2008/10/01/#ubuntu-mozillateam.txt

[00:01] <fta> well, looks fine to me
[00:08] <asac> fta: yeah.
[00:09]  * asac  remove --purge libnspr4-0d from chroot
[00:13] <fta> jdstrand, what else is needed in the bug?
[00:14] <jdstrand> fta: what's the bug number again? (backscroll has got me down...)
[00:14] <fta> bug 276437
[00:14] <ubottu> Launchpad bug 276437 in seamonkey "security upgrade of seamonkey 1.1.12" [Undecided,Fix committed] https://launchpad.net/bugs/276437
[00:15] <jdstrand> looking...
[00:15] <asac> i guess the .diff.gz and .dsc
[00:17]  * asac off doing something else
[00:18] <jdstrand> fta: yeah, I'll need the full source. the debdiff looks good (need to look at it more closely)
[00:18] <fta> jdstrand, full source is already there
[00:18] <jdstrand> fta: once I have the dsc and diff.gz, please just document in the bug your testing
[00:19] <jdstrand> fta: I'm sorry, I meant full source package
[00:21] <jdstrand> fta: once I have the complete source package, and the testing is documented, I'll upload
[00:22] <fta> jdstrand, hm, i've made the intrepid one complete (-sa), and the hardy one as -sd, so intrepid must be uploaded first
[00:22] <fta> unless you prefer the opposite
[00:23] <jdstrand> fta: well, I do for complicated reasons dealing with the security buildd...
[00:23] <jdstrand> fta: sorry about that
[00:23] <jdstrand> fta: currently, we have a different buildd for security
[00:24] <jdstrand> fta: and that buildd doesn't have intrepid yet, cause it's not released
[00:24] <fta> so, both could include the orig.tar.gz ?
[00:24] <jdstrand> fta: so I need to build hardy in the security buildd, push it to LP, then I should be able to build intrepid
[00:24] <asac> jdstrand: intrepid?
[00:24] <asac> why do you need to care for intrepid at all?
[00:24] <asac> ;)
[00:25] <jdstrand> asac: well, I don't, if you or someone else uploads it
[00:25] <fta> i thought i would push to intrepid myself
[00:25] <jdstrand> that's cool-- I'll need -sa for hardy though
[00:25] <asac> jdstrand: yeah. fta will upload that
[00:31] <asac> ppa publishing takes ages today :/
[00:31] <fta> yep, often longer than the build
[00:32] <asac> in this case unfortunate ... everything  is there except the binary-all packages :)
[00:32] <asac> which come from i386
[00:32] <asac> and are not yet published
[00:32] <fta> take it from the lp +archive page
[00:32] <asac> fta: thats the problem ;)
[00:32] <asac> fta: as soon as the .deb is there it will be quite soon in the archive index
[00:32] <asac> but its not there yet :(
[00:33] <fta> oh, i thought librarian had it immediately
[00:33] <asac> i thought so too. most likely that has changed
[00:34] <asac> amd64 bits took 15 minutes to show up after build
[00:35] <asac> anyone interested in fixing ifupdown to honour a global nm setting that would make ifupdown stop from auto upping interfaces?
[00:35] <asac> i hate that source ;)
[00:35] <asac> its one file ... and its written in tex ;)
[00:36] <asac> last time i touched it i thought it was fun ... but now i think that one time is enough for a lifetime :/
[00:36] <asac> http://people.ubuntu.com/~asac/ifupdown.nw ;)
[00:37] <fta> i should stop caring about that : http://www.sofaraway.org/ubuntu/tmp/karma.png  it's too volatile
[00:37] <asac> thats what i am saying ;)
[00:37] <asac> but still nice experiment to keep running ;)
[00:38] <asac> maybe also comparison of bug, bzr and whatever ;)
[00:38] <fta> i'd be curious to see yours
[00:39] <asac> i'd rather not ... when i do the whole night bug triaging and then see my karma not moving at all i would certainly get depressed :-P
[00:39] <fta> i am
[00:40] <asac> yeah. simple cure: dont look at it ;)
[00:40] <asac> fta: do you have the script?
[00:40] <asac> i can add that to my crontab ... or even you can run it on your own i guess :-D
[00:41] <asac> so ... anyone has a C snippet to parse keyfile formats? e.g.
[00:41] <asac> [group]
[00:41] <asac> key1=value
[00:41] <asac> key2=value2
[00:42] <fta> and do what with it?
[00:42] <asac> fta: i need to add a code snippet to http://people.ubuntu.com/~asac/ifupdown.nw that parses /etc/NetworkManager/nm-system-settings.conf
[00:43] <asac> and stops auto upping interfaces when there is
[00:43] <asac> [ifupdown]
[00:43] <asac> managed=true
[00:43] <asac> ;)
[00:43] <pwnguin> asac: iniparser?
[00:43] <asac> pwnguin: yeah. but i guess i cannot use glib :)
[00:44] <pwnguin> ?
[00:44] <pwnguin> oh, gnome probably doesn't have such a thing in their libs
[00:44] <asac> pwnguin: libglib has
[00:44] <asac> pwnguin: but i cannot use it
[00:44] <pwnguin> http://ndevilla.free.fr/iniparser/
[00:44] <asac> people would kill me to pull in libglib dependency on ifupdown
[00:45] <pwnguin> heh
[00:45] <fta> jdstrand, all the files are in the bug now
[00:45] <asac> pwnguin: is that packaged?
[00:45] <pwnguin> i think people would kill you for changing ifupdown file syntax just the same
[00:45] <asac> hehe yeah. thats not my plan - i hope
[00:46] <pwnguin> in gentoo
[00:47] <asac> quite a big thing for this tiny ifupdow tex thing
[00:48] <asac> pwnguin: unfortunate that the tarball doesnt have a license info
[00:48] <pwnguin> the bigger challenge is that it's probably not secure
[00:49] <pwnguin> asac: the website suggests MIT license
[00:49] <asac> pwnguin: yeah websites can tell a lot of things ;)
[00:49] <asac> tarballs too of course ;)
[00:50] <pwnguin> but the author seems to be alive at least
[00:50] <pwnguin> so you could mail him and ask for a better clarification
[00:50] <pwnguin> wait
[00:50] <pwnguin> what's LICENSE
[00:51] <asac> he?
[00:51] <pwnguin> in the tarball
[00:51] <asac> hehe
[00:51] <asac> my fault i guess
[00:52] <asac> i also found libini now ;)
[00:53] <asac> pwnguin: why isnt iniparser secure?
[00:53] <pwnguin> good question; i was thinking that sort of thing should be fuzzed
[00:53] <pwnguin> but i guess if it's a parser, those are normally built to reject invalid input
[00:53] <asac> pwnguin: no
[00:54] <asac> pwnguin: its only about a single key in one config file ;)
[00:54] <asac> pwnguin: yeah. and that file is only writable by root usually
[00:55] <fta> asac, only a few lines of C are needed for that
[00:56] <pwnguin> or perhaps a regex ;)
[00:57] <fta> in perl, it's piece of cake
[00:57] <pwnguin> in fact, i think ini is fairly regular; just sections and key pairs
[00:59] <asac> true.
[00:59] <asac> there are also comments though
[00:59] <pwnguin> touche
[00:59] <asac> not sure how the comment delimiter can be escaped
[00:59] <pwnguin> i dont think it can
[01:00] <asac> i think even if multiline values are possible, we probably dont need this for our use case
[01:00] <pwnguin> new line usually terminates comments
[01:00] <fta> asac, could I push sm to intrepid, or should i wait for someone to confirm the bug ? (motu-something?)
[01:00] <asac> pwnguin: no .. i mean if you want a value with a ; in it ;)
[01:00] <asac> e.g.
[01:00] <asac> value=my;thing
[01:01] <asac> fta: i doubt that anyone will complain ;) ... but you never know :-P
[01:01] <pwnguin> well, you've got the source code to a couple implementations ;)
[01:01] <asac> pwnguin: yeah
[01:02] <asac> I'll take the MIT one i think ;)
[01:02] <asac> at least its pure C
[01:02] <fta> asac, so?
[01:04] <asac> fta: upload :)
[01:05] <asac> fta: it will be stuck in upload queue until beta is released (or in case a release manager clears the queue)
[01:05] <asac> but thats ok
[01:05] <asac> in case RMs dont have time it will automatically go in after beta
[01:06] <fta> ok
[01:06] <asac> you can probably also kindly ask RMs after 24 hours or so
[01:29] <fta> Bug 246820
[01:29] <ubottu> Launchpad bug 246820 in prism "Google Gears does not install on Prism" [Unknown,Confirmed] https://launchpad.net/bugs/246820
[01:31] <fta> [ubuntu/intrepid] seamonkey 1.1.12+nobinonly-0ubuntu1 (Waiting for approval)
[10:57] <fta2> http://launchpadlibrarian.net/18102344/buildlog_ubuntu-intrepid-hppa.seamonkey_1.1.12%2Bnobinonly-0ubuntu1_FAILEDTOBUILD.txt.gz
[11:17] <asac> fta2: only hppa?
[11:17] <asac> thats fine then
[11:17] <asac> most likely imagemagick didnt built or is outdated
[11:17] <asac> you could retry in 24 hours (e.g. use launchpad "retry" button)
[11:34] <Kamping_Kaiser> hi all
[11:34] <asac> hi Kamping_Kaiser
[11:35] <Kamping_Kaiser> asac, evening
[11:56] <armin76> asac: thats okay? die :P
[12:03] <asac> why do you have that much sympathy for hppa?
[12:06] <Kamping_Kaiser> hppa? isnt that just an alpha wannabe?
[12:06] <armin76> i have sympathy for all arches :P
[12:28] <fta2> asac, yep, only hppa
[12:29] <fta2> asac, how come ff/tb are not listed in the CVE pages on lp ?
[12:29] <fta2> asac, eg: https://bugs.edge.launchpad.net/bugs/cve/2008-4060
[12:35] <fta2> asac, you said you fixed ff3.0 yesterday, is it in yet?
[12:35] <fta2> doesn't seems to be
[12:36] <fta2> -s
[12:53] <asac> fta2: huh?
[12:53] <asac> fta2: firefox and thunderbird usually dominate the USN page ;)
[12:53] <asac> fta2: http://www.ubuntu.com/usn
[12:54] <fta2> maybe but lp tracks CVEs and ff/tb are not listed, that's too bad
[12:54] <asac> fta2: not sure if the security team uses that feature seriously
[12:54] <asac> jdstrand: ^^ ?
[12:58] <fta2> Bug 154749
[12:58] <ubottu> Launchpad bug 154749 in seamonkey "seamonkey installer bin will not oper in ubuntu desktop ver 7.10" [Undecided,New] https://launchpad.net/bugs/154749
[13:01] <fta2> doesn't seem to be the ubuntu package
[13:11] <fta2> asac, where is your fix from yesterday? the ff-3.0 missing symlink
[13:12] <asac> fta2: isnt that in .head?
[13:13] <asac> *sigh*
[13:13] <asac> will do that after lunch
[13:13] <asac> fta2: feel free to fix it
[13:13] <asac> if you cant wait ;)
[13:14] <asac> or wait a sec ;)
[13:14] <fta2> i wanted to do it yesterday, you said in #u-desktop that you already did it
[13:15] <asac> i know what i said
[13:15] <asac> its on my laptop ;) ... which is in my bag
[13:20] <fta2> should i wait? or what?
[13:21] <asac> fta2: no ... go ahead ... its such a minimal change that i dont feel any loss :)
[13:26] <fta2> did you do something smart for the 3.1 merge?
[13:26] <fta2> or just a symlink
[13:29] <asac> fta2: 3.1 merge?
[13:29] <asac> fta2: we have the _VER variable. i used that
[13:30] <asac> i think i added
[13:31] <fta2> nm, i'll manage
[13:32] <asac> k
[13:32] <asac> fta2: can you fix that in .head pleaes?
[13:33] <fta2> sure
[13:33] <asac> 3.0.head i mean
[13:33] <asac> in theory there should be no difference in the code
[13:33] <asac> we have the _META ifdef for that
[13:39] <gnomefreak> wasnt 3.0.3 released a while ago? i just got email about release email date was 9/26/08 i thought we had it a while before than
[13:43] <asac> gnomefreak: we released a bit early yeah.
[13:43] <asac> (wasnt that a great idea ;))
[13:43] <gnomefreak> i liked it ;)
[13:44] <jdstrand> fta, asac: the CVE database in LP leaves a lot to be desired-- it is in no way surprising that ff and tb aren't listed
[13:44] <jdstrand> basically, LP imports the CVEs
[13:44] <jdstrand> then you either have to file a bug, then 'Link to CVE' in the bug
[13:45] <jdstrand> if a comment has CVE-XXXX-YYYY in it, then LP will do the 'Link to CVE' step for you
[13:45] <jdstrand> (regardless of if you want it to or not)
[13:46] <asac> jdstrand: ok ... so a CVE: XXXX-xxx changelog parsing feature is missing ;)
[13:46] <fta2> no, it worked for sm
[13:46] <jdstrand> asac: I think it'll see it when things are Fix Released and posted in the bug
[13:47] <jdstrand> (cause it shows up as a comment)
[13:47] <asac> james_w: do we have hardy backport packages for current bzr and bzr-builddeb?
[13:47] <fta2> just plain CVE-XXXX-YYYY, not CVE-XXXX-YYYY..ZZZZ
[13:47] <jdstrand> fta: correct
[13:47] <james_w> asac: no, not yet
[13:47] <asac> james_w: my auto builder doesnt like the --result-dir option hardy chroot :(
[13:48] <james_w> asac: I'll bump it up my todo list
[13:48] <jdstrand> really, the usefulness of the CVE functionality is that *if* a CVE happens to be linked to the bug, you can click and get to the CVE description
[13:49] <jdstrand> we don't try to keep it in sync cause the API just isn't there yet. we use our own ubuntu-cve-tracker
[13:49] <asac> thanks for clarifying
[13:49] <jdstrand> np
[13:49] <asac> james_w: latest builddeb wont work with hardy bzr?
[13:50] <asac> james_w: or could i just checkout builddeb in .bazaar/plugins or something?
[13:50] <james_w> asac: what's the error?
[13:50] <james_w> or is it the depends?
[13:50] <asac> james_w: i havent tried ... i only saw that --result-dir isnt known
[13:50] <asac> in default hardy
[13:51] <james_w> that's nothing to do with the bzr version, the name of the option just changed
[13:51] <james_w> at your request I believe
[13:51] <james_w> --result will work in both
[13:54] <fta2> --result doesn't work for me, i want the results in build-area as before, not above, but i get raise Error, "`%s` and `%s` are the same file" % (src, dst)
=== thunderstruck is now known as gnomefreak
[14:08] <asac> james_w: result works?
[14:08] <asac> hmm
[14:08] <james_w> fta2: please upgrade to 2.0.1
[14:09] <asac> ok giving that a spin
[14:09] <asac> lets see ;)
[14:10] <gnomefreak> 2.0.1 is in intrepid unless he missed updates he should be running it
[14:14] <fta2> i'm using 2.0.1
[14:16] <james_w> fta: file a bug with the backtrace then please
[14:16] <james_w> if there isn't a backtrace then add -Derror to the command line and run it again
[14:28] <asac> fta2: we also need a versioned simlink from _NAME_OTHER
[14:28] <asac> fta2: abrowser-3.0 wont work now i think
[14:29] <fta2> james_w, that's what i get: http://paste.ubuntu.com/52846/
[14:30] <fta2> asac, i wanted to ask you if i can move all the abrowser stuff to the abrowser branding deb
[14:31] <asac> fta2: i think the links are fine in the -common package (which currently i firefox-3.0)
[14:31] <asac> fta2: in that way things that use the abrowser commands will still work with firefox and vv.
[14:32] <fta2> well, in /usr/bin, i don't like to add stuff that are not needed, like /usr/bin/abrowser* for people using plain ff3
[14:33] <asac> fta2: its a matter of compatibility. those scripts dont hurt
[14:34] <asac> fta2: same argument would be true for abrowser. whipping firefox links everywhere but abrowser links not will make "abrowser" more a second class citizen then it already is
[14:38] <asac> fta2: imo we should review how the branding split is packaged for jaunty
[14:41] <fta2> it's now clean for the desktop file, and for icons. i just don't like all those links in usr/bin and usr/lib
[14:42] <fta2> i don't expect that many users to switch to abrowser anyway, so 2nd class citizen is more than probable.
[14:42] <asac> fta2: well. its a political thing
[14:42] <asac> both need to be similar and all those links dont really hurt
[14:42] <asac> they only serve for compatibility
[14:42] <asac> users will only see the .desktop files
[14:47] <fta2> http://paste.ubuntu.com/52853/
[14:48] <fta2> no idea how representative it is
[14:49] <asac> fta2: its not ... ffox is installed by default ;)
[14:49] <thunderstruck> 2nd class my ass i use it and like it ;)
[14:49] <asac> but as i said. its not a matter of popularity.
[14:50] <asac> the bidirectional links need to be in the -common package
[14:50] <thunderstruck> asac: and you are right abrowser-3.0 doesnt work
=== thunderstruck is now known as gnomefreak
[14:51] <fta2> we have 125 times more people using the ff branding just in intrepid
[14:52] <gnomefreak> ??
[14:52] <asac> that doesnt change the fact that both packages should be 100% equivalent
[14:52] <asac> thats only achieved if abrowser links exist everywhere
[14:52]  * gnomefreak has no way of finding that percentage out
[14:53] <gnomefreak> awesome-browser??
[14:53] <gnomefreak> is that abrowser?
[14:53] <fta2> 6941  firefox-3.0-branding            4263  1255     0  3006     2 (Unknown)
[14:53] <fta2> 33650 abrowser-3.0-branding             34     9     0    25     0 (Unknown)
[14:53] <fta2> 4263/34
[14:53] <fta2> ~125
[14:53] <gnomefreak> where did you find that?
[14:53] <fta2> popcon
[14:54] <gnomefreak> oh
[14:54] <asac> ok ... lunch and then coding NM
[14:54] <gnomefreak> fta2: abrowser still gives you firefox-3.0
[14:54] <fta2> 26018 firefox-3.1-branding             118    36     0    82     0 (Unknown)
[14:54] <fta2> 44115 abrowser-3.1-branding              7     1     0     6     0 (Unknown)
[14:54] <fta2> 22665 firefox-3.1                      197    32    45   120     0 (Unknown)
[15:10] <gnomefreak> asac: care to tell the people on bug 87101 to back off. we fix it one way it pisses others off, cant win them all
[15:10] <ubottu> Launchpad bug 87101 in thunderbird "when opening an attachment, it should be written to tmp as a read-only file" [Wishlist,Fix released] https://launchpad.net/bugs/87101
[15:17] <thunderstruck> this is starting to piss me off
=== thunderstruck is now known as gnomefreak
[15:17] <james_w> fta2: thanks, should be fixed in the next upload
[15:29] <gnomefreak> !info libflashsupport intrepid
[15:29] <ubottu> libflashsupport (source: libflashsupport): Support library for sound output of Flash 9+ with PulseAudio. In component universe, is extra. Version 1.9-0ubuntu2 (intrepid), package size 8 kB, installed size 64 kB
[15:29] <gnomefreak> asac: do you plan pon removing libflashsupport from archives anymore?
[15:31] <gnomefreak> s/pon/on
[15:33] <asac> gnomefreak: i am waiting for feedback from ltsp on whether they still need that
[15:34] <gnomefreak> k
[15:34] <gnomefreak> people are making frigging wiki pages on how to fix the issue and i have strong feeling we are gonna get comments that that isnt the fix (because people dont know any better
[15:35] <gnomefreak> )*
[15:38] <gnomefreak> asac: i told guy on bug 255297 i would ping you
[15:38] <ubottu> Launchpad bug 255297 in xulrunner-1.9 "please include debian patch to enable old certificates UI" [Undecided,New] https://launchpad.net/bugs/255297
[15:41] <gnomefreak> !info flashplugin-nonfree hardy-backports
[15:41] <ubottu> flashplugin-nonfree (source: flashplugin-nonfree): Adobe Flash Player plugin installer. In component multiverse, is optional. Version 10.0.1.218+10.0.0.525ubuntu1~hardy1+really9.0.124.0ubuntu2 (hardy-backports), package size 18 kB, installed size 164 kB (Only available for i386 amd64 lpia)
[15:41] <asac> gnomefreak: when?
[15:42] <asac> gnomefreak: i mean i am always in chpe's epiphany channel ;)
[15:42] <asac> he could easily ping me there :)
[15:42] <gnomefreak> i dont really know him but i said it on bug. my mail might have not gotten there yet
[15:43] <asac> ah ok
[15:55] <thunderstruck> asac: did you get my last post about tbird notification?
[15:56] <asac> huh?
[15:56] <asac> i dont think so
[15:56] <asac> at least not here
[16:01] <thunders1ruck> fuck this
[16:02] <thunders1ruck> asac: in case you missed it
[16:02] <thunders1ruck> .:10:50:35:. <      gnomefreak > do we have notification for tbird restart  after update? if not we should really think  about this
[16:04] <asac> thunders1ruck: yeah we dont have it yet
[16:05] <thunders1ruck> k
[16:05] <asac> but makes sense i agree
=== thunders1ruck is now known as gnomefreak
[16:26] <thunderstruck> asac: do we have master bug on the the exec: 118: /usr/lib/firefox-3.0.3/..... people are trying it with all different browser links. like sendible-browser
[16:27] <thunderstruck> x-www-browser gets a different error
[16:38] <asac> thunderstruck: thanks
[16:38] <asac> thunderstruck:  we should make a bug out of that
[16:42] <thunderstruck> i opened the one i filed and marked all as dup and explained in bug what links fail
[16:42] <thunderstruck> asac: bug 273907
[16:42] <ubottu> Launchpad bug 273907 in firefox-3.0 "abrowser does not start: :$pkglibdir/abrowser link missing" [High,Confirmed] https://launchpad.net/bugs/273907
[16:43] <thunderstruck> updated description
=== thunderstruck is now known as gnomefreak
[16:44] <gnomefreak> ok i really have to work on connection now. its getting old fast
=== asac_ is now known as asac
[17:05] <asac> jcastro: did you reping blizzard?
[17:06] <jcastro> asac: not yet, I will do so today
[17:06] <jcastro> sorry I've been running around like a chicken with my head cut off.
[17:08] <asac> jcastro: too points. ask him if he thinks that anyone would come. most likely those will come during UDS and not fosscamp (as its business day for them), right?
[17:08] <asac> jcastro: i am trying to sort out if i need to be at fosscamp :)
[17:11] <[reed]> asac: get mconnor to come? :)
[17:19] <asac> [reed]: i asked him and he said he planned to be in MV around that time and would then consider to come
[17:20] <[reed]> cool
[18:02] <bfiller> asac: where can I find a version of firefox-3.0 with debug symbols?
[18:03] <asac> bfiller: you can find debug symbols
[18:03] <asac> bfiller: https://wiki.ubuntu.com/MozillaTeam/Bugs#Crashes
[18:03] <asac> adapt apt lines for intrepid accordingly
[18:03] <bfiller> asac: thanks
[18:05] <asac> bfiller: if you need a real debug build you need to build on your own
[18:05] <asac> (e.g. with debug output and without optimization)
[18:12] <asac> so now i became a ~ubuntu-mir team member :/
[18:56] <asac> james_w: http://paste.ubuntu.com/52925/
[18:56] <asac> james_w: thats the error i referred to the other day
[18:57] <asac> building -sa complains about binary changes missing sometimes
[20:27] <fta> asac, i tried to make head or tail from the gnash package, it's really awful
[20:52] <fta> asac, http://ubuntuforums.org/showthread.php?t=935231
[21:12] <asac> fta: he?
[21:12] <asac> i updated gnash.ubuntu.head today
[21:12] <asac> should work
[21:13] <fta> nm, it was a few days ago, i gave up and updated nf-flash instead :(
[21:58] <fta> mozilla bug Bug 368428
[21:58] <ubottu> Error: Launchpad bug 368428 could not be found
[21:58] <fta> mozilla bug 368428
[21:58] <ubottu> Mozilla bug 368428 in Networking: JAR "XUL FastLoad cache corruption when application running while upgrading" [Normal,Resolved: fixed] http://bugzilla.mozilla.org/show_bug.cgi?id=368428
=== `6og is now known as Kamping_Kaiser