2008/11/28/#ubuntu-mozillateam.txt

[00:50] <fta> asac, found the prism bug. it's caused by mozilla bug 462774
[00:50] <ubottu> Mozilla bug 462774 in General "drop JSON.jsm" [Normal,Resolved: fixed] http://bugzilla.mozilla.org/show_bug.cgi?id=462774
[00:51] <fta> like for epiphany, restricting prism to xul 1.9.0.* helped
[09:55] <gnomefreak> i have sound :)
[10:14] <gnomefreak> anyone know nafun@jabber.ccc.de is?
[10:42] <asac> gnomefreak: huh?
[10:42] <asac> gnomefreak: i am on jabber.ccc.de
[10:42] <asac> no idea who nafun is ;)
[10:42] <gnomefreak> asac nafun wanted me to add it so making sure its not someone from here
[10:43] <rZr> asac: hi hey you're member of ccc ? I planned to move to ccc camp coup of years ago
[10:44] <gnomefreak> isnt ccc.* open to all?
[10:45]  * gnomefreak not sure who i am on jabber
[10:45] <asac> rZr: not member ... just jabber
[10:45] <gnomefreak> oh crap today is the first :(
[10:46] <gnomefreak> oh no it isnt
[11:04] <gnomefreak> asac: any idea on why IMAP has so many problems?
[11:06] <gnomefreak> bug 302320
[11:06] <ubottu> Launchpad bug 302320 in mozilla-thunderbird "Thunderbird Template is deleted when right clicking on message body" [Undecided,Incomplete] https://launchpad.net/bugs/302320
[11:13] <gnomefreak> asac: did you change your mind on bug 130255?
[11:13] <ubottu> Launchpad bug 130255 in firefox "Wish: Please default-accept CAcert.org SSL Certificates" [Wishlist,Won't fix] https://launchpad.net/bugs/130255
[11:13] <gnomefreak> nevermind the idiot wasnt reading before posting
[11:38] <asac> gnomefreak: no
[11:38] <asac> ok
[11:38] <asac> gnomefreak: CAcert has known weaknesses in their certification process
[11:38] <asac> so its not really trust-worthy
[11:38] <gnomefreak> i agree
[11:38]  * directhex is trustworthy, offers to sign peoples' certificates for beer
[11:38] <gnomefreak> i dont like this bug :( one of your wontfix bugs ;)
[11:39] <gnomefreak> directhex: i finished that package
[11:39] <gnomefreak> its in my PPA
[11:39] <directhex> gnomefreak, which package, sorry?
[11:40] <gnomefreak> asac: bug 266946 should stay as wont fix IMHO but this will chnage with the QT port for FF
[11:40] <ubottu> Launchpad bug 266946 in firefox-3.0 "firefox installs unnecessary packages when installing on KDE" [High,Confirmed] https://launchpad.net/bugs/266946
[11:41] <gnomefreak> directhex: you asked about startupmanager on 11/25 im fairly sure it was you
[11:41] <directhex> did i?
[11:42] <gnomefreak> yes but please dont make me look it up in logs ;)
[11:43] <gnomefreak> asac: i'm debating on flashgot since it NEEDS a download manager we have a few AFAIK it just seems alot of work however people do use it alot
[11:43]  * gnomefreak smoke
[11:52] <asac> gnomefreak: for the KDE thing the most important issue is ubufox. it currently pulls in half of gnome
[11:52] <asac> gnomefreak: the bug in general is valid though ... after a bunch of discussions i admitted that
[11:55] <gnomefreak> ok but are we porting ubufox for QT?
[11:56] <directhex> kubufox?
[11:58] <gnomefreak> sounds good to me bug its gonna be a pain in the ass since it has so many deps
[11:58] <gnomefreak> s/bug/but
[11:58] <directhex> ubufoKs!
[11:58]  * gnomefreak needs to eat and head to show for an hour or so to log in new work so gary can start today
[12:01] <gnomefreak> ok ill be back
[12:08] <asac> directhex: gnomefreak: its not really ubufox. its apturl that needs to better support kde
=== asac_ is now known as asac
=== rZr is now known as RzR
[14:01] <fta> asac, how come build-system.tar.gz is no longer in hardy?
[14:01] <fta> http://launchpadlibrarian.net/20012919/buildlog_ubuntu-hardy-i386.prism_0.9.1%2Bsvn20081126r20278-0ubuntu1~fta2~hardy_FAILEDTOBUILD.txt.gz
[14:14] <asac> fta: i thought this was introduced in intrepid
[14:15] <asac> fta: most likely you had a backport from .head branch in your ppa? and now that archive is higher you see this regression?
[14:16] <asac> http://blog.mozilla.com/ligong/2008/11/18/firefox-china-edition-released-today/
[14:17] <asac> i guess that means we need livemargins packaged
[14:17] <asac> aka Juice
[14:18] <asac> non-free
[14:18] <asac> great that mozilla releases non-free stuff now
[14:19] <asac> or isnt that an official mozilla product?
[14:21] <asac> asked on blog comment?
[14:21] <asac> s/?/!/
[14:26] <asac> seems to be released by mozilla bejing
[14:26] <asac> at from http://blog.mozilla.com/ligong/about/
[14:28] <asac> oops
[14:28] <asac> http://www.mozilla.org.cn/thunderbird/
[14:29] <asac> http://www.mozilla.org.cn/firefox/
[14:29] <asac> ^^ bumb
[14:29] <asac> armin76: ^^
[14:30] <asac> fta: whats going on with 1.9.1? are the jemalloc issues sorted?
[14:31] <fta> yes
[14:31] <asac> uploaded?
[14:31] <fta> not yet, i wanted nspr/nss, or maybe it could wait
[14:32] <asac> fta: as long as it doesnt need those versions at build time i would not wait for them
[14:32] <asac> fta: have you tested nspr/nss?
[14:32] <fta> not yet.
[14:33] <asac> fta: for me stuff works fine. shlibs-control file was intentionally dropped because it doesnt work with "unversioned" sonames as it seems
[14:33] <asac> so we just have symbols
[14:33] <asac> but that seems to work during build time here
[14:33] <asac> also symbols are properly auto-regenerated
[14:33] <asac> and i upgraded just binaries and saw no regressions
[14:34] <asac> only test missing is whether our xulrunner built against that really works with upstream nspr/nss
[14:34] <asac> but i tested that last time afair
[14:37]  * asac spins xul
[14:38] <fta> i'm pushing xul/ff. we'll see later on.
[14:39] <asac> cool
[14:39] <asac> i think archive admins might need a bit until they approve it
[14:40] <asac> we can try to get pittis attention before if we want it
[15:00] <fta> asac, xul uploaded
[15:02] <asac> good ... ffox too?
[15:02] <asac> both with bug in changelog?
[15:02] <fta> i'm on it
[15:02] <asac> (pleaes bug id)
[15:02] <asac> thanks
[15:02] <fta> yes
[15:02] <asac> james_w: bzr: ERROR: Could not find .changes file: ../builds/network-manager_0.7~~svn20081018t105859-0ubuntu1.8.10.1~0hardy~nm1_i386.changes.
[15:02] <asac> thats what i am usually getting when a -S -sd build succeeds
[15:02] <asac> in consequence, the results are not copied to result-dir
[15:03] <james_w> what command are you running?
[15:03] <asac> james_w: http://paste.ubuntu.com/77759/
[15:03] <asac> thats the start and then end ;)
[15:04] <james_w> you need a "-S"
[15:04] <james_w> a second one, outside of the build command
[15:04] <asac> james_w: ok i understand. but why? couldnt you just say: copy all .changes that were created?
[15:05] <asac> e.g. always test for source + binary changes ... and copy those over that exist
[15:05] <james_w> you could
[15:05] <james_w> but what if you had a build command that put the changes somewhere other than ..
[15:06] <james_w> and you had an old bust _source.changes in ..
[15:06] <asac> james_w: as long as you find a matching one, thats not a problem imo
[15:06] <asac> so: none found -> error
[15:06] <asac> one or both found -> use that
[15:06] <james_w> then it would copy the bust one and you might not realise
[15:06] <asac> debuild ends with error code -> fail
[15:06] <james_w> and both found means you don't really know
[15:06] <asac> james_w: but how is that now different?
[15:07] <asac> james_w: you never know that the .changes is the one produced by the build do you?
[15:07] <james_w> yes, it's not ideal if you have a build command that puts things elsewhere than ..
[15:07] <james_w> bug this is conservative at least
[15:07] <asac> james_w: personally i think its not a problem
[15:07] <asac> james_w: build commands should just do normal things.
[15:07] <asac> but my case is quiet normal
[15:08] <asac> james_w: if build command doesnt put results in the build-dir its all busted anyway
[15:09] <asac> james_w: does dpkg-buildpackage or debuild support to put results elsewhere?
[15:09] <james_w> no
[15:09] <james_w> pdebuild does it
[15:10] <asac> yeah. but still this bug seems unrelated
[15:10] <asac> e.g. if you do it now, and have old files in build-dir then the behaviour will do bad things too
[15:11] <asac> you could do feasibility tests by looking at timestamps
[15:11] <asac> (for this issue)
[15:12] <asac> but then, people often mess up time-syncs in local nets
[15:12] <asac> so i think there is really not much you can do for users that specify a custom output dir in build command
[15:12] <fta> i wish we had sqlite >= 3.6.4
[15:12] <asac> james_w: oh. another idea is to look at result-dir before starting
[15:13] <asac> if there are matching changes -> remember timestamps and then check whether they were modified after build
[15:13] <asac> fta: i wish we had whatever version moz uses now again ;)
[15:14] <fta> it's the same wish then :)
[15:14] <asac> fta: what was the 3.1 bug?
[15:14] <asac> fta: they backed out for b2
[15:14] <asac> probably relanded for b3pre
[15:15] <fta> bug 274187 ?
[15:15] <ubottu> Launchpad bug 274187 in ubuntu "FFe - firefox 3.1 and xulrunner 1.9.1 for intrepid/universe" [Wishlist,Won't fix] https://launchpad.net/bugs/274187
[15:15] <asac> ok let me remove FFe
[15:15] <asac> ;)
[15:15] <asac> and open jaunty task
[15:16] <asac> done
[15:17] <fta> there's a bunch of new in-source libs for <video> that we don't have system wide (or that we have but not fresh enough) :(
[15:18] <fta> basically, moz is working with upstream to they are importing snapshots every few days
[15:19] <asac> fta: yeah. makes sense to not do the same for syslibs, but keep using the in-source ones until this settles
[15:19] <fta> yep, that's why i didn't spend a second on that
[15:20] <asac> if you ask me its unlikely that 3.1 will become final in time for jaunty
[15:24] <asac> fta: do you see this random menu auto-activiation bug in 3.1?
[15:24] <asac> i thnk ijust saw it once
[15:24] <asac> in 3.0
[15:24] <asac> when clicking on bookmarks it opened "bookmark this page ..."
[15:24] <asac> and i am 100% sure i didnt even move the mouse
[15:25] <asac> because i just a thinkpad mouse pointer ...
[15:25] <asac> and it doesnt move at all untill i touch it
[15:25] <fta> i was with b2pre, but strangely enough, with b2 i don't. i just didn't force the extensions (yet)
[15:26] <asac> hmm
[15:26] <asac> so maybe extension related
[15:27] <asac> but i think o nbug people claim that even all extensions disabled dont help
[15:28] <fta> so my issue may be a different one. i have 26 extensions.. difficult to tell
[15:29] <asac> fta: for now i would think its not extension related
[15:29] <asac> fta: you saw the same issue with bookmark this page right?
[15:30] <fta> with all the menus in the bm toolbar
[15:30] <asac> fta: do you have mouse gestures?
[15:30] <fta> no
[15:30] <asac> (extension)
[15:30] <asac> ok
[15:30] <asac> fta: menus in the bm toolbar?
[15:30] <asac> i dont have a menu there
[15:31] <asac> ok reproduced here with all extensions disabled (3.0)
[15:31] <fta> a left click selected an item (the 1st one?) while it should have opened the menu, while a click + hold did the right thing.
[15:31] <asac> with history
[15:31] <asac> i have about 1:20 (fail:success)
[15:32] <asac> fta: ah ... you say it happens for bookmark folders
[15:32] <asac> interesting
[15:32] <asac> so there have to be even two bugs:
[15:33] <asac> (either, or i think)
[15:33] <asac> 1) button release gets converted to click event even when button release is over a different UI element than the previous button press
[15:33] <asac> or
[15:34] <asac> 2) event bubbles to menu, and menu doesnt properly process "out-of-window" events
[15:34] <asac> (that could be a missing negative integer check)
[15:35] <asac> since i am sure my pointer doesnt intersect with menu action i would go for 2) for now
[15:38] <fta> https://edge.launchpad.net/ubuntu/jaunty/+queue
[15:41] <fta> mozilla bug 449443
[15:41] <ubottu> Mozilla bug 449443 in Storage "Upgrade to SQLite 3.6.4" [Normal,Resolved: fixed] http://bugzilla.mozilla.org/show_bug.cgi?id=449443
[15:42] <fta> no backout since
[15:45] <asac> fta: i think on release branch
[15:46] <asac> "Heads up for 1.9.0.5... we're backing out the SQLite upgrade."
[15:46] <asac> seems only on 1.9.0
[15:47] <fta> for jaunty, i think 3.6.4 is a good idea. upstream says "Version 3.6.6.2 of SQLite is recommended for all new development. Upgrading from versions 3.6.4 through 3.6.6 is strongly recommented."
[15:55] <fta> asac, i just asked pitti
[16:02]  * fta going out for ~1/2h
[16:07] <fta> gone
[16:43] <fta> asac, http://launchpadlibrarian.net/20030832/buildlog_ubuntu-jaunty-i386.xulrunner-1.9.1_1.9.1%7Eb2%2Bbuild1%2Bnobinonly-0ubuntu1_FAILEDTOBUILD.txt.gz
[16:44] <fta> you broke it :(
[16:46] <fta> asac, my bad for not testing your last commit :(
[16:57] <fta> too bad :(
[16:57] <fta> fixed
[17:04] <asac> fta: the patch rename?
[17:04] <asac> dumb me
[17:04] <asac> argh
[17:04] <asac> heeh
[17:04] <asac> ;)
[17:04] <fta> you renamed the patch, but forgot the series
[17:04] <asac> fta: just reupload the same changelog
[17:04] <asac> with version bumped
[17:04] <asac> yeah
[17:04] <fta> too late
[17:04] <asac> fta: bad
[17:05] <asac> initial uploads should have the full changelog ;)
[17:05] <asac> but well
[17:05] <fta> people don't care about xul, they will read ff
[17:06] <asac> fta: yeah. things that dont get published not need a new changelog ... for future ;)
[17:06] <asac> but ok
[17:06] <fta> ok, i'll try to remember that
[17:06] <asac> i sometimes forget it too
[17:08] <asac> fta: once we have nspr/nss up we need to respin everything ;) ... good for upload count
[17:09] <asac> too bad that most rdepends seem to be in main
[17:10] <asac> libgeier0 ;)
[17:19] <fta> lol
[17:33]  * asac merges NM 0.7
[17:33] <asac> fina
[17:33] <asac> l
[17:34] <asac> really bad that most fixes landed after our release ;)
[17:34] <asac> and first thing they did before they fixed things, they changed API/ABI
[17:34] <asac> so just upgrading to 0.7 isnt possible
[17:35] <asac> and backporting is similar painful
[17:35] <asac> because they switched from using struct members directly to accessor functions everywhere
[18:03] <fta> gasp
[18:03] <fta> http://launchpadlibrarian.net/20032359/buildlog_ubuntu-jaunty-i386.xulrunner-1.9.1_1.9.1~b2%2Bbuild1%2Bnobinonly-0ubuntu2_FAILEDTOBUILD.txt.gz
[18:03] <fta> g++ -o nsNSSCertificate.o -c -I../../../../dist/include/system_wrappers -include ../../../../config/gcc_hidden.h -DNSS_ENABLE_ECC -DXPCOM_TRANSLATE_NSGM_ENTRY_POINT=1 -DMOZILLA_INTERNAL_API -D_IMPL_NS_COM -DEXPORT_XPT_API -DEXPORT_XPTC_API -D_IMPL_NS_COM_OBSOLETE -D_IMPL_NS_GFX -D_IMPL_NS_WIDGET -DIMPL_XREAPI -DIMPL_NS_NET -DIMPL_THEBES  -DOSTYPE=\"Linux2.6\" -DOSARCH=Linux -I/usr/include/nss -I. -I. -I../../../../dist/include/nspr -I../../../../
[18:03] <fta> dist/include/xpcom -I../../../../dist/include/string -I../../../../dist/include/necko -I../../../../dist/include/uriloader -I../../../../dist/include/pref -I../../../../dist/include/docshell -I../../../../dist/include/caps -I../../../../dist/include/dom -I../../../../dist/include/intl -I../../../../dist/include/locale -I../../../../dist/include/profile -I../../../../dist/include/windowwatcher -I../../../../dist/include/js -I../../../../dist/inclu
[18:03] <fta> de/widget -I../../../../dist/include/layout -I../../../../dist/include/content -I../../../../dist/include/xpconnect -I../../../../dist/include/unicharutil -I../../../../dist/include/pipboot -I../../../../dist/include   -I../../../../dist/include/pipnss -I/usr/include/nspr   -I/usr/include  -I/build/buildd/xulrunner-1.9.1-1.9.1~b2+build1+nobinonly/build-tree/mozilla/dist/sdk/include    -fPIC   -fno-rtti -fno-exceptions -Wall -Wpointer-arith -Wover
[18:03] <fta> loaded-virtual -Wsynth -Wno-ctor-dtor-privacy -Wno-non-virtual-dtor -Wcast-align -Wno-invalid-offsetof -Wno-long-long -pedantic -g -fno-strict-aliasing -fshort-wchar -pthread -pipe  -DNDEBUG -DTRIMMED -Os -freorder-blocks -fno-reorder-functions    -DMOZILLA_CLIENT -include ../../../../mozilla-config.h -Wp,-MD,.deps/nsNSSCertificate.pp nsNSSCertificate.cpp
[18:04] <fta> nsNSSCertificate.cpp: In member function 'virtual nsresult nsNSSCertificate::GetAllTokenNames(PRUint32*, PRUnichar***)':
[18:04] <fta> nsNSSCertificate.cpp:913: error: 'PK11_GetAllSlotsForCert' was not declared in this scope
[18:04] <fta> make[6]: *** [nsNSSCertificate.o] Error 1
[18:04] <fta> make[6]: Leaving directory `/build/buildd/xulrunner-1.9.1-1.9.1~b2+build1+nobinonly/bui
[18:05] <fta> how come i didn't come into this earlier ? b2 *is* in my ppa
[18:06] <fta> asac, ^^
[18:08] <fta> --with-system-nspr --with-system-nss ??? eehhhhh ?
[18:09] <fta> hmm
[18:09] <asac> fta: thats why i asked if it builds with old nss/nspr
[18:09] <asac> lets see if its in new symbols but not in old
[18:09] <fta> i thought i was using in-source nss, but it's just ff, not xul !?
[18:11] <fta> time to push nspr/nss. trying in my ppa 1st.
[18:12] <fta> looks like i have the opportunity to fix the changelog after all
[18:15] <asac> fta: fix changelog?
[18:15] <fta> what you proposed earlier
[18:16] <asac> what did i propose?
[18:16] <asac> ah
[18:16] <asac> yeah ;)
[18:16] <fta> <asac> fta: just reupload the same changelog
[18:16] <asac> in xulrunner 1.9.1
[18:16] <asac> thought you want to fix nss ;)
[18:16] <fta> i think i should fix the nss/nspr deps too
[18:16]  * asac is a context-less bot ;)
[18:16] <asac> fta: yes. fix the test
[18:17] <fta> yep
[18:17] <asac> so it uses in-source ... we will respin anyway when nss/nspr is up
[18:17] <asac> just so that it really uses the new libs and not the links
[18:17] <fta> building your stuff this time...
[18:18] <asac> my stuff?
[18:18] <fta> nss/nspr
[18:19] <asac> in ppa ?
[18:19] <asac> ok
[18:19] <fta> i need to check what pkg-config reports now
[18:19] <asac> hopefully the right upstream versoin ;)
[18:21] <asac> Name: NSS
[18:21] <asac> Description: Mozilla Network Security Services
[18:21] <asac> Version: 3.12.0.3
[18:21] <asac> Name: NSPR
[18:21] <asac> Description: The Netscape Portable Runtime
[18:21] <asac> Version: 4.7.1
[18:22] <asac> hmm ... somewhat wrong i guess ;)
[18:22] <asac> but as long as it was bumped its ok i guess
[18:22] <asac> oh i have the old ;)
[18:22] <asac> on this machine
[18:23] <asac> Name: NSS
[18:23] <asac> Description: Mozilla Network Security Services
[18:23] <asac> Version: 3.12.2
[18:23] <asac> Name: NSPR
[18:23] <asac> Description: The Netscape Portable Runtime
[18:23] <asac> Version: 4.7.3
[18:23] <asac> makes more sens
[18:23] <fta> both sent to my ppa
[18:24] <fta> I hope i'll not be flooded by emails like for the subpixel change in xul
[18:24] <asac> fta: so to test: 1st -> test that old apps work with new binaries
[18:25] <asac> 2nd -> test that building rdepends creates proper lower bounds on the new binary versoins
[18:28] <asac> fta: we need to find a smarter pattern to guess the right maxversion/minversion at build time
[18:28] <asac> something like looking at pkg-config version and then guessing the right min/max version thing
[18:29] <asac> doesnt make sense to fix ephy upstream to 1.9 ... it works well with 1.9.1 if its build against 1.9.1
[18:29] <asac> ok ... now i probably loose network
[18:29] <asac> testing something
[18:34] <fta> asac, are you sure ephy is fine when built with 1.9.1 ?
[18:45] <asac> fta: yes. worked here quite good
[18:45] <asac> at least my dev tree
[18:45] <asac> if package doesnt work it must be something related to packaging
=== fta_ is now known as fta
[19:02] <fta> <fta> asac, http://launchpadlibrarian.net/20032926/buildlog_ubuntu-jaunty-hppa.epiphany-browser_2.24.1-1ubuntu2_FAILEDTOBUILD.txt.gz
[19:02] <fta> <fta> checking whether we can compile and run XPCOM programs... no
[19:02] <fta> <fta> configure: error: Cannot compile and run XPCOM programs
[19:13] <asac> fta: ah ... i think i remember
[19:22] <asac> fta: did you commit your uploaded stuff yet?
[19:22] <fta> not yet, i'm still testing nss
[19:22] <fta> seems fine
[19:23] <asac> fta: please push 1.9.1 then ;)
[19:23] <asac> you can go on testing
[19:23] <asac> fta: to bzr i mean ;)
[19:27] <fta> done
[19:27] <fta> seems all fine for epy, ff3, ff3.1
[19:50] <asac> good
[19:50] <asac> fta: is seb still there?
[19:50] <asac> otherwise i can upload that
[19:51] <fta> upload what?
[19:51] <asac> rphy
[19:51] <asac> maxVersion stuff
[19:51] <fta> seb did it already
[19:52] <fta> i'm ready to re-push xul
[19:52] <asac> k
[19:52] <asac> DO it
[19:52] <asac> ;)
[19:52] <asac> most likely pitti wont be back till monday ;) ... so binary new might need to wait a bit
[19:54] <fta> done
[19:54] <fta> i hope it's good this time :P
[19:58] <fta> nice page: https://edge.launchpad.net/ubuntu/+archivemirrors
[19:59] <fta> https://edge.launchpad.net/ubuntu/+ppas :)))
=== fabrice_sp_ is now known as fabrice_sp