/srv/irclogs.ubuntu.com/2007/12/29/#ubuntu-mozillateam.txt

asacarmin76: ok, then pulling in the SetUnichar function and using that instead hopefully fixes it00:05
Ubuletteasac, you should retrigger a build of epiphany, it has been built against nspr/nss from xul (not system) so it's linked against .so, not against .so.[01].d00:12
asacsystem xul is not yet in archive :)00:19
Ubulettedamn00:20
asaconce its in everything needs a respin00:20
Ubuletteremember you missed nssutil00:20
asaci asked you if that causes any issues00:21
asaci don't see any00:21
Ubuletteit did for me in xul trunk weeks ago, that's how i found out00:22
asacwhich symptoms?00:32
Ubulettemissing symbols00:32
Ubulettemoved in nssutil00:32
asacfor which operations? it works here well00:33
asacfor xul 1.9 + ffox + gtkmozembed00:33
Ubulettebuild time00:33
Ubuletteoh, i see. epiphany is broken because you did   -L/usr/lib/xulrunner-devel-1.9b1/lib -lplds4 -lplc4 -lnspr400:43
Ubulettethere're only .so in there even if system nspr is also installed00:44
Ubulettetoo bad00:44
asaci don't see the point you are trying to make00:51
Ubulettenm00:52
asacunfortunately ephy needs to link nspr directly ... which causes all this00:53
Ubuletteit could have worked with xul b1 + regular nspr using a different linking order00:54
Ubulettebut nm, it's not my mess^Wbusiness00:56
extractedany one here know anything about subnetting ?04:51
tonyyarussoextracted: um, some, but why asking here?05:45
armin76asac: shrug, so i'm going to do that stuff10:39
armin76asac: seems to work fine, let me paste you the patch11:50
readyxgive it to me baby - aha - aha!11:54
armin76asac: http://rafb.net/p/gOCIKj67.html12:08
asacarmin76: is the GetUnichar part needed?12:34
asacanyway: rock!12:34
armin76asac: http://rafb.net/p/Pf6NJO18.html <- like that?12:44
=== \sh_away is now known as \sh
asdsadsda:D12:49
asacarmin76: no idea :) ... does it work?13:05
asaci can upload to debian ... finaly!!!13:12
asac(again)13:12
CheGuevaracongrats :P13:25
asacarmin76: :) ... looks good ... if you don13:30
asact get bus errors we should update the bugzilla bug13:31
=== asac_ is now known as asac
armin76asac: the latest patch seems to work fine14:39
asac@time14:40
ubotuCurrent time in Etc/UTC: December 29 2007, 14:40:10 - Next meeting: Edubuntu meeting in 3 days14:40
asacarmin76: the one without GetUnichar?14:40
asacgreat!14:40
armin76yeah14:40
asacplease test a bit longer ... who knows if the <<get>> bus error might be harder to trigger14:40
armin76sure, doing that atm14:41
armin76although without the patch it takes some seconds, maybe 3014:41
asacno idea if it makes a difference, but try to visit unicode sites too14:43
armin76sites with special characters like üéíóú?14:46
asacdepends ... they might use latin encoding ... at best go for some example unicode sites14:47
asachttp://www.alanwood.net/unicode/#links14:47
asachttp://www.alanwood.net/unicode/thai.html14:48
asachttp://www.alanwood.net/unicode/cjk_compatibility.html14:48
armin76http://www.i18nguy.com/unicode-example.html <- i was using that14:48
armin76those links work fine as well14:49
asacok looks good then ... do you use pango?14:49
armin76yup14:50
asacUbulette: btw, i have permission to update kaze in debian14:52
asacdid we go ahead in ubuntu? or did we wait for debian? (can't remember exactly)14:53
armin76asac: do you use external cairo?14:54
asacfor 2.x? yes!14:55
asacfor 2.x: pending14:55
asac(waiting for someone to let new nss into the archive)14:55
asacs/2.x: pe/3.x: pe/14:56
asaccairo in ubuntu is ready for system-cairo on ffox 314:56
armin76remember to test my patch :P14:56
asacsorry i forgot :(14:56
asacwhich?14:57
asacis there a bug?14:57
armin76lame :P14:58
armin76mozilla bug 40997614:58
ubotuMozilla bug 409976 in GFX: Thebes "Cairo in trunk gives SIGBUS on SPARC" [Normal,New] http://bugzilla.mozilla.org/show_bug.cgi?id=40997614:58
asacno idea how to test :)14:59
armin76well, just apply to your firefox tarball, compile with internal cairo, and try14:59
asacwe can include the patch if you want ... but we appear to have close to zero sparc desktop users14:59
asacthe "and try" is my problem14:59
asaci have no access to sparc desktop14:59
welpiewell, but test on whatever you can15:00
asacah15:00
welpiei know it works on sparc, but i don't want to break other arches15:00
asacwelpie: ?15:00
asachehe15:00
welpieyeah, it's me15:01
=== welpie is now known as armin76_
asaci thought so :)15:01
armin76_testing ff on sparc :P15:01
asacarmin76_: i think attaching the actual patch might help :)15:03
armin76nod15:04
* asac out shopping + food et al15:05
CheGuevarahas anyone tried building xulrunner against latest glib15:21
Ubulettehi15:37
UbuletteCheGuevara, which glib ? i use the lastest in hardy, ie 2.15.015:39
armin76asac: i attached the patch15:40
Ubuletteasac, about epiphnay, even with a respin, it will ftbfs15:41
CheGuevaraUbulette: yeah15:41
CheGuevaradoes it build fine?15:41
CheGuevarabecause we just encountered a problem when building kmplayer15:41
CheGuevarawhich should affect glib as well15:41
CheGuevarai mean15:41
CheGuevaraxulrunner as well15:42
CheGuevarahttp://svn.gnome.org/viewvc/glib?view=revision&revision=618515:42
UbuletteCheGuevara, sure. http://launchpadlibrarian.net/11086860/buildlog_ubuntu-hardy-i386.xulrunner-1.9_1.9~b2%2Bnobinonly-0ubuntu1~fta2_FULLYBUILT.txt.gz15:42
Ubuletteoh, that looks like my bug 17911915:43
ubotuLaunchpad bug 179119 in glib2.0 "glib 2.15 not clean with -pedantic" [Undecided,New] https://launchpad.net/bugs/17911915:43
CheGuevaravery likely :P15:43
CheGuevarai am building an updated glib right now15:44
CheGuevaraUbulette: do you mind looking at debdiff when its done building, since i am not exactly the packaging master15:46
Ubulettesure, but remember i can't sponsor you15:48
CheGuevarayeah i know :)15:49
CheGuevaranow at least it closes a bug :P15:50
Ubuletteasac, for epiphany, you need to patch m4/gecko.m4, or improve you xul1.9 patch a bit further15:50
=== \sh is now known as \sh_away
CheGuevaraUbulette: http://launchpadlibrarian.net/11089726/glib.debdiff15:55
Ubulettelooks good. i would have named the patch differently but it's ok15:58
CheGuevaraok thanks Ubulette16:00
Ubuletteasac, what about kaze ?16:13
Ubulettebtw, 0.5.1 has been released today16:14
asacUbulette: build error?16:25
asac(ephy)16:25
Ubuletteasac, you will get that:16:32
Ubulettechecking whether we have a gtk 2 gecko build... yes16:32
Ubulettechecking whether we have a gecko debug build... no16:32
Ubulettechecking whether we have a xpcom glue... yes16:32
Ubulettechecking for gecko version... 1.916:32
Ubulettechecking nspr in gecko... no16:32
Ubulettechecking nspr in system... yes16:32
Ubulettechecking whether we can compile and run XPCOM programs... no16:32
Ubuletteconfigure: error: Cannot compile and run XPCOM programs16:32
Ubuletteit's because the last test never uses -I/usr/include/nspr detected earlier16:33
asacUbulette: ah right16:34
asacok16:34
asaci think thats fixed here already16:34
asacUbulette: you started to port kaze once ... can you give me what you got so far?16:35
Ubulettehmm, i can't find my branch, neither on lp nor locally16:41
Ubulettewtf?16:41
asacUbulette: http://people.ubuntu.com/~asac/ephy.ubulette.patch16:54
Ubuletteupdated patch ?16:54
asacyes16:54
asacits not clean16:54
asacthe interdiff is:16:54
asachttp://paste.ubuntu.com/3091/16:55
asacthe important part for you is most likely line 3016:55
asacand line 51-16:55
asacif you xul has the fixed mozilla-nspr patch from the mozillateam branch then line 30 should be enough for the nspr thing16:55
asacthe rest was just some dirty hacking i did before i decided to fix xul .pc16:56
asac(no clue which patch seb used ... i think he used something between this and the patch in gnome (which is the old.patch in the diff)16:57
asacanyway ... replacing the full patch with this new one should be fine for system-nspr16:57
asacwell ... dpatch hell of course :)16:58
asacaka no fold avail16:58
asacno idea why interdiff strips whitespaces though :/16:59
Ubulettehttp://paste.ubuntu.com/3093/17:00
Ubulettebetween seb and your new patch17:01
asacyou didn't use interdiff, right?17:02
Ubuletteno17:02
asacyes ... but should be ok i guess17:02
asacthough unsure why seb does have the doc/reference/Makefile.am patch17:03
asaci remember that i developed that for him in order to build the docs17:03
asacdoc/reference/tmpl/ephy-embed.sgml -> garbage of course17:03
asacwe need a fix for mozilla bug 40879117:04
ubotuMozilla bug 408791 in Password Manager "nsLoginManagerPrompter.js does not work with TestGtkEmbed prompt." [Normal,New] http://bugzilla.mozilla.org/show_bug.cgi?id=40879117:04
asacso if you have some splendid idea :)17:05
armin76http://dev.gentoo.org/~armin76/ffdialog.png <- i need a fix for that when compiling against xulrunner :P17:09
asaccould be everything17:10
asacxul 1.9 + ffxo 3?17:10
armin76yep17:10
armin76however building it without a shared xulrunner, works fine17:10
armin76weird, huh?17:11
asacarmin76: do you have all our patches?17:11
armin76for xul? i have a lot :P17:12
CheGuevaraasac: is it possible for you to sponsor a glib2.0 upload, or do you only stick to mozilla stuff17:12
asacCheGuevara: depends on what kind of fix it is17:13
CheGuevaraasac: bug 17911917:13
ubotuLaunchpad bug 179119 in glib2.0 "glib 2.15 not clean with -pedantic" [Undecided,New] https://launchpad.net/bugs/17911917:13
asac+-      // g_print ("LOOPSTATE: subtest_running=%d subtest_io_pending=%d\n", subtest_running, subtest_io_pending);17:15
asacis that really C99 ?17:16
CheGuevara/ comments are not allowed in C9917:16
CheGuevaraumm // i mean17:16
asacyou mean "not allowed in C95" =17:17
asac?17:17
asacaeh whatever was before 99 :)17:17
CheGuevarai am honestly not sure17:18
CheGuevarai always thought that // was only dis-allowed in C99 though17:18
CheGuevaraoooh i am talking rubish17:20
asacwell ... i think its just the ;17:20
CheGuevara // was introduced in C9917:20
asacthe rest is glib code-convention17:20
CheGuevarai was getting it the other way around17:20
asacyeah ... thats what i mean ... but i am not even sure about that17:20
CheGuevaraC99 standard says it is :P17:20
CheGuevara/ style comments, ala C++, are now available in C99. A comment of this form starts with // and ends at the end of the line.17:20
CheGuevarahttp://www.comeaucomputing.com/techtalk/c99/#slashslashcomments17:20
asacok17:20
CheGuevarakmplayer fails to compile with17:21
CheGuevara/usr/include/glib-2.0/glib/gtestutils.h:108: error: expected identifier or '(' before '/' token17:21
CheGuevara/usr/include/glib-2.0/glib/gtestutils.h:109: error: expected identifier or '(' before '/' token17:21
asacprobably was available everywhere before it was standardized :)17:21
asacthey use -ansi ?17:21
CheGuevarayeah17:21
CheGuevaramost weir thing kmplayer compiles perfectly fine locally with pbuilder17:22
CheGuevarabut fails on ubuntu build servers17:22
asachmm ... maybe you still have old compiler?17:23
asacand 4.2 became more anal about this?17:23
CheGuevaraits a hardy pbuilder chroot17:24
CheGuevaraso it should be same as there :P17:24
CheGuevarabut the patch is basic and straight from upstream svn, so it shouldn't do any harm17:25
Ubulettepedantic is anal about C using // which is supposed to be C++ only17:25
CheGuevarayep17:26
asaci wait a few minutes to see if someone from desktop team replies17:26
asacotherwise i will sponsor this thing17:26
CheGuevaraweirdest thing we don't use pedantic in kmplayer17:26
CheGuevarabut it still fails17:26
CheGuevarathats what puzzles me most17:26
asacwell ... pedantic in c++ appears to choke because of ;17:26
Ubulettewhat other flags do you have ?17:26
asacplain C appears to be pedantic when using -ansi17:26
CheGuevarahttp://launchpadlibrarian.net/11089417/buildlog_ubuntu-hardy-i386.kmplayer_1%3A0.10.0c-0ubuntu1_FAILEDTOBUILD.txt.gz17:27
asacUbulette: could build ephy?17:27
asacwith pre3?`17:27
asacb3pre17:27
CheGuevarayeah it must be -ansi then17:27
Ubuletteasac, locally yes. I've pushed it to my ppa17:29
CheGuevarawhich still on't explain why its doesn't fail locally :P17:29
asacUbulette: fine ... does it work?17:31
CheGuevaradoes gcc 4.2 set ansi by default?17:35
armin76asac: i get unaligned access on ia64 with 3.017:40
asacarmin76: in cairo?17:40
armin76no17:40
armin76cairo is okay, tested on ia64 and alpha, checking with amd6417:40
armin76but it doesn't matter, on ia64 it doesn't segfault17:41
armin76http://rafb.net/p/y1747Y58.html17:41
asacCheGuevara:  ls -l /etc/alternatives/c8917:45
asacmaybe your alternative points somewhere else? or CC=c89 is used on buildd by default ... otherwise: no clue :)17:46
CheGuevaralrwxrwxrwx 1 root root 16 2007-12-29 13:58 /etc/alternatives/c89 -> /usr/bin/c89-gcc17:46
CheGuevara$magic++17:47
CheGuevara:P17:47
CheGuevaraasac: let me know when you upload please17:48
asacCheGuevara: sure ... doing a testbuild of ice*17:51
CheGuevarakk thanks :)17:52
asacCheGuevara: uploaded18:34
asaclets see how quick the idle buildd's kick in :)18:39
Ubulettehope it's faster than ppas18:43
Ubulette1h15 avg in ppa just to enter the queue18:43
CheGuevaraw00t18:45
CheGuevarathanks a lot asac18:45
Ubuletteasac, http://ubuntuforums.org/showthread.php?t=65291718:47
asacno idea ... didn't change a thing18:54
Ubulettecould be anything18:57
asacUbulette: does ephy work?19:19
Ubuletteasac, http://www.sofaraway.org/ubuntu/tmp/epihany.png19:23
=== \sh_away is now known as \sh
asacUbulette: does the homepage work?19:24
Ubulettehomepage ?19:24
asacdo forms work? e.g. password manager19:25
asacwell ... you have the ubuntu homepage ... i don't have that because locale is not properly recognized19:25
asacbut my build might just be too old and %l is fixed upstream most likely19:25
Ubulettei can login on lp19:27
asacwell ... does it remember your credentials?19:29
Ubulettelp, yes.19:29
Ubulettebut seb complained about htaccess, not cookies auth19:30
Ubulettehmm self signed certificates.. I get the same error as ff3 but how do I add the exception in epiphany ??19:32
Ubulettecrash19:34
asacUbulette: yes he complained about http auth ... but auto password in forms didn't work either19:36
Ubulettestart epiphany-browser, Edit / Personal data => click on properties => crash19:36
asacmost likely fixed by microb backout ... which is why i asked19:36
Ubulettebackout was post b2, i'm still running b2 here19:37
asacgreat chpe landed a bunch of the xul fixes19:37
* asac updating patch19:37
Ubulettestrange, update manager still wants to upgrade epiphany-gecko19:39
Ubulettehttp://www.sofaraway.org/ubuntu/tmp/Update-Manager.png19:40
Ubuletteover and over again19:40
asacmaybe let him do it?19:43
Ubulettei did19:44
Ubulettesame for apt-get19:44
UbulettePreparing to replace epiphany-gecko 2.21.4-0ubuntu2+fta2 (using .../epiphany-gecko_2.21.4-0ubuntu2+fta2_i386.deb) ...19:44
asacno idea ... if it doesn't go away by doing the upgrade19:49
asacdamn thing ... try to convince someone who didn't even try to build against xr 1.9 that we have no real choice but use the glue :)20:53
Ubulettewho ? mike ?21:16
=== \sh is now known as \sh_away
asacno chpe21:37
asache has no idea ... at least caillon stated that he wants -embedding in the end as well ... but fedora needs to figure out some things for lib32/lib64 mixed setups21:37
Ubulettemiro is damn too slow.21:41
Ubulette15~20sec of intense disk activity to have the 1st window21:41
asacwith the miro patch i provide?21:42
Ubuletteyes, but even without it. it's by design21:44
asacok21:44
asacthen i feel good21:44
asacUbulette: tbird is up ... sorry for the branch mess btw, but you started on top of a lame branch that didn't have all revisions21:50
asachttp://paste.ubuntu.com/3105/21:52
asacyou didn't have any missing symbols in pre?21:53
Ubulettedamn, i was about to push the pedantic thing21:53
asache?21:53
asacthe pedantic thing isn't needed anymore21:53
Ubulettewas test building from scratch21:53
asaci didn't include it21:53
Ubuletteyou could have told me you were working on that21:53
asacUbulette: what do you mean?21:53
Ubulettei've lost 1h once again21:54
asacon what?21:54
Ubulettetb21:54
asacsorry ... don't get what you were working on21:54
asacdropping -pendantic change?21:54
Ubuletteyes, and testing it21:55
Ubuletteall for nothing :(21:55
asacok ... sorry about it ... i said above that i will do test builds before sponsoring ... you probably missed that :(21:55
Ubuletteno, i didn't21:56
Ubuletteyou never mentioned tb21:56
asacyeah ... i now see that that comment was in motu (because CheGuevara pinged me there)21:56
Ubulettei've waited for glib to hit the repo, then i've updated & rebuilt tb, and i was about to push21:57
asacyeah ... but the branch was lost anyway, because it started in the past21:58
UbuletteI used your most recent public branch21:58
asacwhich could have been prevented if you'd told me that you start working on a new tbird branch :)21:58
Ubuletteif you don't push your stuff, how people are supposed to know ?21:59
Ubulettegrrr, i'm pissed21:59
Ubulettenm21:59
asacyour work has been reused ...  i cherry-picked the patches et al21:59
Ubulettebut once again, you took ownership while i requested sponsorship. tell me if you don't want to sponsor me.22:01
asacand remember that i would have preferred to just use your branch ... so i suffered already from my lameness of not pushing the stable branch22:02
asaci really don't get what you complain about ... your contributions are clearly acknowledged in the changelog22:25
asacmaybe its launchpad that leads you to the false impression that contributions within one changelog are worth less or something. this is a deficiency of launchpad for sure. Usually in team-maintenance people commit and add their changes to changelog. the one who uploads signs the release off by using dch -r22:29
Ubuletteit's not the 1st time that i complain about the exact same thing. if you're not happy with something, you should tell it publicly and give a chance to the contributor to fix it, instead of redoing it silently on your side and at the same time gaining ownership of almost everything. that's frustrating.22:31
Ubulettei feel spoiled22:32
asaci don't get ownership of anything22:32
Ubulettewhere are my commits ?22:32
Ubuletteeven from the changelogs, you did 2.0.0.922:32
asacand?22:33
Ubuletteand nothing. i just don't like that kind of practice22:34
Ubulettei won't work on that ever again22:34
asacso you would have preferred if i had told you that you need to redo everything based on the branch i forgot to push?22:35
Ubuletteyes22:35
Ubuletteat least ask me if i would like to do it22:35
asacbut again ... the name in the changelog just means: uploader22:36
asacwhat is done by whome is annotated _in_ the changelog22:36
UbuletteI remember i've asked you to sponsor it, you never said the branch was wrong or anything22:36
asacyes ... because i didn't want to waste your time with these messy things22:37
Ubulettebut forget it, you clearly protect your job, i can understand it22:37

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!