[00:30] asac: Pong! [00:30] lol, that's latency ;) [00:31] Not mor than a day :P [00:32] Ah, awesome. Hurray for irssi's hilight window. [00:32] I'll check out the xulified ppa package shortly :) [00:43] mozilla bug 404493 [00:43] Mozilla bug 404493 in Widget: Gtk "Native GTK style for arrows" [Normal,Resolved: fixed] http://bugzilla.mozilla.org/show_bug.cgi?id=404493 [00:43] mozilla bug 405561 [00:43] Mozilla bug 405561 in File Handling "nsGnomeVFSService.cpp has new libgnomevfs/gnome-vfs-utils.h system header dependency" [Normal,Resolved: fixed] http://bugzilla.mozilla.org/show_bug.cgi?id=405561 [00:44] mozilla bug 402742 [00:44] Mozilla bug 402742 in ImageLib "Stock icons will not display if libgnomeui is not present" [Major,Resolved: fixed] http://bugzilla.mozilla.org/show_bug.cgi?id=402742 [00:46] damn, can't find it [01:04] RAOF: hey [01:04] Ubulette: fun with bugzilla? [01:04] or bonsai + bot? [01:04] ;) [01:07] looking into the gnome integration. reed mentionned tons of patches in queue. Can't find any [01:18] asac, ff3 on kde http://img241.imageshack.us/my.php?image=snapshot1ms0.png [01:18] ff2 from the same guy: http://img99.imageshack.us/my.php?image=snapshot2jh0.png [01:31] search for need-checkin [01:31] look on the tinderbox? [01:31] bugzilla [01:32] yay ... thats what kde users deserve ;) [01:32] mozilla bug 408119 [01:32] Mozilla bug 408119 in Widget: Gtk "Firefox should clear gtk icon cache on theme change" [Normal,New] http://bugzilla.mozilla.org/show_bug.cgi?id=408119 [01:32] what was the gnome integration bug again? [01:32] mozilla bug 407044 [01:33] Mozilla bug 407044 in Download Manager "Download manager should use more Linux stock icons" [Normal,Resolved: fixed] http://bugzilla.mozilla.org/show_bug.cgi?id=407044 [01:33] more :) [01:33] mozilla bug 406982 [01:33] well thats not the right one either [01:33] Mozilla bug 406982 in OS Integration "Toolbar separator margin too big" [Normal,Resolved: fixed] http://bugzilla.mozilla.org/show_bug.cgi?id=406982 [01:33] what was the one that landed the first few patches ;) [01:33] mozilla bug 406883 [01:33] Mozilla bug 406883 in General "New style for Preferences, Addons and PageInfo header" [Normal,New] http://bugzilla.mozilla.org/show_bug.cgi?id=406883 [01:34] mozilla bug 406676 [01:34] Mozilla bug 406676 in Widget: Gtk "GTK Menu Separator is not painted natively" [Normal,Resolved: fixed] http://bugzilla.mozilla.org/show_bug.cgi?id=406676 [01:34] mozilla bug 406672 [01:34] Mozilla bug 406672 in OS Integration "Please add 'back' and 'forward' stock icons to help viewer under Linux." [Normal,New] http://bugzilla.mozilla.org/show_bug.cgi?id=406672 [01:34] ahah [01:35] mozilla bug 402742 [01:35] Mozilla bug 402742 in ImageLib "Stock icons will not display if libgnomeui is not present" [Major,Resolved: fixed] http://bugzilla.mozilla.org/show_bug.cgi?id=402742 [01:35] thats the one [01:35] https://bugzilla.mozilla.org/attachment.cgi?id=292112 [01:35] yeah [01:36] Ubulette: do we build with --enable-gnomeui? [01:36] ff3 yes, xul no [01:37] hm [01:37] no to both [01:37] we need to [01:37] --enable-gnomevfs in ff3 [01:37] but we have to take extra care that we do the gnome-support split [01:37] hmm [01:38] Ubulette: so whats his problem? [01:38] does he have the gnome theme package installed? [01:38] what's bothering me is that even gnome users have issues [01:39] kde was expected [01:39] mozilla bug Bug 406868 [01:39] mozilla Bug 406868 [01:39] Mozilla bug 406868 in ImageLib "Fallback scheme for -moz-icon" [Normal,New] http://bugzilla.mozilla.org/show_bug.cgi?id=406868 [01:39] thats what we need [01:39] otherwise kde users will die [01:40] Ubulette: the gnome users that complained in forum were cured by "real" restart [01:41] donno [01:41] Ubulette: do you have mozilla bug 406037 already? [01:41] Mozilla bug 406037 in Toolbars "'back', 'stop', 'reload' icons missing..." [Normal,Resolved: duplicate] http://bugzilla.mozilla.org/show_bug.cgi?id=406037 [01:42] no [01:42] oh [01:42] :) [01:42] its a dupe of the other [01:43] anyway ... try to build with gnomeui [01:43] i read https://bugzilla.mozilla.org/show_bug.cgi?id=402742#c74 [01:43] 402742 is supposed to be in b2 [01:43] Mozilla bug 402742 in ImageLib "Stock icons will not display if libgnomeui is not present" [Major,Resolved: fixed] [01:43] Ubulette: yes ... read that comment [01:43] try with --enable-gnomeui [01:44] "otherwise [01:44] icons may not appear in the navigation toolbar and other places" [01:44] yep [01:44] but it sounds buggy to me [01:44] but take care that the xulrunner-1.9 package is still not depends contaminated [01:44] why? [01:48] oh [reed] is back [01:48] <[reed]> and still broken, though [01:48] solved your X ? [01:48] <[reed]> nope [01:48] <[reed]> since xorg creates xorg.conf.failsafe and uses it, it overwrites the log files [01:48] <[reed]> so, I can never figure out why xorg isn't processing my xorg.conf [01:49] <[reed]> because I have no log files! [01:49] you should kick bryce :) [01:49] i have zillions of backups of my xorg.conf [01:49] in ubuntu-devel :) ... tell him that you cannot read the log because of fail safe ;) [01:51] --disable-gnomeui Disable libgnomeui support (default: auto, optional at runtime) " [01:51] hmm [01:51] default auto.. auto disabled or what ? [01:52] damn, I have 57 tabs [01:53] <[reed]> meaning it's enabled by default [01:53] <[reed]> but if you don't have gnomeui, it'll disable it [01:53] checking for libgnomeui-2.0 >= 2.2.0... yes [01:53] checking MOZ_GNOMEUI_CFLAGS... -DORBIT2=1 -pthread -I/usr/include/libgnomeui-2.0 -I/usr/include/libart-2.0 -I/usr/include/gconf/2 -I/usr/include/gnome-keyring-1 -I/usr/include/libgnome-2.0 -I/usr/include/libbonoboui-2.0 -I/usr/include/libgnomecanvas-2.0 -I/usr/include/gtk-2.0 -I/usr/include/gnome-vfs-2.0 -I/usr/lib/gnome-vfs-2.0/include -I/usr/include/orbit-2.0 -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/inclu [01:53] de -I/usr/include/libbonobo-2.0 -I/usr/include/bonobo-activation-2.0 -I/usr/include/libxml2 -I/usr/include/pango-1.0 -I/usr/include/freetype2 -I/usr/include/gail-1.0 -I/usr/include/atk-1.0 -I/usr/lib/gtk-2.0/include -I/usr/include/cairo -I/usr/include/libpng12 -I/usr/include/pixman-1 [01:53] checking MOZ_GNOMEUI_LIBS... -pthread -lgnomeui-2 -lSM -lICE -lbonoboui-2 -lgnomevfs-2 -lgnomecanvas-2 -lgnome-2 -lpopt -lbonobo-2 -lbonobo-activation -lart_lgpl_2 -lgtk-x11-2.0 -lgdk-x11-2.0 -latk-1.0 -lgdk_pixbuf-2.0 -lm -lpangocairo-1.0 -lpango-1.0 -lcairo -lgconf-2 -lgmodule-2.0 -ldl -lORBit-2 -lgthread-2.0 -lrt -lgobject-2.0 -lglib-2.0 [01:54] #define MOZ_ENABLE_GNOMEUI 1 [01:54] ok, so we have it. [01:54] <[reed]> Ubulette: so, I know I have a working xorg.conf that I can use [01:54] <[reed]> but it's failing and going to failsafe still [02:00] so you're in failsafe or in console ? [02:02] <[reed]> failsafe [02:02] <[reed]> but I've been in console a lot, too [02:02] <[reed]> failsafe is annoying at 800x600 [02:02] <[reed]> :( [02:02] <[reed]> but at least my networking works under failsafe [02:02] <[reed]> it doesn't work under console [02:05] you're running ubuntu ? [02:05] <[reed]> ... [02:05] <[reed]> yes [02:05] <[reed]> :) [02:06] gutsy ? hardy ? [02:06] <[reed]> gutsy [02:06] try to start without xorg.conf [02:06] let Xrand do everything for you [02:06] <[reed]> I might run the latest Firefox builds, but I'm not *gutsy* (pun intended!) to run the latest Hardy [02:06] <[reed]> :) [02:07] :) [02:10] when you know when not to accept an upgrade, you're on the safe side. I've been running dev OS for 10+ years, I never had anything major [02:10] <[reed]> I do upgrade to beta releases of ubuntu, though [02:15] did you try to start X without conf ? [02:16] <[reed]> not yet [02:16] <[reed]> I'll try now [02:16] <[reed]> was waiting to see if bryce had ideas [02:16] <[reed]> :/ [02:21] http://bugs.freedesktop.org/show_bug.cgi?id=10301 [02:21] Freedesktop bug 10301 in freetype font backend "LCD filtering patch" [Normal,New] [02:22] ubotu: can you include that in your cairo packages? [02:22] and see if things improve ;) [02:23] or does system cairo still have regressions? [02:25] i'm back to no system cairo as it corrupted big pages [02:26] but i'm still running cairo 1.5.4 for everything else [02:26] no issue so far [02:26] ubotu: use 24x8 [02:26] Sorry, I don't know anything about use 24x8 - try searching on http://ubotu.ubuntu-nl.org/factoids.cgi [02:26] to fix big pages [02:26] Ubulette: ^^ [02:27] or do you do that already? [02:28] no [02:28] seems i also need a patched fontconfig [02:28] it should fix the pig page corruption [02:28] for what? for the patch? [02:28] http://bugs.freedesktop.org/show_bug.cgi?id=13566 [02:28] Freedesktop bug 13566 in library "Fontconfig options for freetype sub-pixel filter configuration" [Enhancement,New] [02:28] (LCD ...)? [02:28] yes [02:28] Ubulette: the last patch doesn't need that [02:28] from what i read in bug [02:29] "I've created bug 13566 with the modification patch to fontconfig that allows the cairo patch to apply." [02:29] you might need an unpatched fontconfig though :) [02:29] Launchpad bug 13566 in file-roller "Faulty drag and drop behavior for fileroller (dup-of: 13199)" [Medium,Invalid] https://launchpad.net/bugs/13566 [02:29] Launchpad bug 13199 in file-roller "Drag n' Drop directories from File-Roller is somewhat tricky" [High,Fix released] https://launchpad.net/bugs/13199 [02:29] lol [02:29] really ... thats what the bug claims: ubuntu fontconfig tried to do the fix in 13566, but failed miserably ;) [02:29] so here a patch that works with pristine fontconfig [02:30] ;) [02:30] are we reading the same bug ? [02:30] comment 23 and 25 [02:31] https://bugs.freedesktop.org/show_bug.cgi?id=10301#c21 [02:31] Freedesktop bug 10301 in freetype font backend "LCD filtering patch" [Normal,New] [02:32] but well [02:32] the comment maybe you really need the patch [02:37] we have two lcd patches in fontconfig [02:37] * debian/patches/03_ubuntu_lcd_filter.patch: [02:37] - Allow LCD filter configuration for when subpixel rendering [02:37] * debian/patches/04_ubuntu_monospace_lcd_filter_conf.patch: [02:37] - Use the legacy LCD filter for Monospace fonts smaller than 12px. [02:38] I assume they should be dropped [02:38] no idea if both needs to be replaced [02:38] prepare a debdiff ;) [02:38] and submit [02:38] who added those patches? [02:39] scott [02:40] doko did the last merge from debian two weeks ago [02:41] yeah scott is the right one to ask then [02:45] Ubulette: you have any idea why symbols that are on .o might not be available if those .o files put into a .so? [02:45] at 3:45am, no :) [02:46] thats rather strange: [02:46] i do [02:46] ar x /usr/lib/xul*/libxpcomglue.a [02:46] then [02:47] g++ -shared -fPIC -lnspr4 *.o -o /tmp/libxpcomglue.so [02:47] but when linking some simply thing just invoking one extern "C" function ... its not there anymore :( [02:48] <[reed]> Ubulette: whee... that at least got me 1600x1200 on my laptop [02:48] asac, so you want a glue.so no ? [02:48] <[reed]> but I also disabled fglrx [02:48] <[reed]> hmm [02:49] so what did you do ? [02:50] <[reed]> I removed xorg.conf [02:50] Ubulette: i am playing around if its doable :) [02:50] cool [02:50] <[reed]> but I kinda need fglrx to work :) [02:50] asac, don't tell benjamin, he'll kill you [02:51] Ubulette: i will not do it without permission [02:51] i want to resolve this mess [02:51] i want it to be included upstream ;) [02:54] ha ... was obvious ;) [02:55] VISIBILITY_FLAGS are not good if they hide everything ;) [03:08] strange, none of the patches are clean [03:08] fontconfig 2.5.0 both [03:26] i've patched fontconfig, i'll do cairo tomorrow [03:26] night [03:35] night === asac_ is now known as asac === Ubulette_ is now known as Ubulette [14:49] bug 116236 [14:49] Launchpad bug 116236 in libcairo "evince shows a mostly blank pdf" [High,Fix released] https://launchpad.net/bugs/116236 [14:49] bug 173861 [14:49] Launchpad bug 173861 in libcairo "[CVE-2007-5503 security fix regression] firefox crashes after upgrading to libcairo2 1.4.10-1ubuntu4.1" [High,Fix released] https://launchpad.net/bugs/173861 [14:54] hi [14:54] btw, i am on holidays now :) [14:55] though it shouldn't make a big difference [14:55] lol [14:56] how many days/weeks per year do you have ? [14:56] not many [14:56] 20 [14:56] 20 days per year ? [14:56] yeah [14:56] waawww [14:56] imo doesn't matter much if its 0, 10 20 30 40 [14:56] its all a bargain ... you can also take unpaid holidays [14:57] I have 9 weeks, paid [14:57] yeah [14:57] but if you would have 0 weeks you could earn 9 weeks more money :) [14:57] no [14:57] and then take 4 weeks unpaid ... which would give you 5 more weeks payment [14:58] think about it ;) [14:58] its all a compensation thing [14:58] there's a law against that in france [14:58] yeah [14:58] thats a problem then :) [14:58] you have 9 weeks minimum? [14:59] we have 4 weeks minimum in germany ... but default is 6 weeks [15:00] 5w min + 20 days thanks to the 35h/w law [15:00] ha [15:00] well [15:00] i find it good to have minimal holidays and guaranteed insurance if you are a dependent worker [15:00] but i don't feel like a dependent worker ... and after all i am not employed, but a contractor [15:00] oh [15:01] which would usually mean 0 holidays :) [15:01] e.g. it means you are free to go whenever it suites the schedule [15:02] 35h/w law is crazy ... why was that done? to get more people employed or what? [15:02] yes [15:02] didn't work that well [15:03] yeah ... great ... you can take the unemployed living on the street to fill the IT jobs :) ... i already see that ;) [15:08] anway, if i ever want to be employed again ... i will go to france ... thats for sure ;) [15:08] lol [15:08] unfortunately your language is too hard for me ... why can't you speak spanish? [15:08] and lots of people don't speak english :( [15:09] right [15:09] here most people can speak english ... but their pronounciation really sucks [15:09] i feel ashamed whenever i here a german speak [15:09] (except a few that speak like mother tongue) [15:10] is that 9 weeks holiday something you expect to go away soonish? [15:12] i take 1 week from time to time, and 1 month about now [15:13] well ... 9 weeks means every month one week, except 3 month [15:13] thats amazing [15:16] you can also store 2w max per y in an account for later [15:16] and either get them paid or take them in one shot [15:17] or retire earlier [15:21] asac, that 24-8 cairo stuff, I don't find a patch in the moz tree. [15:24] he? its a configure option according to vlad [15:25] retire earlier? ... you mean like 5 years earlier if you stocked up 250 weeks? [15:25] :) [15:27] at a rate of 2w/y, that's just 125y :) [15:28] we have to work 40y, soon 42.5y, that won't work [15:36] asac, which configure ? moz ? cairo ? [15:38] cairo [15:39] fun [15:39] why the stupid langpacks have the branding? [15:39] armin76: that has always been the case [15:39] you have to strip it :) [15:39] so gentoo doesn't ship official branding? [15:40] it does [15:40] OK [15:40] but maybe i'll do what you say [15:40] although that means redo the langpacks [15:41] i don't see the problem if you ship official branding anyway? [15:41] yeah, but i just added the iceweasel icons [15:41] thing is that if you install a langpack, it says mozilla firefox :D [15:41] armin76: cool ... you should just use the debian source for iceweasel and langpacks [15:42] but the debian source doesn't have the mozilla branding [15:43] i don't understand your problem :) [15:43] either you ship mozilla branding ... or you ship ice branding ... why would you want both? [15:43] we have now three brandings: mozilla, bon echo and iceweasel [15:44] thing is that with the mozilla branding you can't distribute binaries, but since we're a source distro, that's not a problem [15:44] why can't yu distribute binaries with mozilla branding? [15:45] if they are signed off by mconnor it should be ok [15:45] because we use patches [15:45] we do too :) [15:46] i have no idea, it's just that mozilla doesn't like gentoo and how we do things [15:46] yeah ... like it was for debian [15:46] so you want to allow users to choose between mozilla and ice branding? [15:46] then just replace the bonecho branding director imo [15:46] and provide two sets of langpacks [15:47] why bother [15:47] if you want to do icedove two: http://people.debian.org/~asac/icedove-branding-2.0.0.x.tar.gz [15:47] atm if you don't use the mozilla branding and install a langpack, it says mozilla firefox, not bon echo [15:47] yeah [15:48] it won't work if i just remove the branding files from the langpacks? [15:48] i think you will end up without a icon at some places [15:48] (but not sure) [15:48] at least at some points i had users without branding :) [15:49] i'll try :) [15:49] give it a try .. exactly [15:49] ok i am out getting some food [15:49] http://dev.gentoo.org/~rbu/icegentoo.png [15:49] bye [16:04] asac, http://lists.freedesktop.org/archives/cairo/2007-November/012115.html [16:28] ok, i'm done with cairo too [16:29] asac, i've left out two moz patches. the assert-not-fatal one, and a max-font-size one [16:29] not sure if i should put that in main cairo [16:33] Ubulette: take the trunk snapshot [16:33] all patches should be in there [16:33] e.g. not 1.5.4 [16:35] any chance such a debdiff is accepted ? [16:36] did 24x8 work? [16:37] i need to rebuild xul [16:37] ok ... anyway, i think since you have the package ready, just submit the debdiff [16:38] and i will bug seb to include it :) [16:38] you mean 1.5.4 or 1.5.*~git ? [16:38] if xul with system-cairo really works well ... thats a big argument and i guess we get it through [16:38] trunk? [16:38] == ~git i guess [16:39] ok, i'll try with git/trunk [16:39] cool [16:40] 1.5.5~git [16:44] the assert patch is not there [16:45] neither is the max-font-size patch [16:46] maybe the assert patch is no longer useful, they change the code to ignore everything except out of memory [16:49] asac: removing the branding files gives a xul error [16:49] i'll just remove the lines that say something about firefox [16:49] gives a XML error, to be exact [16:50] yeah [16:51] so in the end you have the same problem as debian [16:51] either patch locales with bon-echo branding [16:51] or completely switch [16:53] hrm... [16:57] Ubulette: who switched what to ignore everything? cairo? [16:57] yes [16:57] ok [17:30] asac, and ubuntu too [17:40] asac, well, trunk is borked. ftbfs in cairo-directfb. I don't want to chase bugs in there too. 1.5.4 is recent enough (9 days old) [17:59] ping asac [18:10] asac: u there? [18:11] sure [18:11] i commited the fixes for gconf and the preference dialog into the WORKING branch [18:11] however, i couldn't get the non-debug version of it to build [18:12] i only have the debug version building correctly [18:12] the xpcom libs didn't get compiled somehow [18:13] do you have any idea why the libxpcom.so files are not compiled in non-debug builds? [18:14] i've been working on it yesterday, but still don't have a working build, maybe you can point me in the right direction :) [18:15] you have a build failure? [18:15] yes [18:16] because the gconf stuff needs to link the xpcom shared libraries, and they are not there when compiling [18:19] paste the build error please [18:19] hold on [18:20] http://paste.ubuntu.com/2752/ [18:21] isn't libxpcom.so in dist/bin/lib? [18:22] aeh dist/bin i mean [18:22] no, it is not there [18:23] it doesn't get build and copied over [18:24] you did a complete clean distclean before building? [18:24] asac: Pat McGowan from your company really want to demo midbrowser with FF3.0 in CES and he wants the package by no later than next monday. I really appreciate it if you can help Jimmy resolve this. [18:25] i removed the objdir, and then did a make -f client.mk clean [18:25] then i went in the objdir and did a make -j4 since i have a quadcore system [18:29] cwong1: next monday? date? [18:29] he wants the package by next Monday 12/17/2007 [18:31] well ... given the fact that he first told me on 12/11 ... and i am on holiday since today this isn't fun [18:32] cwong1: unfortunately I will be leaving for vacation for 2 weeks starting tonight... :( [18:32] he? [18:32] u2? [18:32] i am official on holiday today ...and will officially be back on jan 2 [18:32] oh crap [18:33] i can help though, but this idea of presenting the browser next monday isn't a great one i must admit [18:33] He need the package by monday but CES is really on Jar 7th [18:33] s/jar/Jan/ [18:34] strange [18:34] I really appreciate if you can help get a package built [18:34] yeap [18:34] i pinged him [18:34] k [18:36] jimmy_: ok ... i think that your build failure is something else ... i did a build here as well and got lots of unresolved symbols during linking instead of the "cannot fine -lxpcom" thing [18:37] hmm, what's your .mozconfig file like? [18:39] jimmy_: http://paste.ubuntu.com/2754/ [18:39] a debug build builds fine [18:39] i already tried that last night [18:43] right [18:43] i tested it too [18:44] but non-debug is somehow messed up [18:44] let me build again using ur configs [18:46] jimmy_: you have anything else in your config? [18:48] this is what i have http://paste.ubuntu.com/2755/ [18:50] jimmy_: thats the same [18:51] jimmy_: sure you did a clean build when switching from debug to non-debug? [18:51] jimmy_: try rm -rf objdir and build again [18:53] ok, i might know why you are seeing different issue than me [18:53] ? [18:55] http://paste.ubuntu.com/2756/ [18:55] Ubulette: ok if its 9 days old then it should be fine [18:55] jimmy_: why don't you start with what is in git? [18:56] please keep the versioning for now [18:56] i did [18:56] he? [18:56] then why is your autoconf.mk.in differnt? [18:56] anyway ... those shouldn't affect the build in the stage it fails for you. [18:56] (i guess you don't try to build a package) [19:00] well, i just merged in the extensions/pref directory from the working copy i have into the git branch [19:01] but other than that, there's no differences [19:01] but autoconf.mk.in isn't in extensions/pref :) [19:02] please do a git diff on top level [19:02] and check whatelse is different [19:02] otherwise do a clean build [19:02] i just did [19:03] i am using the exact same version as in the branch now [19:03] and i see the same -lxpcom error [19:03] how fast is your machine? [19:03] i mean you built it in a minute? [19:04] core2duo quadcore 2.6ghz? [19:04] make xpcom/ [19:04] :D [19:05] no, i mean i just did a clean build before you told me to [19:05] jimmy_: i am doing a build now as well ... lets see [19:05] and i just did a git diff, and there's no differences [19:05] jimmy_: i think that's called core2quad, not core2duo quadcore :P [19:05] ok building: make -f client.mk MOZ_OBJDIR=objdir/ build [19:06] now aborting and boosting [19:06] cd objdir [19:06] make -j8 [19:08] armin76: haha, i kept thinking it is a 2 core with 2 dies [19:08] or 2 dies with 2 cores [19:34] jimmy_: ok i think you have to say FORCE_STATIC_LIB [19:34] at that time the shared lib is not yet build [19:34] and because you are building a LIBXUL_LIBRARY you ask the build system to statically link you in [19:34] to libxul.so [19:35] does it actually build then? [19:35] let me try it on mine [19:37] no idea ... the build system has shifted ... better start with the makefiles from the pref service that ships in firefox [19:37] replace the source files with the ones you have and then see whats going on [19:37] but the idea is to build it as a shared lib and punch it into libxul ... yes. [19:39] so... start with pristine Makefile.in ... and see whats missing ... instead of replacing everything and then finding the bug :) [19:39] look how other extensions are built [19:40] i remembered i tried that, but i think the gconf patch also modfied the way how to extension is built, that's why i am kept playing around with the build parameters [19:43] do you think it is due to the way how FF 3.0 shifted from the internal linkage to the frozen linkage? [19:44] http://developer.mozilla.org/en/docs/XPCOM_Glue [19:44] jimmy_: ok [19:44] here the diff that stops it from failing [19:46] jimmy_: partly yes ... you have to be part of libxul if you want to use internal linkage [19:46] but in this case it is more a general shift in the order of when what is build [19:47] http://people.ubuntu.com/~asac/system_pref.diff [19:47] try that [19:47] if it builds for you two that way, I can commit it [19:51] asac: Is it possible that you can build a package for patm after this build issue is resolve? [19:51] cwong1: i talked to him [19:51] and what did he say? [19:51] told him to ping us mid-end next week [19:52] point is that he will be out of the office ... which is why he wanted a package now [19:52] cwong1: we need to resolve the bookmark issue before as well [19:52] I will be out starting tonight and will have no internet connection because I will be on a ship for a week and half [19:52] asac: ok, let me try it [19:53] asac: wat bookmark issue r u talking about? [19:56] bookmarks not working ... at least they don't appear in the menu [19:56] but since PLACES is somehow fucked up in beta1 it might be an upstream issue [19:57] asac: This is just an demo, I dont think we have to fix everything. I think Pat is interest in showing off the Page zoom feature. [19:58] ok, lets wait if gconf works :) [19:58] cwong1: you will be here for another few hours, right? [19:58] for antoehr 1:30 [20:01] asac: Can you build a package for Pat with what we have? We owe you big time...:) [20:02] cwong1: are we still going to do the zoom buttons for Pat? since it needs to be done by Monday I suppose? [20:04] Jimmy_ Jason is going to send me the icons but giving the time crunch, I don't think so unless you or asac has time to put it in. [20:05] but asac is out too next week, right? [20:06] asac: I know. so unless you want to do it. I would say no. Pat will just have to use menu or ctr++ to show off that feature [20:06] i will be here [20:06] not 100% ... but still be here [20:07] so if i can get it done on monday, but that's a maybe [20:07] we can do the icons if there is time left [20:07] so good to have them i guess [20:08] jimmy_: pat is happy with mid or end next week. so lets do as much as possible .) [20:08] ok, i am still here next week [20:08] but will be gone after that [20:08] thanks guys [20:09] asac: Can you build a package once everything is in? [20:10] cwong1: yes ... should be fairly simple ... its like firefox 3 :) [20:12] asac: you are the man.... :) Please send the package to Pat when ready. [20:12] btw he needs the lpia build for menlow [20:14] hmm [20:14] i planned to build using ppa ... they provide lpia bits [20:14] no idea if that is menlow [20:15] the lpia bits is menlow so it should be ok [20:16] Jimmy, I can show you how to put 2 new icons in the toolbar. You can put that in next Monday. Jason will supply me the graphic. [20:21] cwong1: ok [20:21] cwong1: r u going to the event thing today? [20:22] i am going to lunch now, be back in a bit [20:27] k [20:27] Ubulette: gnome bug 503657 [20:27] Gnome bug 503657 in Build "xulrunner 1.9 support" [Enhancement,Unconfirmed] http://bugzilla.gnome.org/show_bug.cgi?id=503657 [20:27] jimmy_: I am not going to the event. [20:27] rebooting, brb [21:13] asac, subpixel seems fine :) [21:14] and 24x8? [21:14] large pages? [21:21] both fine [21:22] i'll push that to my ppa and ask for feedbacks [21:32] asac: do you have problem checkout source from WORKING branch? [21:33] cwong1: no ... why? [21:33] (well not tried the last few hours) [21:34] asac: I did a checkout of the main branch than I did a checkout -b WORKING origin/WORKING and it gave me an Error [21:36] asac: how do you checkout the WORKING branch? [21:36] i currently have a hint file: .git/remotes/moblin.master3 [21:36] with content: [21:36] URL: ssh://asac@moblin.org/home/repos/projects/mobile-browser.git [21:36] Push: moblin.master3:refs/heads/WORKING [21:36] Pull: refs/heads/WORKING:moblin.master3 [21:36] then i just say: [21:36] git pull moblin.master3 [21:37] then git checkout moblin.master3 [21:38] I will give that a try [21:39] asac: are u able to build the non-debug version? [21:40] yes [21:40] you not? [21:40] i got some compilation errors in nsCompressedCharMap.cpp after i patched with your changes [21:40] he? [21:40] which place is that? [21:41] thats somewhere else isn't it? [21:41] nvm, it's not that, that was just a warning [21:41] let me paste it [21:43] http://paste.ubuntu.com/2758/ [21:45] oh [21:45] at least libxul tries to pull in system pref [21:45] strange that i didn't have any issue here [21:45] ;) [21:46] but probably needs a clean build to show up [21:46] jimmy_: for me there was a ../../staticlib/components/libsystem-pref.a [21:46] which is good [21:46] but lets see [21:46] doing a clean build now [21:47] but i did do a clean build [21:47] yeah ... i don't doubt it [21:48] http://paste.ubuntu.com/2759/ ... thats what i had in statilib/components [21:48] now doing a clean build [21:49] jimmy_: please try to go to the extensions/pref directory [21:49] hit make [21:49] and see if after that the build finishes properly [21:49] (that would mean that its just a sequence issue) [21:50] ok [21:54] no, it still doesn't build [21:54] i don't have the libsystem-pref.a [21:56] i see that i have libsystem-pref_s.a built, but it doesn't get copied over to staticlib/components [21:59] do I need to define this in the gconf/Makefile.in? SHARED_LIBRARY_LIBS = ../libsystem-pref_s.a [22:03] asac: jimmy_ hey guys I have to go. Thank you both for working on this. [22:03] cwong1: nice holiday! [22:03] jimmy_: I will send you email later on how to add the icons. [22:04] thx you 2 [22:04] ok, have fun :) [22:08] asac: was that the only change you make in the diff file you sent me? there's nothing else changed outside the extensions/pref directory? [22:08] yes [22:08] jimmy_: but it might totally be true that it doesn build from scratch [22:09] have you tried if going to objdir/extensions/pref [22:09] run make there [22:09] and then run make from top level works? [22:16] yes [22:16] i tried that, it builds it [22:16] ok [22:16] then when i go to the root level to build again, it still fails [22:16] then we just have to fix the order [22:16] nonono [22:18] http://paste.ubuntu.com/2760/ [22:18] this is what i get after i build it from the top level [22:19] jimmy_: ok [22:19] jimmy_: fix the libxul-config.mk [22:20] it refers the the old lib name [22:20] (without -gconf) [22:20] then you end up in a build error because the GCONF_LIBS are not linked into libxul [22:23] hmm [22:24] ok, building now [22:38] wait, libxul-config.mk doesn't point to gconf, doesn't it? i thought you meant to modify system-pref to system-pref_s, am I doing something wrong? [23:03] asac: still doesn't work [23:40] -COMPONENT_LIBS += system-pref [23:40] -COMPONENT_LIBS += system-pref [23:40] +COMPONENT_LIBS += system-pref-gconf [23:40] jimmy_: ^^ [23:40] in linux-config.mk [23:41] libxul... i mean [23:41] i tried that, that doesn't build [23:41] yes it fails because of missing gconf libs [23:41] i think we need to link system-pref, not system-pref-gconf [23:41] yes [23:41] i already fixed that [23:42] but it doesn'change a thing [23:42] -EXTRA_DSO_LDOPTS += $(LIBS_DIR) $(EXTRA_DSO_LIBS) [23:42] +EXTRA_DSO_LDOPTS += $(LIBS_DIR) $(EXTRA_DSO_LIBS) $(MOZ_GCONF_LIBS) [23:42] thats what i did in toolkit/library/Makefile.in [23:42] (to test) [23:42] then there is only one symbol left missing: [23:42] sStaticXULComponents.o:(.data.rel.ro+0x308): undefined reference to `nsSystemPrefModule_NSGetModule(nsIComponentManager*, nsIFile*, nsIModule**)' [23:43] i think that has something to do with how we build pref [23:44] i think that symbol should be defined by nsSystemPrefService.cpp [23:44] no idea why it isn't though [23:45] hmm its defined by nsSystemPrefFactory.cpp [23:45] ok, i tried moving the EXPORT_LIBRARY = 1 to system-pref [23:45] NS_IMPL_NSGETMODULE(nsSystemPrefModule, components) [23:46] and then i moved FORCE_STATIC_LIB = 1 from system-pref into gconf [23:46] both should be static imo [23:46] wait a sec [23:46] i think we need to export system-pref, not system-pref-gconf [23:46] SHARED_LIBRARY_LIBS = ../libsystem-pref_s.a [23:46] thats what i have in gconf/Makefile.in [23:46] do you have that as well? [23:46] i think we need to export both :) [23:47] but in the end its the gconf thing that provides the lib [23:49] no i don't have that [23:49] ok i manged to build libxul :) [23:49] don't ask me how ;) [23:49] lol [23:49] lets see if i can reproduce it [23:49] you have better luck than me :) [23:51] good it works [23:52] ok here the full patch again (against current git head) [23:53] http://people.ubuntu.com/~asac/full.diff [23:54] jimmy_: ^^ [23:54] cool [23:54] let me test it