[00:47] * thumper primal screams [00:47] thumper: fwiw. supermirror on staging is revno 8386. does that help? or you need app as well. ?? [00:47] spm: app [00:48] bleh [00:48] my great plan to speed something up has been derailed [00:48] * thumper needs to rethink [02:14] * thumper heads to class [02:15] spm: can you poke stub when he arrives and point him at his build failures? === thumper is now known as thumper-afk [02:15] thumper: sure [02:15] ta === thumper-afk is now known as thumper [06:28] thumper: as I'm sure you've noticed, no stub yet. tho he was sprinting last week, so may be in transit or similar. === henninge is now known as henninge-bbl [08:36] good morning [08:44] hi adeuring [08:44] hi jtv! [08:52] Good morning, people of Launchpad. === henninge-bbl is now known as henninge [09:00] hi henninge! [09:00] and hi gmb, as well :) [09:00] :) [09:00] hi jtv! [09:01] henninge: how are things coming along with the new model class? [09:02] jtv: it's still Monday morning here! [09:02] henninge: I'm not expecting you to have code, just asking if you have any ideas about it yet. :) [09:03] henninge: (in other words, whether you have any interest in a pre-imp call or whether you want to look at it some more) [09:03] jtv: no pre-imp yet ;-) [09:03] ok :) [09:07] Um, why are bugnotificationrecipients being archived, but not the actually useful bugnotifications? [09:21] wgrant: the idea was to archive both, so stub probably forgot to archive the BugNotification table, or intend to do that later. [09:21] stub: ^^^^ [09:22] Morning :) [09:24] BjornT: Or maybe the intention is to not prune bugnotification, as that isn't too supermassive at the moment. [09:27] wgrant: the intention was to prune bugnotication and have a cascading delete that delets the bugnotificationrecipients records as well. [09:27] BjornT: Ah. [10:21] I hate software. [10:26] gmb: so do I [10:26] gmb: I hate hardware too [10:27] thumper: Hmm. I also hate ISPs over whose network a connection to Canonical IRC seems astonishingly laggy. [10:27] (As does any connection to my IRC proxy. I wonder if that's because both are over SSL). === mrevell changed the topic of #launchpad-dev to: Launchpad Development Channel | Week 3 of 2.2.8 | https://dev.launchpad.net/ | Please use #launchpad for support. | https://launchpad.net/~launchpad-dev | Get it: https://dev.launchpad.net/Getting | http://people.canonical.com/~herb/ | http://paste.ubuntu.com/ [10:41] BjornT, FTR, I'm not able to connect to the internal IRC server reliably ATM for the morning call.. [10:41] s/call/meeting [10:42] gmb: ok. does skype still work? [10:43] BjornT: Yes. === danilo-afk is now known as danilo === danilo is now known as danilos [12:58] * gmb lunches === mrevell is now known as mrevell-lunch [13:26] gooooood morning everyone [13:27] hiya beuno :) [13:27] morning beuno [13:27] how's it going noodles775? [13:28] beuno: yeah ok, glad to be back to doing some actual dev work :) (had a busy OCR day last thursday). And you? [13:29] noodles775, sprinting with the Ubuntu One guys [13:29] beuno: ah, great! Hope it's lots of fun! [13:31] noodles775, it is :) === mrevell-lunch is now known as mrevell [13:44] Does anyone know when is edge going to update? It's been *ages*! [13:45] allenap: yeah, it's a bit annoying. staging is up-to-date, though, if you need to test stuff [13:45] allenap, let's stalk flacoste as soon as he comes up [13:45] intellectronica: Cool. [13:45] beuno: Yeah :) I want to *see* new stuff already :) [13:46] me too! [13:48] beuno: not urgent but just a reminder, that we still need to figure out a nice hover image for pages with large clusters of edit icons [13:48] intellectronica, yes, it's on the top of my list [13:48] I will try and get something out today or tomorrow [13:48] coolio [14:06] herb, question: did my CP of 9107 and 9030 really go onto the codehosting server? === barry is now known as barry-away === barry-away is now known as barry [14:15] hey leonardr. My branch to put all of the zope code as distributions as assembled by buildout is waiting for review from flacoste. I know that was a prerequisite for migrating launchpad to the newer versions of lazr.restful and friends, but I don't remember why--and particularly, if there was something else that needed doing before it could be merged in. Do you remember? [14:18] gary: i don't remember. i think it had something to do with the interplay between launchpad, lazr.restfulclient, and launchpadlib [14:18] but at this point i think the best thing to do is just try it [14:21] leonardr: ok. I might give that a brief whirl sometime today to just get a feel for the problems. [14:24] barry: btw, Jim Fulton now has a predefined set of Zope packages ("Known Good Set") that work together and with Python 2.5 and 2.6. (the improvement over the previous situation is that each package was supposed to be ready for 2.5/2.6, but there was no automated testing to show that the individual packages all worked together reasonably). This should hopefully help us in our migration path. [14:25] gary_poster: fantastic! [14:26] leonardr: lazr.restful.simple isn't in the released version of lazr.restful yet, is it? [14:26] barry: no, but if you need it i can do a release [14:27] leonardr: if you can do that this morning it might be cool. i think i can at least do a partial simplification of my code before i give my lazr.restful talk tonight [14:27] barry: ok, i'll do that now, should just take a minue [14:28] leonardr: awesome, thanks! === kiko-afk is now known as kiko [14:32] barry, can you review this trivial diff: https://pastebin.canonical.com/21213/ [14:32] sure [14:32] r=me [14:35] jtv: ping [14:35] henninge: pong [14:35] jtv: ah, you dropped of the internal IRC [14:35] jtv: Is there an existing example for something like "TranslatedMessage" ? [14:36] henninge: the new model class? Not that I can think of, and that's why we are having these problems. TranslationMessage sort of fakes it right now by keeping track of the "browser_pofile." [14:37] jtv: I see. [14:37] I'm getting the error "Sender not authorised to commit to branch lp:~launchpad-pqm/lazr.restful/trunk" when I try to pqm-submit a lazr.restful branch. Any ideas what I'm doing wrong? [14:38] henninge: you could see ProductSeriesLanguage as a recent example of the general construct though. [14:39] jtv: let me look at that. Yes, I was looking for the general construct. [14:39] henninge: at its heart it's a "tuple" of two proper model objects." [14:39] adeuring: i would try specifying the bzr+ssh url instead of the lp one [14:39] BjornT: thanks, I'll try that [14:40] jtv: a productseries and ... a pofile ... a language? [14:40] henninge: in that case, a productseries and a language. There can be multiple POFiles, otherwise there'd be little point. :-) [14:40] jtv: ah, of course ... [14:41] adeuring: yes, using the bzr+ssh protocol should work. however, it didn't work for me last time i tried for independent reasons, so let me know if it works for you or not [14:41] jtv: there is a ProductSeriesLanguage and a -Set. [14:41] intellectronica: sure... just isuued the pqm-submit again... [14:42] jtv: so analogous a TranslatedMessageSet would be potmsgsetset (a potemplate) bound to a pofile. [14:43] henninge: no, that's your TranslatedMessage AFAICS. If you want a -Set, that's basically the POFile. [14:43] henninge: sorry, I misread. What you said is basically a POFile. [14:45] henninge: or maybe I'm confused about your terminology, since the "Set" in POTMsgSet is in no way related to the -Set in TranslatorSet, POTemplateSet etc. [14:45] barry: ok, lazr.restful 0.9.3 is out [14:45] jtv: yes, I used "SetSet" there ... [14:46] * jtv cries softly into his keyboard [14:46] jtv: so I a aware of the different meaning of set [14:46] * henninge gets jtv something to wipe his keyboard with . [14:46] leonardr: thanks! [14:47] jtv: We will need a DummyTranslatedMessage, too. [14:48] won't we? [14:48] henninge: yes, I think we need one for dummypofiles. [14:49] henninge: but it may depend on the details... a TranslatedMessage referring to a DummyPOFile may be good enough. [14:49] jtv: ok, details later ;-) [14:49] I certainly wouldn't introduce a DummyTranslatedMessage unless it's necessary. [14:49] jtv: ok [14:50] jtv: hm [14:51] jtv: I just realized I am not sure what we are trying to accomplish with the new model class. [14:51] jtv: please don't cry!1 [14:52] henninge: you're trying to build a model class that we can attach the "one translatable message plus translation & suggestions" view to. [14:52] jtv: A TranslationMessage is already specific to a POFile. Why can't we work with TranslationMessage directly [14:52] oh! [14:52] jtv: So it is "All TtranslationMessages for this POTMsgSet for this POFile" ? [14:53] Pretty much, yes... I'd say "the translation state for this POTMsgSet as seen in this POFile." [14:53] jtv: ok, or that ;-) [14:53] As you can see, the TM is "one step too far down" to do this well. [14:54] jtv: Yes, I see that now. I was only confused for a short time. [14:54] Also, right now, there's the weird problem that when you look at an individual message now, you're basically looking at a TranslationMessage that may be shared between POFiles. So... how do we get the sequence number? [14:55] jtv: So I need to provid acces to 1. The current message 2. the imported message 3. the shared message 4. all local suggestions 5. the external suggestions. [14:56] jtv: POFile gives us POTEmplate [14:56] jtv: doesn't it? [14:56] oh no, it's the other way round. [14:56] henninge: there's no reference to either (except if the TM is diverged, but let's not count on our luck :) [14:56] jtv: so the model class needs a concept of which potemplate it refers to. [14:57] henninge: exactly, that's the other bit of information you'll be holding onto in your new class. [14:57] so init would be (POTMsgSet, POFile, POTemplate) [14:57] henninge: POTemplate == POFile.potemplate [14:57] henninge: but otherwise, yes, that's it. [14:57] jtv: oh, it does [14:58] * henninge has to look at jtv's great diagram again ... [14:59] jtv: that's what I meant when I said [14:59] jtv: POFile gives us POTEmplate [15:00] henninge: and you were right there, not wrong :) [15:00] jtv: so init would just be (POTMsgSet, POFile) [15:00] henninge: yup [15:01] that's the basic information you're holding together, and everything else will follow from that. [15:01] jtv: let me draft the interface to see what I might be missing. [15:02] jtv: btw, should we stick with "TranslatedMessage"? [15:03] henninge: it's going to suck typing that, but I guess we'll have to get used to it. I can't think of anything better. (Even if the message is untranslated, it's basically just a TranslatedMessage without a current translation) [15:07] henninge, jtv: TranslatableMessage is what it is [15:07] \o/ [15:07] OMG ugly: notfound-traversals.txt [15:08] danilos: The probelm is just that "Set" is already used in different places for a "List" ... ;-D [15:08] henninge: oh well, I believe 'TranslatableMessage' is good, except that I don't like our choice of 'TranslationMessage' then :) [15:09] bac loot at the firefox project as sample person in dev to see a converted layout. The sidebar uses all the conventions [15:09] * henninge just noticed the "able" instead of "ed" [15:09] but that's ok. [15:09] barry: hey, at least it's not spending all its time in sleep() any more! [15:09] bac: mozzila is also a good one to look at [15:09] danilos, henninge: I initially coined -able, but I think -ed is just fine; less confusion possible with POTMsgSet [15:10] * henninge halts in the middle of renaming [15:10] jtv: POTMsgSet is English/SourceMessage :) [15:10] danilos ? [15:10] danilos: don't go there today... just... don't. :-) [15:11] jtv, henninge: there's no solution possible that works well for everything, but I like Translat*able*Message much more than anything else [15:11] * barry wants to spend all his time in sleep() [15:11] danilos: you're the boss [15:11] one could say that we never fixed this because we couldn't come up with a proper name :) [15:12] ...but changing names every few years is almost as good, right? ;-) [15:12] barry: you can instead look at the incremental diff for the review you did for me on Friday (thanks!), per your request, I made it another 800 lines :) [15:12] barry: one of those things in life we shouldn't automate [15:12] ooh! pretty footers [15:12] jtv: heh, yeah, definitely, that helps a lot keep the momentum [15:13] jtv: oh trust me, i'm not assigning that task to my clone army, i'm keeping that one for myself. sure, i'll let a clone eat for me now and then, or shave or stuff like that [15:13] oh, here comes a review ... [15:14] barry: I wouldn't want to automate eating either, but frankly, "hungry" is a problem I already fixed. Why does it keep regressing? [15:16] sinzui, https://staging.launchpad.net/ubuntuone-client [15:16] TraversalError: (, 'specications')
[15:16] beuno: if someone would only review my fix [15:16] form friday [15:17] sinzui, I should of know you had already fixed it :) [15:17] flacoste, hi [15:17] jtv: i don't know, but my unit test for that is always dumping core [15:17] flacoste, where's my edge rollout? :) [15:17] beuno: The Involvement portlet will work with everything as soon as I can land it [15:17] hi beuno! [15:17] beuno: good question [15:18] barry: that's invariable a sign of bad input [15:18] jtv: bzr rm dairy.txt [15:18] barry: please don't take this analogy to "bzr push"... :) [15:20] :-D [15:21] beuno: should be back later this morning [15:22] flacoste, merce beaucoup [15:23] rockstar: have you guys seen something like https://code.edge.launchpad.net/~danilo/launchpad/bug-410579/+merge/10179 (OOPS-1325EA163) before? [15:23] beuno: hay no problema [15:25] flacoste: I think you want plural there (i.e. 'problemas') :) [15:26] actually, just invert the words: no hay problema [15:26] oh well, I guess I have no clue :) [15:26] but what about the Spanish? [15:26] danilos, you need more sprints in Argentina [15:27] beuno: we'll have them, I am all for it :) [15:29] beuno: do you know the runes to get those vertical lines I needed in the two-row-headed table on the DSPR mockup I showed you? [15:29] rockstar, hi [15:29] bigjools, css rules? [15:29] jtv: I have a review on the list now. What was that about your time? [15:30] henninge: I'm past EOD, I need to grab dinner, and I've got a call coming up. [15:30] beuno: I would hope so, but I don't know much about how to do that sort of thing [15:30] henninge: so find another sucker. :-P [15:30] jtv: ok, talk to you tomorrow, then. I'll grab the other sucker, then... ;-) [15:30] bigjools, border-right-style: solid;? [15:30] henninge: Hals- und Beinbruch. :) [15:31] * jtv runs [15:31] lol === jtv is now known as jtv-brb [15:32] sinzui, beuno ping [15:32] Hi barry [15:33] barry, pong [15:33] beuno, sinzui: hi. so on /people, there is a small action portlet to view projects, view distributions, view people, view meetings, register a team, merge accounts. i'm making this page w/o side portlets. how much of those functions to we want to migrate inline [15:34] beuno, sinzui i have already moved merge accounts inline. is it useful to keep the 4 view actions and teh register a team action? [15:34] barry: That is an excellent question. [15:35] barry: For content objects, that menu would be rendered as a related pages section; the last portlet in main content. [15:35] sinzui, barry, would you guys like to joins us in the UI call today? [15:35] beuno: sure what time [15:35] sinzui: yeah. i really don't want to add a portlet just for those actions [15:36] mars, rockstar, EdwinGrubbs, noodles775, intellectronica, jtv-brb, barry, sinzui, UI call in 25' [15:36] yup. [15:36] barry: beuno: the top-level collections are not common content objects, may be should consider a sidebar for them with an action style menu [15:36] i'm just not sure how useful the 4 'view' links are. registering a new team, yeah i can see that, and i could easily add that inline [15:36] sinzui, maybe that's the right thing to do, yes [15:36] beuno: cool [15:36] beuno: sure. cheers for the reminder [15:37] barry: the two new menu presentations work with NavigationMenus. [15:37] beuno, sinzui so, add an action style menu for this page and move those links into that? [15:38] beuno, sinzui and put that menu in a sidebar? [15:38] barry, probably, but a screenshot would give you a definite answer :) [15:38] barry: It is possible to define one menu in browser/__init__, then use it on all the top level collections. [15:38] sinzui: they are slightly different among the three top level pages [15:38] (currently) [15:39] beuno: gotcha ;) [15:39] beuno, hi [15:39] barry: should they be? We can use enable to enable some links. [15:39] rockstar, I've been wondering what's up with answers? [15:40] sinzui: if we move some of the actions inline, then they would share the 'View Thing' links. there are some admin functions in the sidebars currently, but like with /people, i think we can move those inline [15:40] sinzui, beuno, should https://edge.launchpad.net/codeofconduct/1.0.1/+sign be locationless? [15:40] sinzui: so shared would be view projects, view distributions, view people, and view meetings [15:41] salgado, I think so, or "Launchpad.net"? [15:41] beuno, what do you mean? [15:41] barry: these pages could be the first legitimate use of a context and view navigation menu. Consider that (+) Register is in a action menu, and the other collections are inline in a related searches menu [15:41] rockstar, I haven't seen any landings? [15:41] beuno, it's because I haven't yet landed anything. [15:41] beuno, "Launchpad.net"? [15:41] sinzui: that makes sense [15:42] rockstar, any reason for it? as in, are you going to land everything in one chunk? [15:42] beuno, I have a pipeline here where Answers is being ported. It's getting along. [15:42] sinzui: okay cool. let me see what i can mock up [15:42] salgado, I think the location for everything that's not in a context is "Launchpad.net". sinzui? [15:42] barry: I think we need to move the browse/search links inline if they exist [15:42] beuno: salgado: yes. [15:43] rockstar, I'm asking because we're tracking progress in https://devpad.canonical.com/~mars/conversions.html [15:43] rockstar, and answers is at zero, and I get asked about it a lot :) [15:43] beuno, I understand this. If I wasn't filling in for abentley on CHR today, we'd probably see changes. [15:43] sinzui: /people doesn't have those links [15:43] beuno, why anyone is asking you I don't know. [15:43] beuno: everyone. do not just look at the number of templates converted. thumper ans I are also deleting pages [15:44] sinzui, we don't have a way of tracking that, do we? [15:44] sinzui, good point. I've deleted probably 8 templates so far. === deryck_ is now known as deryck [15:45] barry: true, but a related pages menu should be shared by all the top-level collections, that Is why I mention that we want a common menu in __init__.py and that is related, not action [15:45] rockstar: you rock === jtv-brb is now known as jtv [15:45] beuno, sinzui, I'm not following you. what's this "Launchpad.net" location? is it just another name for the locationless macro? [15:45] sinzui: yes, i think so [15:45] beuno: we needed to save the count of total pages when we started [15:45] sinzui, :) I'm not done yet. [15:45] sinzui, generic-edit is frakkin awesome. [15:46] yep [15:46] salgado, yes [15:55] henninge: I've filed a https://blueprints.edge.launchpad.net/rosetta/+spec/pofile-translate-page and the first bug for you (bug #414856, linked from the blueprint as well) [15:55] Bug #414856: Provide a model class for TranslatableMessage [15:59] danilos: cheers, now I find purpose in life! [15:59] ;) [16:01] henninge: there you go, I am always happy to be so helpful :) [16:03] barry, UI call! [16:05] kiko's # [16:05] yay [16:05] hola kiko [16:12] beuno: ah dang. calling in now [16:15] beuno: edge updated and will be updated daily from now on [16:15] flacoste, danke [16:22] gary_poster, flacoste: ping [16:22] sidnei: pong [16:23] gary_poster: we found an issue that seems like it might affect launchpad too [16:23] sidnei: what's up? [16:23] gary_poster: request.supportsRetry() does a sleep() with a random number in the version of zope3 we are using [16:23] heh, uh...weird [16:24] sidnei: is that as insane as it sounds? [16:24] gary_poster: yeah :/ [16:25] gary_poster: seems like this was reported as #401586 [16:25] Bug #401586: Zope sleeps during tests [16:25] except it's being papered over during tests, and might still affect production [16:26] sidnei: looks like it's only being papered over for one test, if I read that bug report correctly. have you seen if this is in trunk for that package? [16:27] gary_poster: about to check [16:27] me too ;-) [16:27] sidnei: yes [16:28] sidnei: not an env; apparently supposed to be used as a monkey patch... [16:28] changed I mean [16:29] gary_poster: yeah. also, the sleep() should probably be in retry() not in supportsRetry() right? [16:30] sidnei: yeah, I would sure think so. have you done an annotate yet to see who did this thing? I went to the web interface. [16:31] sidnei: stevea did it, with Jim following along behind... [16:32] gary_poster: indeed [16:32] sidnei: you wanna follow up on list/IRC/steve, or me? [16:33] maybe we should look at doctests really wuickly [16:33] quickly [16:35] sidnei, gary_poster: my guess would be that if two requests collides with each other, causing a retry, they shouldn't be retried at the same time, causing them to collide again. not sure how feasible that use case is, though... [16:35] sidnei: I don't see anything explaining it in doctests, interfaces, or tests [16:35] BjornT: yes, but why in supportsRetry rather than retry? [16:36] gary_poster: actually, it was jim that added it, and stevea just cleaned up 1/True [16:37] sidnei: ah, ok, I was not paying attention to the revids closely :-/ [16:37] gary_poster: yeah, that seems a bit crackful... could it be that the transaction gets started before retry() is called? [16:38] nope, doesn't look like that would be the case [16:38] BjornT: no don't think so either [16:39] sidnei: you want me to contact Jim about this? Or are you going to pursue? [16:40] gary_poster: let me ask therve, he's the one that found it. we are about to leave for lunch. [16:40] sidnei: ack [16:42] gary_poster: so, if you can ping jim that would be great, as you know we are resource constrained. :) [16:42] sidnei: ok will do. [16:50] beuno: colgroup FTW -> http://people.canonical.com/~ed/dspr_mockup4.png [16:50] bigjools, NAIS [16:51] that page still has issues, but getting there [16:51] beuno: I am only aware of the file size one now ...? [16:52] bigjools, yes, and things are kind of squished? [16:52] fonts are wonky [16:52] it renders differently in Konq vs FF [16:53] danilos, hi [16:53] Ursinha: hi, how's it going? [16:54] danilos, good, good [16:57] flacoste, https://dev.launchpad.net/UserInterfaceReviewNotes [16:58] beuno: https://dev.launchpad.net/ReviewerSchedule [16:59] leonardr: ping [16:59] barry, yo [16:59] leonardr: there's a critical bug in lazr.restful 0.9.3. can you do a quick update to 0.9.4? simple.py does not import traceback [17:00] dammit [17:00] sure === matsubara is now known as matsubara-lunch [17:00] let me make sure my test works with that change... [17:02] leonardr: yep. my test fails but for unrelated reasons. adding traceback to simple.py will fix the problem [17:02] flacoste, I have a user asking me why we're using SSL. What should I tell him? [17:03] leonardr: rs=me [17:03] leonardr: before you make another release... [17:03] ... [17:03] maybe you should remove this 2.5-ism [17:03] File "/home/gary/canonical/lp-sourcedeps/eggs/tmpm4lpr9/lazr.restful-0.9.3-py2.4.egg/lazr/restful/example/wsgi/root.py", line 44 [17:03] self.schema = ('https' if config.use_https else 'http') [17:04] gary_poster: 2.6'ism? [17:04] barry, heh, I couldn't remember :-) . I went for being confident in the hope that at least I'd communicate my point ;-) [17:04] :-D [17:05] kfogel, any news on the LP badge? [17:12] deryck, description editing on edge, woooooooooooooooooo [17:12] beuno, excellent!!! [17:14] deryck, what do you think about making the descriptiuon text larger? [17:14] beuno, totally fine with that. I did wonder if it was small once. Early on. :) [17:14] :) [17:16] beuno, I hate to be too lazy, but do you mind opening a bug on that for me? [17:17] deryck, damn, I was about to ask you the same thing [17:17] heh [17:17] you win for being slightly less lazy [17:17] leonardr: i'm going to get some lunch. ping me if you need a review. would love to get 0.9.4 out today (for my talk tonight) [17:17] * barry -> lunch [17:17] beuno, and that's the grown-folks version of the little kids "not it!" played out in IRC. [17:17] gary: since barry just left, would you review https://pastebin.canonical.com/21222/ ? [17:17] deryck, LOL [17:21] leonardr: yes. looking [17:21] leonardr: r=gary [17:21] yay === beuno is now known as beuno-lunch === salgado is now known as salgado-lunch [17:23] gary: argh, i forgot this part [17:23] === modified file 'src/lazr/restful/version.txt' [17:23] --- src/lazr/restful/version.txt 2009-08-17 13:38:04 +0000 [17:23] +++ src/lazr/restful/version.txt 2009-08-17 16:23:45 +0000 [17:23] @@ -1,1 +1,1 @@ [17:23] -0.9.3 [17:23] +0.9.4 [17:24] gimme an ok on that and i'll do the release [17:24] leonardr: +1 [17:27] barry: 0.9.4 is out [17:27] leonardr: awesome thanks [17:30] sinzui, could I get you to look at https://answers.edge.launchpad.net/launchpad/+question/80132 ? [17:33] rockstar: I am replying. I need to read the code first [17:33] sinzui, great, thanks. [17:39] beuno-lunch, sinzui: I made a few tweaks: http://people.canonical.com/~ed/dsp_mockup_with_linkage.png [17:39] any comments? [17:53] bigjools: If you review my involvement branch, I can fix a big error on edge and you get a one-line replacement for the dsp involvement portlet. [17:54] sinzui: can its context be set differently to the pillar? [17:54] sinzui: I can review it, but I have to disappear for 2h first, it's food + kids bed time [17:54] bigjools: I was *very* clever in my branch, I adapt to the pillar, read the official apps, then enable only those menu items [17:55] sinzui: in my case, it needs to work with a DSP, not a D [17:55] bigjools: I was thinking of you when I wrote the branch [17:55] even though the pillar is a D (distribution) [17:56] anyway, bbiab [17:56] bigjools: I still do not understand th latest releases portlet. How often does *all* that information change. It looks like package details which I expect to be in the main content. [17:57] * sinzui is still an idiot about anything that has a sourcepackage in its name [17:57] sinzui: it *can* change on any upload, whether it does or not is another matter [17:58] bigjools: for example, I always expect the maintainer and version to be the first portlet in the top-left of the main content like project and product [17:58] * bigjools has to go, I can chat about this later! [18:01] * gmb -> food === matsubara-lunch is now known as matsubara [18:06] leonardr: getting some basic failures from attempt to switch to distributions for lazr.* . One: "import lazr.restful.testing.layers" fails. Indeed, that module does not exist in 0.9.4. Where is it now? (See lib/canonical/lazr/testing/layers.py for this particular usage) [18:10] gary: sorry, i gotta go pick up lunch. i have questions for you as well; we'll swap when i come back [18:11] leonardr: cool, I'll do the same === gary_poster is now known as gary-lunch === leonardr is now known as leonardr-lunch [18:14] leonardr-lunch: thanks! [18:16] dang, edge.lp just got even sexier with the inline description editing and new description layout [18:23] bigjools, in the portlet [18:23] "Uploaded by:" should be in a separate line than the person's name === beuno-lunch is now known as beuno === salgado-lunch is now known as salgado [18:29] rockstar: sorry, forgot to hit ENTER before getting distracted by something else: SSL, we are using to secure logged in connection, we should drop it for anonymous users [18:29] there is a bug open about that [18:30] flacoste, well, I think the user was confused. He was calling it an "SSL leak" because Google could index our bugs. [18:30] hmm, yeah, that is confused [18:40] salgado, how are those breadcrumbs going? === leonardr-lunch is now known as leonardr [18:45] beuno, I need to talk to flacoste about them, but if he's ok with the idea of using canonical_url(obj, rootsite='mainsite') for the links there, I should have it ready for review RSN [18:45]