=== micahg1 is now known as micahg === flacoste_afk is now known as flacoste [02:26] hggdh: That's a bug in the greasemonkey scripts, caused by a change in the links three weeks ago. [02:26] wgrant, can you expand, please? [02:27] what I see is an error (encrypted), and a status==0. If I understand what is wrong, I may be able to correct it [02:28] hggdh: Person links recently had the "bugs." dropped from them. This is quite intentional. Your problem might be that the browser security is kicking in and refusing to let you make a cross-domain request. [02:29] wgrant, I am rather new to this(in fact, my very first JS script ever). How could I bypass the browser security? [02:29] hggdh: You can't. You have to make requests to the same domain. [02:30] so we are pretty much dead with the GM scripts against bugs.edge? [02:30] hggdh: What you could do is construct a relative URL. This probably means taking everything after the '~' from the person URL, and prepending '/~'. [02:30] hggdh: No, you just have to make the requests against bugs.edge. [02:30] hggdh: Link me to the script, and I will fix. [02:30] (some of LP's JS broke the same way, so I know how to fix) [02:32] cool! [02:32] https://code.edge.launchpad.net/~hggdh2/launchpad-gm-scripts/hggdh2 [02:32] it's the lp_karma_suffix.user one [02:32] and, to boot, I will learn more! [02:33] I understood I could not change the host in a XMLHttpRequest... [02:34] hggdh: You just have to change the URL that you give it. [02:34] hggdh: Is that the canonical copy of that script? [02:38] wgrant, this is the change I proposed to add in the ability to follow a commeter's bug work (for bugs-control) [02:39] I do remember I *did* change the URL, and still got hit by the status=0 -- so I guess I did it wrong, out of the depths of my ignorance [03:55] wgrant, ping [03:57] wgrant, I think I got it. I changed the links as here: http://paste.ubuntu.com/262862/, and it is working now. I *guess* this is what you menat [04:00] hggdh: Sorry, got distracted. I don't see how that could work, but if it does, great. [04:01] wgrant, sir, if YOU don't know, who am I ;-) [04:01] so this was not quite what you meant [04:02] hggdh: What is 'link' there? [04:02] https://bugs.edge.launchpad.net/ubuntu/+source/coreutils/+bug/416981 [04:02] Launchpad bug 416981 in coreutils "coreutils: du(1) doesn't summarize correctly" [Medium,Incomplete] [04:02] in this case [04:03] so the fully-resolved link gets to be https://bugs.edge.launchpad.net/ubuntu/+source/coreutils/+bug/416981/~/https://edge.launchpad.net/~rrossouw/@@+portlet-details [04:03] hggdh: Ah, now I've actually looked at the code... [04:03] people_cache[person]['team_link'] = '/~' + person + "/+participation"; [04:03] yes [04:03] That should do it. [04:03] cool. Thank you very much. Bug-control bows to the master [04:04] I will add it in the bug, and I guess it will be invalid for all tasks except gm-launchpad-scripts [04:05] hggdh: Does my suggestion work (also applying it to karma_link)? [04:09] wgrant, yes, both of them work fine [04:10] that's a weird redirection, at least to me, though [04:10] hggdh: Why? [04:10] hi! there seems to be a package that is marked among bugs related to me, even though I'm not subscirbed to it anymore. how do I remove the association with me? [04:11] Q-FUNK: Where are you seeing this bug? [04:11] https://bugs.launchpad.net/planner/+bug/24281 [04:11] Launchpad bug 24281 in planner "(Breezy) Planner Gantt view vertically misaligned" [Unknown,Confirmed] [04:11] I have previously commented that bug, ages ago. [04:11] wgrant, please keep in mind that -- until now -- I had never programmed for web, or Javascript. The tilde in the middle is what I failed, er, fail, to understand [04:12] Q-FUNK: A comment will make it show up on your related bugs page until it is closed. [04:12] hggdh: Much like normal paths, putting '/' at the start of a link makes it absolute. [04:12] hggdh: So the path my code constructs is 'https://whatever.domain.we.are.on.now/~wgrant/+participation [04:14] wgrant which is quite different from the path I got to work (even given that I did not quite get your instruction) [04:15] hggdh: I would really like to know how that worked. It doesn't make sense. [04:15] I see that it does work, though. [04:16] yes... I cannot answer you, but please rest assured that I will keep testing [04:29] er, another question: is there a way to retract a merge proposal on LP? [04:31] hggdh: you can delete it [04:33] oh [04:33] how? [04:39] wgrant: ah, ok. thanks [04:41] hggdh: look for the trash can icon [04:44] wouldn't it delete the whole branch? [04:46] ah, I can just recreate it [04:55] not on the merge proposal page [04:58] yes, I understand now. thank you [04:58] rather smallish icon, BTW [05:13] yeah, noone ever seems to find it unaided [05:13] our designers are working on it [05:13] (i hope!) === jtv1 is now known as jtv [05:35] hello, The founder of a LoCo team has passed on the leadership to me. Does he have to do something on Launchpad to make this change "official"? [05:46] does anyone know the answer... he doesn't speak english, so he doesn't get very well with launchpad [05:54] spm, i'd like some help with some pqm stuff this afternoon [05:55] not right now [05:55] thumper: still here? [05:55] Legendario: more an ubuntu question than a launchpad question really, isn't there #ubuntu-loco or something? [05:55] Legendario: Does the founder own the team on Launchpad? [05:56] poolie: sure [05:57] wgrant, he owns. But this is the question. Does he have to give me this ownership? [05:58] Legendario: He doesn't have to, but it would make sense. [05:58] Legendario: So, you should ask him to. [05:58] poolie: yes, although I was going to make dinner [05:58] poolie: whazzup? [05:58] thumper: we're going to upgrade our branches on lp soon [05:59] ok [05:59] what sort of timeframe? [05:59] probably by upgrading the pqm master branch, pushing that to a new location on lp [05:59] hours [05:59] soon, in an hour [05:59] ok [05:59] right, this arvo [05:59] setting pqm to primarily host on lp [05:59] cool [05:59] making that new branch the dev focus [05:59] ie leaving lp:~bzr/bzr/trunk as it is, with stuff stacked on it [05:59] * thumper nods [06:00] this also should get pqm going into lp more smoothy [06:00] smoothly* [06:00] spm, that's what i was going to say to you :) [06:00] we can talk more about the details [06:00] thumper, any comments or concerns? [06:00] no concerns [06:01] should be magic [06:01] ish [06:01] wgrant, the problem is that he doesn't speak english, so he doesn't know how to work with launchpad well. [06:01] * spm prepares the cockrel and obsidian knife [06:01] I think people who are going to push to LP need to know that they need their local repos in the 2a format [06:01] so email to bzr-dev or whatever [06:01] is the branch sufficiently reconciled? [06:02] Legendario: The page needed is https://launchpad.net/~someteam/+reassign [06:02] Legendario: Just get him to enter your username there, and it will be done. [06:04] wgrant... thanks a lot man [06:06] thumper: yes, but i'l check first [06:06] kk [06:07] spm, i'll do a check here first and send mail, then you can slice away [06:08] poolie: cool, I'll start lighting the candles on the hexogram (pentagrams are so 19th century) [06:20] i'll leave. thanks guys [07:46] spm, so, shall we do it? [07:46] should i call you? [07:49] poolie: if you prefer? [07:49] either way [07:49] did you see my mail? [07:49] Hi folks - is there anyone here who was involved with savannah <> launchpad stuff? [07:51] heh. I like the way that mentioning 'savannah' kills any conversation [07:57] Elwell: heh. in what sense involved? [08:00] spm: I've been hacking on some CLI tools for working with savannah.cern.ch for getting patch/bug status etc, posting comments, changing fields (python twill/beautifusoup stuff) and wondered if it'd be any use to launchpad [08:00] no I haven't looked at the launchpad code *at all* yet so don [08:00] ''t know how it works [08:04] Elwell: hrm. good question indeed. LP makes an API available which quite a few folks abuse.. err *use*. So there may be overlap/resuse scope of some sort. [08:06] yeah, and savannah is such a POS^W historically 'interesting' design I now know why its not been integrated before 1.22 [08:06] lack of savannah API is a pain [08:06] Bugs people might be interested in that. [08:07] They'll be around in a couple of hours. [08:07] Elwell: perhaps the best way - join the lp dev list; and post a summary of what you've done and ask for interest/pointers???? [08:08] ok wasn't sure where the main discussions took place (ml/irc/other) [08:08] all of the above :-) [08:09] my thinking via list, is that way you'll hit the widest possible audience [08:09] yeah, but then I'd have to de-cruft my code if it hit public scruting :-) [08:09] hahahahahahaha [08:10] Affero GPL == GPL for web services, so you have to release code? [08:11] kinda yes. I'm not all that up on the detail of the difference [08:11] Elwell: fwiw, the team/list in question is this one: https://edge.launchpad.net/~launchpad-dev [10:39] hi I've try to suppress a bunch of pkg from the telepathy ppa and I get Error ID: OOPS-1340ED158 [10:39] https://lp-oops.canonical.com/oops.py/?oopsid=1340ED158 [10:40] (37 pkgs) [10:41] mmm it looks like the pkg have been removed [10:41] bigon: nice bug though, would you mind filing a new bug describing how you caused that please? [10:41] (on soyuz) [10:42] ok [10:42] thanks [10:45] bigjools: bug #422459 [10:45] Launchpad bug 422459 in soyuz "Error while removing pkg from ppa" [Undecided,New] https://launchpad.net/bugs/422459 [10:46] bigon: thanks - I saw you were using a filter as well, can you describe a bit more about using that? I think it's what's caused the oops. [10:47] oh right [10:48] done [10:49] thank you === mpt_ is now known as mpt === soeb_ is now known as soeb [11:34] Hi! Concerning PPA's, say the build failed and I fixed a problem in the rules file and want the build queue to rebuild my package. If I 'dput -f' the same package (version still the same) again, the build server rejects the package with: 'File blabla.diff.gz already exists in bla, but uploaded version has different contents'. The help page ist not very helpful, as I did a 'debuild -S' before the second upload. Any ideas? [11:35] flohack: you can't upload the same version more than once [11:35] flohack: you have to increment the version as the source is already published. [11:35] yeah, what he said :) [11:36] flohack: I'd recommend using pbuilder to check your package locally before uploading - that way you can be sure that it will work. [11:36] I wouldn't go so far as to say "be sure", but it should increase your confidence at least :P [11:37] idnar: heh, ok :) [11:39] Why does the project download page say when each file was last downloaded? [11:39] If I point wget recursive at the page, will it change everything to "today"? [11:41] mpt: I don't think the registry guys are around until later. [11:42] mpt: btw, jtv had a question about a style-rule in styles.css to which you might know the background - on lp-dev email list. [11:42] mpt: a case where making labels nowrap works out badly. [11:43] mpt: Yes, and other spiders like to do it too. [11:43] mpt: So it's fairly pointless. [11:43] jtv, sent how long ago? The only message I see from you is on warthogs [11:43] mpt: launchpad-dev, 09:14 UTC [11:43] oh, as in a couple of hours ago? that's why [11:44] mpt: unless your spam filter dev-nulls "considered harmful" in subject lines... :-) [11:44] haha [11:44] ah yes, I know the answer to this one [11:47] bigjools: Thank you guys! [12:14] jtv, replied [12:15] mpt: thanks, and hi btw :) [13:11] why are the /+faq/1234 links showing questions but not their answers?? i don't see the point if the answer is not there [13:15] fta: Works fine for me. Got an example? [13:15] https://answers.edge.launchpad.net/chromium-browser/+faq/674 [13:17] hm, when i created that faq from a question, i thought it was title + question + answer, apparently, it was question + answer + comment for the initial question [13:17] right? [13:18] I admit I've never created an FAQ. [13:18] * wgrant tries. [13:19] that was my 1st.. [13:23] fta: So, it defaults to using the question description as the content. The content of the FAQ is meant to be the answer. This doesn't make much sense. [13:25] indeed === matsubara changed the topic of #launchpad to: Launchpad is now open sourced: https://dev.launchpad.net/ | Help contact: - | Join https://launchpad.net/~launchpad-users | Channel logs: http://irclogs.ubuntu.com | #launchpad-dev is the developer channel === matsubara changed the topic of #launchpad to: Launchpad is now open sourced: https://dev.launchpad.net/ | Help contact: matsubara | Join https://launchpad.net/~launchpad-users | Channel logs: http://irclogs.ubuntu.com | #launchpad-dev is the developer channel === henninge_ is now known as henninge-bbl [14:08] hi [14:08] I upload a .deb to my ppa, and dput says Successfully uploaded packages., but in launchpad I cant see the packages, some wrong? [14:09] Hi shakaran, what's the address of your ppa? [14:09] https://launchpad.net/~shakaran/+archive/ppa [14:09] I cant see a package called "tivion" [14:09] shakaran: and how long ago did you upload the package? [14:10] 15 minutes or so [14:11] You can see my success upload: http://paste.ubuntu.com/263093/ [14:11] shakaran: and you haven't received any rejection emails yet? [14:12] no emails yet [14:12] shakaran: the key id that you've used to sign the package appears to be different to the key on LP? [14:13] yeah, is different [14:13] I need sign with my launchpad key? [14:13] shakaran: yes, launchpad needs to know that the package really is from you, and it can only do that if it's signed with a key attached to you account. [14:14] ok, I will try again with my lauchpad key [14:14] shakaran: yeah, or added the other key to your LP account... [14:15] how to add it? [14:16] shakaran: https://launchpad.net/people/+me/+editpgpkeys [14:16] ok === kiko__ is now known as kiko [14:23] ok, I added the fingerprint and I recieve the email, but how to confirming the key? [14:23] (I have installed FireGPG) [14:23] For Gmail [14:25] shakaran: the instructions should be in the email? (I'll look for the help page for you) [14:25] https://help.launchpad.net/ReadingOpenPgpMail (this not work) [14:27] shakaran: Hmm... (did you use FireGPG to confirm the other key in your LP account?), if it's a hassle, it migth be easier to sign your package with the other key (I'm not sure why FireGPG wouldn't be signing your email correctly). [14:28] well...I sign with the other key [14:31] I get a error: http://paste.ubuntu.com/263108/ [14:33] shakaran: It seems you don't have that other key on that machine. [14:35] umn, How to retrieve again to this machine? === yofel_ is now known as yofel [14:36] shakaran: you need to copy the secret key from another machine that has it. [14:36] Getting FireGPG working is almost definitely easier. [14:37] umn, the only machine that should have the key is my current machine [14:37] I open the gmail message, select the text, and dont appear any menu for FireGPG === matsubara_ is now known as matsubara [15:08] is it possible to add some formatting to a ppa board? bold, underline, italics.. [15:09] fta: do you mean for the description? https://bugs.edge.launchpad.net/soyuz/+bug/392123 [15:09] Launchpad bug 392123 in soyuz "Allow Markup in PPA Descriptions" [Low,Triaged] [15:11] noodles775, yes, thanks [15:12] Can a Launchpad admin please add me as a maintainer to the lp:drupal project? The current maintainer seems to be MIA. [15:14] noodles775, i was trying to make https://edge.launchpad.net/~chromium-daily/+archive/ppa look nicer, some markups would sure help [15:14] fta: yes definitely - it would be great to have ReST support. [15:14] (across LP generally). [15:15] davidstrauss, can you file a question in https://launchpad.net/launchpad/+addquestion requesting that? [15:18] matsubara: Done. https://answers.launchpad.net/launchpad/+question/81605 [15:28] davidstrauss, thanks. I assigned it to one of the admins [15:28] matsubara: thanks [15:28] np [16:03] I'm trying to automate the submission of source packages to my PPA via our build system -- I'm having trouble with signing the packages at the end because there is no /dev/tty. I know I can pass --passphrase to gpg to get around this problem, but I don't see any obvious way to do this via debuild or dh_builddeb. [16:03] any help would be awesome [16:12] cprov, can you help Box ^? [16:15] matsubara: I think I've got it figured out; if I pass -uc -us to debuild I can sign the files manually at the end [16:15] Box, cool :-) [16:15] Box: good idea [16:16] okay, cool, glad it's not a boneheaded thing to do for whatever reason [16:18] Box: I'd be scared to have the password of *my* key hanging around in public machine. After all, if it's a machine-key it shouldn't require a password, then you make sure it is kept protected. [16:19] Box: I'm just thinking aloud, don't take it too serious ... [16:20] cprov: it's cool, it's actually on our internal network so it's not literally a public machine. But educate me, what's a machine-key? [16:21] Box: a term I've just invented apparently (sorry), but I meant a key that is used to sign automatically generated content. [16:22] Box: e.g. the PPA signing keys hosted in LP, they don't have/require password to sign stuff [16:23] Box: they were generated with one specific purpose -> to be used for script to sign repository manifests [16:23] Box: they are not used by humans, so they don't need a password. [16:24] cprov: I see; that sounds like what I should be using. I assume the no-password-ness is a feature of gpg I should look for? [16:25] Box: yup, I'm not sure you can reset the password (make it empty, actually) of an existing key, though. [16:26] cprov: nice, thanks a lot [16:26] Box: you're welcome. === kiko is now known as kiko-fud [16:38] hm, in the new https://edge.launchpad.net/builders/, it's no longer obvious which builders are doing PPA vs official builds [16:42] The build daemon used for PPA's on launchpad, is that a separate project or is the code in launchpad itself? [16:45] Or does launchpad simple integrate buildd from debian? [16:54] cprov, ^ another one for you :-) [16:56] flohack, anything needed for Launchpad to work is in the tree [16:58] beuno: I'm thinking about running the build daemon only, that's why I'm asking. Do you know if buildd from debian is used? Can the build daemon used by launchpad be run without launchpad itself? [16:58] I don't === henninge-bbl is now known as henninge [17:01] beuno: Thanks! === salgado is now known as salgado-lunch === deryck is now known as deryck[lunch] [17:06] flohack: it's part of the Launchpad code, it's basically a heavily modified sbuild === beuno is now known as beuno-lunch === bd__ is now known as bd_ [17:16] flohack, wgrant documented the process of setting up soyuz locally: http://williamgrant.id.au/f/1/2009/soyuzness.html === matsubara is now known as matsubara-lunch === yofel_ is now known as yofel === deryck[lunch] is now known as deryck [17:47] hello [17:52] anyone know how to register a meeting? [17:52] I can't seem to find(!) where it would be [17:59] jcastro, a meeting? You mean a sprint? [18:00] yeah [18:01] a sprint, I need to make one for UDS. [18:03] jcastro: https://edge.launchpad.net/sprints [18:04] right, but I don't see where I can create a sprint [18:05] jcastro: https://edge.launchpad.net/sprints/+new [18:05] looks like the link is gone with the redesign?... [18:05] sinzui: any ideas? ^^^ [18:06] jcastro: I do know how to register a sprint...and I am adding it to the series and milestones pages right now [18:06] I just created it [18:06] now that I know the URL [18:07] sinzui: would it not make sense to have an action link to the add form in /sprints ? [18:07] intellectronica: I am reporting that bug right now [18:07] sinzui: cool, thanks [18:07] ditto an edit button [18:07] it seems to be missing after I create the sprint [18:08] intellectronica: I think the intent was that it should be done form a project/series/milestone, but that does not make sense since meetings can be broader than a project. === beuno-lunch is now known as beuno === salgado-lunch is now known as salgado === EdwinGrubbs is now known as Edwin-lunch === matsubara_ is now known as matsubara === kiko-fud is now known as kiko === jkakar_ is now known as jkakar === Edwin-lunch is now known as EdwinGrubbs [20:28] Is someone else having problems too with the replacement networkmanager for the plasma widget? I'm getting segmentation faults since the last update from the PPA Backport [20:28] flohack, this is probably not the place to ask [20:36] Backtrace: http://pastebin.com/d6859b27 (Couldn't find the dbg-package for /usr/lib/kde4/solid_networkmanager07.so) [20:36] Maybe this helps someone. Maybe not :-) [20:37] Sure is that I'm going to sleep now ;-) [20:37] Good night! === ripps_ is now known as ripps [21:15] tormod, i suggest asking intellectronica all your launchpad bug-related questions [21:15] leonardr, tormod: i can has context? [21:16] intellectronica, tormod is wondering how to modify the content-type of a bug attachment using the web service [21:16] losa ping: anyone available to help me out with some staging QA? [21:16] i showed him the code to do this, but it's possible the code won't actually work because of a fix to bug 271051 [21:16] Launchpad bug 271051 in malone "Do not allow changing of content of attachments" [Medium,Fix released] https://launchpad.net/bugs/271051 [21:17] he was wondering how it works in the web interface [21:17] gmb: hello [21:18] gmb: I can help you with that [21:18] Chex: Cool. So, pretty much as before, can you start by running the following in `make harness` on staging please: http://pastebin.ubuntu.com/263394/plain/ [21:20] leonardr, tormod: on the bug page you have a list of attachments (on the right hand side). for each attachment you can edit there's an edit link which takes you to a form where you can change the mime type [21:20] yes I know, but how can I do this with launchpadlib? [21:21] seeing that leonardr's example does not work, how can it be done? if the web interface can, there must be a way, right? [21:22] tormod: i think it likely that the feature has been disabled in the web service because of the security concerns thekorn raised [21:22] tormod: what happens when you try to run my code? [21:22] leonardr, haven't tried, I meant "seeing from your discussion that" [21:23] tormod: i'm not sure it won't work. all i know is that bug 271051 was closed 'fixed' [21:23] Launchpad bug 271051 in malone "Do not allow changing of content of attachments" [Medium,Fix released] https://launchpad.net/bugs/271051 [21:23] if it was closed, then maybe no one thought about what it meant to have the same feature available on the website [21:25] on the web site you can only delete attachments, new ones will be in a new comment. but content-type can be changed freely. [21:26] ok, it sounds like the website is reaching directly into the librarian and modifying the content type [21:26] whereas the design of the webservice doesn't allow that [21:27] ok, just to show the others your example: http://pastebin.ubuntu.com/263382/ === EdwinGrubbs is now known as Edwin-afk [21:33] kfogel: doing that wishlist was -hard- [21:37] lifeless: your "three wishes"? [21:37] leonardr, ok, your example raises an 405 HTTPError, so it's all good :) [21:37] kfogel: yeah. Required though. [21:38] thought. === kfogel is now known as launchpad === launchpad is now known as kfogel [21:39] lifeless: you know what's really going to be hard? What I have to do tonight, which is count and summarize that thread :-). [21:39] nah thats easy. [21:39] throw it all away. [21:39] write down *your* three wishes. [21:56] leonardr, so in the end we can not change the mime type with launchpadlib? [21:56] kfogel: I think that bug is invalid for Ubuntu. [21:57] sinzui: accident [21:57] sinzui: no idea how it got in ubuntu, I didn't mean for it to [21:57] sinzui: also no idea how to change it now, UI not obvious about that -- still looking [21:58] kfogel: you cannot [21:58] sinzui: AAAAARgh :-) [21:58] kfogel: choose (+) Also affect project [21:58] sinzui: is this on purpose?? [21:58] then mark the bug invalid for ubuntu [21:58] kfogel: I do not think so [21:59] sinzui: I am pretty sure I entered "launchpad" into the project field when filing. Wasn't recording, so no way to find out what happened, but because my mouse is having problems right now, I'm paying special attention to what happens on the screen. [21:59] kfogel: but you clearly did not file that bug while in the launchpad project, because it wont switch the project once the context is set [21:59] sinzui: "while in the launchpad project" ? I just went to bugs.launchpad.net and clicked on "report a bug" === cprov is now known as cprov-afk [22:00] kfogel: I think you are mistaken. [22:00] kfogel: The form does not prompt for a target [22:01] kfogel: I think you were at the bugs site: https://bugs.edge.launchpad.net/ ... and I think it is insane to offer a link to file a bug from that page [22:01] sinzui: sure it does [22:01] https://bugs.edge.launchpad.net/launchpad/+filebug [22:02] sinzui: I'm looking at https://bugs.edge.launchpad.net/bugs/+filebug right now [22:02] kfogel: ^ that is launchpad.net, not launchpad the project [22:02] kfogel: The page even says "launchpad.net" [22:02] sinzui: I typed "launchpad" into the box next to Project. I distinctly remember clicking on the radio button next to project too, but with my mouse problems, I could have accidentally reset it back to ubuntu [22:03] sinzui: I do not understand what you're saying. [22:03] sinzui: go look at the URL I pointed to, do you see what I see? [22:03] the "Where did you find the bug?" part of the page is what I'm looking at [22:03] Did you check the (*) Project [22:03] sinzui: see what I wrote above === salgado is now known as salgado-afk [22:04] sinzui: so the only explanation I can think of is that after I hit the radio button next to Project, I brushed the mouse pointer back over Distribution and clicked that, and then submitted with "launchpad" text in the Project text field but with the Distribution radio button checked. That ought to be an error, but apparently it is not. [22:05] I think workflow should get you to the project before filing a bug. [22:06] kfogel: This form is limited too. The forms in the project allow you to specify more details. [22:06] sinzui: virtually every bug I've filed in Launchpad was done the way I described above. I'm not sure what "workflow" would be otherwise -- go to launchpad project home page and then click "Report a bug" ? That might work. [22:07] kfogel: yes, that is how most people do it. The first search the project for bugs before reporting one [22:07] sinzui: I depend on the dup-finding code. [22:09] kfogel, sinzui: there's been lots of talk of getting rid of the top-leve +filebug and only keeping the route through the project/distro/package. if you ask me that's the best solution [22:09] the top-level route is an open invitation for wrongly targeted bugs [22:09] kfogel: I think you are victim of a flawed workflow. I see a lot of bugs reported in the registry, blueprints and answers, and I think most came from that form. [22:10] sinzui: and now, when I updated the project description to note that the Ubuntu target was a mistake, it's hanging on the submit. So to recap: we found a bug in the UI for merging accounts; when I went to file the bug on that, we found another bug in the bug-filing UI; and when I went to take care of that, we found another bug in the bug mis-filing-correction UI. It wonder if this kind of recursion ever ends? [22:10] intellectronica: yes, I think we are accommodating Ubuntu in this case and we should just offer a fast way to Ubuntu [22:11] kfogel: don't file another bug [22:11] sinzui: au contraire. for ubuntu we're going to remove _all_ routes to filing bugs from the web interface [22:11] we just need to add the other launchpad target [22:11] sinzui: What I would like to know is, how many bugs misfiled in Ubuntu had text in the "project" name field? Because that means a mistake 100% of the time. [22:11] intellectronica: ^^ [22:11] intellectronica: bug via desktop? [22:11] leonardr: Hi, as I missed your UDW session, can I ask you a question about the LP API framework? I've tried to use python-launchpadlib in a multi-threaded script to speed up the fetching of many objects (at least that's the idea) but failed as python-httplib2 isn't thread-safe. Do you know if there is something in development to fix this (or a workaround to make python-launchpadlib usable in a [22:12] multi-threaded script)? [22:12] sinzui: yes, it seems that's the only way ubuntu can get high quality bugs [22:13] kfogel: i'm not sure i understand what the bug is. is it that entering text into the textbox doesn't do what you expect, or just that the choice between distribution and product bugs makes this path error prone? [22:15] intellectronica: I entered text into the project name textbox, and clicked the project radio button (thus taking radio focus off Distribution). But then, we theorize, I clicked back on Distribution, although of course leaving my text in the project name box. This should have resulted in an error: if the Distribution button is clicked, but there's text in that other box, then the user is clearly not doing what they intended to. [22:15] kfogel: i don't remember the exact numbers, b.t.w, but last time i looked not many people were using the top-leve route [22:15] intellectronica: instead, my bug got submitted against ubuntu -- IOW, the project name text was simply ignored. [22:16] kfogel: i think that's how most comparable user interfaces work, but that probably means that this type of user interface (a hybrid of radio and text) is not so good and we should avoid using it [22:18] kfogel: so i think that rather than change the behaviour (in a way that is, at least imo, not obviously better) we should get rid of that form or at least re-design it in a way that makes more sense [22:18] intellectronica: well, this is not a case where what the majority does should be followed, because it's not a habituation or familiarity question. Even if every other site in the world behaved the same way, we'd still be making an improvement if Launchpad fixed this. [22:19] intellectronica: if we get rid of the form, then no issue, obviously. If we keep it, it will be strictly better if it protects against this kind of mistake. [22:20] kfogel: if it's up to me then this form is gone. if it means that getting to the correct +filebug is painful, we're better off at improving that [22:21] intellectronica: +1 === micahg1 is now known as micahg [22:38] hey guys - [22:38] is there a place to send in a suggestion? [22:39] one of our translators wants to send in their suggestion to offer better pagination links on the translation pages, but isn't sure where to post his suggestion [22:40] WiserEarth: a bug report on bugs.launchpad.net/rosetta probably [22:40] think they'll mind having suggestions mixed in with bug reports? [22:57] mwhudson, thanks for the answer :) [22:58] geser: i don't know of anything. you'd have to use a thread-safe http client instead of httplib2 [22:59] do you know if any exists for python? [23:00] urllib [23:00] / httplib [23:00] or twisted's httpClient [23:00] or pycurl [23:00] Twisted's isn't threadsafe, but it can do concurrent requests in one thread. :) [23:00] httplib2 is the only one I've heard of that isn't threadsafe [23:01] exarkun: callFromThread :P :> [23:01] exarkun: did you get your merge issue sorted? [23:01] urllib is actually only sort of mostly thread safe [23:01] lifeless: I cherry picked everything I needed [23:01] exarkun: ok cool. [23:01] exarkun: are you guys considering migrating ? [23:02] It'll be a while yet. [23:02] I think everyone mostly agrees that we'll switch to bzr eventually. [23:03] awesome [23:04] thanks, I'll try to see how hard it will get to use a thread-safe http lib within python-launchpadlib [23:07] oh also [23:07] it may be obvious [23:07] but bzrlib.transport.get_transport(url) [23:07] is _love_ [23:07] fast [23:08] threadsafe (per-object caches, shared only if you clone between them) [23:08] and lots of bug fixes to deal with odd servers and so on [23:32] gmb: Okay, it's fixed. [23:40] mkanat: Awesome, thanks. I'm about to hit the sack but I'll check it out first thing in the morning [23:40] gmb: Okay. The sooner the better. [23:41] mkanat: Actually, give me a second, I'll run the very quick version of the QA... [23:41] gmb: Awesome. [23:43] mkanat: Looks good! Can you make sure that this fix is rolled out to the test instance as well for the purposes of our QA? [23:43] gmb: It's on there. [23:43] Awesome. [23:44] mkanat: In that case I'll catch some sleep. Thanks for your help :) [23:44] gmb: Welcome!