[01:05] Hey I think we have an ubuntu-related firefox bug - namely, it doesn't happen with the firefox 3.0 bundle from mozilla.org but it does happen with the ubuntu 8.04 default firefox [01:05] https://bugzilla.mozilla.org/show_bug.cgi?id=444472 [01:05] Mozilla bug 444472 in Prism "Prism Refractor (v0.2) breaks firefox 3.0 on Ubuntu 8.04" [Normal,Unconfirmed] [01:06] I couldn't figure out the proper place to enter this in launchpad but I reported it to mozilla because it's related to the prism extension which isn't an ubuntu package [01:07] even though it's not an ubuntu package the problem only occurs with the ubuntu firefox package and not with other distributions of the same software. [01:22] Added to launchpad: https://bugs.launchpad.net/ubuntu/+source/firefox/+bug/247132 [01:22] Launchpad bug 247132 in firefox "[Hardy] The Prism Refractor Extension (v0.2) causes file downloads to kill firefox 3" [Undecided,New] [01:25] irc://irc.mozilla.org [02:11] twentyafterfour: try libnss3-1d and libnspr4-0d from hardy-proposed and reinstall the extension after that [02:45] asac: that failed to install [02:45] "E: /var/cache/apt/archives/libnspr4-0d_4.7.1+1.9-0ubuntu0.8.04.2_i386.deb: trying to overwrite `/usr/lib/libnspr4.so', which is also in package libnspr4" [09:26] hi! [09:26] twentyafterfour: wait for next upload. that should fix it [09:26] or uninstall libnspr4 [09:26] manually [09:26] hi XioNoX [09:26] just arrived? [09:26] hi asac [09:26] yep [09:42] asac: is next upload happening soon or should I go ahead and try manually installing it? [09:49] twentyafterfour: the upload has just happened in intrepid [09:50] hardy upload is happening right now (but will take a bit as it requires archive admin approval) [09:50] twentyafterfour: i will upload the upload to the mozillateam ppa as well [09:52] twentyafterfour: ok upload to mozillateam ppa is happenening [09:52] twentyafterfour: could you please stay in this situation and the do the upgrade from PPA? [09:56] asac: ok [09:58] this url? https://launchpad.net/~mozillateam/+archive [10:05] twentyafterfour: yes [10:05] twentyafterfour: add that to sources and upgrade [10:05] and see if its fixed [10:06] ok the upgrade worked, I'll test it now [10:07] http://-kol.deviantart.com/art/Grass-Blades-73597063 <- ff3 ubuntu does not open it. It works on XP using FF3 [10:07] Is this known? [10:07] twentyafterfour: ok, please confirm that the PPA version fixed bug #244439 [10:07] Launchpad bug 244439 in nspr "missing symlinks break binary compatibility with native upstream components" [High,In progress] https://launchpad.net/bugs/244439 [10:08] elmargol: weird, I have the same trouble [10:08] and#245122 [10:08] and bug #245122 [10:08] Launchpad bug 245122 in nss "libnspr4-0d missing conflicts+replaces libnspr4?" [Critical,Fix released] https://launchpad.net/bugs/245122 [10:08] elmargol: thats a known non-bug [10:08] its an illegal url [10:08] asac, why does it work oon FF windows? [10:08] that windows accepts it is a bug on its own because ffox uses windows parser [10:09] Ah ok Windows has the bug :D [10:09] yeah [10:09] Windows has many bugs [10:09] just because windows accepts httpx://----good damn dog shit.com doesnt mean that they are right [10:10] asac: I can confirm 245122 fixed [10:10] twentyafterfour: please comment on bug and name the package version as well as the where you got it from there [10:10] thanks! [10:10] asac: did you get new nvidia packages by chance? [10:11] gnomefreak: i have no nvidia hardware ;) [10:11] oh that might be a good reason why you didnt :( [10:11] i really dont want to use his PPA [10:14] twentyafterfour, do you have a paper or something to prove that a host should not start with a -? I fill a bug on debiantart [10:14] elmargol: search for URL RFC [10:14] its illegal [10:15] there are other bugs in launchpad which discuss this [10:15] no time to dig them up though [10:15] asac: commented [10:15] thanks! [10:16] asac: now I'll test my bug ;) [10:16] hehe [10:20] brb [10:27] asac: unfortunately my bug persists! [10:29] twentyafterfour: what is your bug? [10:30] seems like it might be related to 244439 ... firefox crashes when I try to download a file but only if I have the Prism Refractor extension installed [10:30] bug 244439 [10:30] Launchpad bug 244439 in nspr "missing symlinks break binary compatibility with native upstream components" [High,In progress] https://launchpad.net/bugs/244439 [10:31] twentyafterfour: try to uninstall the extension and install again [10:31] twentyafterfour: you have a link for that extension? [10:32] ok i thin i have it installed now [10:32] how to reproduce the crash? [10:32] yeah.. my problem is bug 247132 and the extension is https://addons.mozilla.org/en-US/firefox/addon/6665 [10:32] Launchpad bug 247132 in firefox "[Hardy] The Prism Refractor Extension (v0.2) causes file downloads to kill firefox 3" [Undecided,New] https://launchpad.net/bugs/247132 [10:33] just install extension, restart firefox and then try to download something like a pdf (any filethat is handled by an external app rather than rendered in the browser) [10:33] for example try to download this file once the extension has been installed: http://people.mozilla.com/~mfinkle/prism/prism-0.9.en-US.linux-i686.tar.bz2 [10:34] official mozilla build doesn't crash but ubuntu's firefox package does [10:42] twentyafterfour: prism doesnt have a binary component [10:42] so it has nothing to do with binary compatibility [10:48] asac: weird... it definitely seems related to prism for me though. did you manage to reproduce the bug? [10:48] twentyafterfour: yes. [10:49] i am chatting with the prism dev atm [10:49] ok [10:49] but its not of binary nature [10:49] I thought it strange that it didn't seem to produce a crash report [10:50] so I guess it's just a software glitch rather than a bug in the compile ... still weird how the mozilla distro doesn't have that bug ... I only reported it because other people were complaining that prism didn't work with ubuntu so I thought I would help track down the problem. [11:03] asac ? [11:05] XioNoX: one second ;) [11:05] I've look into the loop, and there are strange things. For exemple at each new loop, another table is made, and if i just add another row, it display at the same line as the rest. [11:29] ok prism crash solved [11:30] XioNoX: not sure. i didnt do the layout refactoring [11:30] ok [11:30] maybe tr was overloaded to not be a block element in some CSS? [11:30] asac: should the strace command show jemalloc stuff with those patches you gave me? [11:30] armin76: no [11:30] armin76: those link jemalloc into the binaries statically [11:30] I'm starting to du it directly in xul, whereas html [11:31] with the advice of gandalf [11:31] armin76: if you grab the patches one revision before you get the same, but with shared libs [11:31] XioNoX: feel free ;) [11:31] all improvements appreciated [11:31] asac: ah..then it doesn't crash [11:31] XioNoX: we could also move away from manually assembling xul in javascript to rdf datasources [11:32] but thats more advanced [11:32] armin76: good [11:32] later :D [11:32] armin76: do you compile with FORTIFY? [11:32] forti what? [11:32] my crash is somewhere in libselinux.so [11:32] armin76: gcc hardening option [11:32] we have [11:32] CPPFLAGS=-U_FORTIFY_SOURCE [11:32] export CPPFLAGS [11:32] previously we had -D_FORTIFY_SOURCE=2 [11:33] maybe try those two ;)( [11:33] k, i will [11:34] ill see if we still need the "normal" fortify or if we can go for 2 [11:37] *sigh* ETOOMANYBUGSOPEN [11:47] yeah, and the ones i didn't file :P [11:48] armin76: can you do my bug work? [11:48] hppa, ppc segfaults, sparc bus error :P [11:48] ;) [11:48] i think you could follow your passion and at least go through all bugs and ask that they "bumb" :) [11:48] i have enough :P [11:48] lol [11:48] and report back if bumb fixes it ;) [11:49] armin76: i can write you a python scripts that sends a message "please bumb" to all bugs ;) [11:49] Please bumb: [11:49] + either bumb to intrepid [11:49] + or to mozillateam PPA [11:49] + or hardy-proposed [11:50] haha, and why don't you do it instead of me? :P [11:51] armin76: because i am not in a position to cause that much traffic ;) [11:51] and my bug karma is already scary enough ;) [11:52] armin76: people would complain: "those canonical guys dont even read my bug before replying" :) [11:52] armin76: i would make the python script just send the bumb message and set bug to "invalid" ... asking them to reopen if they still care ;) [11:53] my experience shows if you do that you will end up having about 50% less open bugs afterwards [11:53] use any of your padawans then :P [11:53] e.g. close everybug and ask submitter to reopen if he still sees it/cares ;) [11:53] yeah, tipical ubuntu users :P [11:53] lol [11:53] armin76: i did something similar for debian two [11:53] there were even less bugs reopened ;) [11:54] gentoo users directly don't file bugs :D [11:54] i think i did that for network-manager about a year ago ;) [11:54] it worked :-D [11:55] but some people really got angry ;) [11:56] i always have an excuse: if you don't like how i do things, take over mozilla, i don't want to maintain this stuff :) [11:56] you don't *g* [12:00] asac, will I see what I've done into intrepid ? [12:00] noes, pay first [12:01] ha [12:01] ok [12:01] how much ? [12:03] 231 eur [12:03] ok, give me an address and I send the cash [12:03] :D [12:04] asac@gentoo.org [12:05] XioNoX: yes [12:05] XioNoX: you are working for intrepid now [12:05] he's bumbing? [12:06] XioNoX: and we can upload the new ubufox to hardy-backports as soon as it rocks [12:06] cool :D [12:07] i am stupid ... i fail to edit patches directly :( [12:07] fail [12:07] yeah ... because of that i fail to bumb [12:11] success [12:12] spinning with maximum pressure using -j1000 [12:12] armin76: intrepid will get 3.0.1 today ;) [12:13] not released [12:13] build1 [12:14] ricer version [12:34] asac: when is your next planned 3.0 push? [12:35] and did you have plans of fixing the about: saying 1.9 [12:35] i see nspr was pushed [12:37] gnomefreak: the about: page is really a minor issue [12:37] we should fix it [12:37] not sure if we should do it in hardy though [12:37] maybe not in 3.0.1 ;) [12:37] remember that we still need things that we can fix in 3.0.2 ;) [12:38] bug 247132 [12:38] Launchpad bug 247132 in xulrunner-1.9 "[Hardy] The Prism Refractor Extension (v0.2) causes file downloads to kill firefox 3" [High,Triaged] https://launchpad.net/bugs/247132 [12:38] asac: we have plenty of bugs to fix in 3.0.2 [12:38] ;) [12:38] just wanted to know since i am gonna start gutsy 3.0 [12:39] god i hate nss nspr names [12:40] wth is nss named :( [12:40] gnomefreak: hehe [12:40] SONAME version [12:40] libnss3-0d [12:40] is the keyword hjere [12:40] libnss3-1d [12:40] id? [12:40] opps [12:40] 1d? [12:40] gnomefreak: let me induce you [12:40] please write this down;) [12:41] 1. dont backport anything except xulrunner-1.9 and firefox-3.0 [12:41] asac: i know but i cant if i upload to PPA [12:41] 2. try to build xulrunner-1.9 as it is in gutsy [12:41] i need nspr and nss for PPA [12:41] gnomefreak: no [12:41] you dont need that [12:41] oh [12:41] well ... the packages that are named in Depends: should be there [12:42] we do auto magic to use in-source nss/nspr if the versions available are too low [12:42] same goes for cairo [12:42] if the xulrunner-1.9 build fails let me know [12:42] ok will o [12:42] do [12:44] !info libhunspell gutsy [12:44] Package libhunspell does not exist in gutsy [12:44] !info libhunspell-1 gutsy [12:44] Package libhunspell-1 does not exist in gutsy [12:45] gnomefreak: hmm [12:45] !info hunspell gutsy [12:45] hunspell (source: hunspell): spell checker and morphological analyzer (program). In component universe, is optional. Version 1.1.5-6 (gutsy), package size 68 kB, installed size 192 kB [12:45] whats the problem? [12:45] ok [12:45] thats ok [12:45] just making sure [12:45] gnomefreak: look at jdongs changelog for the previous backport [12:45] i couldnt remember if it was gutsy that got it [12:45] it should list all required changes you need [12:45] gnomefreak: asac wants you to bumb the bugs [12:46] we need a bug bumper ;) [12:46] asac: i didnt know he did a backport already [12:46] gnomefreak: he did for previous builds [12:46] i think last done is b5 [12:46] armin76: asac if i get done with ff i will do bugs [12:46] gnomefreak: we have 864 open :( [12:46] asac: you left it in our build right? [12:46] in firefox-3.0 [12:46] we have to get back to 200 ;) [12:46] asac: thats not an issue i have seen some dupes [12:47] gnomefreak: not an issue? [12:47] bug count explodes again :( [12:47] asac: i can do them (as in not an issue) [12:47] all is lost :( [12:47] * gnomefreak will start on new today [12:48] xulrunner breaks epiphany-gecko? [12:48] we have 118 confirmed/triaged/fix committed bugs [12:48] that means that 650 are incomplete or new [12:48] most will be incomplete im sure [12:48] thats a shame [12:48] we have to get them invalid or confirmed much quicker [12:49] asac: well people need to work on bugs ;) [12:49] who do we have besides me and you that do bugs? [12:49] gnomefreak: my bug karma is at 15k. when i clean up my current mail backlog it will be at 20k i am sure [12:49] ubotu :D [12:49] that number indicates that i waste far too much time on bugs already [12:50] * gnomefreak hasnt gotten over 4800 no matter how many i do [12:50] gnomefreak: changing states gives more points ;) [12:50] bleh, why nobody commited yet my hppa patch :( [12:50] gnomefreak: marking dupes gives massive points [12:51] asac: might i suggest you read your mailing lists mail that you have laying around. once i get tags/states fixed people will know what to do ;) [12:52] hmmmm they were pushed this morning [12:52] gnomefreak: right. i have to slip that in between doing security updates :) [12:52] thats fine [12:52] gnomefreak: can you look at the bug triager handbook page? [12:52] and once answers on that i can finish 3 wikis [12:52] i did quickly [12:53] * gnomefreak doesnt have links thanks to bookmark exporting being screwed up [12:53] asac: [12:53] bumb? [12:53] * Shindig has quit () [12:53] Good day [12:53] eh? bumb? [12:53] what does that mean? [12:53] i'm tired of explaining it! :P [12:54] armin76: create a wiki page ;) [12:54] * gnomefreak thought he couldnt spell bump :( [12:55] gnomefreak: https://wiki.ubuntu.com/MozillaTeam/Bugs/TriagersHandbook [12:55] gnomefreak: i think thats good enough. i just have to do the MozillaTeam/NormalizedBugFormat page [12:55] which basically describes how the summary/description of a bug must look like [12:57] # [12:57] Crashes: strace + backtrace [12:57] why both why not apport logs [12:58] How to Triage (Ubuntu Bug): [12:58] 1. [12:58] if reproducible or a reasonable evaluation of the bug exists set to TRIAGED [12:58] i have a feeling thats gonna bite us in the ass [12:59] but lets see how it goes [13:00] hmmmmmm 2.0.0.12 is in gutsy right? [13:09] gnomefreak: ok i added crash bugs form [13:09] we can change that once we have more experience [13:10] asac: java plugin amd64 gutsy [13:10] maybe we give a short explanation for each of the items in the Description template [13:10] asac: and wht the hell is Guadalinex V4 [13:10] s/wht/wth [13:10] gnomefreak: spanish derivative [13:10] so upgrade is supported? [13:10] gnomefreak: used in almost all spanish schools accross the country [13:10] upgrade? [13:10] from that to ubuntu [13:11] gnomefreak: not supported from our side [13:11] gnomefreak: if guadelinex supports that is their decision [13:11] and the user should ask them [13:11] ok so after upgrade ff crashes i should let them ahndle it [13:11] handle [13:12] gnomefreak: tell the guy to install ubuntu from scratch if he wants support. we cannot deal with guadelinex to ubuntu upgrades imo [13:12] gnomefreak: he can also try the usual things: uninstall extensions/disable extensions, plugins etc. [13:12] i am [13:14] gnomefreak: ok, what can we do about the bug format? [13:16] dont know not reading it atm [13:16] define what you mean by to do to format [13:18] https://wiki.ubuntu.com/MozillaTeam/Bugs/TriagersHandbook [13:18] refers to [13:18] https://wiki.ubuntu.com/MozillaTeam/NormalizedBugFormat [13:18] in order to get bugs to state Confirmed [13:18] asac: how well do you know apache? [13:18] gnomefreak: depends :) ... i used to know most basics and some advance stuff [13:19] i know the pages i was looking at them already but i have 15+ tabs open atm [13:19] When trying to access the server Apache HTTPD locally, via http://localhost port 80 [13:19] with ff3 [13:19] Though the site seems valid, the browser was unable to establish a connection. [13:20] thats what i get just browing to the link [13:20] he states it worked in fff 2.0.0.13 [13:22] that looks fairly good but what do we need Type (optional): [13:22] for [13:24] the reporter would have that info (as for memory) if that is what you are asking and most of time they dont know how to check or that it might be important unless its a memory caused crash (not so sure that is relevant to most crashes) [13:24] gnomefreak: its optional [13:25] its pre-evaluation [13:25] good to estimate severity [13:25] but i dont expect that incomplete triagers will figure that [13:25] so we can remove it [13:25] doing that now [13:27] i feel like an upload-dog [13:28] bumbing everything everywhere ;) [13:28] armin76: bumb [13:28] lets see how i feel about breaching alpha-2 freeze ;) [13:28] i think i breached it already by uploading nss/nspr to intrepid this morning :-P [13:29] lets hope our RM is long enough asleep that he misses it ;) [13:32] -j1000 build runnning again [13:38] asac, where does the windows get the pluginInfo.name for exemple ? [13:41] who has gutsy with ati? [13:41] asac: i still need java info on amd64 in gutsy (what package it is) the java plugin package? [13:49] XioNoX: in the Wizard.js [13:49] yeah, i've find that [13:50] but where come from the informations present in pluginInfo ? [13:50] XioNoX: from the datasource.js [13:50] XioNoX: i gave you the line yesterday ;9 [13:50] yep [13:50] where pluginInfo is assembled [13:50] I know [13:50] ? [13:50] but on this line [13:51] informations present in the pluginInfo array come from somewhere [13:51] i've find some python files, which seem to interogate the apt database [13:52] XioNoX: yeah [13:52] XioNoX: if you want to improve the website/database part thats fine [13:52] XioNoX: if you dont, we can just define how the rdf element will be called [13:52] and i will fix it [13:52] But the information line that I have to add will not come from this "database" [13:53] XioNoX: no. thats why i said you should display a default description "No description available yet." in case pluginInfo.description is null [13:53] i've add the line description: getPFSValueFromRDF("description"), in the var pluginInfo = { array [13:53] XioNoX: yeah thats right [13:53] ok [13:53] (unless it throws an exception there) [13:54] haven't try yet [13:54] but what should I do for that getPFSValueFromRDF("description") give me back a sentence ? [13:55] getPFSValueFromRDF() function is strange [13:59] XioNoX: dont understand the question about "sentence" [13:59] for the moment getPFSValueFromRDF("description") return nothing [14:00] how can i do for that he return something (and define this something) ? [14:00] XioNoX: dont implement the fallback in the datasource [14:00] XioNoX: implement that in the Wizard.js [14:00] thats where it belongs [14:01] the datasource should just be stupid and know nothing about fallbacks and such [14:01] ok [14:01] (and shouldnt have any strings) [14:01] Wizard.js is clearly UI so strings have to be done there [14:01] (make it translatable if possible) [14:01] so where can I put a default sentence like "No description available yet" ? [14:01] XioNoX: use a translatable properties entry and resolve that in Wizard.js [14:02] and in wizard check whether pluginInfo.description == null [14:02] if thats the case use the sentence as the description [14:02] makes sense? [14:02] yep [14:02] XioNoX: cool [14:03] XioNoX: Ill see if i can fix the webservice in bzr today [14:03] would have to tell you how to setup though [14:03] but maybe you can document it during that ;) [14:04] ok, so it is done [14:04] XioNoX: maybe say "No description found in plugin database. Report a bug to ubuntu-mozillateam@lists.ubuntu.com" :) [14:04] XioNoX: that makes it a bit longer and maybe helps to test the visualization of line breaks [14:04] ;) [14:05] to learn something you could even make a template out of that string [14:05] ho, there are no line break [14:05] I don't know why [14:05] it is the only bug [14:05] e.g. so we can say: "no description found for X in plugin database. ...." [14:05] XioNoX: fix that ;) [14:06] XioNoX: in html it would be just filling the text in the i guess [14:06] (e.g. without label) [14:07] for template: where X is the pluginInfo.name ;) [14:08] maybe we also want to display a link to a homepage [14:08] which is just left out if pluginInfo.homepage is null [14:12]