. [10:39] maxb: Also, the distinction between official and PPA, and between the architectures, is going to become much more hazy soon. [10:40] maxb: This new layout makes that much easier. === mpt_ is now known as mpt [10:57] Morning. [10:57] Hi deryck :) [10:58] deryck: btw, thanks for the TextAreaEditorWidget - it couldn't have been easier to integrate :) [10:58] noodles775, excellent! good to hear. [12:42] wgrant: around? [13:06] bigjools: I am. [13:06] wgrant: do you know who's the LP MOTU contact? [13:06] bigjools: mok0 and myself. [13:06] fabulous [13:07] wgrant: I grant you three wishes for Soyuz improvements [13:07] heh, and you're not allowed to choose things that you're implementing yourself ;) [13:07] * wgrant shall put it to the list. [13:22] wgrant: https://dev.launchpad.net/VersionFourDotO/Soyuz [13:23] bigjools: Thanks. [13:23] wgrant: I kept the items from the 3.0 pages on there as suggestions, feel free to re-use them [13:24] and now, food === henninge_ is now known as henninge-bbl [14:04] maxb: ping? [14:18] abentley: btw, I talked with Francis about the approach we talked about for bug 113993. He thought it sounded good, and had some suggested refinements. My focus right now is on a couple of other things, but I can maybe get to it next week or the week after. Alternatively, I can share what Francis told me with you, and you can do it. Let me know. [14:18] uh, mup? bug 113993 ? [14:18] https://bugs.edge.launchpad.net/launchpad-foundations/+bug/113993 [14:19] gary_poster: Thanks for following up on that. [14:19] np [14:20] gary_poster: For right now, I don't think I'll tackle it, but I'll let you know. [14:20] ok cool [14:21] I'll keep it on my medium-term to do list [14:42] maxb, it was because I was installing the karmic version, which dos not install in python2.4 [14:58] salgado: Are you still working on bug 419742? Do you want to drop it for 3.0. [14:58] Bug #419742: downloads page should include some data about the releases [14:59] sinzui, I've fixed that too; just need to do a couple minor tweaks that martin requested [14:59] salgado: okay === matsubara_ is now known as matsubara [15:48] anyone know why I could be getting this error? http://paste.ubuntu.com/263176/ [15:50] danilos, looks like something to do with translations ^ [15:54] beuno: libgettextpo isn't building... Problem with your rf setup. [15:55] * jtv is not being very helpful [15:55] jtv, care to give me any tips? [15:55] :) [15:56] beuno: slow but easy to try: move your lp-sourcedeps/sourcecode out of the way and run rocketfuel-setup [15:58] jtv, I can do slow and easy (right after I finish with my skype calls) [15:58] thanks [16:25] rockstar, ping [16:26] deryck, yo! [16:26] beuno: ping [16:26] barry, hi [16:27] beuno: hi! Please verify the rules here https://dev.launchpad.net/VersionThreeDotO/UI/Conversion#Heading%20rules [16:27] beuno: the coding guidelines haven't yet been updated, but please let me know if i got any of the ui rules wrong [16:28] * beuno looks [16:28] barry, I'm secretly on a call, but will get back to you about this [16:29] beuno: no rush. i have to run up to the local computer store. my wireless ap died last night. i'm going to combine that with lunch so will be back in 60 minutes or so [16:29] beuno: just pvtmsg me with your feedback and we can chat when i get back [16:29] barry, will do. Have fun computer-shopping [16:32] beuno: it's always fun shopping, never fun paying :) [16:32] barry, you can make it more exicting if you don't... === kiko is now known as kiko-fud [16:34] beuno: just be sure to respond to that facebook "bail me out" app request [16:34] :) [16:38] beuno: not sure why you'd get that, it's about compiling gettext python library... is this a new system? [16:38] danilos, upgraded to karmic [16:39] beuno: hum, haven't tried that yet, perhaps I will as soon as the new laptop gets here [16:48] beuno, have you given any thoughts on breadcrumbs for https://code.launchpad.dev/~name12/firefox/main/+bug/5/+delete ? [16:48] beuno: I think I need to design the product and distro series pages at the same time to ensure they are consistent [16:49] salgado, same as the bug, but an extra "Delete this bug"? [16:49] sinzui, I agree [16:50] beuno, that page is for deleting a bug-to-branch link [16:50] salgado, ah (I don't have a running rf) [16:51] salgado, than "delete link to branch"? [16:52] beuno, right, but the previous items should be identical to the bug's breadcrumbs? nothing pointing to the branch(es)? [16:52] salgado, I don't think so, no [16:53] that should be an ajax action anyway [16:53] fair enough === henninge-bbl is now known as henninge === salgado is now known as salgado-lunch === deryck is now known as deryck[lunch] [17:12] beuno: huh. How did you end up with the karmic version, not the karmic-ppa version? [17:12] maxb, by being stupid :) === beuno is now known as beuno-lunch === bd__ is now known as bd_ [17:14] oh, that reminds me, does a member of ~launchpad fancy copying subversion from ~maxb/+archive/launchpad to ~launchpad/+archive/ppa ? === matsubara is now known as matsubara-lunch === deryck[lunch] is now known as deryck [18:03] gary_poster: Just noticed your ping of 4 hours ago - I'm semi-around now === beuno-lunch is now known as beuno [18:10] maxb: np. hey. had two checkins of interest for you yesterday [18:10] first should get rid of zope branch completely [18:11] second switches to lazr.* distributions [18:12] oh that's excellent, I shall be sure to get another testrun going overnight then [18:12] that *might* fix the karmic buildout [18:12] I'd love to know if it does === salgado-lunch is now known as salgado === EdwinGrubbs is now known as Edwin-lunch [18:30] gary_poster: hmm... no luck on the "caused the installation of lazr.uri 1.0" buildout issue. [18:30]