[00:54] off [03:08] <[reed]> asac: ping? [08:35] [reed]: yes? [08:58] debian bug 470128 [08:58] Debian bug 470128 in icedove "xulrunner: FTBFS with libnss3-dev=3.12.0~beta2-1" [Serious,Open] http://bugs.debian.org/470128 [10:16] [reed]: are b4 rc2 builds looking good? [10:36] asac: that bug is old [10:36] yes [10:36] i know [10:37] its just that some guy told me that i should fix that bug id in debian [10:37] just wanted to know what bug that is [10:38] ah [10:52] * asac hates if people just refer to bug ids without including at least a title ;) [11:11] i really go crazy if i read the bug 178558 [11:11] i mean ... i posted a test package and are there any comments about that recently? no, of course not [11:11] those folks don't deserve to get a fix ;) [11:17] answer: https://bugs.edge.launchpad.net/ubuntu/+source/xulrunner-1.9/+bug/178558/comments/36 [11:17] hope that is still "nice" enough [11:18] (in line with the ubuntu etiquette) [11:22] Launchpad bug 178558 in xulrunner-1.9 "Firefox 3.0 makes everything annoyingly huge" [High,In progress] https://launchpad.net/bugs/178558 [12:00] haha [12:02] yeah :) ... thats how things work ;) [16:07] asac: ping [16:18] cwong1_: hi [16:19] asac: Couple of things I want to discuss with you: [16:19] asac: 1) Do you have any changes that you want to check in? [16:19] yes. sorry that i was rather in active in the recent past [16:19] got dragged in lots of other things [16:19] np [16:20] we all do :) [16:20] cwong1_: i have a hald ported ffox 3 beta3 thing [16:20] s/hald/half/ [16:20] but maybe we should go directly for beta 4 which is about to be released today or so [16:20] at least the CVS is tagged already [16:20] whatelse? [16:20] that sounds good to me. I was going to suggest the same. [16:21] 2) I want to hold off the switch to Xulrunner [16:21] great [16:21] hold of? [16:21] thought we wanted to use a mobile specific fork [16:22] I have a lot of bugs to work on and I dont think I can get to it in the next week or 2 [16:22] unless you have the time :) :) [16:22] i can look [16:22] maybe we can spin both binaries from the same source [16:22] (as we already have all in there) [16:22] but lets first do beta4 [16:23] is jimmy avail? [16:23] ok beta4 first [16:23] he also looking into bugs. [16:23] btw do you know of a plugin that is equivalent to window media plugin [16:23] ? [16:23] for that i would need to know what "window media plugin" is :) [16:24] what does it do compared to totem? [16:24] it plays window media files and totem doesn't [16:24] <[reed]> asac: yep, releasing today [16:25] [reed]: thanks [16:25] cwong1_: totem doesn't? [16:25] <[reed]> unless something disastrous happens [16:25] <[reed]> but hopefully not ;) [16:25] cwong1_: you have a test file? [16:25] <[reed]> asac: can you comment in mozilla bug 418885 about what you think? [16:25] [reed]: i think i will release the bits in a few hours then :) [16:25] ubotu: wake up [16:26] asac: I will send u a test site shortly... [16:26] cwong1_: i thought it was .wmv files. to play those its just a matter of codec (unless you are talking about DRM) [16:26] asac: is there a wmv codec for totem? [16:27] Bug 418885 – Firefox 3 shouldn't require GTK+ 2.10 (edit) [16:27] [reed]: that sounds lame ;) [16:27] <[reed]> I want to WONTFIX it [16:27] <[reed]> ;) [16:27] [reed]: reading the title i would say "yes" [16:27] but let me read the content in a few more minutes [16:28] cwong1_: i am not sure if it plays all ... at least i have been watching wmv files happily for quite some time [16:28] <[reed]> we're not changing our minimum (of 2.10), but there is a possibility we'll add some checks to at least allow 2.8 to work somewhat [16:28] <[reed]> even if they can't print [16:28] cwong1_: other players that might play more files are vlc + mplayer [16:28] cwong1_: those should ship all the codecs they support out of the box [16:29] cwong1_: i think to get all available codecs you need to install: [16:29] gstreamer0.10-plugins-bad-multiverse [16:29] gstreamer0.10-plugins-ugly-multiverse [16:29] gstreamer0.10-plugins-ugly [16:29] gstreamer0.10-plugins-bad [16:29] asac: I knew about mplyaer, but never heard of vlc. I will give them a try and see. [16:30] gstreamer0.10-plugins-good [16:30] I did install gstreamer0.10* and still couldn't play some of the files [16:30] I will find the site and let you know in a few minutes.. [16:31] cwong1_: gstreamer0.10-ffmpeg [16:31] then i don't know [16:31] only thing i can imagine is that those files are DRM enabled ... for which you are unlikely to find a free implementation i guess [16:31] cwong1_: thanks ... i will be travelling in a few, but will catch up on work later - so probably a better time for you :) [16:32] ok [16:32] later [16:34] [reed]: hmm ... if the patch work ten i don't know why not to take it :) ... it would allow us to provide backports of firefox 3 for dapper for instance [16:34] which is still supported for quite some time [16:34] <[reed]> asac: well, see caillon's comments [16:34] <[reed]> where he plans on backporting Fx3 back to RHEL 2.1 [16:37] yes ... just read that [16:37] i doubt that he will succeed on a gtk 1 port ;) [16:39] [reed]: http://steelgryphon.com/blog/?p=101 [16:39] i always understood the phrase "Older distros will be able to have build-time support/workarounds as necessary, but Mozilla will not ship or test builds for older platforms. " [16:40] that those tweaks should go the mozilla CVS ... though disabled during compile time [16:40] note that there is a conflict here: distros are not allowed to ship arbitrary patches without getting review on them first [16:41] huge patches because of old gtk+ should thus be reviewed in bugzilla [16:41] we could keep them unapplied then, but that makes it harder to reuse those. [16:41] Mozilla bug 418885 in OS Integration "Firefox 3 shouldn't require GTK+ 2.10" [Normal,Assigned] http://bugzilla.mozilla.org/show_bug.cgi?id=418885 [16:41] Sorry, I don't know anything about wake up - try searching on http://ubotu.ubuntu-nl.org/factoids.cgi [16:41] for instance caillon will create patches for supporting the old gtk [16:41] maybe he submits them to mconnor who either trusts him or tries to get a review [16:42] from someone else [16:42] i guess it would end up: post a bug and get a review+ [16:42] but then its hard for other distros to find all the required pieces and all that [16:42] ... so in the end a hard question [16:42] but for now i agree, that 2.8 doesn't need to be supported [16:42] at least not at runtime [16:43] i am fine with wontfixing this === Ubulette_ is now known as Ubulette [17:59] hi [17:59] wontfix for me too [18:04] asac, i didn't add libsqlite3-dev on purpose [18:05] it's not used and will not be in hardy anyway [18:57] right ... but it doesn't hurt. otherwise its not even really optional === asac_ is now known as asac [19:28] asac, could we stop naming nspr with the gecko version ? [19:29] Ubulette: why? [19:29] now, nspr claims #define PR_VERSION "4.7.1 Beta" [19:30] or a last put the full version in front [19:30] like 4.7.1~b~1.9~b4 [19:31] Ubulette: how does the version read atm? [19:31] 4.7.0~1.9b3-0ubuntu2 [19:31] ie, it doesn't respect the full nspr version [19:32] same for nss [19:34] Ubulette: i have no problem to squeeze in a ~beta~ [19:34] i think the main point was that there was no real version at some point [19:34] so mike just choose somethign reasonable [19:35] Ubulette: is there a tag that reads liks 4.7.1 beta now? [19:35] i think the problem was that they didn't tag the version with an nspr'ish version... but just with the firefox release tag [19:37] doesn't matter much, we can still fetch it using a gecko tag but name it like we want [19:39] NSPR_4_7_1_BETA1 [19:39] http://paste.ubuntu.com/5549/ [19:41] Ubulette: yes ... but it wasn't really clear how to name it at that point [19:41] i think the error was to use 4.7.1~1.9 ... instead of 4.7.1~~1.9 [19:41] so yes ... go for 4.7.1~beta~1.9+... if you have a better feeling about that [19:42] k [19:44] NSS_HEAD_BEFORE_RFC4507BIS [19:58] ** The format of the version string is [19:58] ** ".[.] []" [19:58] it's even documented [20:11] asac, it still doesn't solve how I can name my snapshots to keep them intertwined with the official releases we do here [20:12] nspr_4.7.1~beta~~cvs20080310t1054.orig.tar.gz is no good [20:15] indeed [20:16] why not 4.7.1~beta~1.9b4~cvs... ? [20:17] how could mozclient identify 1.9b4 just from nspr sources ? [20:22] asac: hey [20:24] RainCT: hi [20:24] RainCT: i have a question :) [20:24] why did you remove the upstrema sources from your branch? [20:24] (the extension?) [20:25] asac: because I find it easier to work with it that way [20:25] how easier? [20:25] what steps are easier? [20:25] or is there some problem with that? [20:26] well ... i would love to have a single approach for all extensions ... if you really want to keep it that way, then i am fine [20:26] point is that extensions are so tiny that I'd prefer not to have patch system in them [20:26] dunno.. I don't remember the actual reason, but I also find it cleaner and that [20:26] ok. [20:27] RainCT: so what do you want? [20:27] shouldn't be a problem to add it again though.. [20:28] well.. help on getting it working :P [20:28] maybe consider that. its easier for new contributors to just have the upstream sources in there ... and so on. [20:28] anyway ...what doesn't work? [20:28] asac: I tried installing it but I don't see it in Firefox 3 [20:29] hmm ... did you try to use the xpi.mk i introduced recently in mozilla-devscripts? [20:29] (don't know if it works with Firefox 2 as I don't have it installed anymore) [20:29] that should automize all [20:29] no, looking at that :) [20:29] you just need to specify an extension name + a build command [20:29] (that expects a .xpi to be produced) [20:29] RainCT: https://wiki.ubuntu.com/MozillaTeam/Firefox3Extensions/Packaging [20:30] thats the document i used at my extension packaging session [20:30] it should work for all trees that can produce a proper .xpi [20:30] oh, forgot to read that at the end.. :P [20:30] s/that/the packaging season log [20:31] is xpi.mk already in Hardy? [20:31] yes [20:31] you can look at some other extension [20:31] e.g. http://bazaar.launchpad.net/~jetsaredim/firefox-extensions/mozgest.ubuntu/annotate/jgreenwa%404lom-20080227214222-fruivuu7tfxwbcka?file_id=rules-20080227041249-6uckd8mo559wu2sf-6 [20:31] asac, you can remove the mention of your ppa, it's no longer needed [20:31] where? [20:31] RainCT: https://wiki.ubuntu.com/MozillaTeam/Firefox3Extensions/Packaging [20:31] ah [20:31] Ubulette: its in hardy, right? [20:32] !info mozilla-devscripts [20:32] Package mozilla-devscripts does not exist in gutsy [20:32] !info mozilla-devscripts hardy [20:32] mozilla-devscripts (source: mozilla-devscripts): Collection of dev scripts used by Ubuntu Mozilla packages. In component universe, is optional. Version 0.05 (hardy), package size 14 kB, installed size 108 kB [20:32] yes [20:32] Ubulette: ok done [20:32] thx [20:33] btw ... if all goes well mozilla-devscripts will go into main [20:33] to produce the translation .xpis [20:34] you can remove the whole line, it's already listed above [20:34] oh :) [20:34] ok done :) (2nd) [20:38] asac: Vcs-Bzr is accepted now (no XS- necessary) [20:39] * RainCT says this because of XPI.TEMPLATE [20:39] oh :) [20:39] thanks [20:39] np :) [20:40] also, mozilla-devscripts in hardy is version 0.05 but the template says 0.5~ [20:40] thanks [20:40] done [20:41] RainCT: right .. but thats a lower bound and should be ok [20:41] i have another great candidate: all-in-one-sidebar [20:41] its tri-licensed [20:42] :) [20:42] and appears to be 3.0 compatible already [20:42] asac: will xpi.mk autodetect the .xpi if it's created on build time and inside debian/? [20:43] yes [20:43] oh [20:43] no [20:43] RainCT: why is it in debian/ ? [20:43] oh, there's a BUILD_COMMAND option [20:43] yes [20:43] BUILD_COMMAND ... and you can also explicitly name the .xpi to pack-up [20:43] I've this right now: perl ./create_xpi.pl debian/adblockplus.xpi [20:44] yeah ... just use adblockplus.xpi [20:44] clean will auto delete it form there [20:44] should I change it to this? MOZ_XPI_BUILD_COMMAND = perl ./create_xpi.pl adblockplus.xpi [20:44] yes [20:44] makes sense imo [20:44] if you have issues let me know :) ... this xpi.mk certainly doesn't cover all cases yet :) [20:45] use :) [20:45] aeh := [20:45] MOZ_XPI_BUILD_COMMAND :perl ./create_xpi.pl ... [20:45] MOZ_XPI_BUILD_COMMAND := perl ./create_xpi.pl ... [20:45] will it automatically add the dependencies for unzipping and all that? [20:47] yes [20:47] everything thats needed by itself will be added [20:47] if you need anything else because of perl you need to add that [20:48] oh wait :) [20:48] no right :) ... oyu confused me ;) [20:48] all shold be fine ... just add mozilla-devscripts to build-depedns [20:49] ok, cool :) [20:49] create_xpi.pl needs zip; I've to add this one manually, or? [20:50] RainCT: yes [20:50] xpi.mk just needs unzip [20:53] asac: http://paste.ubuntu-nl.org/59182/ [20:55] RainCT: you need to drop your .install files [20:55] and your .links [20:55] that happens automatically now [20:56] or don't you have any? [20:56] oops, right [20:56] :$ [20:56] good [21:05] asac: adblock-plus should also be compatible with those: iceweasel (>= 1.5) | seamonkey-browser (>= 1.0) | iceape-browser (>= 1.0) | midbrowser (>= 0.2) [21:06] what should I do with them? just remove that from debian/control? [21:06] no you can add more applications to xpi.mk [21:07] RainCT: look in xpi.mk file for MOZ_XPI_MOZILLA_DIRS [21:07] default is firefox-addons i guess ... you can add whatever you want [21:07] so add iceape seamonkey midbrowser iceweasel [21:07] and firefox + firefox-addons [21:09] i use it with prism too but it's tricky [21:09] as prism doesn't show the red sign [21:10] RainCT: is midbrowser in upstream install.rdf? [21:10] that would be cool ;) [21:10] asac: yes [21:20] Ubulette: how about renaming the mozclient branch? [21:20] you think its mature enough to get a real name ;) [21:20] hehe [21:20] yes [21:21] will u do? [21:21] * Ubulette thinking of a name not too long [21:22] i'm up since 6am so my mind is blank [21:23] Ubulette: hehe :) ... how about mozilla-devscripts for now? [21:23] locally, i will call it mozilla-devscripts.dev [21:23] or without .dev is good too [21:23] can we cann it that way on launchpad? [21:24] i think .dev is good for the development branch [21:24] s/cann/call/ [21:24] https://code.edge.launchpad.net/~mozillateam/mozilla-devscripts/mozilla-devscripts [21:25] we just have one branch so it's okay like that [21:25] i am fine with that ... but atm we don't even have nont . branches for other things [21:25] thats ok [21:26] maybe we should once again rename the branches for all the packages :-D [21:26] remove .dev ... rename .head -> .dev [21:26] :) [21:26] and releast branches get .gutsy and so on [21:26] maybe another day later this week [21:27] but rather push that forward till hardy+1 :) [21:27] or during UDS ;) [21:27] yeah ;) [21:27] branch clean up session :) [21:34] I have to update my buildbot now. [21:34] update to what? [21:34] woody? [21:34] ? no [21:34] it uses mozclient [21:35] just kidding [21:35] ah ... cool [21:36] you were supposed to comment on my ff3 sdk bug last week [21:39] thanks for the reminder. completely dropped off of my radar [21:52] we should start a "package 5 extensions a day" [21:53] yeah :) [21:53] the problem is the list of viable extensions though [21:54] so "investigate 5 extensions a day" would be extremely helpful though [21:55] if we#d had a long list of extensions that are ready (license + compatible) we could initiate an "extension packaging marathon" :) [22:04] blog about it on planet [22:07] ur right [22:15] asac: great, adblock-plus works now (after switching to xpi.mk) :D [22:16] asac: what did you say, does it need a FFe? [22:24] RainCT: no it doesn't just open a bug and refer in it to the general feature freeze exception for ffox extensions [22:25] subscribe mozillateam next [22:25] and once Ubulette or me confirmed you can upload :-D [22:25] (we agreed to ack those to make universe-release team happy) [22:25] asac: ok. upload to revu or link the branch? [22:26] RainCT: you have upload rights? [22:26] yes [22:26] just link the branc [22:26] h [22:26] allright [22:26] cool [22:27] Ubulette: you get mozillateam subscribed bug mail? [22:27] ubotu, bug #191954 [22:27] (e.g. thats != mozilla-bugs) [22:27] ubotu: you're slow :/ [22:27] has been dead for me today [22:27] oh [22:28] bug 191954 [22:28] ubotu: you there? [22:28] most likely a dead-lock :) [22:28] seems like he is ill... yesterday we had an 1 hour lag in #ubuntu-youth lol [22:28] (1 hour, really :)) [22:32] asac: ok, subscribed [22:33] asac: btw, what do you want to do with mozilla-firefox-adblock? remove it and add a dummy package to adblock-plus or convert mozilla-firefox-adblock itself into a dummy package, or just remove it? [22:36] a dummy is needed for the transition [22:38] yes [22:39] yes, but were? in adblock-plus, or? [22:39] yes [22:40] you add dummy packages depending on mozilla-firefox-adblock in there [22:40] so you get an upgrade [22:40] and you add conflicts/replaces to adblock main package [22:40] (verseioned conflicts/replaces) [22:41] okay. a last question, should I add the dummy package now or once mozilla-firefox-adblock has been removed? [22:41] right from the beginning [22:41] removal happens after migration [22:42] oh ... take care that the version is higher than the on of mere -adblock [22:42] maybe you need to add an epoch for that [22:42] why? [22:42] otherwise -adblock users will not be auto transitioned [22:42] ahhh right [22:42] (as the dummy package version would be lower ... so it couldn't pull in the new package as a dependency) [22:43] that's evil :) [22:43] ah no, it's ok :) [22:43] 0.7.5.3-0ubuntu1 (new) VS 0.5.3.043-4ubuntu1 (old) [22:44] asac, "You received this bug notification because you are a member of Mozilla Team, which is a direct subscriber." so i guess it's ok [22:47] yes great [22:47] RainCT: yes ... you have luck [22:50] and you should stop the sync/merge from debian [22:51] if any [22:52] Launchpad bug 191954 in ubuntu "[needs-packaging] Adblock Plus" [Wishlist,Triaged] https://launchpad.net/bugs/191954 [22:52] Sorry, I don't know anything about you're slow :/ - try searching on http://ubotu.ubuntu-nl.org/factoids.cgi [22:52] Sorry, I don't know anything about you there? - try searching on http://ubotu.ubuntu-nl.org/factoids.cgi [22:52] :) [23:04] maybe i should drop my nick and use fta instead [23:08] RainCT: so you will upload it to debian too ? [23:13] Ubulette: is mozilla-devscripts in Debian? [23:13] no [23:13] then not [23:13] and our paths for mozilla products are different too [23:14] then you should target hardy in debian/control [23:14] true.. :) [23:15] and drop iceape-browser [23:15] and iceweasel too? [23:15] and iceweasel too [23:15] yep [23:15] ok [23:17] * RainCT had added those as there are some people which use Ubuntu but install the mozilla packages from Debian.. but now that he thinks about it, those aren't probably many :P [23:17] :) [23:17] you can rename the branch too then [23:17] otherwise, it looks ok to me [23:18] I like rules files so simple :) [23:18] * RainCT too :D [23:19] that was the whole purpose of mozilla-devscripts [23:30] Ubulette: if you leave an ack on the report I'll upload it tomorrow :) [23:31] well .. keeping iceweasel + iceape is sane [23:31] we should contribute it to debian at somepoint [23:31] but i don't care hard [23:32] it's still targeting unstable, i was waiting for at least this commit before giving an ack [23:32] yes [23:33] Ubulette: yes, my connection is slow :/ [23:33] it's up now [23:33] RainCT: you can use dch -r to prepare changelog for release [23:33] and please use 0ubuntu1 ... in case you didn't yet [23:33] I did :) [23:33] still unstable? [23:33] strange [23:34] unless you are on debian of course :) [23:34] rev 12? [23:34] good here [23:34] should be: adblock-plus (0.7.5.3-0ubuntu1) hardy; urgency=low [23:34] yes [23:35] there's a storm here [23:37] we had enough storms for this year already [23:45] good night :) [23:45] 'night