[12:35] night all === uid [n=suid@a83-132-170-181.cpe.netcabo.pt] has joined #launchpad [12:44] hello [12:44] i want to contribute [12:44] into a project [12:44] with who i have to contact? [12:46] uid, it depends on the project [12:46] ubuntu-server [12:46] uid: having a launchpad account is a good start [12:47] i already created [12:47] :p [12:55] uid, you should contact the admins of that project [12:56] im already doing that [12:56] thank you [12:56] :) === Seveas [n=seveas@ubuntu/member/seveas] has joined #launchpad [01:40] moing [01:56] lifeless: my merge failed on the TestImportKeyRing test -- I'm wondering if we're seeing the crash bug some others (Bjorn, cprov) were seeing [01:56] (the output just stops at that test) [02:02] Dec 14 16:58:49 pqm [46912504432336] INFO: running precommit hook: make check_merge [02:02] *** glibc detected *** double free or corruption (!prev): 0x0000000005989200 *** [02:04] jamesh: ^^ [02:04] that's it [02:04] it's something I haven't been able to reproduce locally [02:06] I wonder what's different between bellany and chinstrap? [02:06] amd64 [02:06] and breezy [02:06] I can run manual tests for you [02:06] if you tell me what to do, I shall do it [02:07] lifeless: BjornT was able to reproduce the crash with "python test.py -vvf canonical.launchpad.utilities" [02:08] it'd be useful to get a copy of gpgme with symbols [02:08] Znarl: ^^ [02:08] elmo: ^^ [02:08] if thats a reasonable thing, I'll open an rt for it [02:08] also python2.4-dbg too, if it's not there [02:09] I half wonder if it's worth rewriting the portions of pyme that we care about [02:09] without the use of swig [02:09] I would talk with upstream [02:10] we have permission to feed the changes back === jamesh wishes we had fedora-style debuginfo packages [02:11] what do they offer? [02:12] dettached debug information [02:12] they are generated automatically as a side effect of the package build process [02:12] for every package that contains executables [02:12] (or shared libraries) [02:17] theres a trivial debhelper facility for that [02:17] with compat >= 5, dh_strip --dbg-package==foo-dbg [02:17] sure [02:17] the reason it's useful is that there is debuginfo packages for pretty much everything [02:18] rather than only the cases where a package maintainer bothered turning on the feature [02:19] fair enough [02:19] that's more a matter of policy though? === poningru [n=poningru@n128-227-82-53.xlate.ufl.edu] has joined #launchpad [02:22] LarstiQ: yes. That doesn't change the fact that I wish we had them :) [02:22] debian has some reasonable size concerns about this ;) [02:23] ok theres a config build running [02:23] that's what mdz said when I mentioned it to him [02:23] when it finishes I'll do that test for you [02:24] thank you [02:25] jamesh: I understand the craving :) [02:28] on the archive size issue, the fedora archives have the debuginfo packages in their own subdirectory, so a mirror could easily exclude them === stub [i=stub@sweep.bur.st] has joined #launchpad === siretart [i=siretart@ubuntu/member/siretart] has joined #launchpad [03:16] jamesh: [03:16] pqm@balleny:~/tests/launchpad$ make build [03:16] python2.4 utilities/shhh.py make -C sourcecode build PYTHON=python2.4 \ [03:16] PYTHON_VERSION=2.4 LPCONFIG=default [03:16] pqm@balleny:~/tests/launchpad$ python test.py -vvf canonical.launchpad.utilities [03:17] Configuration file found. [03:17] Running FUNCTIONAL tests at level 1 [03:17] Running FUNCTIONAL tests from /home/pqm/tests/launchpad/lib [03:17] Parsing ftesting.zcml [03:17] testCheckTrustDb (canonical.launchpad.utilities.ftests.test_gpghandler.TestImportKeyRing) ... ok [03:17] testEmptyGetKeys (canonical.launchpad.utilities.ftests.test_gpghandler.TestImportKeyRing) ... ok [03:17] testImportKeyRing (canonical.launchpad.utilities.ftests.test_gpghandler.TestImportKeyRing) ... ok [03:17] testPopulatedGetKeys (canonical.launchpad.utilities.ftests.test_gpghandler.TestImportKeyRing) ... ok [03:17] testSetOwnerTrust (canonical.launchpad.utilities.ftests.test_gpghandler.TestImportKeyRing) ... *** glibc detected *** double free or corruption (!prev): 0x000000000196c490 *** [03:17] Aborted [03:17] jamesh: have you tried with valgrind ? [03:33] lifeless: yeah. I haven't been able to reproduce locally (either on my AMD64 system, or my x86 laptop) [03:34] jamesh: ok. keep feeding me things to do, I don't have cycles to debug but I do to do stuff for you [03:35] lifeless: what would be most useful for me would be a stack trace generated with a gpgme with symbols [03:35] jamesh: ok. I can't do that trivially until we get admin time. [03:35] what does your amd64 run? [03:36] breezy [03:36] what packages do you have, should we compare versions of gpgme, etc ? [03:36] btw, scarey thing: that result is from rocketfuel HEAD. [03:36] 1.0.2-1build1 [03:37] lifeless: would it be possible to build gpgme and run the tests with LD_LIBRARY_PATH? [03:37] jamesh: not without the build depends [03:37] lifeless : What's up? [03:37] Znarl: background: we've had a intermittent C error occuring on some folks machines for launchpad [03:38] 13:16 < lifeless> testSetOwnerTrust (canonical.launchpad.utilities.ftests.test_gpghandler.TestImportKeyRing) ... *** glibc detected *** double free or [03:38] corruption (!prev): 0x000000000196c490 *** [03:38] 13:16 < lifeless> Aborted [03:38] this is occuring on balleny, which makes newpqm, well, useless. [03:38] so we need to get enough info together to fix this bug asap. [03:38] Znarl: we'd like to be able to test with a libgpgme11 with debug symbols [03:39] (preferably with the same compile options though -- the debug build option for some packages turns off all optimisations, which could hide the bug) [03:39] installing the gpgme build deps might be easier though, and just build gpgme from source [03:40] Sounds like you don't have a clear plan? [03:41] Znarl: well its a locked down machine, so my basic plan is to give jamesh whatever stuff he asks for - right now hes asking for a stack trace with line numbers for gpgme [03:41] (libgpgme11) [03:43] the traces I got from cprov and bjorn looked like some of the frames came from functions whose symbols had been stripped [03:43] ok [03:43] its 3am for Znarl [03:44] he's going to do dedicated stuff with use when he has slept [03:44] which is now :) [03:44] jamesh: you know how to tweak a package to generate debug symbols ? [03:45] jamesh: I think it would help znarl to have a patch to the gpgme11 source to make a -dbg package [03:45] which we can just get installed [03:47] lifeless: looks like rebuilding the package with DEB_BUILD_OPTIONS=nostrip should do [03:48] jamesh: thats different [03:48] jamesh: that gives a variation of the package. [03:48] adding a control and rule for -dbg gives us debuginfo style flexability [03:49] if you don't know how, I'll do that in the next 4-5 hours [03:49] I don't know how. I haven't gotten round to learning debian packaging in detail [03:49] ok [03:49] I'll do it, and you can learn from the diff ;) [03:50] in fact, I'll cheat, I'll see if a motu has some cycles to do it === tambaqui [n=patricia@200-183-97-151-mns.cpe.vivax.com.br] has joined #launchpad [04:00] haven't done it before, but I'll try. [04:31] http://www.magicalforest.se/~nafallo/gpgme1.0_1.0.2-1ubuntu1.debdiff <- jamesh Znarl thats the recipe to make detached symbols for gnupg [04:31] thanks heaps to Nafallo [04:31] cool. [04:31] I'll have to teach myself about this sometime ... [04:32] jamesh: just read the diff [04:32] I did [04:32] except for s/version/symbols/ its a good example; [04:33] there is actually one other thing which the fedora debuginfo packages do: include source code, and rewrite the debuginfo to point at the location the source code gets installed to [04:33] ah, thats either great or diabolical [04:33] but this should hopefully give the needed info [04:34] (I'm thinking 1Gb source packages) [04:34] they usually don't turn out as large as you might think [04:34] no really [04:34] openoffice is a gb [04:34] hehe :-) [04:34] cut all the build infrastructure, documentation, images, etc [04:35] ah [04:35] well X used to be huge too [04:35] but yeah, I can see the difference [04:36] it pulls the list of source files to include from the actual debug info [05:13] Excellent, I can actually bzr push to my sftp server. [05:14] spiv: congratulations [05:18] jamesh: Your openssh/paramiko sftp hack caused me a little bit of grief last night, until I remembered it. :) [05:19] spiv: the original one, or the version that eventually got merged? [05:19] Whatever is in bzr snapshot deb from yesterday. [05:20] okay. [05:20] my original version was a bzr plugin that monkey patched the sftp transport [05:20] and now? :-) [05:20] which like many monkey patches is quite fragile [05:20] Because I needed to teach my SFTP server to treat a request to exec 'sftp' as if it were a request for subsystem 'sftp'. [05:21] (Previously I just refused all exec requests). [05:21] Nafallo: I refactored it as a patch against bzr, which got merged [05:21] Which was pretty simple, once I figured out what was needed :) [05:21] spiv: it should be requesting the sftp subsystem [05:21] jamesh: ah, nice :-) [05:22] jamesh: Hmm. The version here isn't. I wonder why not. It doesn't really matter, both ways work fine now... [05:23] spiv: the command it runs here is: ssh -oForwardX11=no -oForwardAgent=no -oClearAllForwardings=yes -oProtocol=2 -oNoHostAuthenticationForLocalhost=yes -s chinstrap.ubuntu.com sftp [05:23] jamesh: Same here. [05:23] the "-s" argument is to request a particular subsystem === spiv double-checks [05:23] "The subsystem is specified as the remote [05:23] command." [05:23] (according to the man page) [05:23] Yeah, it does that. [05:23] But conch is seeing that as an exec request. [05:24] what does it think of the command line sftp client then? [05:25] It runs: /usr/bin/ssh -oForwardX11 no -oForwardAgent no -oClearAllForwardings yes -oProtocol 2 -s chinstrap.ubuntu.com sftp [05:25] The command line sftp works. [05:25] essentially the same command [05:26] Hmm, it's possible I got confused in my testing. Ah well. [05:27] The more annoying problem is that openssh insists on using my real home directory, rather than the one I specify in $HOME :) [05:28] But I've figured out a way around that as well, thankfully. [05:29] "ssh -F configfile" ? [05:32] jamesh: Tricky when I don't directly control the invocation of ssh -- it's in bzrlib :) === Nafallo is now known as Nafallo_away === Nafallo_away is now known as Nafallo === Nafallo is now known as Nafallo_away === kagou [n=kagou@84.5.165.142] has joined #launchpad [08:10] hi [08:17] i want to merge 2 accounts on launchpad. Before an error. To use https://launchpad.net/people/+requestmerge i have to be logged in the duplicated account . Am I right ? [08:19] kagou: you should be logged in as the account you want to keep [08:19] ok jamesh [08:19] kagou: and then select the account you want to merge [08:20] kagou: you'll receive some emails to confirm the merger [08:20] ok jamesh . do you know how to delete an account on the wiki too ? [08:20] kagou: the wiki uses the launchpad user database [08:22] ok thanks jamesh [08:38] jamesh, i'm trying to delete a page on the wiki : https://wiki.ubuntu.com/VetselPatrice2 [08:39] but i must be wrong === carlos [n=carlos@71.Red-83-53-49.dynamicIP.rima-tde.net] has joined #launchpad [08:39] morning [08:44] how can i delete a page in the wiki. I'v empty it and put "DeletePage" in ... [08:46] kagou, you don't need to remove the content [08:46] kagou, there is an option to delete it [08:46] in the same place where you can select to edit it [08:47] kagou, look at the 'More Actions' entry [08:47] carlos, so sorry ... this morning my eyes are not in the good way :/ [08:47] the last option you have is 'Delete Page' [08:48] Morning all [08:48] kagou, no worries [08:48] carlos, i had to do 2 others things ... First change the name of my account on ths wiki [08:48] sivang, morning [08:48] kagou, ubuntu's wiki? [08:49] yes [08:49] morning carlos , how are you? [08:49] kagou, go to your launchpad's page : https://launchpad.net/people/kagou (change 'kagou' with your launchpad id) [08:49] i'm in [08:50] hmmm [08:50] https://launchpad.net/people/vetsel-patrice [08:50] kagou, seems like we don't have a way to change the wiki name.... you should ask salgado when he wakes up [08:51] sivang, fine, thanks, and you? [08:51] wait [08:51] found it [08:51] ;-) [08:51] ok carlos . he last problem is to change the "Registered Details [08:51] Wiki: VetselPatrice2 " on https://launchpad.net/people/vetsel-patrice [08:51] https://launchpad.net/people/vetsel-patrice/+editwikinames [08:52] great carlos [08:52] kagou, the links to edit that is at the end of the page on the right [08:52] we changed the layout recently and I didn't see it [08:53] kagou, you need to select first 'Edit details' [08:53] fine thanks ! [08:54] carlos: fine, apart from being too busy at work the last couple of days. Say, do you know anything about jblack's rocketfuel-get script? [08:54] sivang, other than he's working on it?.... [08:54] I read that yesterday [08:55] carlos: well, as it's not working I'll go back in the wiki revision and attempt the checkout the "old" way :) [08:55] carlos: yes, I forgot it's still WIP [08:55] ;-) === robitaille [n=robitail@ubuntu/member/robitaille] has joined #launchpad [09:00] lifeless, hi, around? [09:01] yes [09:02] lifeless, I think that either pqm.ubuntu.com points to the wrong machine or pqm.ubuntu.com mails are not reaching the new pqm server.... [09:04] procmail is not being processed it looks lik [09:05] X-Amavis-Alert: BAD HEADER Non-encoded 8-bit data (char C3 hex) in message header 'From' [09:05] From: ...erello@canonical.com (Carlos Perell\303\263 Mar\303\255n)\n [09:05] thats not fatal, just FYI [09:05] yeah, IT's a problem with the mail command [09:05] my local name is using UTF-8 chars [09:06] I don't know how to fix that [09:06] hi [09:06] moin [09:06] and as I'm only using it to send pqm requests... I didn't expend too much time on it [09:06] SteveA, morning [09:06] hi carlos === bradb [n=bradb@195.182.78.95] has joined #launchpad [09:14] hey lifeless , SteveA [09:17] lifeless: re. the test crash, could you try something for me? [09:17] lifeless: try applying this patch to the generated gpgme_wrap.c file run make in sourcecode/pyme: https://chinstrap.ubuntu.com/~dsilvers/paste/file6eLTpE.html [09:18] then rerun the failing test [09:22] https://chinstrap.ubuntu.com/~dsilvers/paste/fileebwN4K.html [09:24] carlos, lifeless: That header is badly encoded - the mail client being used is broken. Could be worth filing a bug. [09:24] stub, it's a problem with the Linux 'mail' command [09:25] stub, it takes it from the unix's user database [09:25] carlos: File a bug - Ubuntu was supposed to be UTF-8 everywhere with the Hoary release. [09:25] stub, I thought it was a problem in my side more than a problem with Ubuntu... [09:28] carlos: Either the user database is ASCII only, in which case you shouldn't have been allowed to set that name, or the user database is in some other known encoding, in which case 'mail' should encode the From: header correctly. Or the user database is in an arbitrary, unknown encoding which would also be a bug. [09:29] Anyway - mail is spitting out non-RFC compliant messages which is bad [09:29] stub, I'm not sure what's the problem [09:29] because [09:29] If I use non ascii and non UTF-8 chars [09:30] GNOME fails to show my name [09:30] if I set it as UTF-8 it works [09:30] but I don't see anyway to specify a concrete kind of encoding for the user database [09:31] so perhaps is the 'user database is in an arbitrary, unknown encoding which would also be a bug.' === carlos files a bug and lets Ubuntu developers decide... [09:31] It needs to be a known encoding or you can't interoperate - if you don't know the encoding, you can't read it as text (just as binary gibberish) [09:32] stub: it is generally interpreted as being in "the user's encoding" [09:32] jamesh: does that help ? [09:32] even if different users use different encodings [09:32] jamesh: That doesn't help, as the system needs to read it too [09:32] lifeless: it looks pretty much the same as the results on my box (where it doesn't crash) [09:33] no unmatched free() [09:33] eh [09:33] is soo [09:33] (at least at that level) [09:33] oh, I see, [09:33] overlapped, not mismatchd [09:34] I'm amazed that valgrind does not report any naughtiness to you [09:34] does it do so for you? [09:34] dont have valgrind there [09:34] ah [09:35] I've got a suppressions file that helps for Python programs on amd64 [09:35] (the one that comes with Python is missing some entries for 64-bit) [09:35] gpgme_wrap is very dirty [09:35] gcc is -not- happy about it [09:36] interesting [09:37] I added a 'after release' print [09:37] stub, http://bugzilla.ubuntu.com/show_bug.cgi?id=21033 [09:37] done [09:38] Ubuntu bug #21033: The 'mail' command is not encoding the user name correctly Product: Ubuntu, Component: mailx, Severity: normal, Assigned to: debzilla@ubuntu.com, Status: UNCONFIRMED http://bugzilla.ubuntu.com/show_bug.cgi?id=21033 [09:38] jamesh: https://chinstrap.ubuntu.com/~dsilvers/paste/fileP7bCoV.html [09:50] jamesh: approved sitemap [09:50] SteveA: thanks. [09:50] some comments in the email [09:51] jamesh: so its definately in the gpg_release routine [09:51] SteveA: the new PQM machine is exhibiting the pyme related crash Bjorn ran into (which I still haven't been able to reproduce locally) [09:51] lifeless: yeah. That's what the stack traces I've seen so far indicate [09:51] jamesh: could it be a race in gpgme ? [09:52] i.e. create two, release two ? [09:52] lifeless: possibly, although the context objects are intended to avoid these problems: they wrap up all the global state [09:53] I have to go shopping [09:53] or I wont have dinner [09:53] back soon as possible [09:53] so if you have two bits of code using gpgme, they'll be safe if they use different contexts === lbm [n=lbm@cpe.atm4-0-1301006.0x50a0824e.vgnxx6.customer.tele.dk] has joined #launchpad [10:01] Znarl: for when you arrive: procmail in the pqm account isn't; can you please install gdb and valgrind in the base os only for debugging this problem; can you please apply this http://www.magicalforest.se/~nafallo/gpgme1.0_1.0.2-1ubuntu1.debdiff to the libgpgme11 package and install the resulting library, headers and -dbg into the base os [10:01] I'm now gone grocery hunting, will be back in ~60 I suspect === ddaa [n=ddaa@nor75-18-82-241-238-155.fbx.proxad.net] has joined #launchpad [10:07] stub: ping? whiteboard data? [10:07] SteveA: pong? Wha? [10:07] stub: i emailed you yesterday about getting some data from production to see how people are using the whiteboard [10:08] Hmmm.... flagged as read, but I don't remember reading it :-/ [10:08] weird X files situation [10:18] BjornT_: (trying to understand some conflicts here) Do you remember changing the section "Privacy and Launchpad Admins" in bugtask.txt, to query tasks maintained by sample person instead of by foo bar? [10:19] bradb: no, that doesn't seem right. let' me take a look, i do remember changing something around there. [10:19] ok [10:24] bradb: ah, misread your question. yes, i did, so that bug 1 would show up in the list (i see now that i forgot to rename a variable) [10:24] Malone bug #1: Microsoft has a majority market share In: Ubuntu, Severity: Critical, Assigned to: Mark Shuttleworth, Status: Accepted https://launchpad.net/bugs/1 [10:31] BjornT_: why would a whole package changelog end up in a status whiteboard? [10:31] bug 2157 [10:31] Malone bug #2157: openafs-modules-source does not build with current (2.6.12-8-686) kernel In: openafs (Ubuntu), Severity: Normal, Assigned to: MOTU, Status: Fixed https://launchpad.net/bugs/2157 === BjornT_ takes a look [10:33] or not... my nameserver seems to be down. what's the ip of launchpad.net? [10:35] launchpad.net has address 82.211.81.179 [10:36] BjornT_: (sorry, got distracted for a second there) I guess that was to do with removing the upstream ubuntu task, right? because foo bar was in the team that maintains upstream ubuntu. [10:36] bradb: yes [10:37] ok, thanks for helping clarify [10:38] SteveA: my guess is that he thought it was a comment box. the same changelog is added as a comment a minute later [10:39] thanks [10:40] BjornT_: a fix for this would be to have an actual comment box on the editstatus page [10:40] brad and i are discussing this now [10:41] jamesh: yes, i think that will happen some day. there's a bug that discusses that i think [10:41] BjornT_: maybe you want to come over today, and also discuss malone stuff? [10:41] i think it will happen soon ;-) [10:41] SteveA: sure. maybe after lunch? [10:42] there's the launchpad meeting later. how does your lunch figure in that? [10:42] ah, true. i will have lunch in 30-60 mins [10:43] at your place? [10:43] yeah. or when do you guys plan to go lunch? [10:43] probably 1 - 1.5 hrs [10:43] somewhere in old town [10:44] actually, to make sure we're back for the meeting, 1 hr or so [10:44] so, you could come to my place for the meeting [10:45] also, we could meet in the old town for lunch in 1 hr [10:45] ok, then i could just as well join you for lunch then [10:45] s. valentinos? [10:46] sounds good. around 12:45 then? [10:46] 12.45? [10:46] the vilnius posse is in the house! === BjornT [n=bjorn@84.32.140.183] has joined #launchpad === Seveas [n=seveas@ubuntu/member/seveas] has joined #launchpad === koke [n=koke@ubuntu/member/koke] has joined #launchpad [10:58] daf, carlos: saw my reply? [10:59] jordi, yes [10:59] a pity it was only for a single language... [11:00] carlos: re what you were talking before [11:00] what do you mean "user database"? [11:00] if you mean the geckos info in /etc/passwd, it's in arbitrary encodings [11:00] jordi, /etc/passwed [11:00] jordi, /etc/passwd [11:00] if someone calls adduser with a latin1 locale, adduser will write latin1 [11:01] jordi, isn't it broken? [11:02] jordi: good reply [11:02] everything should be UTF-8 [11:02] daf: just not the want we wanted me to write :P [11:02] s/want /one / [11:02] yeah :( [11:03] I think this was sabdfl's check to see if I am paying attention :P [11:03] does anyone have a silly script to make a pot from a po file? [11:04] or, would rosetta accept a po file for a pot? [11:04] jordi, msgfilter [11:04] hmm, I think with TranslationUploads merge, it will accept it [11:04] but you can try [11:04] just remember to rename it to end in .pot [11:04] carlos: how do I use msgfilter for this? [11:05] msgfilter can filter the translations [11:05] jordi, just apply a filter that returns '' [11:06] I suppose you will need to manually fix the header (msgid "") [11:06] nod [11:06] it's for these kernel translation guys [11:07] I wonder who bothers with make config these days :) [11:07] they do ;-) [11:07] jordi, anyway... doesn't it go against our current policy? [11:08] will kernel guys get the .po files from Rosetta? === ddaa finds the thought of doing smoke tests with pot vaguely amusing === Seveas [n=seveas@ubuntu/member/seveas] has joined #launchpad [11:24] ddaa: Oh cool [11:26] ddaa: I sent email about it already, but to start pulling I need an url with which to start scanning. The url can be empty but must exist. [11:28] -ECONTEXT [11:28] BTW, there's something I just though about in bed yesterday, about spiv's acceptance tests [11:29] He mentions generation of a mapping file for the apache rewriting rule as part of the push-sftp functionality. Does that mean that the mapping for pulled branches is generated separatedly? [11:29] ddaa: You want a supermirror scanner up and running. [11:29] For that, I need the url to parse. [11:30] ID branchsource [11:30] ID branchsource [11:30] didn't I tell you... [11:30] I have http://ubuntu.launchpad.net/+newbranchlist or somesuch. [11:30] let's look at it [11:30] Unfortunately it dosen't exist [11:30] I'm pretty sure I never said that... [11:30] Rob gave me that. === cprov [n=cprov@200-171-140-32.dsl.telesp.net.br] has joined #launchpad === ddaa checks the zcml === BjornT heads off to old town [11:31] morning dudes [11:31] mornign cprov. [11:31] ddaa: The url isn't important to me. Thats just a push and an rsync. [11:32] jblack: hey buddy [11:32] jblack: www.launchpad.net/+supermirror-pull-list.txt [11:32] should only be accessible from within the DC [11:33] daf: you can set a facet on a view anyway [11:33] daf: on the view's class, in the init, say self.__launchpad_facetname__ = 'foo' [11:33] luckily vostok has lynx [11:33] Ruh roh. [11:33] vostok can't see that [11:33] Mh... [11:34] chinstrap neither... note that most DC systems have w3m instead of lynx [11:34] I wonder what continent elmo is on [11:34] SteveA: ping! [11:34] hello ddaa [11:34] i'm leaving for lunch in 2 minutes [11:35] SteveA: ah, so if I add a view class to the generalform, I can fix it? [11:35] any clue what could be missing for the /+supermirror-pull-list.txt page to work? (the private LAN page that's defined in branch.zcml) [11:35] oh, it has a class already [11:35] daf: yes [11:35] ddaa: to work? [11:35] ddaa: He'd have to change the firewall I suppose. [11:36] mh... probably another port [11:36] InternalHTTPLayer is almost certainly not 80. [11:36] ddaa: ask stub please [11:36] eh? [11:37] where can I contact the InternalHTTPLayer of Launchpad.net? [11:38] ddaa: gangotri.ubuntu.com:9000 [11:39] ddaa: If you can't reach that, there is a port block in place. [11:39] I can reach it... But I get a 404 [11:39] for "w3m http://gangotri.ubuntu.com:9000/+supermirror-pull-list.txt" [11:40] blah! [11:40] no plus... [11:40] yay! [11:40] stub: thank you love! [11:41] jblack: [11:41] No good [11:41] it works from chinstrap... [11:41] it's only accessible from within the DC [11:42] Elmo's always been a bit paranoid about vostok. Wel... not paranoid. Careful. [11:42] if there's a port block from vostok, please RT about it [11:43] is there something that does the equivalent of "bzr switch"? === Nafallo_away is now known as Nafallo [11:44] daf: "bzr push ../daf/launchpad/da-branch ; bzr pull --overwrite ../da-other-branch" [11:44] thanks [11:44] there was a bug with pull --overwrite not doing anything if there was nothing missing from the other branch, but it has been fixed recently. === SteveA + brad -> lunch with bjorn [11:46] ddaa: For this one, I'll stalk. [11:46] I have a whole put of coffee on my desk. I can outwait him. :) [11:47] I may as well rt@ as well though... Even when he does stuff he likes a record [11:48] hey there [11:48] meeting in 1:15 [11:53] is pqm working again? [11:54] no [11:55] (see topic :)) [11:55] jblack: so, what do you know about the mapping generation for the supermirror (that I asked you about earlier)? [11:56] I emailed that to you as well. :) [11:56] but.. [11:56] let me look it up [11:56] oh, I thought it had only minor problems with mail.. [11:56] http://bazaar.canonical.com/~$person/$product/$branch/ [11:57] http://bazaar.canonical.com/~$person/+junk/$branch/ [11:57] http://bazaar.canonical.com/~$team/+product/$branch/ [11:57] jblack: ? [11:57] More information on https://wiki.launchpad.canonical.com/SupermirrorFilesystemHierarchy. [11:57] Oh, you mean how is it sorted on the filesystem? [11:57] no [11:57] additional topics to MeetingAgenda please [11:57] I mean about which component is in charge of generating the mapping file. [11:58] spiv put that as part of the acceptance tests for the push-sftp [11:58] I thought that was you. [11:58] For the scanner, its off in magicland. It just gets it. [11:58] he sent me the tests to review them [11:58] I don't think push-sftp knows. [11:59] so I'm concerned that nothing would generate the mapping for the pull branches [11:59] OH. [11:59] You mean the mod_rewrite stuff. [11:59] yes [11:59] sync. [11:59] ddaa: http://bazaar.launchpad.net/ :) [11:59] jblack: ^ [12:00] jblack: bazaar.canonical.com is the bazaar project website :) [12:00] ddaa: I don't know who's doing that. [12:00] ddaa: whats the question ? [12:00] I can't, because I don't have the lp data. [12:00] At LAST one part of launchpad that does not suffer from information overload!! === WaterSevenUb [n=WaterSev@azevedo.astro.up.pt] has joined #launchpad [12:01] lifeless: spiv mentioned generation of the mapping file for the rewrite rule as part of the acceptance test for the push-sftp. What component is in charge of generating the mappings for pull branches? [12:02] ddaa: what do you mean by that? How are pull branches different for this? what do you mean by component ? [12:02] lifeless: Ok. People want to get http://bazaar.launchpad.net/person/product/branch [12:02] Pull branches are not handled by the sftp system. [12:02] But its stored on the filesystem as /xx/xx/xx/xx/xx or somesuch. [12:02] jblack: ddaa has the question, please let him explain [12:02] jblack: I need to pin down the exact semantic he is asking about [12:03] ddaa: with you so far. [12:03] If I understand correctly, they are handled by another subsystem, that pulls from third party branches and from the private sftp filesystem. [12:03] into the filesystem of bazaar.launchpad.net [12:04] ddaa: right. pull AND sftp branches are handled identically for the 'publish' step. [12:04] ddaa: they are both copied from some 'master' location by jblacks script, and published from their id-based location on disk. [12:05] I was just wondering about whether the mapping file generation spiv mentions in the acceptance tests, that is used by the mod_rewrite thingy, was wrote to be used by the publisher script instead of by the push-sftp. [12:05] yes, push-sftp has nothing to do with publishing [12:06] he wrote a separate script that maps all branches at once in one hit [12:06] Since it was documented as part of he push-sftp acceptance tests, I wanted a clarification that the mapping file for pull branches has not fallen through the cracks. [12:06] ddaa: thats implemented and rolled out already === ddaa paranoia is relieved [12:06] ddaa: in the context of publishing there are no 'push' or 'pull' branches, just 'branches'. [12:07] fine [12:07] jblack: thanks for letting ddaa spell it out, I thought it would be what it was. [12:07] just wanted to sanity check the issue. As I told jblack, it occured to me in bed yesterday after sending the review for the push-sftp acceptance tests. [12:07] No problem. [12:07] You hurt my feelings though. Now I need a hug. [12:08] (just kidding) === niemeyer [n=niemeyer@200.103.133.2] has joined #launchpad [12:08] Got to hug partner right now btw. [12:10] lifeless: Can you do me a favor please? [12:10] its a quick one [12:10] Never mind [12:11] jblack: that was easy :) === beyond [n=beyond@200-171-140-32.dsl.telesp.net.br] has joined #launchpad [12:11] Never mind .I still want the favor. === matsubara [n=matsubar@200-171-140-32.dsl.telesp.net.br] has joined #launchpad [12:12] good morning! [12:12] I don't seem to have the second reply I sent you last night. Can you bounce me a copy? [12:12] I think mutt on my system has a problem when there's a lot of Ccs involved [12:13] lifeless: ^ === raphink [n=raphink@gra86-1-82-239-88-236.fbx.proxad.net] has joined #launchpad [12:15] hmmmm [12:15] "All methods should be named fooBar, and all other attributes, properties and functions should be named foo_bar" [12:15] reviewers, check your queues [12:16] lifeless: "vrooom vrooom" ? [12:16] ok, now I'm so happy to use lots of properties with the new POMsgSetView or I should rewrite it again to follow the new policy.... [12:16] mentally, I was doing that [12:16] don't be _too_ literal [12:17] foo-bar is not a very expressive name, regardless of case variations [12:17] Bom dia! [12:18] Hello [12:18] jblack: title ? [12:18] niemeyer: Bom dia! === salgado [n=salgado@200-171-140-32.dsl.telesp.net.br] has joined #launchpad [12:19] "Re: Works in progress, new rocketfuel.*" [12:20] lifeless: BTW, got the launchpad fix for the new bzr [12:20] jblack: I did not recieve that [12:20] ddaa: cool! [12:20] the test suite just finished, I'm going to push it right now [12:20] trivial fix for the WorkingTree refactoring [12:20] jblack: last email I have with 'works' in the title was dec 01 === raphink [n=raphink@gra86-1-82-239-88-236.fbx.proxad.net] has joined #launchpad [12:21] ddaa: could you please do a 'make check_merge' === ddaa cries [12:21] ddaa: for my sanity ? I can't do that on pqm at the moment [12:21] "make check" took 27 mins already... [12:22] jamesh: I have that error in gdb [12:22] jamesh: shall we chat out-of-channel ? [12:22] lifeless: okay [12:22] uh... [12:22] pardon? === jblack sits back for a moment [12:22] lifeless: BTW your branch is robertc/bzr/dev, right? Not robertc/bzr/importd as you told me. [12:22] Yeah, you got them. You replied to them. :) [12:23] Thats ok [12:25] ddaa: right bzrtools is import [12:25] importd [12:25] jblack: was it a list ? [12:25] lifeless: ha right, I need to update bzrtools as well! [12:26] lifeless: Its ok. :) [12:26] jblack: I store threads by list [12:26] lifeless, is pqm back already? if so, what's the address that I should send the merge requests? [12:26] jblack: tell me the list, I'll bounce it for you [12:27] salgado: ' PQM is being upgraded, patience please! | launchpad.net | developer meeting: Thur 15 Dec, 1200UTC (wiki:MeetingAgenda) | launchpad-users@lists.canonical.com ...' [12:27] It would have gone to launchpad@ [12:27] But its ok. Really. :) [12:28] jblack: its no bother. one minute [12:28] done [12:31] thanks === Nafallo is now known as Nafallo_away [12:36] salgado: the PQM machine is suffering from the pyme crashes BjornT and cprov experienced [12:36] salgado: we're working on it [12:36] ouch. that's really bad. :-( [12:37] sheesh [12:37] jamesh, thanks for letting us know [12:37] jamesh: glibc corrupted ? === mpt [n=mpt@200-171-140-32.dsl.telesp.net.br] has joined #launchpad [12:37] cprov: yeah [12:37] cprov: the double free abort [12:38] Goooooooooooooooooooooooooooooooooooood morning Launchpadders [12:42] morning people. bah [12:42] hallo [12:43] hey jordi [12:43] mpt, morning [12:43] cprov: we have a handle on it though. Getting closer. [12:43] mpt, dude, have you seen?: https://launchpad.net/people/carlos/+editwikinames [12:43] mpt, I thought it was only on our development branch... [12:44] lifeless: good, thx [12:44] carlos, I have a fix for that already [12:44] salgado, ok. Is it for the whole launchpad or just the people's page? [12:44] (and all other person edit pages that have the same problem) [12:44] salgado, because we have the same problem with the Rosetta translation form [12:45] I'm talking about the overlapping of text and portlets [12:45] carlos, AFAICT, this will be a problem in all pages that have fixed widget sizes that are larger than the available space in the middle column [12:46] salgado, so it's something that we should fix per page [12:46] I might be wrong, though [12:46] mpt, could you confirm, please? [12:46] if that's the case I will try to fix it for Rosetta [12:46] salgado, that man speaks the truth [12:48] carlos, text fields which need to be as wide as possible should have style="width: 100%", not width="n" [12:48] mpt, SteveA asked me to remove the style tags from the templates and move them to the .css files [12:48] mpt, do we have a class for that? [12:48] a css class [12:49] hmm, no [12:49] we probably should [12:49] could you add one? [12:49] sure [12:49] ddaa, ping [12:49] mpt, thank you [12:50] mpt: pouet [12:50] ddaa, other than yours, do you know of any bugs reported about confusion with projects? [12:51] ** meeting in 13 minutes, take a workrave now if you need it ** [12:52] mpt: I'm not sure either way... I may have asked confused users to file bugs way back (like 6+ months ago) [12:53] stub, got time for a quick database query? [12:53] thanks ddaa === ddaa is cooking noodles may be a few mins late [01:01] Launchpad meeting time! [01:01] who's here [01:01] ? [01:01] you are [01:01] Here [01:01] me [01:01] me === niemeyer me [01:02] niemeyer is here [01:02] me [01:02] here [01:02] We're the knights who say "me" [01:02] me [01:02] me [01:02] bjorn is just booting up [01:02] stub: ? [01:03] mpool sends apologies [01:03] ddaa, Kinnison and daf will be a few minutes late [01:03] kiko? [01:03] jamesh: ? [01:03] Yo [01:03] here [01:03] cprov: ? [01:03] sorry. was distracted === daf is here [01:04] meeting time! [01:04] == Agenda == [01:04] * Roll call [01:04] * Agenda [01:04] * Next meeting [01:04] * Activity reports [01:04] * Items from last meeting [01:04] * Production / staging (stub) [01:04] * Renaming projects to product groups, to avoid confusion with people used to the way SourceForge, Freshmeat, and gnome.org use the term "project". (MatthewPaulThomas) [01:04] * Change of review requirements to add mandatory test coverage. (ChristianReis). [01:04] * Error reports on chinstrap. (SteveAlexander) [01:04] * Confirm method naming convention. (SteveAlexander) [01:04] here [01:04] * Keep, Bag, Change [01:04] * Three sentences [01:04] [01:04] Wow [01:04] The next meeting: same time next week? [01:04] wow? [01:05] SteveA, works for me [01:05] Large agenda.. let's go! [01:05] Can I wear a Santa suit to the next meeting? ;) [01:05] so [01:05] only four non-routine things, niemeyer [01:05] have there been no conclusions from the meeting times wiki page? [01:05] same time next week? [01:05] == Next Meeting == [01:05] yes. === ddaa arrives [01:05] same time next wek? [01:05] yes [01:05] Fine with me. [01:05] kiko: you will be on vacation, no? === ..[topic/#launchpad:SteveA] : PQM is being upgraded, patience please! | launchpad.net | developer meeting: Thur 22 Dec, 1200UTC (wiki:MeetingAgenda) | launchpad-users@lists.canonical.com (wiki:MailingLists) | Channel logs are here: http://tinyurl.com/72w39 === hpk [n=hpk@200.103.133.2] has joined #launchpad [01:06] I may be able to make it [01:06] that will be Festivus-eve [01:06] that's okay [01:06] == Activity reports == [01:06] The following people were lax with activity reporting last week: [01:06] * spiv, lifeless, jblack, stevea, [01:06] who's up to date this week? [01:06] I am up to date! [01:06] not me, unfortunately, again [01:06] i am [01:06] I am === ddaa raises a fork [01:06] (And I caught up with the missing ones from last week, too) [01:06] I am [01:06] I am up to date [01:06] me [01:06] I'm up to date because I restarted the reports last Monday [01:07] I am [01:07] me too === jamesh is not === daf sends yesterday's [01:07] I'm up to date [01:07] Same as last week: I'll be up to date in approximately five minutes [01:07] I'm to days late :( [01:07] s/to/two/ === cprov is up [01:07] daf: is Kinnison with you? [01:08] no, he isn't [01:08] but will be up to date today in the afternoon [01:08] I'm expecting him at any moment [01:08] ok [01:08] * Items from last meeting [01:08] * RobertCollins to set up a bzr-for-launchpad meeting at a different time to the launchpad developers meeting. [01:08] lifeless: ? [01:08] * MatthewPaulThomas to set up the MeetingTimes wiki page, and encourage people to fill in their information. [01:08] hi [01:09] mpt: you did that, right? [01:09] did everyone fill in their meeting times? [01:09] yes, within a couple of hours of the meeting [01:09] uhm, so had a rather nuts week, did not get that done. Will do tomorrow, cross heart [01:09] (ok, *now* I'm up to date) [01:09] lifeless: okay, MeetingAction for this meeting [01:09] * Kiko to set up a launchpad community meeting. === niemeyer didn't.. [01:09] kiko: ? [01:10] * Steve to announce the production of last week's summary on the launchpad-users mailing list. [01:10] I didn't. [01:10] i didn't, as the summary isn't really finished [01:10] If you don't fill in your row in MeetingTimes or keep it up to date, that just means meetings are more likely to be held at a time that's crappy for you :-) [01:10] * Daf to work on the PythonStyleGuide. [01:10] I've really been too busy to do this [01:10] but I'll do it [01:10] daf: ? [01:10] done [01:10] I'm awaiting feedback [01:10] kiko: we can put it off until the new year, if the planning is done now. [01:10] thanks daf [01:10] * DanielSilverstone to write SQL patch for adding correct extra archs to production DB [01:10] Kinnison: when you arrive... [01:10] * JamesH to add to his work queue removing `__len__` from SelectResults, as discussed. [01:10] jamesh: ? [01:10] I got the sql patch from Kinnison [01:11] cool === niemeyer notes that there's a discussion going on in python-dev about PEP-8 [01:11] SteveA: sorry. I got side tracked with the sitemap code and some reviewing work [01:11] niemeyer: can you summarize it? === BjornT [n=bjorn@195.182.78.95] has joined #launchpad [01:11] jamesh: please make sure this is up to date in the appropriate bug, with what was decided last week [01:11] * Bzr people and Kiko to come up with a way that works for all of them to record the priority and status of bzr-for-launchpad issues. [01:11] SteveA: okay [01:11] Nothing serious.. it's mostly about missing points getting included, and clarifying existent ones [01:12] this is mainly a bzr people thing. lifeless / jblack ? [01:12] niemeyer: okay. would you say when / if pep-8 has been updated? [01:12] SteveA: I'll write a mail to the launchpad-dev mailing list with a summary after the discussion stops. [01:12] thanks [01:12] Stevea: Isn't that a malone thing? [01:13] SteveA: we've - kiko and us - been insane, no progress [01:13] jblack: options discussed were wiki or using malone [01:13] If not, I can put up a page [01:13] okay, it remains as an agenda item [01:13] i don't personally care whether it is malone or a wiki. [01:13] i just want something that works, and for the bzr folks to be clear to the (other) launchpad folks on how to use it [01:13] we didn't really get around to that, SteveA -- possibly because I wanted us to agree on using Malone and therefore wanted to talk about the keywords/status whiteboard thing [01:13] to express the current issues etc. [01:14] Noted and marked as a priority item [01:14] kiko: we shouldn't delay capturing good bzr feedback from the launchpad team on this [01:14] so, if jblack can set up an effective wiki page, let's just do that [01:14] and move it to a better "malone or whatever" solution later [01:14] perhaps we can capture things on the wiki and file bugs afterwards? [01:14] can we PLEASE consider using malone FIRST? [01:14] too late. it's been a week, and it was an item from last week. [01:14] I am totally amazed at us even suggesting a wikipage [01:14] we can move it later. [01:14] Malone is too hard to use --> people start making lists of stuff elsewhere [01:14] when we MAINTAIN a bugtracker [01:15] no, seriously [01:15] use malone [01:15] this issue has had a week. [01:15] a wiki for tracking bugs is ridiculous [01:15] we can change from a wiki page to malone later. [01:15] it isn't necessarily bugs. [01:15] you don't have time to push this through. [01:15] kiko: we were using malone, then you said you wanted something different [01:15] no [01:15] well... not really [01:15] the issue was [01:15] we just wanted a way to indicate priorities for launchpad [01:15] that there is no way to get the launchpad team to set a priority in malone [01:15] and keep track of their statuses [01:16] although one way to do this is to have a bzr-for-launchpad product [01:16] I have suggested using the status whiteboard [01:16] that is in the launchpad project [01:16] Guys, I suggest that the two of you discuss it after the meeting. [01:16] and set status there [01:16] Email me the results of the discussion and I'll act appropriately. Ok? [01:16] kiko: will you discuss it today, immediately after this meeting? [01:16] yes. [01:16] either this gets resolved today, or we go for the wiki solution. [01:16] look [01:16] okay [01:16] I added a meeting agenda item [01:17] it was /for this specific purpose/ [01:17] to explain how to use malone for this [01:17] so I suggest we do it ;) [01:17] brad and i have moved in a different direction on the whiteboard [01:17] so, no [01:17] moving on... [01:17] other notes: [01:17] * Andrew's jury duty didn't happen. [01:17] * Only eight days left to Festivus. [01:18] Festivus? [01:18] * Production / staging (stub) [01:18] yeah, what's Festivus? [01:18] daf, ddaa, http://en.wikipedia.org/wiki/Festivus [01:18] "A festivus for the rest of us" [01:18] Business as usual on both production and staging. Staging is still being used for Gina and publishing testing. Production rollout happened yesterday. [01:19] stub: next production update? [01:19] Are there any patches that need to be rolled out next week? [01:19] yes [01:19] working site map [01:19] various UI fixes [01:19] Have they landed? [01:19] stub, TranslationUploads [01:19] sitemap is on its way through pqm [01:19] I have that branch ready to merge since two weeks ago [01:20] ?lpnotification=... being disabled [01:20] but failed to be merged because conflicts or bugs in our test suite [01:20] SteveA: That is still up for review. [01:20] who was reviewing? lifeless ? [01:20] SteveA, is pqm working again? [01:20] lifeless: is pqm working? [01:20] This list sounds like too many for 'need' to roll out, especially as they won't all get through pqm in time. [01:20] SteveA, I was told that is not working [01:20] SteveA: yes but: [01:21] * procmail is being fixed right now [01:21] but you say that sitemap is there or did I misunderstood you? [01:21] * merges are failing because of the glibc bug [01:21] hello, I was in another meeting, sorry guys [01:21] SteveA: I was not reviewing, I put stuff in my queue today === Kinnison makes incredibly rude noises about broken down busses. Sorry I'm late [01:21] SteveA: should get reviewed tomorrow [01:21] stub: so, can we do an out-of-phase rollout, say taking a patchlevel on monday, and rolling out sometime later next week? [01:22] thanks lifeless [01:22] SteveA, not too late, please [01:22] SteveA: speaking of reviews [01:22] SteveA: I am up-to-date [01:22] if something breaks... will be broken too much time [01:22] SteveA: can I make a small mention later of the review team changes [01:22] lifeless: yes [01:22] SteveA: I suspect just skipping next week would be best - nothing mentioned sounds like 'we must have this landed to work'. Am I wrong? [01:23] stub sounds about right. [01:23] stub: are you planning to work over the holiday? [01:23] in particular because PQM isn't even working right now === SteveA can't remember stub's plans [01:23] okay fine, skip a week [01:24] * lifeless: small mention on review team changes [01:24] right [01:24] so the review team general queue is now being purged by me daily [01:24] reviewers: please check your queues once a day [01:24] SteveA: I'll be working over the christmas break. I *might* need to take 1st-5th Jan off, but I havn't cleared that with anyone and it is a bit up in the air. [01:24] reviewees - you should get better service we hope. [01:24] thanks SteveA [01:25] (1st is Sun, 2nd and 3rd are public holidays) [01:25] lifeless, so you're assigning branches to reviewers within 24 hours of them arriving? [01:25] mpt: roughly, yes [01:25] lifeless, cool, thanks [01:26] lifeless, unless that provides motivation for reviewers to be faster, I think that will actually result in longer average waiting times [01:26] (which is why, for example, many banks have switched from one-line-per-teller to one-line-for-all-tellers queues) [01:26] mpt: we've also added in a feedback loop on how busy people are [01:26] mpt: and I'm tracking the review status [01:27] there's a teller-manager [01:27] mpt: so we're not like a bank. we can move people between queues [01:27] ok [01:27] without getting beaten up by irate grandmothers [01:27] moving on... [01:27] * Renaming projects to product groups, to avoid confusion with people used to the way SourceForge, Freshmeat, and gnome.org use the term "project". (MatthewPaulThomas) [01:27] mpt: there was a sense that the general queue was too big and there was no 'I should do one' feeling [01:28] ok [01:28] ddaa reported a bug about projects being confusing, and I realized that they were [01:28] because Freshmeat, SourceForge, and gnome.org all use the term "project" for what we call "product" [01:28] general meeting meta-notice: for items that you propose on the agenda, before the meeting, prepare a few sentences on it, like with your three sentences. then paste these in to introduce the topic. === SteveA will update the agenda page with this advice [01:29] That wouldn't be that bad by itself, except that Launchpad uses "project" to mean something else [01:29] mpt: I stumbled across this yesterday when helping somebody get something set up in Rosetta -- I'll write up notes to the list [01:29] and if you look through the list of projects , you can see that maybe half of them actually should be products [01:29] So, I propose renaming projects to product groups [01:29] it confused the heck out everybody until projects became essentially a hidden feature [01:30] Just in the UI for now, and if it works well, the URLs later [01:30] whichthey are no longer [01:30] we can't pass a proposal like that at this meeting. [01:30] arrrrg [01:30] so, i'd like a mailing list discussion on it [01:30] ok :-) [01:30] for the record, I am in support of this -- just would need to think when to do it === bradb suggests "Projects" and "Project Groups" to use words that users use when they talk about these things, but that sentence has expended 98% of the energy I'm willing to put into debating on this. [01:30] and some specific examples of the projects that should be products [01:30] This has previously been discussed on IRC - it would be purely UI change at this time === ddaa gives procuration to mpt to voice a loud swearing +1 to this proposal [01:31] Internally, we would still have projects [01:31] it's a nomenclature thing [01:31] this issue has *history*, and needs a focused call with mark. before such a call, we should examine the issue, and collect information. [01:32] stub, can you get the number of projects that have 0 products, and the number that have 1 product? [01:32] so, AgendaItemForNextMeeting: see how the project -> somethingelse discussion is going [01:32] mpt, I did that a while back -- it really was a minor number [01:32] time marches on [01:32] * Change of review requirements to add mandatory test coverage. (ChristianReis). [01:32] kiko, half those I'm looking at have only 1 [01:32] mpt, kiko: use EMAIL [01:33] I don't think I can discuss that at length right now [01:33] use me, use me, use me baby! [01:33] mpt, those come from the old freshmeat import where all products had a project even when they were the only product [01:33] kiko: okay, AgendaItem for next meeting [01:33] or the next meeting you'll be at anyway [01:34] * Error reports on chinstrap. (SteveAlexander) [01:34] so, the /errors page is gone from launchpad [01:34] no it's not [01:34] it lives on [01:34] when someone gets an error from a web page, they see a page with an OOPS-xxxxx code in it [01:34] daf: that's an error :-/ [01:34] this is being discussed in a bug [01:35] the xxxxx code can be retrieved from the approrpriate directory on chinstrap [01:35] the appropriate directory depends on whether it is production or staging, and the date [01:35] so, /srv/gangotri-logs/date/.... [01:35] bug 5789 [01:35] Malone bug #5789: /errors page should be removed In: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: New https://launchpad.net/bugs/5789 [01:35] within a date directory, error reports have a filename timestamp.oopscode [01:35] timestamp is seconds thru the day [01:36] the directories on chinstrap are rsynced every 10 mins or so [01:36] Can they be made Web-accessible somehow? [01:36] so, you may have to wait a few minutes to debug an oops report [01:36] mpt: perhaps. [01:36] daf: /errors must go. [01:37] a way to view OOPS reports through the web may replace it. [01:37] * Confirm method naming convention. (SteveAlexander) [01:37] is this in the PythonStyleGuide now? [01:37] yes [01:37] basically, foo_bar_baz style for everything except methods [01:37] and the chosen method is the not-too-vile one? [01:37] which use fooBarBaz style [01:38] * Keep, Bag, Change [01:38] SteveA: I assume it's fooBarMethod, foo_bar_property and FooBarClass ? [01:38] Kinnison: yes [01:38] SteveA: good oh [01:38] can I raise a Keep: launchpad weekly reports from kiko [01:38] sure [01:38] Keep: christmas. Bag: stress. Change: year. === Kinnison finds then very useful/interesting and I'd like to thank kiko [01:38] hurrah for reports! [01:38] I will keep them coming [01:38] weekly reports ++ [01:38] kiko's paypal account is... [01:38] this week's report will be sad unless we unwedge the pqm wedgie [01:39] Keep: meeting summaries === ddaa points at his BazaarMeeting reports [01:39] mmm yes, they're really useful too [01:39] ddaa: where did they go? [01:39] That directory is badly named, but we can fix that when the next production server goes online [01:39] ddaa: I didn't know about them [01:39] ddaa: reports should be announced to the luanchpad list or the launchapd users list [01:39] They are. [01:39] cool [01:40] onwards then [01:40] https://wiki.launchpad.canonical.com/SupermirrorTaskList?action=fullsearch&context=180&value=bazaarmeeting&titlesearch=Titles [01:40] * Three sentences [01:40] go for it [01:40] DONE: Vast quantities of niggly publisher fixes [01:40] TODO: More pool-related fixes before the end of tomorrow [01:40] BLOCKED: Nada [01:40] DONE: Soyuz testing, HCT discussions, planning, minor coding [01:40] TODO: More Soyuz testing, staff review feedback, Vacation! [01:40] BLOCKED: not ultimately blocked on anything [01:40] 2DONE: supermirror install. rocketfuel/pqm 1st draft. bzr community [01:40] DONE: Spec'ing in Vilnius with Steve. Landed a few bug fixes. [01:40] TODO: supermirror running. rocketfuel/pqm 2nd draft. bzr community [01:40] BLOCKERS: Firewall issues for supermirror [01:40] TODO: More spec'ing in Vilnius. Possibly land the status changes. Maybe get time to land IBC. [01:40] DONE: lang bugzilla-import branch / code reviews / NewPageLayout sitemap / other bug fixes [01:40] TODO: SelectResults.__len__ / pyme crash bugs / code reviews [01:40] BLOCKED: no [01:40] DONE: fixed bug of non-ascii passwords blowing up launchpad, bug reports [01:40] TODO: finish error report, fix bugs, catch up with bugmail [01:40] BLOCKED: nope [01:40] BLOCKED: No. [01:40] DONE: Supermirror SFTP work -- can successfully "bzr push" now. Some other supermirror-related stuff. [01:40] TODO: Iron out the bugs in supermirror SFTP, but it's looking good. [01:40] BLOCKED: no [01:40] DONE: supermirror install. rocketfuel/pqm 1st draft. bzr community [01:40] DONE: pqm move to balleny, mucho bzr reviewing. [01:40] DONE: Code review, random fixes, implemnted ProperSignUpWorkflow, started InactiveMembershipDeletion and MirrorManagement [01:40] TODO: make it work, baz2bzr custom build, submit-merge plugin, lp test work [01:40] BLOCKED: SteveA's zope3 update (week 4) [01:40] TODO: finish MirrorManagement, restart InactiveMembershipDeletion [01:40] BLOCKED: No [01:40] DONE:finished up my email-ui error messages branch, still some to come later, though.. fixed a few bugs in the email system, mainly gpg related one. reviews. [01:40] mpool: DONE: bzr development, many merges, some development of new storage (didn't finish) [01:40] DONE: much communication around Bazaar integration and company admin, little code [01:40] TODO: Vacation [01:40] BLOCKERS: None [01:40] mpool: TODO: finish storage development, plan for api stability, 0.7 release [01:40] mpool: BLOCKED: no [01:40] TODO: fix a few more bugs in the email interface. start on SupportTrackerTweaks. reviews. [01:40] DONE: Helping Soyuz deployment, maintenance of Smart, Grumpy and Gantry discussions [01:40] TODO: Do something on Soyuz, release Smart 0.41, inicial thoughts for Gantry, get a USA visa [01:40] BLOCKED: Nope [01:40] DONE: Cleanup and rewrites on POMsgSetView and POFileView to implement PoMsgSetPage spec, fight to merge TranslationUploads [01:40] BLOCKED: no [01:41] DONE: Implementing LaunchpadCapitalization, writing MaloneSearch [01:41] TODO: finish LaunchpadCapitalization and cleanup, layout tweaks, return to NZ [01:41] BLOCKED: no [01:41] TODO: Finish PoMsgSetPage implementation and get TranslationUploads merged [01:41] BLOCKED: no [01:41] DONE: bug 2230, bug 57871, misc Soyuz UI work, PythonStyleGuide and other docs, pyflakes integration [01:41] Malone bug #2230: /distros/ubuntu/+sources will need to present something useful In: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: New https://launchpad.net/bugs/2230 [01:41] Error: Error getting Malone bug #57871: Bug does not exist [01:41] TODO: more Soyuz UI work, land fixes, more docs work [01:41] DONE: some malone work with brad in vilnius, UI wrangling work [01:41] TODO: more malone with with brad, management, zope3-at-last [01:41] BLOCKED: no [01:41] BLOCKED: BradB for LaunchpadHackingFAQ revision [01:41] daf, you've seen make lint, yeah? [01:41] daf: I shouldn't be blocking you on that. [01:42] daf has been fixing the future [01:42] 55000 bugs in the future [01:42] DONE: breezy-autotest (almost done), publisher fixes [01:42] TODO: dapper-autotest [01:42] BLOCKED: None [01:42] ddaa, jblack: if you say BLOCKERS and not BLOCKED then you risk being excluded from the summary [01:42] sorry. [01:42] daf: To repeat, my interest was only in a simple LP coding guidelines document. [01:42] lifeless, http://en.wikipedia.org/wiki/Queue_area [01:42] BLOCKED: Firewall issues for supermirror [01:42] BLOCKED: no [01:42] DONE: Odds and sods mainly [01:42] TODO: fti.py updates and cachine [01:42] BLOCKED: Nope [01:42] I have the following blocking issues: jblack BLOCKERS: Firewall issues for supermirror [01:42] lifeless BLOCKED: SteveA's zope3 update (week 4) [01:42] daf BLOCKED: BradB for LaunchpadHackingFAQ revision [01:42] daf: And, on a fairly quick skim, the PSG seems to achieve that. [01:43] jblack: are these issues being addressed? [01:43] bradb: ok -- my impression from the last meeting was that I needed to get your input [01:43] lifeless: yeah, see my TODO. sorry it is taking so long [01:43] stevea: Yes. [01:43] kiko: I hadn't seen that [01:43] steva: Well, I'm attempting anyways. I'm behind pqm. [01:43] bradb: http://www.psg.fr ? [01:43] kiko: I still think my scripts are cool [01:43] ddaa: PythonStyleGuide :) [01:44] any other blocking issues? [01:44] daf, well, check it out [01:44] anyone know where i can get a long metal pole in vilnius? [01:44] The document might be better named LaunchpadCodingStandards, but that's it [01:44] SteveA: a strip bar? [01:44] okay, we've had 45 minutes [01:44] nice idea brad, we have plenty of those [01:45] although, i do want a clean one [01:45] MEETING ENDS [01:45] thanks everyone [01:45] hopefully there is soap in vilnius too [01:45] thanks SteveA [01:45] stevea: kiko: When the two of you are done discussing the launchpad/bzr priority issue, email me the result? [01:45] They're lax on replacing the soap in my hotel room [01:45] er, TMI [01:45] jblack: willdo. [01:45] ok. [01:45] jblack: maybe via the launchpad list though, okay? === ddaa goes to finish lunch [01:46] Both, please [01:46] k [01:48] mpt: btw, my dads finished-but-not-submitted masters is in queueing theory [01:49] Sounds like you could survive living in london. :) [01:49] mpt: but thanks for the link/ [01:49] ok, I'll trust you then lifeless :-) [01:50] ok [01:50] pqm should be recieving mails now [01:50] but they will still fail for a little [01:51] stevea: my lines, things are a bit crazy here right now :/ [01:51] DONE: new imports, lots of mailing; TODO: process pending emails, wiki updates; BLOCKED: still wanting launchpad-experts feautre so I can edit projects myself [01:51] kiko: ^^, any news? [01:52] jamesh: so, think we can fix this tonight? [01:52] jordi, I suck? [01:52] jamesh: want to walk through it with me ? [01:52] oh man [01:52] kiko: but only slightly. [01:52] yeah [01:52] MaloneExperts and RosettaExperts are specced, LaunchpadExperts is not, correct? [01:52] I'm embarassed but my situation right now is that of chaos [01:52] kiko: I love when you come running out in the cold streets of the planet [01:52] mpt, correct [01:53] it's a rogue patch [01:53] that may prove controversial [01:53] oh, it's a patch [01:53] mpt: it's... a secret branch [01:53] because we assumed that anyone needing to hack registry stuff would be a Launchpad admin [01:53] (Launchpadmin?) [01:53] kiko: I know of anoither of your branches which might be controversial too :) [01:54] mpt, which has proven, well, constraining [01:54] kiko, only in jordi's case? [01:54] mpt: it's good to test perms that I'm not promoted to a full fledged admin [01:54] mpt: for now, yes. [01:55] has anyone got a suggestion for a URL in the librarian that'll never (or is very unlikely) to be GC'ed ? [01:55] We might get launchpad contributors which we might want to give those rights in the future as well, ie a registry janitor team [01:55] Permissions useful for testing permissions aren't necessarily permissions useful for anything else [01:55] elmo, any bug attachment, I'd say. want me to find one? [01:55] jordi, so what shouldn't you be able to do that a Launchpadmin can do? [01:55] mpt: I am not sure. Edit people details, for example? [01:55] mpt, LA is kinda like god. [01:55] edit stuff in dists? [01:56] etc [01:56] edit people, teams, etc [01:56] kiko: pls; preferably something small tho [01:56] ok [01:56] I need it to check the librarian's availability in nagios [01:56] elmo, sure. [01:56] so this is really RegistryExperts [01:56] mpt: my current tasks have sometimes the need to transfer ownership of products to people. I can't do that. === mpt returns to hacking on his multi-hundred-file branch [01:56] there's probably other corner cases === hpk [n=hpk@200.103.133.2] has left #launchpad ["Leaving"] [01:58] elmo, http://librarian.launchpad.net/1185577/Pornlet.png [01:58] salgado: Looking at http://localhost:8086/people/name16/+assignedbugs, I click on the "Advanced..." button and nothing happens. === kiko looks at jordi [01:58] er, except that time [01:58] now, repeatly clicking "Simple...", nothing happens [01:58] elmo, is that fitting? [01:59] kiko: it'll work [01:59] thanks [01:59] These buttons are behaving as though they have an "active" area and an "inactive" area [01:59] bradb, by "nothing happens" you mean that the form isn't even POSTed, right? [01:59] sure [01:59] Clicking near the edge of the button depresses it, but causes no action. [01:59] salgado: yeah [02:00] kiko: that is good. :) [02:01] salgado: Seems like it's a more general problem with the UI. I'll file a bug. [02:01] bradb, I've seen this with some buttons in auto-generated forms. [02:01] there's even a bug filed on that, IIRC [02:02] yes, there is, one I couldn't reproduct [02:02] e === Nafallo_away is now known as Nafallo [02:02] salgado: I don't experience this problem at localhost:8086/ [02:02] mpt: Easy to reproduce at the URL above [02:02] er, the +assignedbugs one, that is [02:03] Erm, speaking of form buttons, any idea when we're gonna fix most of the forms to have the right tab order? [02:03] bradb, eh? you first said you experienced the problem while looking at http://localhost:8086/people/name16/+assignedbugs [02:03] salgado: Correct. [02:04] It's easy for me to reproduce it there. I haven't found another page on which I can reproduce the problem yet though. [02:04] bradb, works for me [02:04] the task edit page works as expected, lp home page, etc. [02:04] mpt: are you synched up to rf? [02:05] bradb, as of yesterday afternoon, yes [02:05] though the button does get smaller when I click it [02:05] now why is that happening [02:05] mpt, IT'S ALL YOUR FAULT [02:05] we used to blame it on plone.css [02:05] and now, you've been unmasked [02:05] mpt: so, if you click right near the edge of the "Advanced..." button, in between the last period and the edge of the button, it submits the form? [02:06] the evil button-shrinker of peru [02:06] (on the +assignedbugs page) [02:06] kiko, there is still a large chunk of plone.css in launchpad.css that I haven't killed yet [02:06] bradb, yes it does [02:06] weird [02:07] mpt, excuses excuses excuses. meanwhile, hundreds of buttons are getting shrunken daily [02:07] mpt: what browser? I'm on FF 1.0.7. [02:07] epiphany 1.8.2 [02:07] ah :) [02:07] trying with Fx now [02:07] I don't expect it to be different ... [02:07] I also just reproduced it with: https://launchpad.net/products/malone/+bugs [02:08] clicking on "Search" [02:08] and it made the button shrink [02:08] yeah, I get the button shrinkening all right [02:08] err, shrinking === kiko throws mpt to the lions [02:11] By rights it should be an :active rule somewhere, but the only :active rules in launchpad.css are for links === Kinnison preps to head off to buy a fuse and drive home [02:12] ciau [02:13] HA [02:14] AHAH? [02:14] when I comment out the "UNCHANGED FROM PLONE.CSS" section, the problem goes away [02:14] that was obvious === carlos -> lunch [02:14] s/UNCHANGED FROM PLONE.CSS/BEHOLD THE CRUD/ [02:15] so, narrowing it down ... === auth00_ [i=auth@fiji.grd.sgsnet.se] has joined #launchpad [02:18] that's bizarre [02:19] input {font-family: "Lucida Grande", "Trebuchet MS", sans-serif;}, and nothing else, causes the bug [02:19] report it. [02:20] I guess this is bug 2635 [02:20] Malone bug #2635: When reporting a bug, the "Add" button moves around In: launchpad (upstream), Severity: Normal, Assigned to: Matthew Paul Thomas, Status: New https://launchpad.net/bugs/2635 [02:20] mpt: ^^ [02:24] Doesn't happen in a minimal HTML document, though :-/ === SteveA wonders about priority [02:28] for annoying issues that aren't really critical such as this, i recommend the following [02:28] bug priority, SteveA? [02:28] oh, this priority [02:28] recognize that it is something worth doing, but only somewhat [02:29] allocate N minutes to looking into it [02:29] say, 20, 30, 45, 60, whatever [02:29] and then stop [02:29] set an alarm. no, really. [02:29] work on it, until the alarm goes off [02:29] then totally stop. [02:29] ensure work to date is documented in the bug. [02:29] ok [02:30] i know i have spend hours and hours trying to tie down something wrong but really not all that important === jbailey [n=jbailey@modemcable139.249-203-24.mc.videotron.ca] has joined #launchpad [02:32] What's the right way to specify multiple revision control systems in launchpad for a product? [02:33] And why is bzr not one of the options? =) [02:33] jbailey: eh? [02:33] https://launchpad.net/products/cdbs/+series/trunk/+source?lpnotification=fcTDxydmS9PqnAI2UHACV4xh6XE [02:33] thats a product series [02:34] just add a new series for each thing you want imported [02:34] cprov, what's DistributionMirror.pulsesource and who's supposed to be allowed to change it? [02:34] stuff in bzr does not need to be imported, goto 'code' and 'add branch' [02:34] salgado: changes by the owner [02:34] lifeless: Imported to where? [02:34] salgado: same person who changes the pulsetype [02:34] jbailey: to bzr [02:34] lifeless: I'm just trying to indicate that the old trunk was in X place until revision X. The new one should be located at X. [02:34] jbailey: on bazaar.launchpad.net [02:35] cprov, and what a pulsesource is? [02:35] lifeless: When I click on "trunk" , I'm clicking on "Edit Source". Nothing there implies import to me, nor do I care to have it done. [02:35] Is says We have the following revision control system details on file for the branch:, and I'm trying to update the details. [02:36] salgado: the url "poll" pulse should use, IIRC [02:37] lifeless: I'm likely to start hacking on cdbs again soon, so I figured that perhaps I should take over the LP record and use it to track my development. [02:37] cprov, can you find out what it is for me? I need to add some text explaining what a user is supposed to enter in the pulsesource entry === jinty [n=jinty@205.134.224.215] has joined #launchpad [02:39] Ah, I suspect part of what I need is to have someone with priviledges go to https://launchpad.net/products/cdbs/+reassign for me. Is there a mailing list where I ought to request this? [02:40] salgado: """An URL to be used in "pulse pull" method.""" is enough for a while [02:40] jbailey: I can do it, I think. /me checks [02:41] bradb: Thanks! [02:41] jbailey: There, I've made you the owner. [02:42] bradb: Pretty. Thanks. [02:43] np [02:44] jbailey: FWIW, I don't know of any more "official" way to ask for someone to do that kind of admin work than just to ask in here, in future. [02:44] cprov, who should be allowed to register a new mirror? [02:44] salgado: public, logged in LP [02:45] cprov, right, and why the history of probes is not visible by the owner? [02:45] s/by/to [02:46] salgado: we may not publish the probe history if we don't want too [02:46] bradb: Cool, thanks. [02:52] Is it a bug that hitting add on "https://launchpad.net/products/cdbs/+addbranch" switches my context so that I'm in "Jeff Bailey" instead of cdbs? [02:52] The result URL is https://launchpad.net/people/jbailey/+branch/cdbs/cdbs-trunk [02:52] Which confused me when I wanted to go back and set more cdbs things. === bradb thinks ddaa and niemeyer might be able to help [02:54] ddaa: ^^ [02:56] jbailey: that's a known issue [02:57] ddaa: Thanks. [02:57] but no solution has been proposed as of yet [02:57] It's a feature! [02:57] ddaa: I think the only real solution is inter-page context tracking. [02:57] I really do not know how to fix it [02:57] So that there's history of where you came from. [02:57] I think the right person to give hell to is mpt... === ddaa checks whether there's a bug already [02:59] jbailey: maybe getting context from cookie would be a way to fix it. But it appears that launchpad is trying to avoid dependences on cookies. [03:00] I would understand that as it seriously impairs the reproductibility of the UI. [03:00] and tracking context using a blob in the URL is not really great either... [03:00] mh... [03:00] ddaa: Right. You could mangle the URI to include +context-packages:cdbs/ or something. [03:00] Or do that just in the absence of cookies, sort of like session handling is usually done. [03:01] yeah, I guess a cleartext sort of context tracking like that would be the least bad solution. [03:01] ddaa: Considering that our login requires cookies, it's a bit late to avoid dependence on them. :) [03:01] ddaa: cookies != request state [03:01] bradb: You don't fall back to URI mangling in the absence of cookies? [03:01] what lifeless says [03:01] ddaa: we have notifications & redirects [03:01] jbailey: Nope. [03:02] (redirect as the result of a form submit [03:02] In practice, I haven't heard of that being a problem (cookies for logins.) [03:02] And the Kinni said "Let there be fuses." and there was *light* and he saw that it was goooood [03:02] lifeless: the request is for something more involved than notifications... [03:03] lifeless: but for example, how does the breadcrumb functionality of moin works? [03:03] (the recent visited pages) [03:03] I guess it's storing the pages in a cookie [03:04] I suspect so, it's persistant between browser windows. [03:04] the issue with branch pages is that we do not want the same branch to have two different pages. [03:04] but since a branch can live under a product and under a person, there's a problem... [03:05] What problem are you trying to solve with the only-one-page? [03:05] So that visited links can be highlighted correctly? [03:05] adding steps in the url with e.g. +context-product:cdbs would break the "one page per branch" requirement [03:05] Right. But what does that requirement serve? [03:05] jbailey: not sure... that's a requirement from SteveA. Apparently doing otherwise confuses search engines. [03:06] ... [03:06] You're letting search engines index the site? =) [03:06] jbailey: we're google friendly :) [03:06] Frightening. =) [03:06] Okay, so that makes more sense then. [03:06] ITYM sexy [03:07] jbailey: The best search UI to Launchpad is Google, at the moment. :) [03:07] I guess that a url argument (after a "?") would be more search-engine friendly [03:07] but I'm no authority [03:07] ddaa: hello. i felt a disturbance in the force [03:07] lifeless: sexy if they actually bother. They often don't seem to actually index entire sites. [03:07] ddaa: Not really, sadly. [03:07] like 50000 bugs all got assigned high priority at once [03:07] SteveA: jbailey complains that going to a branch changes his context away from the product he was in. [03:08] right, it does that [03:08] ddaa: noo, ? bad [03:08] We're brainstorming trying to find a way to fix that. [03:08] i see [03:08] i think something brad prototyped might help here [03:08] SteveA, ddaa: To LPs credit. With the new UI it was fairly immediately obvious that it had happened. === ddaa got himself confused countless times with the old UI [03:08] It just didn't facilitate me going on and finishing configuring the product. [03:09] So I hit the back button 4 times and went on. [03:10] someone please submit a merge [03:10] SteveA: about that, more tell me, master jedi [03:10] lifeless, submitted [03:11] salgado: thanks [03:11] lifeless, can I submit another one? [03:11] so, brad's prototype was of a "here's where you've been" box [03:11] that tells you the contexts you've been in recently [03:12] salgado: knock yourself out [03:12] so? [03:12] imagine a box [03:12] it says "launchpad location history" or something at the top [03:12] Oh, yes, I imagine it alright. But I do not think that fixes this issue. [03:13] I think that basically a branch page should be able to product context _or_ person context. [03:13] well, it would mean that jbailey could have clicked the item in there [03:13] And that switching contexts should be explicit. [03:14] SteveA: Quicknav back would even be lovely yes. I think it's not the right solution long term, but it at least provides be a way to get back. [03:14] SteveA: The problem being that it can still be a bit hard to find a particular page in LP. [03:14] jbailey: actually, ATM you already have a link to theh product overview. [03:14] There's a link to the product in the detail portlet of the branch page. [03:15] having a branch in two places has its own problems [03:15] ddaa: It wasn't obvious from the page I was sitting at, but it could have blurred in. [03:15] i think this issue needs careful consideration. i can't help with that this week though [03:15] jbailey: it's basically invisible anyway. I tried to force myself to use it, but it was too much pain because I generally want to go back to the code facet of the product anyway. [03:15] as i'm having a malone week, with brad and bjorn in vilnius [03:16] well... I'm on leave until next year starting saturday... [03:16] so it's no emergency [03:16] jblack: still around? [03:16] SteveA: Sorry, I don't mean it to be a distraction. [03:17] SteveA: I had a thought of "It would be nice to write a spec for some of these changes to cdbs" and was trying to follow it through in LP. Got a bit lost on the way ;) [03:17] jbailey: i'm keen on getting UI feedback. it is great that you're using launchpad. i'm just sorry i don't have time to really do anything to help just now. [03:17] jbailey: you're welcome to file a bug though [03:17] assign it to me, for now [03:18] Now that I'm listed as the registrant of cdbs, how do I get it listed under "Packages" under my userid? [03:18] ddaa: Okay. I'm no longer sure that I'm looking at a single bug here. I'll try and detangle them into actual separate bugs. === mgalvin [n=mgalvin@ubuntu/member/mgalvin] has joined #launchpad [03:18] ddaa: Or I could flie a support request and leave it to you guys to detangle? [03:18] A bug would be better I think. We are already aware there is _some_ issue. [03:19] ddaa: Great, I'll try to sort this out. [03:19] It would be nice to have a user describe the issue in his own words. [03:19] Ooo, clever: https://launchpad.net/distros/ubuntu/dapper/+source/cdbs/+changelog [03:19] I wonder how I get it to populate this. [03:20] stub: hello. did you write a wiki page with top 10 bzr issues for launchpad, or something? [03:20] SteveA: Before UBZ I threw together a list of issues that I felt needed to be addressed [03:20] i want to refer to it [03:21] when mailing jblack and the list [03:21] about that issue from earlier [03:22] SteveA: https://wiki.launchpad.canonical.com/ThingsBzrNeeds [03:22] ta [03:22] SteveA: Probably out of date [03:36] stub: https://wiki.launchpad.canonical.com/BzrPrioritiesForLaunchpad [03:36] also mailed the list [03:46] SteveA: That list is quite comprehensive. Are you sure it can be accomplished in a reasonable period of time? =) [03:48] i have the utmost confidence in the bzr team's abilities === koke [n=koke@ubuntu/member/koke] has left #launchpad ["Konversation] [04:08] bradb, when is BugTaskSet's self.title ever used? === Nafallo is now known as Nafallo_away [04:10] or BugWatchSet's self.title? === matsubara is now known as matsubara-lunch [04:14] ThingsBzrNeeds can use some updates [04:16] mpt: Not intentionally. === raphink [n=raphink@gra86-1-82-239-88-236.fbx.proxad.net] has joined #launchpad === cprov [n=cprov@200-171-140-32.dsl.telesp.net.br] has joined #launchpad [04:35] daf, hi [04:36] daf, do you have a moment? [04:38] stub: ping [04:40] stub: when you see this, ssh to newpqm and sudo to pqm [04:40] stub: then strace pid 15558 [04:40] stub: its a test suite run which appears to be hung, looping on librariangc connection of some sort. [04:41] spooky [04:49] stat("/home/pqm/.pgpass", 0x7fffffaf99a0) = -1 ENOENT (No such file or directory) [04:49] socket(PF_FILE, SOCK_STREAM, 0) = 41 [04:49] fcntl(41, F_SETFL, O_RDONLY|O_NONBLOCK) = 0 [04:49] fcntl(41, F_SETFD, FD_CLOEXEC) = 0 [04:49] connect(41, {sa_family=AF_FILE, path="/var/run/postgresql/.s.PGSQL.5432"}, 110) = 0 [04:49] getsockopt(41, SOL_SOCKET, SO_ERROR, [0] , [4] ) = 0 [04:49] getsockname(41, {sa_family=AF_FILE, path=@}, [2] ) = 0 [04:49] poll([{fd=41, events=POLLOUT|POLLERR, revents=POLLOUT}] , 1, -1) = 1 [04:49] rt_sigprocmask(SIG_BLOCK, [PIPE] , [] , 8) = 0 [04:49] sendto(41, "\0\0\0003\0\3\0\0user\0librariangc\0databas"..., 51, 0, NULL, 0) = 51 [04:49] rt_sigprocmask(SIG_SETMASK, [] , NULL, 8) = 0 [04:49] poll([{fd=41, events=POLLIN|POLLERR, revents=POLLIN}] , 1, -1) = 1 [04:49] recvfrom(41, "R\0\0\0\10\0\0\0\0E\0\0\0`SFATAL\0C53300\0Mcon"..., 16384, 0, NULL, NULL) = 106 [04:49] close(41) = 0 [04:49] futex(0x13336f0, FUTEX_WAKE, 1) = 0 [04:49] select(0, NULL, NULL, NULL, {1, 0} [04:49] etc [04:49] night === heyko [n=heyko@tor/session/x-6682f73b27474eb4] has joined #launchpad [04:52] google++ # really good at searching bugs filed in Malone [04:54] bradb, better than our own search form? ;-) [04:54] lifeless, night [04:54] carlos: they have more programmers working on improving our search :-) === poningru [n=poningru@n128-227-82-53.xlate.ufl.edu] has joined #launchpad [04:55] SteveA, we could just integrate google searches into launchpad :-P [04:58] That wouldn't give us operators like reporter:carlos [04:59] but I'm speccing all the good stuff that Google does do in MaloneSearch [05:00] mpt, well, I was joking about requesting the search to google's servers [05:00] mpt, they have a way to "integrate" their search engine into your web application [05:03] Yes, but we wouldn't have the source code === matsubara-lunch is now known as matsubara [05:05] mpt, as I said, it was just a joke === mpt shrugs [05:08] Amazon.com's and IMDb's search is apparently powered by A9 [05:08] and there are plenty of high-profile sites that do use Google for their searching [05:09] stub: around? [05:13] mpt: You get a small discount on the price of books you find and purchase through A9 searches. === bradb wonders if Malone users could be similarly bribed [05:14] only if we put advertising next to the search results [05:14] bradb: We ship them free CDs if they file bugs? =) [05:14] which we could, since that space is largely empty at the moment [05:14] jbailey: Good point. [05:14] monetize the portlets! [05:15] mpt: 10,000 unique page views would translate to about, what, 18 portlet views? :) [05:15] bradb, I prefer the pornlet feature we talked at .ca.... [05:15] Some people seemed to be curious about if it was serious. [05:16] the pornlets idea? [05:16] :-D [05:18] yeah :) === thierry [n=thierry@modemcable031.72-131-66.mc.videotron.ca] has joined #launchpad [05:20] mpt, I have a label that is a link and that should be set 'bold', should I put Foo tags or Foo ? [05:21] bradb, Jordi told me that kiko implemented it... [05:21] but I didn't see it [05:21] carlos, why should it be bold? [05:22] In general, anywhere we use is a bug [05:22] mpt, it's at pofile-portlet-translators.pt [05:22] is missing the link [05:22] and is set as bold [05:22] I'm just adding the link [05:23] I don't understand why that's bold [05:23] is it because it's the translation group, and therefore more special than the other list items? [05:24] but then why does it end in ":"? [05:24] mpt, https://launchpad.net/distros/ubuntu/breezy/+source/pmount/+pots/pmount/es/+translate [05:24] mpt, loot at the right column [05:24] where you have the list of translators members of the Spanish team [05:25] it's the title of the team [05:25] hoo boy, that page is broken in the new layout [05:25] mpt, there could be more than one team === lamont [n=lamont@mib.fc.hp.com] has joined #launchpad [05:26] mpt, dude I told you that already! [05:26] I know, I just didn't see it :-) [05:26] that's mainly because of the
 bug
[05:26]  mpt, anyway, I'm changing it a lot so I will try to fix it at the same time
[05:27]  So why does that portlet have two headings?
=== rbelem [n=rodrigo@ubuntu/member/rbelem]  has joined #launchpad
[05:27]  Why not just an "Ubuntu Spanish translators:" heading, carlos?
[05:28]  oh, because it's a link
[05:28]  phone....
[05:30]  That's invalid HTML anyway, carlos
[05:30]  
      ...
is not allowed [05:30] only
  • ...
      ...
    ...
[05:34] ok, I will fix it at the same time [05:34] why isn't a nested ul allowed? [05:34] oh i see [05:35] wow, that's so cool. I called on Tuesday to IBM because my external CD unit was broken and today I got a new one... [05:35] you need to put it in an item [05:35] only two days!!! [05:36] carlos, as for or , it doesn't really matter which [05:36] mpt, ok [05:36] but I'm bothered that I don't understand what that portlet is trying to say === lbm [n=lbm@x1-6-00-13-10-7a-d1-e4.k233.webspeed.dk] has joined #launchpad [05:36] mpt: idea... [05:37] mpt: some css included in the debug server [05:37] mpt: that causes and other bad things to be brightly highlighted === mpt wonders if that could be done with bad text, as well as bad markup :-) [05:37] mpt, that portlet list the teams on charge of that pofile translation [05:38] there could be more than one team [05:38] and then, it lists all members of the team [05:38] carlos, so there is one (and only ever one) translation group [05:38] but the translation group might contain more than one team [05:38] correct? [05:38] perhaps we could take the same UI we have now for the Overview/bugs/translations... portlet... [05:39] mpt, no [05:39] SteveA, great idea, but what do you mean by the "debug server"? === Wellark [i=wellark@ranssi.paivola.net] has joined #launchpad [05:39] mpt, usually there will be only translation group but you would have more than one [05:39] oh [05:39] s/only translation/only one translation/ [05:40] mpt: on localhost, one port gives you what is seen in production. one port gives you what developers use. [05:40] mpt, and every translation group will list a set of persons (not teams) [05:40] there are particular views / variants registered just for the DebugLayer [05:40] and these are not seen in product [05:40] ion [05:40] only when developing [05:40] carlos, so a template has at least one translation group, and each translation group might have one or more teams, and each team has at least one member [05:40] mpt, the context here is a POFile not a template [05:41] but yes, your afirmation is right [05:41] ok, a PO file [05:41] in this case, as it's a pofile [05:41] we only get one team for the translation group [05:41] the one for the language we are displaying [05:42] carlos, in that case, I suggest making each translation group an
[05:42] then a two-level list after each
[05:42] listing people, and/or teams with members, in that translation group [05:42] ok [05:43] mpt, thank you for your input [05:43] hmm, wait [05:43] stub: ping [05:43] the
will be a link [05:43] mpt, I suppose it's not a problem, right? [05:43] that's ok [05:43] ok === carlos workraves [05:44] thanks [05:44] but in that case, the goes *inside* the
:-) [05:46] bradb: pong [05:49] carlos, 100 validation errors on that page -- about 20 of them are because the "Used elsewhere" suggestion isn't escaped properly [05:57] mpt, I think we should execute an html validation command as part of our tests... === Seveas [n=seveas@ubuntu/member/seveas] has joined #launchpad === sabdfl [n=mark@ubuntu/member/pdpc.silver.sabdfl] has joined #launchpad [06:03] carlos, yeah, the W3C validator can't see that page because it can't log in [06:03] so it complains about HTML errors on the login page instead :-) [06:03] mpt: what are our html errors? [06:03] mpt, well, you can always store the html and upload it to the validator [06:03] are there errors in the main template? [06:04] carlos, that's what I did to get the 100 figure [06:04] SteveA, there weren't in the main template a couple of months ago, I'll look again now [06:05] It wasn't very easy to try because the validator didn't support HTTPS back then, but now it does [06:05] main template's fine [06:05] that's good [06:05] front page has 1 error [06:06]