[00:39] ha ... wikipedia uses ubuntu screenshot for firefox 3 :) [00:39] http://en.wikipedia.org/wiki/Mozilla_Firefox [00:49] nice [01:31] armin76, http://blog.mozilla.com/gen/2008/05/23/firefox-3-utf-8-support-in-location-bar/ === asac_ is now known as asac [10:38] james_w: hey :) [10:39] i face a moderate tricky issue :) ... we have a debian/ only tree ... now i would like to merge that one on top of a full source upstream thing [10:39] technically that should work, but i think that bzr complains [10:39] technically ... e.g. just applying all commits on top of the upstream tree [10:39] james_w: any idea? [10:40] :) [10:43] do you want to keep merging in future, or are you going to throw the debian/ only branch away after that? [10:46] james_w: is there an option for merging in future? [10:47] actually, the way I would think of works in both cases I think [10:47] merging in the future has a few issues I think, but it works. [10:48] james_w: my ears are wide open :) [10:48] give me insight! [10:48] actually, do you version the "debian/" directory, or just it's contents? [10:48] * asac looking [10:49] james_w: http://svn.debian.org/wsvn/pkg-utopia/packages/experimental/networkmanager [10:49] so debian is inside [10:49] ah, so a plain merge should do what you want [10:49] (given that i migrate that branch to bzr) [10:50] james_w: can i do this in both directions? [10:50] as the branches have no history in common then you will need to specify an explicit "merge base", which is the revision to use as a common ancestor. [10:50] e.g. merge upstream on debian ... or theother way around? [10:50] james_w: which revision is that? [10:50] 0? [10:50] yup, you use 0, or rather 1. [10:51] or rather 0, :-) [10:51] i am always confused ;) ... i know i get the initial checkin by bzr diff -c1 [10:51] you use "merge -r 0..-1" [10:51] what kind of hack is that? [10:51] -1 ;) [10:51] -1 means that last revision [10:51] ok so HEAD :) [10:52] I'm not sure how it will do at two way merging. [10:53] james_w: is there alreay a way that bzr understands svn branching? [10:53] most likely not i guess [10:53] you should be able to merge back in to the pkg-utopia one as it's tip revision is in the history of the resulting branch. [10:54] james_w: oh merge back into pkg-utopia ... hmm [10:54] bzr-svn does know something about svn branches, but I don't know how reliable it is. [10:54] that would be great [10:54] the merge would pull in all of the upstream code though. [10:54] james_w: yeah obviously [10:55] I'm just guessing, but if you merge back, and then revert everything except debian/ before committing you will be able to merge debian/ backwards and forwards from then on, but I'm not sure. [10:56] james_w: ok. i think he would be more open to this if he sees all his commits in launchpad ... so i guess i should branch that svn version and then merge the full upstream tree on top of that, right? [10:56] (what a shame that launchpad doesnt show nested revisions) [10:57] well, you could stick a copy of the pkg-utopia branch there as a mirror to show all the revisions. [10:58] but that does sound reasonable. [10:58] thanks. ill see what i can do :) [10:58] lets hope that debian bzr is up-to-date enough to not stop the show :) [10:59] james_w: oh ... if i do merge -r 0..-1 ... wont all the subrevision be lost? [10:59] at least picking smal ranges always eliminated the subrevisions for me [10:59] e.g. bzr merge -r 2..5 => just one flat commit [11:00] actually, I'm not sure. I thought it didn't, but maybe it does consider it a cherrypick. [11:01] james_w: let me test :) [11:01] first setup a mirror :) [11:03] ok ... so one cannot setup a mirrored branch from svn ... thats for sure [11:04] i guess i need a series [11:04] darn ... ill just do it manually [11:16] james_w: help :) [11:16] bzr merge -r0..-1 ../networkmanager.debian.svn/ [11:16] WARNING: the 'deb822' top-level module is *DEPRECATED*, please use 'debian_bundle.deb822' [11:16] bzr: ERROR: Repository KnitPackRepository('file:///tmp/network-manager/.bzr/repository/') is not compatible with repository KnitPackRepository('file:///tmp/networkmanager.debian.svn/.bzr/repository/') [11:16] /tmp/network-manager/ == lp:network-manager [11:16] ah, bzr-svn creates rich-root-pack, your other branch is probably just pack-0.92 [11:17] you will need to "upgrade --rich-root-pack" your other branch [11:17] ok ... so merge upstream on top of debian? [11:17] james_w: err [11:17] wait a second. both are svn branches [11:17] e.g. lp:network-manager == upstream svn [11:17] the other is the pkg-utopia thing [11:17] but upstream is done by launchpad? [11:17] james_w: yes [11:18] they don't use bzr-svn [11:18] can i do it the way they do it? :) [11:19] darn ... so i have to register a series ... and they will refuse to mirror that experimental branch ... i already see what is coming ;) [11:20] james_w: any idea if their sync tool is public? [11:21] asac: no idea, sorry [11:21] you might want to jump in to #launchpad to see about mirroring that branch [11:21] yeah ... just wanted to jump into #bzrlp :) [11:21] on canonical irc ... but lets try open means first [11:58] james_w: good. here i have another nifty topic :) ... i have upstream branch + topic branch + packaging branch (based on upstream branch) ... now i would like something like ./debian/rules bzr-topics-update :) [11:58] is there a cdbs helper already for something like this? [11:58] given that we have a patchsystem that is [11:59] so you want to take the diff from upstream->topic and drop it in to debian/patches? [11:59] james_w: yeah. i want to regularly update a patch (and create it initially, but that can be done manually) [11:59] I don't know of anything existing. [12:01] james_w: ok ... i think a cdbs helper that just looks at debian/bzr.topics with syntax: would be good here. [12:02] that could work [12:02] james_w: hmm but how to figure which upstream branch revision to use for the diff? [12:03] james_w: i think this boils down to a generic commadn that allows you to diff current head against last merged head :) [12:03] like bzr diff -r2772..2752.1.435 [12:04] e.g. 2772 == current head of topic brnach [12:04] 2752.1.435 == revision of latest merge i see in bzr log :) [12:04] given that we assume that thats an upstream branch :( [12:04] ok ... too hard task i guess for me. next topic ;) [15:31] Hi there! [15:31] is there any way to make Thunderbird not notify me about newly arrived Junk mail? [16:09] p3sho: good question. i dont think that such a feature exists [16:09] except if you say "delete junk immediatly" (maybe?) [16:35] <[reed]> asac: http://www.flickr.com/photos/lukasblakk/2588016297/ [16:35] <[reed]> :P [16:40] [reed]: hihi === rzr is now known as RzR [16:49] asac: hi , sorry w/ the delay I was busy w/ tuxguitar-1.0 [17:05] asac: is it ok now ? [17:09] RzR: you still didnt sync with the ubuntu-dev branch [17:09] I have to push it right [17:10] RzR: push what? [17:10] hum I did it [17:11] ok [17:11] I missed something [17:11] RzR: you didnt merge it [17:11] yesterday you said me to drop my branch [17:11] RzR: yeah [17:11] you didnt do that [17:11] but I dont undestand why [17:11] If I merge it [17:11] it should be ok ? [17:11] not really [17:11] I just merged to ubuntu-dev [17:12] from [17:12] not to [17:13] if i dont drop it, it will be rejected then ? since there is a branch cross ? [17:14] no ... i just ask you to drop your branch so we dont have merges of merges and so on [17:14] Ok I blindly trust you :) [17:20] asac: ok I dropped my branch I'll create it again later i have to go [18:53] asac, i've implemented a generic build-system for xul applications [18:54] asac, i've tested it with fennec, seems ok: http://www.sofaraway.org/ubuntu/tmp/fennec.png [19:04] asac, the question is now, should we patch xulrunner so that xul apps stop shipping a full copy of xulrunner, like our firefox [19:06] fta: how do you want to patch xulrunner? [19:08] see what is specific for ff, and make it generic for all xul apps [19:08] not sure what you mean :) how did you solve the issues i mentioned? [19:09] depends if we want xulapps to be standalone or depends on xulrunner [19:10] now, i can build fennec just with: --with-libxul-sdk=$(DEBIAN_XUL_DEV) --enable-application=mobile without any patch [19:11] so i assume i can do it with prism too. and try songbird too [19:11] fta: so how is the build system maintained? [19:12] shipped in xulrunner-1.9-dev and how applied? [19:13] applied by mozilla-devscripts hooks [19:13] (cdbs hooks provided by mozilla-devscripts) [19:17] fta: is the build system derived from upstream one? [19:19] exact copy [19:22] so how would the xulrunner patch look like? [19:31] bug 236613 [19:31] Launchpad bug 236613 in firefox-3.0 "firefox 3 rc with prism extension crashes when download a file." [High,Invalid] https://launchpad.net/bugs/236613 [19:32] now also claims that our prism package crushes :) [19:33] prism extension = upstream xpi, not our package [19:34] fta: no read the last comment [19:34] he claims that our package crushes system :) [19:34] asac: Ok I am back, I created the branch with the same name again, but it seems to be stuck on scan : https://code.edge.launchpad.net/~rzr/firefox-extensions/flashblock.ubuntu [19:36] http://launchpadlibrarian.net/15472047/gdb-firefox-3.0.txt [19:36] asac, donno, i'm using prism from hardy and intrepid every day, no problem [19:36] I'd say driver issue [19:36] dropping buggy_repeat caused instability for some [19:37] bug 214817 [19:37] Launchpad bug 214817 in firefox-3.0 "When I close a tab that has a Gmail page loaded, firefox crashes" [Undecided,Incomplete] https://launchpad.net/bugs/214817 [19:54] Jazzva: hi, are you still on jabbin? [19:56] asac, as expected, the build system doesn't provide a make install target [19:58] RzR: Planning to do it after exams... Would you like to give it a try? [19:58] sure [19:59] well i am bit buzy now [19:59] but I'll find you a few time for you [20:01] Well, go ahead if you want :). Just give me a notice if you start packaging it [20:01] RzR ^ [20:01] I'll do this On debian side [20:02] If I have to do something on jabbin [20:06] Jazzva: can I have a look at your .dsc ? [20:08] RzR: For Jabbin? I don't have it. I started doing it last year, then it got lost when I formatted the drive. I haven't been working on it since then... [20:08] ok [20:09] if you can unassign yourself , I'll may take it If I am too bored ;) [20:09] BTW, as reminder the mozilla metting is this sunday , I was about to forget it [20:10] RzR: Can't you just reassign it if you decide to do it ;)? [20:10] ok [20:10] I would like to do it, if you don't start working on it. I would ask you before, of course... [20:11] are there good alternatives to it ? [20:12] Pidgin, if it would implement voice talk :)... [20:12] there is a jiggle plugin [20:12] under developpement at least [20:14] Really? Cool. Didn't know that... [20:28] fta: shame ;) [20:29] fta: but well ... not a bug problem i guess ;) [20:51] asac, well, it's always the same story.. unix/packages-static [21:20] asac, problem is upstream usually doesn't want to maintain that [21:29] asac, http://paste.ubuntu.com/21663/ => that's what is produced by default now with just http://paste.ubuntu.com/21664/ and http://paste.ubuntu.com/21665/ but i didn't create removed-files and */packages-static [21:30] asac, the xul patch i mentioned earlier would be to get rid of the xulrunner dir [21:36] [reed], is fennec an active project ? [21:36] <[reed]> fta: yep, very active [21:36] [reed], in hg ? [21:36] <[reed]> first release is at the end of 2008 [21:36] <[reed]> yep [21:37] [reed], i don't see a lot of commits in there [21:37] 10 in a month [21:38] <[reed]> under mobile-browser? [21:38] yes [21:38] <[reed]> a lot of the work was happening in svn, but then it moved to hg [21:38] <[reed]> I think people have been busy lately [21:47] [reed], i expected to see that http://dougt.wordpress.com/2008/06/12/firefox-mobile-ui-proposal-aza-style/ [21:47] <[reed]> dunno [21:47] <[reed]> ask #mobile on moznet [21:47] <[reed]> I haven't been paying attention [21:48] http://upload.wikimedia.org/wikipedia/commons/2/25/Lesser_panda_standing.jpg [22:07] asac: my branch is now resurected :) [22:07] [reed], if you're looking for something to watch, try The Middleman [22:07] <[reed]> k [22:07] <[reed]> thanks [22:08] watching the pilot, it's funny [22:36] RzR: ok. i still dont get why you claim that it fixes the crash you are closing without even being able to reproduce ;) [22:36] but i will drop that line from changelog [22:37] asac: because this comment it not related to the original crash [22:38] the guy talk about the whitelist option [22:39] I am pretty sure even without flashblock the swf plugin crash his browser [22:39] ok. but then why close the bug in flashblock upload? [22:40] asac: Just a question about merging from debian... Should we get debian version, resolve the conflicts, then add that to extensions .ubuntu branch? [22:40] because this version is obsolete [22:40] Since I'm gonna do few of them... [22:40] so all its bug should be closed soon or later [22:41] well I understand what you mean asac [22:42] yep [22:42] but not in upload [22:42] the main reason I decided to close it , it was because this bug was a but confusing, I my last comment I invited them to open new bug on upcoming version [22:42] ill upload this now to intrepid [22:42] ok [22:42] RzR: i agree that this bug should be closed. but not fixreleased in upload ... set it to invalid if the user cannot provide the required info anymore [22:43] this makes sense [22:43] I was too lazy [22:44] shame on me :) [22:45] anyway thanks for all asac , excuse me about all the process I am discovering ubuntu workflow everyday [22:45] asac, just got the permission from google to use their logos for prism webapps [22:45] now bedtime ... I'll be back for the meeting === RzR is now known as rZr [22:48] fta: permission granted to who? [22:49] to redistribute in ubuntu [22:49] hard to say if we want that [22:49] ill ask [22:50] that's what i've asked to google. permission to ship google icons within prism packages shipped in ubuntu [22:50] fta: if we want to claim an exception we need permission to distribute it in ubuntu and derivatives [22:51] otherwise it has to go to multiverse [22:52] but good to know that you retrieved an answer :) [22:52] how long did it take? [22:52] Date: Thu, 7 Feb 2008 13:59:14 -0800 [22:53] heh [22:53] i have no mouse cursor in fennec [22:54] am i supposed to run that within something ? [22:54] a mobile env ? [22:54] i dont think so [22:54] but who knows :) [22:54] hildon maybe [22:54] is there any binary module? [22:55] http://paste.ubuntu.com/21663/ [22:57] i doubt that they want hildon. maybe something else, but as i dont see any special binary components i wouldnt think that thats the case [22:58] they ship a debian dir as installer, it contains: fennec.links:/usr/share/applications/hildon/fennec.desktop etc/others-menu/0112_fennec.desktop [22:59] and a dbus service [22:59] which? [23:00] [D-BUS Service] [23:00] Name=org.mozilla.fennec [23:00] Exec=/usr/local/fennec/fennec [23:01] /usr/local/fennec/fennec starts the browser? [23:01] (fixed path obviously= [23:01] ) [23:03] it's their stuff, i didn't use that [23:04] they most likely develop for nokia atm [23:04] nokia has debian + hildon [23:04] but well ... i doubt that this is the problem [23:04] well, it was just a test bed for my xul app build system [23:04] wonder which window manager nokia uses [23:04] maybe matchbox? [23:05] fta: yeah. i dont think that we need a package now. we can package it with volunteers during the next open developer week ;) [23:05] session "packaging xul apps" :) [23:06] fta: try to use xephyr xserver + matchbox window manager [23:06] :) [23:07] someone asked me (and the ubuntu-mozillateam m-l) how to package a xul app today [23:08] fta: yeah ... lets get the build system app ... make it fail safe for beginners [23:08] document in wiki and point them to it [23:08] fta: mz mailing list? [23:08] strange ... didnt get the mail [23:08] ubuntu-mozillateam at lists.ubuntu.com [23:10] asac, where should i write that in the wiki ? [23:10] https://lists.ubuntu.com/archives/ubuntu-mozillateam/2008-June/thread.html [23:10] the mail did not go through [23:10] :) [23:10] anyway. [23:11] fta: https://wiki.ubuntu.com/MozillaTeam/XulApps equiv to Firefox3Extensions [23:11] and https://wiki.ubuntu.com/MozillaTeam/XulApps/Packaging with detailed how to [23:11] package [23:11] seems he wants to package this: http://code.google.com/p/aliwal/ [23:12] .dmg [23:16] so far, i used the full configure.in, meaning a lot of unnecessary build-deps [23:16] of course, each xul app could patch it as it's installed before the patch [23:17] not really inspiring :) [23:17] ? [23:17] to patch configure.in :) [23:17] what do you mean ? [23:17] yep, you have a better idea ? [23:17] wrong word ... not really nice ;) [23:19] not sure. i think the problem is the lack of modularization of configure.in [23:19] i started to do it once for the enigmail build system ... but back then configure.in was much dirtier than nowadays (though its still a heap of shit :)) [23:19] so maybe its worth taking a look [23:19] again [23:20] but before we are putting any serious work into this, we have to figure if configure.in is not dying soon :) [23:40] never read anything about that so far [23:42] bug 211212 [23:42] Launchpad bug 211212 in moblin-ui-framework "Network-manager process hangs after resume from S4" [Medium,Fix released] https://launchpad.net/bugs/211212