[00:02] nickserv has problems [00:02] thats for sure [00:02] ha [00:02] i am now +u again ;) [00:02] nickserv just had a _huge_ delay [00:02] fta: better? [00:03] yep n=asac@debian/developer/asac on both nick [00:04] thx [00:07] hmmm ... why is mysql-common pulled in by build-essential? [00:07] yes, except that xchat doesn't refresh === asac_ is now known as asac [00:09] Will xpi.mk look only in chrome/ for locale, skin and content dirs, or will it parse them from chrome.manifest? the first? [00:10] oh, it will follow the MOZ_XPI_BUILD_COMMAND [00:10] Jazzva: good question ;) [00:10] yeah [00:11] not sure if i added a "generic" command [00:11] but i dont think so [00:11] can't find one... Anyway, can i call build.xml somehow? [00:11] Jazzva: i think there is certainly a need for a unified script that unpacks and xpi to our standard format [00:11] and a script for the other direction [00:11] both could go into mozilla-devscripts.mk [00:12] and we could use one by default for MOZ_XPI_BUILD_COMMAND [00:12] if we dont have a .xpi in top level folder [00:12] Jazzva: yes you can run ant [00:12] :) [00:12] Aren't we going for packaging with unpacked xpi's? :) [00:12] for build.xml style sources the BUILD_COMMAND is just "ant" :) [00:12] Yay. I can clean my rules file :) [00:13] Jazzva: obviously as we want things to be in branches [00:13] Jazzva: you need to build depend on ant of course [00:13] Just asking, cause you mentioned "unpacks and xpi to our ..." [00:13] Jazzva: dont understand your question then ;) [00:14] Jazzva: ah sorry [00:14] didnt read "aren't" :) [00:14] yes we are ;) [00:14] good :) [00:15] Ok then... just to test-build it and test-run it and it will be pushed. I think it might be good for tb and songbird too... it's sort of mentioned in chrome.manifest.packaging... [00:15] * Jazzva googles about c.m.p file [00:16] Jazzva: hmm [00:16] Wow... 4-5 results for that file :) [00:16] Jazzva: are tbird and sonbird in install.rdf too? [00:16] And seems foxyproxy is the only one that uses it [00:16] yes [00:17] have to check if this id is for tb, though [00:18] I think it's flock [00:18] Jazzva: flock is: {a463f10c-3994-11da-9945-000d60ca027b} [00:18] that's the one [00:18] (install.rdf open in another window ;)) [00:18] ok [00:18] Jazzva: try if it works in midbrowser ;) [00:18] add that one in the "ubuntu" branch ;) [00:18] if it works [00:19] last time i tried I couldn't run midbrowser... [00:19] Jazzva: hmm compiz? [00:19] Jazzva: disable desktop effects. then it should start. next upload will fix that [00:19] dunno if I used compiz then... Anyway, since I stopped using it, I'll try it again :) [00:19] {aa5ca914-c309-495d-91cf-3141bbb04115} [00:20] thats the id [00:20] It doesn't show up in install.rdf. What are the min and maxVersion, so I can add them? [00:20] /usr/lib/midbrowser/application.ini [00:20] has the current version [00:21] k [00:21] 0.3.0rc1 [00:21] i think its ok to use 0.3.0 as maxVersion [00:21] hm, my clamav-milter on hardy dies a couple of times per hour :( [00:21] I need to install it first in intrepid chroot, so it'll take a while... ~20 minutes [00:22] Jazzva: midbrowser is small [00:22] should take a second ;) [00:22] But it still has deps ;) [00:22] if you have xul [00:22] ok [00:22] fta: i think thats a topic best dealt with on #ubuntu-server :) [00:23] i know, i'm just complaining ;) [00:23] hehe [00:23] i am using clamav on etch ;) [00:23] that is rock solid. but i have no clue if it ever found a virus :) [00:23] asac: you know if exist a way to be noticed when a new appversion is added here? ( https://addons.mozilla.org/en-US/firefox/pages/appversions ) [00:23] it started to die yesterday for me [00:24] fta: hardware? or was there a sec update or something? [00:24] Volans: why do you want that? [00:25] on Mozilla Addon site you can upload only extension with min e max version declared there as I know [00:25] Volans: you should use 3.0.* [00:25] as maxVersion [00:25] thats what we use [00:26] Volans: for trunk i think that AMO site bumps your maxVersion automatically [00:26] at least reed said so iirc [00:26] that does not happen to my extension... and the new version is still waiting an editorial review to be publicated [00:27] for 3.1a1pre ? [00:28] 1.0.6.1 version had 3.0pre, 1.0.6.2 have 3.0.* [00:28] ok. yeah [00:28] thats good then [00:28] not sure about the AMO procedure. sorry [00:29] no problem [00:34] midbrowser works :) [00:37] now to get another 30MB of ant and it's deps... [00:49] yay, java [00:50] assembler and linker projects were required to be coded in java... don't ask me why... i has no idea [00:52] * Jazzva spends a lot of time on icanhascheezburger, apparently [00:52] *i have no idea :) [00:54] Jazzva: well. imo you could also teach students to write java op-code [00:54] instead of assembler [00:54] is quite similar and is at least usable on all platforms that have a vm [00:54] :) [00:55] true... I think that's a project that I'm gonna do next year [00:55] i doubt that anyone will become a real assembler wizard during class anyway [00:55] to make java op-code :) [00:55] its more about "getting the idea" :) [00:56] I agree. that's what we were doing. The instruction op-codes were different from the real one, only 7 instructions used. Just to get the basic concept :) [00:56] Jazzva: yeah ;) [00:56] then go and use a java VM ... its a simple machine so quite a good thing to learn on [00:56] maybe makes sense then ;) [00:57] well, I'll study it next year, for sure :)... [00:58] Jazzva: in which direction do you want to go? [00:59] if you take too classes on low level programming ;) [00:59] two [00:59] It's not my decision. Everyone has them on this department... [00:59] which department? [00:59] But I found assembling and linking quite interesting [01:00] it would be something like dept of computer sciences [01:00] ok so not a sub-department of computer science ;) [01:00] nope [01:01] Jazzva: you know that we have assembler code in mozilla :) [01:01] Damn, wanted to show you my courses list, but it's available only in Serbian [01:01] you probably wanna help fixing mips :) [01:01] Hehe... dunno if I'm that good :). [01:01] ...dunno if I'm good at all :) [01:02] http://www.etf.bg.ac.yu/index.php?option=com_content&task=view&id=70&Itemid=83 ... maybe you can manage to understand something. [01:02] http://lxr.mozilla.org/seamonkey/source/xpcom/reflect/xptcall/src/md/unix/xptcinvoke_asm_mips.s [01:04] http://lxr.mozilla.org/seamonkey/source/xpcom/reflect/xptcall/src/md/unix/xptcstubs_asm_shle.s [01:04] wow... had no idea :) [01:04] http://paste.ubuntu.com/22738/ [01:04] those are all .s files [01:04] there is also some nested assembler in some .cpp files iirc [01:06] most likely more assembler than in the kernel ;) [01:06] hehe [01:07] I can barely understand code... but have no idea what it does. Is this _asm_shle.s for "go to label, then do these instructions, and call some other function"? [01:08] http://paste.ubuntu.com/22739/ [01:08] :) [01:08] thats the kernel .S files ;) [01:09] Jazzva: a good thing is to look at how syscalls are implemented on various archs [01:09] e.g. how to get messages from user-space to kernel :) [01:09] I'll put it on my todo list:) [01:10] its not that hard to understand [01:10] some platforms even have a C implementation [01:10] others need assembler [01:10] ;) [01:10] linux has c implementation, right? [01:13] it has :) [01:14] Jazzva: ? [01:14] Jazzva: linux support various architectures. i think x86 has pure C implementation for syscall [01:15] others dont have [01:15] I suppose I found that one :) [01:15] Jazzva: good :) [01:16] Jazzva: you also have to look at the libc part i guess [01:16] to get the complete picture [01:17] ok :) [01:17] thanks [01:22] foxyproxy puts it's xpi in ../targets/. Is it ok if I patch build.xml to put it in .? [01:24] looks like i can't... It removes the contents of the that dir before build :) [01:26] Jazzva: yes. if that patch isnt too intrusive [01:26] e.g. if its just a few lines that ok [01:26] One line, though it erases the whole contents, so it's pointles... [01:26] if done on ubuntu branch - but i guess thats what you are talking about ;) [01:26] right :) [01:34] thats painful [01:34] git-core merge looked good. now it fails two test cases [01:34] what a shame [01:35] hmm git-svn [01:35] perl: /build/buildd/subversion-1.5.0dfsg1/subversion/libsvn_ra/ra_loader.c:674: svn_ra_get_dir: Assertion `*path != '/'' failed. [01:52] good night and good work... see you [01:59] foxyproxy is recognized in midbrowser [02:00] but it reports an error that it can't read settings in .mozilla/midbrowser/profile/foxyproxy.xml and that it should be reported to the developers [02:00] i'll check if it exists in fx3 too [02:02] it works. it checks if it's a new installation in fx, if the settings file is not found. but it doesn't do that in midbrowser. [02:03] is "profile" a placeholder in the path you posted? [02:03] yes [02:03] otherwise the dir doesnt exist [02:03] k [02:04] :) [02:04] Jazzva: who dod the ytest i fits a new installation? [02:04] damn ;) [02:05] huh? :) [02:05] how do they test if its a new installation? [02:05] it reports this in case of fx: "FoxyProxy: Unable to read preference extensions.foxyproxy.settings in getSettingsURI(). Checking for new installation." [02:05] I'm looking at the code now [02:06] Jazzva: look in chrome.manifest [02:06] you can configure appid=.... there [02:06] to do special things for special apps ;) [02:06] maybe they do that [02:06] it writes it out here components/foxyproxy.js:341 [02:41] asac, they do set different stuff for different apps in chrome.manifest. I tried just to add another set of settings for midbrowser, but it's not working. I'll just prepare the merge now, without changing this, and then we'll see in the upload after merge :) [03:38] The QA tracker is smarter than I am. What do I click on to find the actual test you want done? [03:47] ScottK, for the first two tests (normal use with new and existing profile) there are no special tests. Just do your normal browsing and report if any glitch occurs. [03:48] ScottK, for the third test, there's a link to , which provides the testplan [03:57] What are the links for the first two? [03:59] ScottK, there are no links to testplan for the first two. The links for reporting section depends on which package are you testing [04:00] The testplan could be "Go around the usual sites you visit everyday for 15 minutes, and see if something bad and different happens." [04:02] Ah. [04:02] OK. [04:02] I can do that. [04:02] Thanks for doing that :) [05:19] FYI, got another flash related crash in dapper. (firefox-bin:28048): GLib-CRITICAL **: g_quark_from_string: assertion `string != NULL' failed [05:19] (firefox-bin:28048): GLib-GObject-CRITICAL **: g_object_notify_queue_thaw: assertion `nqueue->freeze_count > 0' failed. [05:24] This was with the Adobe flash plugin that is installed by the Plugin Finder Server. Again was not reproduced (so again probably not worth worrying about?) [05:40] I don't quite get why Ubuntu keeps the Firefox name while Debian doesn't... is Ubuntu getting special approval from Mozilla for the Firefox modifications it makes? [05:48] hello [09:45] asac: are you here yet? [10:02] * gnomefreak think he bumped his head way too damn hard to want to add 20 more packages to our maintaining list :( [10:06] gnomefreak: sure [10:06] gnomefreak: yeah. thats shirish idea [10:08] what order for nss nspr i cant remember im thinking nspr first than nss [10:10] gnomefreak: you are right! [10:10] yay :) [10:10] * asac hugs gnomefreak [10:10] ;) [10:10] respin firefox as well (atleast should need to) [10:10] * gnomefreak hugs asac [10:11] gnomefreak: if its about the nss upgrade bug, I just want nss built on gutsy [10:11] to reproduce [10:11] no need for ffox [10:12] ok [10:12] ill have them uploaded in next couple of hours i just want to get updates and shit done first [10:13] thx [10:13] np [10:14] fta: firefox-3.0 from your ppa is being kept back (i would assume depends issue) xul went through so i will check after that [10:15] asac: what are we doing with bugs on extensions that we dont maintain? [10:15] gnomefreak: which extension? [10:15] IMHO we should close them or add them as wishlist if user wants them from us [10:15] s3fox [10:15] see bug 242741 [10:16] hmmmmm [10:16] botty [10:16] im asking about him now [10:17] gnomefreak: that bug is "invalid". he should report it to the extension author (unless we have that extension packaged or that extensions is _really_ important) [10:17] ok i will mark and comment as such [10:17] great [10:17] bots ISP is having issues trying to get an alt. in here [10:18] we wont have bot for a while, it should be worked out by end of day just not sure whos day [10:18] ha [10:19] well. i think we can live without a bot for a day ;) [10:19] can we? [10:19] Hey [10:19] yep should beable to [10:19] hi anakron_ [10:19] hi [10:19] someone is of mozillateam?? in LP? [10:19] hi [10:19] :) [10:19] both of us [10:19] ok [10:19] + others as well :) [10:19] i need to know something [10:20] feel free to ask [10:20] cause i wana help like a beta testers for mozilla [10:20] but im using ubuntu intrepid [10:20] anakron_: then you are already a big-beta tester ;) [10:20] anakron_: do you want _more_ recent crack? [10:21] fta: here is the error The following packages have unmet dependencies: firefox-3.0: Depends: xulrunner-1.9 (>= 1.9+nobinonly-0ubuntu3) but 1.9+nobinonly-0ubuntu3~fta3 is to be installed [10:21] gnomefreak: darn. my by [10:21] my bad [10:21] asac: oh thats you? [10:21] :) [10:21] lol [10:21] im looking at http://mozilla.qa.ubuntu.com/ [10:21] for any version for intrepid [10:21] but i cant find one at least! [10:22] * gnomefreak not real sure if we have intrepid packages up there for testing [10:22] anakron_: ah. well. firefox 2 isnt supported in intrepid anymore [10:22] anakron_: have you received the mail? [10:22] :O yeah [10:22] anakron_: there are other options to help out: you can just use virtualbox to install one of the stable releases [10:22] so, what i can do to [10:22] asac: maybe add devel releases to testing in late alpha so our changes are in and we work on them for bugs at that time [10:22] jaja [10:23] i cant use it cause my hardware will get burn xD [10:23] anakron_: chroots :) [10:23] i have a Pentium IV 1600MHZ with 320mb RAM [10:23] so, its like a dream to run a virtualmachine [10:24] i know that i can use chroot, but my hardware is a limit for my dreams [10:24] anakron_: yeah [10:25] crap. someone pplease let me know when Jazzva starts on the extension pages :( [10:25] anakron_: you are brave enough testing intrepid already. so if your hardware doesnt allow stable testing, thats fine [10:26] asac: oh shit i didnt give you important bug you needed to look at :( be right back let me find it [10:26] thanks for the answer [10:26] see you later [10:26] anakron_: cu [10:26] better [10:26] see you soo [10:26] asac: https://bugs.edge.launchpad.net/ubuntu/+source/firefox-3.0/+bug/196202 [10:26] soon [10:27] IMHO its the way it should work :( maybe im confused but sounds like he wants natulis to open ftp only [10:28] gnomefreak: the issue here is that the ftp server was added to gconf [10:28] errr, the ftp handler [10:29] and thats ff issue? [10:29] im gonna assume whom ever did it meant it to happen that way hense not a bug [10:29] gnomefreak: not sure :). but i assume so [10:30] gnomefreak: ask them to figure out when and how the ftp handler ended up in gconf [10:30] next who did it is gonna be hard to find out [10:30] they most likely did it once manually [10:30] ill ask [10:32] i asked [10:33] in bug report atleat [10:33] atleast [10:35] but please ask in -devel if you know who to ask (might want to start with seb.) [10:36] gnomefreak: i think it he cannot reproduce this bug, we can just close it [10:37] the ftp bug? [10:37] y [10:37] reproduce as in "reset the gconf key", then does firefox still take over ftp handling ;) [10:38] but removing the keys doesnt help since everyone would have to [10:38] thats why i added it as wishlist to maybe make user decide what he wants as default [10:38] since i only see 2 or so people in the bug complaining (maybe one) [10:39] 3 people [10:39] out of how many millions [10:39] gnomefreak: well. i doubt that many people see that issue [10:40] its firefox-3.0b4 that is set in gconf [10:40] when i was using ftp nautlis did it for me that was gutsy [10:40] i am pretty sure that it was the user who did that [10:40] its possible i doubt mozilla did it [10:41] gnomefreak: well. let me check ;) [10:41] actually i think that there was a bug asking for that feature ;) [10:41] lol [10:41] e.g. make firefox take over ftp handling [10:41] but let me check in code [10:41] sounds like a few other flip flop bugs [10:41] like offline mode and saving docs to temp in tbird :( [10:44] gnomefreak: ok. it takes over ftp :-D [10:45] due to what? [10:45] ok let me try that again [10:45] due to who? [10:45] gnomefreak: if you allow firefox to check whether its the default application it will take over [10:45] http + https + ftp [10:46] and chrome [10:46] yeah thats why i said add update-alt for ftp it makes it alot easier [10:47] gnomefreak: well. [10:47] ill think about it [10:47] but as default browser it should open them unless you are connecting to your own ftp [10:47] will let you know in 30min :-D [10:47] do you want nss nspr in mt ppa? [10:47] hi [10:47] gnomefreak: nope. in your ;) [10:47] fta: hi sorry but asac said he did it [10:47] hi fta [10:47] asac: ok [10:48] did what? [10:48] ah the dependency problem [10:48] yeah [10:48] now firefox 3 head branch is prepared for preview xul builds [10:50] ok smoke than fix depends for nspr [10:51] i tweak build-deps but not deps, so if you updated deps, that's why it's stuck [10:51] fta: its fixed in .head now [10:51] didnt know that you tweak buld-deps [10:52] we just have to remember that if we need versioned depends on xul package, that we should append ~ to allow preview builds [10:54] asac: how did you screw up deps on fta's package in his ppa? [10:54] ah nevermind [10:54] i just read it [10:54] gnomefreak: he builds from .head branch :-D [10:54] head i thought was branded 3.1 not 3.0 [10:55] if head is devel [10:55] please dont ask me to dig that up it was in mozilla-dev-app mailing list IIRC [10:56] what issues will i run into if i use the -d flag to build nspr/nss? [10:56] there are 2 build-deps that hard higher versioning [10:56] maybe just install them since i wont affect running them [10:57] what are you talking about? [10:58] fta: hardy uses higher versions of 2 build-dep packages than gutsy does [10:59] -d flag should let you build without them but i dont think thats best idea [10:59] debhelper (>= 5.0.61) dpkg-dev (>= 1.14.9) [10:59] forgot that [10:59] fta: about ffox 3 not being installable in your ppa? [11:00] asac, this i understood. and it's fixed here too. [11:00] good [11:00] in which way? you auto patch Depends: now too? [11:02] i just resynced. but i can auto update deps too. it's just that it was tricky before (when we had >= x, << x+1) [11:02] fta: no need to. i should just remember that you append a tilde version [11:02] usually we dont have exact package versions in the build-depends [11:02] so its a special case [11:02] err/build-// [11:02] s/build-// [11:03] i started to auto-update build-deps in order for the ppa to build the packages in the right order, no other reason [11:04] fta: right, build-deps are great, but i dont think you should bother for depends as its really the exception [11:07] dch -i doesnt know what timezone im in :( [11:07] gnomefreak: most likely you are in a chroot then [11:07] oh maybe i was when i ran it [11:07] thanks ;) [11:07] np [11:08] jtv: there? [11:08] what package in dpkg-buildpackage in? [11:08] my uploads to midbrowser still dont get auto approved :( [11:08] asac: here. [11:09] asac: I'll take a look, just a moment... [11:09] jtv: i tested a partly translation upload for the grabanddrag translation in midbrowser [11:09] thats why there is just it-IT.xpi [11:09] (grabanddrag is not in firefox, so we dont get the strings from there) [11:09] jtv: you think it might work? :) [11:09] e.g. just partial .xpi? [11:09] asac: it-IT.xpi won't be auto-approved, but it.xpi will. [11:10] jtv: hehe [11:10] ok [11:10] let me reupload [11:10] jtv: so you rename it? [11:10] yes, I can. [11:10] This next release may also help us fix the problem. [11:11] asac: not sure what you meant about partial translations. [11:11] jtv: no i mean, do you usuallyy just rename the .xpi? [11:11] or do i need to change all the content to use "it" instead of it-IT (i think firefox translations use it-IT too) [11:11] Well, I don't actually need to rename it. Just set the language by hand. [11:11] Ahhh [11:11] jtv: well ... part of the strings are translated in the firefox translation pack we get [11:11] but we have more strings ... i crafted special .xpi's now that have those strings :) [11:11] That's an interesting problem, actually: the auto-approval and what's in the file are completely separate. [11:12] So for auto-approval, just rename the file and that's that. [11:12] Ping [11:12] ping? [11:12] (That was the sound of a thought) [11:12] ah ;) [11:12] We currently just ignore the language codes in the manifest. [11:12] jtv: so you want me to give you the .xpi so you can look inside? [11:12] jtv: ok great. you should filter them at least [11:12] No need. Just name the thing it.xpi in the future and it'll be auto-approved as Italian. [11:13] i could then just upload all-in-one .xpi's multiple times [11:13] Right now, that won't work yet. [11:13] sure [11:13] but would be great [11:13] :) [11:13] ok ... let me check what other locales i have [11:13] Yes, but it doesn't mesh well with the whole structure of the code for now. [11:14] jtv: http://paste.ubuntu.com/22820/ [11:14] which do i need to rename? [11:14] i guess everything except zh-XX ? [11:15] asac: getting hardys version to build on gutsy is gonna be a bitch due to build deps but im going for smoke right now maybe ill look at it more later [11:15] asac: remove the country codes from all but zh-XX and en-XX. Then, for zh-XX, replace the hyphen by an underscore. The hyphen in en-US is hard-coded, so leave that as it is. [11:16] yeah. [11:16] let me try that [11:16] i wont upload en-US [11:17] jtv: err, how about pt-BR? [11:17] asac: oh, and pt-BR becomes pt_BR, just like cn-TW becomes cn_TW [11:17] :-P [11:17] asac: cs.xpi fi.xpi ja.xpi nl.xpi ru.xpi zh_CN.xpi da.xpi en-US.xpi fr.xpi ko.xpi pl.xpi zh_TW.xpi de.xpi es.xpi it.xpi pt_BR.xpi tr.xpi [11:17] jtv: ok. ill try that [11:18] jtv: that that i have you here .... [11:18] ? [11:18] jtv: 1. when will the "product" export be of the same structure as the distro export [11:18] 2. can we get the en-US.xpi in the export for "product" too? [11:19] its always a game to hunt down the en-US.xpi that was uploaded :) [11:19] if 1. is not going to happen, let me know. i have to think how to properly deal with that then [11:20] jtv: also, i think we have a problem that might make all this to die. all firefox translations are tri-licensed and we probably cannot relicense them to BSD :( [11:21] GPL/LGPL/MPL [11:21] asac: the structure of the product exports are determined by the paths associated with the individual files. Unfortunately, when you upload a single file, we don't have much of a path. [11:22] As for 2, isn't the en-US.xpi included when you select an export that includes the template? [11:22] jtv: how? [11:22] https://translations.edge.launchpad.net/midbrowser/trunk/+pots/midbrowser/+export [11:22] if i select "Everything" then its not [11:23] asac: that's a bug. [11:25] good :) [11:26] asac: http://pastebin.mozilla.org/468539 wtf am i missing? i have changed build-deps on build-essential and dpkg-dev versions from hardys version to gutsy version and got this as well as geting rid of the >versins alltogether [11:27] gnomefreak: i'd say something is broken in your chroot [11:27] gnomefreak: why depends on build-essential? [11:27] and dpkg-dev [11:28] those are implicit build-depends [11:28] you dont need to change dependencies on them [11:28] build-essential [11:28] |Depends: libc6-dev [11:28] Depends: [11:28] libc6-dev [11:28] Depends: gcc [11:28] Depends: g++ [11:28] Depends: make [11:28] Depends: dpkg-dev [11:28] thats b-e [11:28] jtv: darn. so we need install.rdf? [11:28] in translations? can it just be empty? [11:29] http://pastebin.mozilla.org/468540 has dpkg-dev deps [11:29] asac: that may work, yes... why? [11:29] jtv: because the imports failed [11:29] jtv: i just crafted special .xpi's [11:29] with chrome.manifest + .jar [11:29] (same structure as in en-US.xpi ... just a partial tree) [11:30] asac: all we really take from install.rdf at the moment is the contributors list, I think—and even those we end up ignoring unless we can parse them as email addresses. [11:31] asac: here is the broken package(s) and i cant really uninstall them since they are needed to build the package [11:31] asac: http://pastebin.mozilla.org/468541 [11:31] jtv: ok lets see. uploaded with empty install.rdf [11:32] asac: I guess we could provide a default install.rdf if there is none on input. [11:35] asac: about the licensing... as long as we mark the upstream translations as coming from upstream, they remain under their upstream license. Modifying and extending them may be a problem though. [11:35] hi, I'm getting problems with dates returned form javascript in FF 3 [11:35] all dates returned from javascript hava a 24+ hours offset [11:35] asac: ill rebuild the chroot to see [11:37] be back reboot [11:38] hk47xx: is that ubuntu specific? [11:39] jtv: hmm [11:39] jtv: i think its ok to inject BSD licensed strings into the DB and use them to update hte upstream translations [11:39] the exported, aggregated translation pack will be upstream license again [11:39] jtv: makes sense? [11:40] asac: yes i'm dual booting windows and this problem didn't show up [11:41] e.g. I translate something to improve upstream translation. the string itself i BSD, but when exporting (as .xpi for instance) it will be included in a combined work, which basically relicenses the string in that export [11:41] hk47xx: interesting [11:41] hk47xx: if you type "date" in a X-terminal [11:41] is that ok? [11:42] asac: yes it shows: mié jun 25 06:07:12 VET 2008 [11:42] hk47xx: how do you get the date in javascript? [11:42] asac: yes, makes sense. Only problem is, there's no way to tell the system that translations made in the UI are "upstream." [11:43] jtv: ah ;) ... yeah. a check box during upload would help i guess [11:43] asac: so if you import the whole thing... yes, I think that could work. [11:43] if I type in the address bar for example javascript:alert(new Date()) [11:43] asac: i get Thu Jun 26 2008 06:08:05 GMT+1930 (VET) [11:44] hk47xx: for me its accurate [11:44] hk47xx: what is VET? [11:44] im from venezuela [11:44] asac: we have the checkbox in some upload forms and not in others... IIRC the theory is that upstream translations are normally uploaded in batch, as a tarball for the whole project. [11:44] GMT+1930? [11:44] ouch [11:44] jtv: right [11:44] jtv: still the problem exists for "products" in launchpad [11:45] but i understand the issue there [11:45] asac: actually the terms are used more or less interchangeably in practice: I was really talking about products. [11:45] jtv: actually, does anyone understand if single strings are actually licensable? [11:45] for me it sounds more like only the combined export can have a license ;) [11:45] not each and every single string [11:46] strange idea :) [11:46] asac: here is wednesday but the javascript date shows Thu [11:46] jtv: think about someone licensing "Welcome!" :) [11:46] i doubt you can have a copyright on that ;) [11:47] asac: generally, I agree with you. The problem is that it's never quite that clear-cut: it's more a question of enough text being copied. [11:47] asac: indeed, a short text like that can't be copyrighted. The assumption in copyright is that the expression is one that you can't reasonably expect other people to reproduce accidentally. [11:48] hk47xx: date --rfc-2822 on console? [11:48] is that -0530 ? [11:48] gnomefreak: can you run the javascript command above please [11:49] gnomefreak: javascript:alert(new Date()) [11:49] asac: is Wed, 25 Jun 2008 06:14:12 -0430 [11:49] just type that in location bar and hit return :) [11:49] hk47xx: yeah right [11:49] i am bad at maths ;) [11:49] have the feeling its a rounding issue [11:49] date --rfc-2822 gives me right date and time [11:49] gnomefreak: no the javascript thing ;) [11:50] with (new Date())? [11:50] it fails [11:50] on token new [11:51] gnomefreak: the complete phrase as i pasted it [11:51] javascript:alert(new Date()) [11:51] javascript:alert(new Date()) doesnt work [11:51] as i said token new [11:51] gnomefreak@Development:~$ javascript:alert(new Date()) [11:51] bash: syntax error near unexpected token `new' [11:52] gnomefreak: don't click has to copy and paste [11:52] * gnomefreak tried that first [11:52] try it in ff [11:52] hk47xx: dont click what? [11:52] Wed Jun 25 2008 06:52:35 GMT-0400 (EDT) [11:53] sorry i have a bunch of terms open so i did it there first [11:53] oh and the above is right time and date [11:53] hk47xx: are you on amd? [11:54] asac: no intel core 2 duo [11:54] asac: i first noted this problem in gmail [11:54] hk47xx: well. 64bit? [11:54] install? [11:55] hk47xx: what version of java? [11:55] not that it should matter [11:55] hk47xx: i thought i might have been a unsigned int issue [11:55] asac: i don't know for sure uname -r gives 2.6.24-19-generic [11:55] hk47xx: but since gnomefreak doesnt have it ... and he is in -XXXX timezone too, i think its something on your system [11:56] * gnomefreak cant test in hardy atm [11:56] asac: java version "1.5.0_15" [11:56] java doesnt matter here [11:56] its javascript [11:56] hk47xx: you should really update to 6.0 [11:56] yeah i know [11:56] as i said it shouldnt matter [11:56] gnomefreak: could you try to set your location to venezuela? [11:56] e.g. so date shows VET as timezone? and then try again? [11:57] with tz... [11:57] no i think with gnome menu [11:57] hk47xx: how did you setup your timezone? [11:57] if by default doesnt that use tzseleact and god i hope gnome menu [11:58] asac: at installation time choose 'America/Caracas' [11:58] ok [11:58] asac: for the time zone [11:58] gnomefreak: yeah. i think you should try that [11:58] hk47xx: ok [11:59] gnomefreak, my ppa is fixed now [11:59] working on it [11:59] hk47xx: are you using the firefox installed by ubuntu right? [11:59] fta: thanks [11:59] fta: cool [11:59] asac: yes from synaptic [11:59] jdhore: the jemalloc build should be in fta archive by now [11:59] Thu Jun 26 2008 06:29:44 GMT+1930 (VET) [11:59] after some hick-ups ;) [12:00] gnomefreak: you get that? [12:00] wow [12:00] that is with new timezone [12:00] asac, boom [12:00] fta@ix:~ $ firefox [12:00] /usr/lib/firefox-3.0/firefox: error while loading shared libraries: libjemalloc.so: cannot open shared object file: No such file or directory [12:00] yes i do [12:00] fta: hmm [12:00] fta: ld.so.conf.d installed? [12:00] in xulrunner? [12:00] yes [12:00] should i try in hardy? [12:00] fta: did you run ldconfig? [12:01] err [12:01] was that run during postinst :) [12:01] it didn't [12:01] it didnt? wow [12:01] asac: i even tried starting ff in safe mode in case some plugin was misbehaving [12:01] Unpacking replacement firefox-3.0 ... [12:01] Processing triggers for menu ... [12:01] Setting up firefox-3.0 (3.0+nobinonly-0ubuntu3~fta4) ... [12:01] Please restart all running instances of Firefox-3.0, or you will experience problems. [12:01] Setting up firefox-3.0-gnome-support (3.0+nobinonly-0ubuntu3~fta4) ... [12:01] that's it [12:01] omg this guy is funny as hell ;) [12:02] fta: the build warns something about "useless ldconfig run in postinst" [12:02] we'd better link it properly rather than depend on that [12:02] fta@ix:~ $ ldd /usr/lib/firefox-3.0/firefox [12:02] fta: link it? [12:02] linux-gate.so.1 => (0xb7f2a000) [12:02] libpthread.so.0 => /lib/tls/i686/cmov/libpthread.so.0 (0xb7efb000) [12:02] libjemalloc.so => not found [12:02] hk47xx: hold on ill be back and ill test on hardy since you are using hardy [12:02] fta: yes, ldconfig was the idea [12:03] we cannot really link it ... there is no static dir for xulrunner in theory [12:04] ldconfig doesn't help [12:04] maybe we should link it statically in xulrunner-stub and xulrunner-bin [12:05] the firefox binary is just the xul stub, right ? [12:05] fta: i think so [12:05] we currently link dynamic jemalloc into that [12:05] for me it works [12:05] with ldconfig [12:06] fta@ix:~ $ sudo ldconfig [12:06] fta@ix:~ $ firefox [12:06] /usr/lib/firefox-3.0/firefox: error while loading shared libraries: libjemalloc.so: cannot open shared object file: No such file or directory [12:06] you sure you have /etc/ld.so.conf.d/xulrunner-1.9 ? [12:06] fta: i saw that message [12:06] fta@ix:~ $ cat /etc/ld.so.conf.d/xulrunner-1.9 [12:06] /usr/lib/xulrunner-1.9 [12:06] well. but ldconfig should help then ;) right? [12:06] at least if LD_PRELOAD=/usr/lib/xulrunner-1.9/libjemalloc.so does [12:07] or LD_LIBRARY_PATH=/usr/lib/xulrunner-1.9/ [12:07] hk47xx: your not on hardy are you? what does lsb_release -a give you [12:07] hardy kernel isnt -19 [12:07] that was gutsy IIRC [12:07] gnomefreak: its -19 now [12:08] $ uname -a [12:08] Linux hector 2.6.24-19-generic #1 SMP Wed Jun 18 14:15:37 UTC 2008 x86_64 GNU/Linux [12:08] asac: oh ok i forgot the 24 [12:08] yeah [12:08] i was thinking 2.6.19 [12:08] gnomefreak: No LSB modules are available. [12:08] Distributor ID: Ubuntu [12:08] Description: Ubuntu 8.04 [12:08] Release: 8.04 [12:08] Codename: hardy [12:08] hk47xx: i got it thanks [12:09] well. please open a bug [12:09] command again please [12:09] gnomefreak can reproduce so its confirmed [12:09] hk47xx: translation pack? [12:09] err [12:09] probably not :) [12:09] asac: not sure yet [12:09] im on hardy atm if you give command i can tell you [12:09] asac, i have a fix [12:10] javascript:alert(new Date())? [12:10] fta: http://paste.ubuntu.com/22824/ [12:10] fta: are you use you are running xul 1.9? not 1.9.1 ;)? [12:10] ohh firefox updates in hardy [12:10] fta: whats the problem in your evalution? [12:11] fta@ix:/etc/ld.so.conf.d $ sudo mv xulrunner-1.9 xulrunner-1.9.conf [12:11] then ldconfig [12:11] hk47xx: apt-cache policy firefox-3.0 what does it say? [12:11] fta: huh? since when does extension matter there? [12:11] the other files dont have any either [12:12] at least x86_64-linux-gnu [12:12] they all does in intrepid [12:12] gnomefreak: firefox-3.0: [12:12] Instalados: 3.0+nobinonly-0ubuntu0.8.04.1 [12:12] hmm [12:12] Candidato: 3.0+nobinonly-0ubuntu0.8.04.1 [12:12] Tabla de versión: [12:12] but there is a .conf [12:12] *** 3.0+nobinonly-0ubuntu0.8.04.1 0 [12:12] 500 http://archive.ubuntu.com hardy-updates/main Packages [12:12] 100 /var/lib/dpkg/status [12:12] 3.0~b5+nobinonly-0ubuntu3 0 [12:12] 500 http://archive.ubuntu.com hardy/main Packages [12:12] fta: ok great. fix rules then ;) [12:12] ok [12:12] hk47xx: thanks i will downgrade and see what happens than [12:13] hk47xx: does xulrunner-1.9 match firefox version? [12:13] gnomefreak: you could reproduce, no? [12:14] gnomefreak: sorry, how can i check that? [12:14] asac: im trying with his versions of packages [12:14] asac: i get the right time and date afaik [12:15] Thu Jun 26 2008 06:44:58 GMT+1930 (VET) is that the right time and date? [12:15] gnomefrak: xulrunner-1.9: [12:15] Instalados: 1.9+nobinonly-0ubuntu0.8.04.1 [12:15] for that CC? [12:15] Candidato: 1.9+nobinonly-0ubuntu0.8.04.1 [12:15] Tabla de versión: [12:15] hk47xx: i have it [12:15] *** 1.9+nobinonly-0ubuntu0.8.04.1 0 [12:15] 500 http://archive.ubuntu.com hardy-updates/main Packages [12:15] 100 /var/lib/dpkg/status [12:15] 1.9~b5+nobinonly-0ubuntu3 0 [12:15] 500 http://archive.ubuntu.com hardy/main Packages [12:16] gnomefreak: well here is wenesday 25 [12:16] hk47xx: correct time and date would be? [12:16] gnomefreak: no its not the right time [12:16] * gnomefreak never heard of +1930 [12:16] gnomefreak: you can reproduce the bug [12:17] ok so what i posted is the wrong time and date [12:17] gnomefreak: yes [12:17] ok cool did you file a bug yet? [12:17] asac: bug in ff or gnome? [12:18] gnomefrak: no [12:18] * gnomefreak just noticed intrepid gnome system menu is backwards it has admin first not prefferences [12:19] asac: its wrong for me as well [12:20] with new_york [12:20] gnomefreak: ff [12:20] gnomefreak: well. not sure [12:20] Wed Jun 25 2008 08:19:40 GMT-0300 (EDT) [12:20] it should be -0400 [12:20] maybe a combination of ff + tzdata or something [12:20] gnomefreak: daylight saving? [12:20] it will be -0500 in oct/nov [12:21] but atm it is -0400 [12:21] wait a minute [12:22] asac: in my case i don't think so, unless i can save 24 h .~) [12:23] ah bots back [12:23] @time utc [12:23] gnomefreak: Current time in Etc/UTC: June 25 2008, 11:23:13 - Current meeting: Bugs for Hugs Day [12:23] hk47xx: no. your bug is confirmed. if the launchpad bug is filed its good [12:23] yeah atm its -0400 [12:23] maybe it is -0300 but i thought -0500 in oct/nov [12:25] xul fix pushed to my ppa [12:31] why did i think it was correct in fta's firefox [12:31] testing now to see but firefox is failing to open [12:31] gnomefreak@Hardy:~$ firefox-3.0 [12:31] /usr/lib/firefox-3.0/firefox: error while loading shared libraries: libjemalloc.so: cannot open shared object file: No such file or directory [12:32] that would be why [12:32] thats hardy fta4 [12:32] gnomefreak, yep, fixed. wait for the new xul fta4 [12:32] ok [12:33] about 40 min [12:33] im going back to intrepid so once update hits i can finish my chroot or wait until i get home this afternoon and fix it [12:41] hmmmm hardy has fta4 and it fails intrepid has fta3 and works [12:41] fta: from your ppa [12:44] well atleast my timezone is right in intrepid wrong in hardy but yours is off in both [12:45] gnomefreak, i seriously doubt it: https://edge.launchpad.net/~fta/+archive/+builds?build_text=&build_state=all [12:46] fta: what do you doubt? [12:46] that hardy has fta4 [12:46] in my hardy it did hence the reason i said its broken [12:47] it's still building, how can you already have it ? [12:47] xul fta4, not ff fta4 [12:48] 3.0+nobinonly-0ubuntu3~fta4~hardy [12:48] that ff [12:48] (hardy)gnomefreak@Development:~$ policy firefox-3.0 [12:48] firefox-3.0: Installed: (none) Candidate: 3.0+nobinonly-0ubuntu3~fta4~hardy [12:49] the fix is in xul, not ff [12:49] thats in my chroot but installed in hardy system [12:49] xulrunner has fta3 [12:49] fta: you mixed them up [12:52] * gnomefreak not upgrading to fta4 in intrepid yet [12:53] no i didn't ;) [12:54] xul fta4 is still building [13:20] gnomefreak: i've found a post in un¡buntuforums that describes a similar situation http://ubuntuforums.org/showthread.php?t=363084 [13:21] gnomefreak: the post is from 2007 [13:25] gnomefreak, xul fta4 is ready (except for amd64) [13:25] asac, ff crashes on shutdown [13:25] asac, apparently not related to jemalloc as it crashed before [13:26] fta: you can reproduce? [13:26] yes [13:26] weird stack http://paste.ubuntu.com/22837/ [13:27] fta: is that a bus error or segv? [13:27] Program received signal SIGSEGV, Segmentation fault. [13:28] selinux? [13:28] do you have that enabled? [13:29] asac: where does firefox get the javascript:alert from? [13:31] i have a hard time thinking it gets it from gnome or google [13:32] they stopped the thread in dec of 07 im wondering why it stopped [13:35] gnomefrak: http://paste.ubuntu.com/22842/ [13:36] gnomefreak: this simple test yields the same result in my machine [13:36] hk47xx: that doesnt tell me wher eit gets it from im sure its not getting it from HTML [13:37] firefox crashed but yet its still open :( [13:37] asac, not that i remember, maybe a new default in intrepid [13:37] im thinking this is bad it even logged the crash it happened during updates [13:38] anyone care to guess why it "crashed" [13:39] * gnomefreak would rather not have to file a bug since its fta's package [13:40] fta: im getting tracebacks when starting fta4 from term [13:41] cant import cStringIO [13:41] pastebinit please [13:41] ok [13:41] asac: I'm not sure if this counts as a regression or not, but with the Dapper packages, tab mix pro failed on it's initial recovery attempt with the new firefox. [13:42] http://pastebin.mozilla.org/468663 fta [13:42] Subsequent ones worked. [13:42] ScottK: new firefox? [13:42] * gnomefreak thought we decided to keep 2.0 there [13:43] .15 is out but i didnt think we upgraded to it yet [13:43] This is test packages. [13:43] ah [13:43] it is .15 right? [13:44] ick im sorry my dapper still has 1.5 [13:44] ScottK: what was your previous version? [13:45] ScottK: as tab mix pro upgraded too (from addons.mozilla.org)? [13:45] s/as/was/ [13:45] gnomefreak, this has been there forever [13:46] asac: I have 0.3.0.5. [13:46] gnomefreak, it's not fatal, and causes no issue as far as i know [13:46] fta: ok [13:48] asac: I see there is a new version available. [13:49] ScottK: ok. my question was more: do you use that dapper firefox regularly? or maybe the profile was used a long time ago last time? [13:49] I use it every day. [13:50] For some reason I tend to use FF on my dapper desktop mostly and konqueror on my hardy laptop. Not sure exactly why. [13:50] asac, you have hardcoded too many 1.9 in xul recently, it makes my life harder to keep merging that in 1.9.1 [13:50] ScottK: ok. so what does "initial recovery attempt" mean in the tab mix case (sorry, i dont use that) [13:50] I can/want to improve that [13:50] asac: The feature that's in FF 2.0 where it will get all your windows/tabs back after a crash is provided by tab mix pro for 1.5. [13:50] fta: except for ld.so.conf, where did i add 1.9? [13:51] ScottK: oh ok. [13:51] hmm [13:51] So tab mix pro ran and tried to recover all the windows, but then FF and/or the extension locked up. [13:51] system-grepref.js [13:51] Killing it and restarting clean gave a working fresh start. [13:51] Subsuquent recovery attempts worked fine. [13:52] hmm. maybe a race? [13:52] let me check something [13:52] Probably. [13:55] darn ;) [13:56] asac: if i change my time zone from 'America/Caracas' to 'America/Asuncion' the date problem don't show up [13:56] hk47xx: yeah [13:56] tzdata issue imo [13:58] asac: but i can not keep that time zone because then my system time is offset by half an hour? [13:59] hk47xx: America/Paramaribo does that work too? [13:59] asac: let me check that [14:00] you really changed your offset in 2008 [14:00] wierd that some countries still sort out their time ;) [14:01] err in dec 2007 ;) [14:02] asac: no problem in firefox but system time off by one and a half hour [14:03] asac: is there a way i can solve this problem? [14:07] hk47xx: i dont know yet. we need a bug [14:07] the nspr code that calculates offsets might be buggy [14:08] ScottK: try to stop firefox such that it would start with multiple tabs open next time [14:08] asac: just looked at a midbrowser export... I see the problem now: you do get the template, but it's the XPIPO export, not the original XPI. [14:08] ScottK: then remove compreg.dat and extensions.* files from your profile and start firefox [14:08] does that reproduce your hang? [14:08] maybe try a few times if its a race [14:09] removing those files should make your extensions be reparsed and so on, which might cause the required delay here [14:09] asac: OK. I'll try that in a bit. [14:09] jtv: in distro export i get both [14:10] asac: takes some special-casing. The code that generates language packs is actually separate from the code that exports (batches of) translation files. [14:12] asac: Alternatively, you can upgrade to the current version of tab mix pro and it just hangs. [14:13] I'd already started down that road before your last comment. [14:13] ScottK: yes, but you cannot reproduce that easily as you cannot upgrade all the time [14:13] removing those files should simulate an upgrade [14:13] if you can reproduce that way, we can at least see if its a regression [14:14] Right. OK. I'll need to get the old version back first. [14:22] thanks for your help, i will file a bug report as soon as possible [14:28] hk47xx: if you have the bug, let me know so i can get it properly started [14:29] ScottK: if it still hangs when upgrading to latest extension version, you could also try to reproduce with that version (by removing the files i mentioned) [14:29] OK [14:45] asac: i'm leaving, here is the link to the bug report https://bugs.launchpad.net/ubuntu/+source/firefox/+bug/242926 [14:45] Launchpad bug 242926 in firefox "Incorrect date in firefox for 'America/Caracas' timezone" [Undecided,New] [14:45] thanks [14:58] asac, http://paste.ubuntu.com/22864/ <= makes sense to you ? [15:00] fta: can we somehow preserve the "fail if not manually updated" use-case so we can reuse it for 1.9.1 and 2.0 and so on? [15:01] we still have: ifeq (,$(findstring $(EM_TRANSLATION_VERSION), $(LOCALE_VERSION))) [15:03] but it's now useless as LOCALE_VERSION is parsed from DEBIAN_VERSION and EM_TRANSLATION_VERSION too [15:03] fta: can we keep the "manual setting" in case we are on a development version? [15:03] i think we have code for "development" version for the branding somewhere [15:04] ifeq (1,$(DEVELOPMENT_VERSION)) [15:04] EM_TRANSLATION_VERSION := 1.9 [15:04] else [15:04] ... [15:04] hm, ok [15:04] not sure ;) [15:04] just thinking loud [15:05] well, no. I have to tweak manually EM_TRANSLATION_* anyway so there's no point in adding that test === jt1 is now known as jtv [15:07] fta: you mean for 1.9.1 you need to tweak it manually? [15:08] just EM_TRANSLATION_* [15:08] as it's not 1.9 [15:09] ok. do what you proposed for now [15:09] maybe i will think about it, but most likely this will just sink in my braindump :) [15:09] the diff i just pasted for 1.9, and other EM_TRANSLATION_* values in 1.9.1, with the same ifeq test in both [15:10] even if the test is useless in 1.9 [15:10] fta: yeah. [15:10] i think thats fine [15:10] its supposed to be useless in 1.9 now [15:11] i always wanted to wait with that translation maxVersion bump until 3.0.1, but fine. lets go for it now [15:11] and hope that reason exists upstream ;) [15:17] hmm, nope. it's bad. EM_TRANSLATION_VERSION should not come from changelog as we tweak the version. it must be vanilla so come from mozilla/config/milestone.txt [15:27] yeah most likely. probably you need "late" evaluation then to make this feasible for tarball layout [15:29] yes, right [15:35] the test is useless in both cases, I just have to set MAX = xul_version that's it [15:36] (in 1.9.1) === ubott2 is now known as ubottu [16:45] newz2000: which flash are you using? [16:45] the latest from adobe [16:46] using their installer [16:46] is there a way to spit out info about what parts are using memory? [16:46] latest 9 or latest 10?` [16:46] 9 [16:46] 9.0 r124 [16:46] good question. not easily id say [16:47] newz2000: is ffox also consuming cycles? or just mem? [16:47] it's using cycles to. bouncing between 5 and 30% [16:47] while doing nothing? [16:47] e.g. just about:blank open [16:47] hmm [16:47] wait, maybe not, let me check [16:48] staying at about 14%. I hav a bunch of tabs open but the window isn't active [16:48] sorry, it's up and down [16:49] newz2000: ok if you want you can install xulrunner-1.9-dbgsym and firefox-3.0-dbgsym and attach gdb to the process [16:49] then dump the threads [16:49] !dbgsym [16:49] dbgsym is https://wiki.ubuntu.com/DebuggingProgramCrash [16:49] ok. Usually closing and restarting makes this go away for a while. [16:50] newz2000: bug you can reproduce it? [16:50] but [16:50] it happens periodically [16:50] deb http://ddebs.ubuntu.com hardy main universe [16:50] deb http://ddebs.ubuntu.com hardy-updates main universe [16:50] deb http://ddebs.ubuntu.com hardy-proposed main universe [16:50] install the dbgsym packages from above [16:50] when it happens you can use [16:50] gdb -pPID /usr/lib/firefox-3.0/firefox [16:50] i guess [16:51] and then [16:51] (gdb) threads apply all bt [16:51] and [16:51] (gdb) threads apply all bt full [16:51] we might be able to see which threads are running and guess from there if its flash [16:52] ok, though firefox is probably the single most important program for my day to day work so if this interfereres I will probably have to revert. [16:52] newz2000: what interferes? dbgsym packages? [16:52] no they wont [16:52] they are not touched unless you use gdb [16:52] ok [16:53] do I also need to install firefox-3.0-dbgsym? [16:53] yes is better i guess [16:57] Hi I tested Firefox 2.0.0.15 on Ubuntu Festy, where do I upload the results? [17:01] lastent, see http://mozilla.qa.ubuntu.com [17:03] lastent, click on the version you tested, then on the test you did (15 minutes normal usage with new profila, or 15 minutes normal use with existing profile, or testplan), then enter your comment in the form. [17:04] You need to register on mozilla.qa.ubuntu.com and log in before entering comments. [17:05] maybe it makes sense to draft a wiki with basic instructions on how to submit tests? [17:05] i think we constantly repeat the same things ;) [17:08] asac, MozillaTeam/QA/Submit is ok? [17:08] (the address to use, nothing so far there) [17:13] Jazzva: yeah [17:13] btw, the address for Mozilla QA tracker in title needs to be updated to ubuntu.com [17:14] The one in the title points to stgraber.org [17:20] armin76: can you run xulmine demo from http://benjamin.smedbergs.us/xulrunner/ [17:20] on ppc? [17:22] asac, https://wiki.ubuntu.com/MozillaTeam/QA/Submit ... I don't think we need a whole page for this :). Maybe we could move it later to /MT/QA... [17:22] (noticed the error in wiki formatting) [17:23] Jazzva: why not keep it that simple. if its hidden among other content it might confuse folks and discourage them [17:23] in that way they instantly see that "this is simple" :) [17:23] Ok :) [17:23] anything else I can help? [17:23] I'll add the link to it to the MT main page in the QA section [17:24] lastent, interested in extension packaging? [17:24] ; [17:24] ;) [17:25] Jazzva, what do you mean by that(sorry not native english speaker), is it something like packing .xpi? [17:26] lastent, preparing deb packages for extensions. The process is documented at https://wiki.ubuntu.com/MozillaTeam/Firefox3Extensions/Packaging [17:27] There are also other ways of helping out :). See https://wiki.ubuntu.com/MozillaTeam#head-1b2cc2b9eb04b5dc2692ce8c95cfab53985b511f [17:29] Jazzva, I could help with the packing [17:31] lastent, great :). Maybe it's the best to package an extension you're using frequently. Check first if it already isn't packaged. All packaged extensions are available on . [17:32] lastent, just follow the procedure from packaging page and ask here if you bump into any problem. Somebody will help you out... [17:32] ok Jazzva [18:34] massive netsplits ... wow [18:35] -dmwaters- {global notice} Good day all, In a few minutes I'm going to be rebooting one of our main us hubs. This will mean some splitting, but things should come back together rather quickly. Thank you for your patience, and thank you for using freenode! [18:35] [18:41] [18:38] asac, are the apport hooks really working ? [18:39] if '/usr/lib/xulrunner' in my_ext: [18:39] section_gre.append((my_ext)) [18:39] elif '/usr/lib/firefox' in my_ext: [18:39] section_app.append((my_ext)) [18:39] this is not right for us [18:40] fta: they somewhat work ... somtimes [18:40] at least i get exentsion info and plugin info for a bunch of submissions [18:40] but not all [18:40] most likely because of this [18:42] and it should not be elif as this is exclusive [18:43] asac: i'll do that later [18:55] * asac off for game [18:58] asac, still here for a sec? [20:20] asac: how i'm supposed to run that? [20:20] root@luna:~# xulrunner-1.9 --app xulmine-0.9.xulapp [20:20] Error: App:Name not specified in application.ini [20:21] * armin76 wonders if he should unpack it [20:22] asac: okay, i've unpacked it, it complained about only supporting 1.8, i forced it, and works fine [20:24] asac: about window segfaults, best times segfaults, custom field segfaults, looks like it segfaults when wanting to show a window? [20:24] asac: changing the level works [20:24] asac: and the game works as well [20:41] asac, seems ff is a pig with jemalloc [20:42] PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND [20:42] 15786 fta 20 0 1010m 436m 39m R 52 21.5 288:09.44 firefox [20:42] while idle [21:17] (firefox:15786): GLib-GObject-WARNING **: invalid unclassed pointer in cast to `GdkDrawable' [23:52] fta: intersting [23:52] for me it worked quite snappy [23:53] otoh, it might just be subjective perception ;) [23:53] i never had problems anyway [23:53] with jist a few tab, it's ok, with 80+ tabs, it's almost unusable [23:53] tabs [23:54] i often reach 100 tabs now [23:56] fta: that suggests that you might consider to review your workflow [23:56] i mean, i rarely have more than 10 tabs open :) [23:56] it was fine before [23:56] LOL :) [23:56] hell, i cannot even remember what i had opened in 4 tabs :) [23:57] my desktop consist of X terminal tabs + 3 firefox tabs. [23:57] by accident that can be two firefox windows :) [23:57] but that kills my brain power :-D [23:59] fta: anyway thanks for pointing out [23:59] now we just need some hard data to provide upstream that they are totally wrong on using jemalloc ;) [23:59] asac_: you need some tester?