[00:34] for me that make check error looks like upstream issue too ... odd - makes me wonder if its really run by moz [00:35] ah i think i know what the problem is somewhat [00:35] ;) [00:36] check-sync-dirs.py needs to somehow ignore all files created during build if no OBJDIR is used [00:36] right ;) [00:36] good [00:36] that check is a source check ... so shouldnt be run post-build [00:38] or we just workaround by using OBJDIR [00:39] just is relative though ;) [00:39] that logic is kind of entrenched in client.mk i think [00:40] fta: ^^ [00:42] maybe passing GARBAGE and adding that to the matched patterns would help in that .py script [00:42] ? [00:43] fta: check-sync-dir complains about files that are really different [00:43] those files are those that are produced during build ... and that script should just ignore them [00:44] that i figured [00:44] right. but you didnt tell that above ;) [00:44] so solution is to either use MOZ_OBJDIR mechanism to do all biuld stuff in build-tree/obj/ or something [00:45] or fix the script to ignore them in a maintainable fashion ... [00:46] MOZ_OBJDIR does not work with just configure + make ... client.mk is needed or some magic [00:47] for the script fix i thought that using GARBAGE from build system might give a good exclude list for built stuff [00:47] but maybe thats not enough [00:48] maybe ALL_TRASH ? [00:49] and ALL_TRASH_DIRS [00:52] asac: you have time to talk about other stuff? === micahg1 is now known as micahg [00:55] micahg: a bit ;) [00:55] what's happening with pyxpcom? [00:55] 20 minutes before i really have to go [00:55] ok [00:55] is it outside of xul code now? [00:55] i am not sure ... upstream made a standalone project out of it afaik [00:56] someone just opened a bug in xulrunner about producing it [00:56] should I convert to a needs-packaging? [00:56] hmmm [00:56] debian seems to still build out of xulrunner [00:57] ah, they take the whole tree I think [00:57] what do they produce [00:57] 1.9.1 seems to have libpyxpcom.so at least [00:57] but gone in .2 and .3 [00:58] yeah, it seems to be gone [00:58] it's on the lucid idea list :) [00:59] yep. saw that [00:59] and i am positive that it should get done [00:59] https://developer.mozilla.org/en/pyxpcom [01:00] then there's this: http://pyxpcomext.mozdev.org/ [01:01] will it be a new source package then? [01:01] http://benjamin.smedbergs.us/blog/2009-08-07/pyxpcom-welcome-todd-whitema/ [01:01] thats the last thing i remember [01:02] yeah, me too [01:02] so, I guess new source then? [01:02] http://hg.mozilla.org/pyxpcom [01:02] not too much development [01:03] well. not much expected i guess [01:03] with some luck it just works [01:03] let me check [01:04] * micahg still needs to upstream a prism patch as well... [01:06] ok [01:07] for 1.9.1 its broken (fails to build) [01:07] i managed to get it further by copying mozilla-config.h to topsrcdir [01:07] but then it chokes on something odd ;) [01:07] lets first check 1.9.2 [01:11] /usr/bin/python2.5 ../../../config/nsinstall.py -R -m 755 libpyloader.so ../../../dist/bin/components [01:11] ../../../config/elf-dynstr-gc ../../../dist/bin/components/libpyloader.so [01:11] make[4]: ../../../config/elf-dynstr-gc: Command not found [01:13] doh! [01:13] so yes. elf-dynstr-gc is missing [01:14] they need to ship that i guess [01:15] asac: I upstreamed the prism patch, do you want to be copied? [01:15] sure. [01:20] nice, [reed] got that bug fixed and pushed to all 3 branches in under 3 hours [01:21] <[reed]> ;) [01:22] <[reed]> it took some work, believe me! [01:22] asac: I've been tagging stuff fixed-1.9.1.6 and fixed-3.5.6 [01:22] <[reed]> involved several arguments, too [01:24] oh, asac, where can I comment on the lucid-ff blueprint? [01:24] s/where/how/ [01:24] whiteboard [01:25] is the best place [01:25] I can put my thoughts there [01:25] until there is a wiki page [01:26] <[reed]> there's a blueprint? [01:26] <[reed]> link? [01:26] https://blueprints.edge.launchpad.net/ubuntu/+spec/desktop-lucid-new-firefox-support-model [01:27] will check if we need to put up another "hardy" blueprint [01:27] we need a way forward for that too [01:28] ... which probably is much more work than what we actually need to do for lucid :( [01:28] hardy-jaunty-blueprint ;) [01:28] hardy-karmic [01:28] i dont think we should cover that in the same document [01:28] yeah ... ;) [01:29] karmic will run into trouble 6 months after 3.6 is released [01:29] but 3.0 is a different beast [01:29] we can use a similar solution [01:29] karmic feels simliar to what we do in lucid now [01:29] of course not identical [01:29] but should be a subset of problems we have to fight in hardy [01:29] whiteboard updated [01:35] micahg: add a date/name mark in whiteboard [01:35] its really a rudementary thign ... you cant see who said what afterwards [01:35] before my stuff? [01:36] yeah somehow [01:36] be innovative ;) [01:36] how's that? [01:37] heh. mabye a colon ... ;) [01:37] but good [01:37] they're supposed to be working on comments for blueprints [01:37] indentation might help separating it better [01:37] but its probably just because its squashed in a tight space there [01:37] in the past whiteboard definitly was wider [01:40] also added one more thing to Ideas [01:40] yep, nice one [01:41] added also one item about testsuite during build [01:41] with [reed] as proposer ;) [01:42] <[reed]> :) [01:42] https://wiki.ubuntu.com/MozillaTeam/Lucid/Ideas [01:43] <[reed]> yay, I see crash reporter as an item [01:44] <[reed]> I got asked about that today [01:45] i have to talk to pitti about that ... its a bit tough to get the symbols out from the builders automatically [01:45] they are not really connected to net etc. [01:46] but should be some way [01:46] the -dbgsym stuff also comes out to some other not so high security server [01:46] somehow [01:47] we also have this firefox kde session scheduled [01:47] <[reed]> again? [01:47] seems like folks claim that suse patched a lot qt/kde stuff ... felt odd [01:47] they want us to check why we cant have the same :( ... [01:47] <[reed]> they just wrote a helper thing [01:47] <[reed]> it's very ugly [01:47] just the file dialog? [01:47] <[reed]> no, other stuff, too [01:47] <[reed]> :( [01:47] hmm ... i thought there was much more [01:47] yeah [01:48] I thought it was going to get integrated upstream [01:48] well ... but kde folks now have firefox as a top level browser offering in kubuntu [01:48] which is also a good win ;) [01:48] <[reed]> they haven't approached us about it [01:48] <[reed]> I would have noticed [01:48] at best that could be further strengthened by giving them at least a bit of an improvement in experience [01:49] [reed]: afaik it got stuck somewhere during upstreaming ... that session is also about finding out what is going on ;) [01:49] let me check [01:49] <[reed]> I don't even know of a bug on it! [01:49] https://blueprints.edge.launchpad.net/ubuntu/+spec/desktop-lucid-firefox-kde-integration [01:50] see whiteboard ... i asked riddel to add information there [01:50] novell bug 170055 [01:50] Launchpad bug 170055 in inkscape "PRINTING: Predefined paper sizes" [High,Invalid] https://launchpad.net/bugs/170055 [01:50] hmm [01:50] thats odd ;) [01:50] oh launchpad [01:50] suse bug 170055 [01:50] <[reed]> slow ubottu [01:50] http://en.opensuse.org/KDE/FirefoxIntegration [01:50] thats the wiki page i got pointed to a few times [01:51] <[reed]> they have not approached Mozilla about getting it upstreamed, as far as I know [01:51] [reed]: I thought you guys have rules against that type of stuff [01:52] has mozilla approached them about not patching firefox to death=? [01:52] ;) [01:52] <[reed]> are they calling it Firefox? [01:52] i would think so [01:52] i havent heart that they started to debrand [01:52] wouldnt that have been in the news? or did opensuse never ship official branding? [01:53] s/heart/heard/ [01:53] <[reed]> dunno [01:54] <[reed]> never used opensuse [01:54] [reed]: anyway, what i tried to say above is that in karmic users click on browser for first time and can then choose which browser they want to use [01:54] a) firefox ... b) konqueror [01:54] firefox became more important (finally) [01:54] <[reed]> cool [01:54] even for kde devs [01:54] kubuntu devs that is [01:55] http://news.opensuse.org/2009/10/27/sneak-peeks-at-opensuse-11-2-kde-4-3-experience-with-lubos-lunak/ [01:55] looks branded to me [01:55] i see branding there ;) [01:55] <[reed]> mconnor: ^ [01:59] asac: I started targeting my bugs for lucid-alpha-2, I figure 2 months is enough time for what's assigned to me [01:59] the opensuse stuff is all with our approval [01:59] fwiw [01:59] it [02:00] good [02:00] s not perfect, but its interesting, and we want them to push it upstream [02:00] mconnor: do you know anything about it? [02:00] ah ok. so waiting [02:01] well, I did the changeset appro val [02:01] omg, this keyboard is tiny [02:01] mconnor: do you know if those things are done in a way that they detect whether user runs gnome/kde dynamically [02:02] or is that a kde-only build time flag? [02:02] some of both [02:02] iirc [02:03] you should talk to wolfgang though [02:03] is wolfgang still involved much? thought its someone else now... [02:03] <[reed]> he's been submitting patches lately [02:04] <[reed]> including one to make "set desktop background" only work on gnome (and not KDE, as it isn't supported there) [02:05] where's that patch? would like to check how smart that detection is [02:06] mozilla bug 266230 [02:06] Mozilla bug 266230 in Shell Integration "disable "set wallpaper" on non-GNOME environments" [Minor,New] http://bugzilla.mozilla.org/show_bug.cgi?id=266230 [02:06] mozilla bug 140751 [02:06] Mozilla bug 140751 in General "Integrate Mozilla with KDE" [Normal,New] http://bugzilla.mozilla.org/show_bug.cgi?id=140751 [02:07] hmm ... no patch on first bug [02:07] desktop one is bug 526717 [02:07] Error: Launchpad bug 526717 could not be found [02:07] maybe a dupe? [02:07] mozilla bug 526717 [02:07] Mozilla bug 526717 in Shell Integration "remove non-working "Set as desktop background" from the UI for unsupported desktops" [Normal,New] http://bugzilla.mozilla.org/show_bug.cgi?id=526717 [02:08] i think DESKTOP_SESSION=gnome does not work [02:09] i remember someone saying that its not true everywhere when i checked that [02:09] oh yeah ;)( [02:09] thats just the name of the session in gdm [02:10] so you could just set it up as "my gnome" [02:10] and it would be "my-gnome" [02:11] will double check that tomorrow [02:11] and comment [02:11] k [04:16] asac_: I hope you're sleeping :) === asac_ is now known as asac [05:35] [reed]: you around? [05:36] [reed]: your rapid mozilla test broke some test suites and they're going to back it out of 3.5.6 [05:37] <[reed]> I'm dealing with it [05:37] ok [05:37] just wanted to make sure you saw :) [09:18] hi [09:23] hi asac [09:23] I fixed 1.9.1.6 daily, bumped cairo to 1.8.8 [09:26] asac: no urgency, but if you have a chance, could you look at bug 476129 [09:26] Launchpad bug 476129 in firefox-3.5 "Update to Karmic fails to remove Firefox-3.0, causing some Firefox problems" [Undecided,New] https://launchpad.net/bugs/476129 [09:27] * micahg is off to bed [09:30] good night! [10:52] hello everybody [10:54] does anyone know what's the "distribution.canonical.bookmarksProcessed" filter [10:54] in the about:config? [10:54] i can't find any information on it and it's name sounds a little bit weird [11:05] reads like a flag that remembers if distribution specific bookmarks were imported or something [11:10] thanks alot [11:11] np [12:24] Howdy all. I have a weird problem with XHTML/XSL where it works as expected in Firefox 2 , but behaves a little weird in Firefox 3. When I click on page links in Firefox 3, I have to click in some cases several times before the browser will actually render the page. Firefox 2 goes to the links and renders the first try, as does Opera. Any clues as to the hit and miss rendering of clicked links? [12:25] drclue: maybe extension issue? [12:27] Well, I have a guinea pig friend who installed Firefox bare footed and got the same result [12:28] also on windows? [12:28] or just ubuntu issue? [12:29] I get the problem on both ubunto and windows [12:29] asac, hi, have you had a chance to look into scheduling the FF translations session we were talking about the other day? [12:30] I've run every page through the W3 validation engines too. [12:33] What's even weirder is that while it may take a bunch of clicks to get a link to render , the browser does notice the title of the new page. but then the render just stals [12:34] I can even type the location into the address bar and get the same result [12:34] or hit the reload button. [12:35] After some random number of tries FF3 wakes up and renders the page [12:35] dpm: https://blueprints.edge.launchpad.net/ubuntu/+spec/desktop-lucid-firefox-translations [12:35] dpm: can you subscribe folks you think are interested in that? [12:36] also maybe dump first thoughts to the whiteboard [12:37] drclue: you can install dailies for firefox-3.6 and firefox-3.7 and use them without risking profile problems. .... checkout https://edge.launchpad.net/~ubuntu-mozilla-daily/+archive/ppa [12:38] would be good to verify that its an issue on 3.7 and then file a bug in bugzilla.mozilla.org i would think [12:38] drclue: hmm [12:38] asac, cool, thanks, yeah, I'll let translators know and add more content on the whiteboard or on a wikipage for the spec. Has it already been scheduled? [12:38] drclue: maybe try to disable IPv6DNS [12:38] in about:config [12:39] change network.dns.disableIPv6 [12:39] to true [12:39] and restart [12:39] if all other pages work normally its probably not the problem [12:39] dpm: just created it and pinged pitti [12:39] i would think it doesnt take long [12:40] ok, thanks [12:40] I might have to wait until my new box gets in to do some of that dailies stuff, but I could certainly try the about:config thing on one of these boxes I have. [12:40] I think I'm just tickling a soft spot with this application/xhtml+xml XHTML/XSL/XHTML stuff [12:43] yeah. please try still [12:43] the behaviour that sometimes things dont load was also reported wih IPv6 [12:43] by some users [12:47] asac: no joy on the network.dns.disableIPv6 thing. I set it as requested , restarted and cleared cache etc , same behaviour [12:49] k [12:49] so yeah. do the dailies and file upstream ... [12:50] I do have a whole server to myself on this one , and could show the site [12:52] It will do it's weirdness even on the about / contact / privacy type pages [12:52] url? [12:52] 207 154 83 202 [12:53] thats not a url ;) [12:53] * asac wants to click :) [12:53] http://207.154.83.202 [12:53] one url for a page that doesnt render [12:53] so to about [12:53] That one one don't [12:54] http://207.154.83.202/about.html? [12:54] http://207.154.83.202/?view=about [12:55] works for me ;) [12:55] navigation, clicking, reloading ... works [12:55] use wget [12:56] and see if you get the full page at all [12:56] Works on my FF2 and my Opera, but not on my FF3 on the ubuntu nor on FF3 on the shop windows box [12:56] yes thats strange [12:56] i run 3.5 ;) [12:57] double check that you have no extensions installed [12:57] firefox extensions [12:57] 3.5.4 on my ubuntu [12:57] The shop box has the latest official release and is totally stock [12:58] i run the latest 3.5.6pre daily ... maybe try that then [12:58] or the 3.5.5 release from the security ppa (will be pushed out today) [12:58] https://edge.launchpad.net/~ubuntu-mozilla-security/+archive/ppa [12:58] Well , I need to be able to sort this out for the masses [12:58] yes. if 3.5.5 fixes it [12:58] it might be that it was a 3.5.4 regression ;) [12:58] check what version you use on windows [12:59] 3.5.4 had a few regressions ... users get auto updated :) [12:59] but not so sure at all [12:59] what your problem is ;) [12:59] ... besides extensions [13:00] Well , in looking around , it's not just me. I would agree that sometimes it feels like I'm the only one pushing this XHTML/XSL/XHTML stuff around , but there are others [13:02] the doctype in the same line as the stylesheet instruction looks odd [13:02] is that valid xml? [13:02] Well, it validates [13:02] also you use XHTML strict in doctype ... so whatever you produce must be strict ;) [13:03] It's strict [13:03] I never use transitional [13:03] ok [13:04] yes. and it works here ... so i would think its either a 3.5.4 only issue or something is wrong on your side ;) [13:04] I've been hand coding this stuff since before Netscape [13:05] I do make mistakes , but I can't find one here in the tagging [13:05] i asked someone else to see if it renders something [13:05] well ... i would suggest to produce a minimal testcase [13:05] really [13:06] but first ensure you tried 3.5.5 first [13:06] I like XML/XHTML and such just because it is so bitchy about the syntax and lets nothing slide [13:06] ;) [13:07] ok got a confirm that this page renders for others in 3.5.4 [13:07] i think its you :) [13:08] network problem could be ;) [13:09] When I get my new PC in here in the next few days I'll put the 3.5.5 on it. I have three different machines here and several different browsers. I also have machines located a couple hundred miles away on another network. [13:10] I've been walking this problem around and it likes to chew on FF3 [13:11] I'm thinking that maybe when the browser got a zillion times faster , some of the application/xhtml+xml stuff missed the boat [13:16] At least my FF 2.0.0.11 , Opera likes the application/xhtml+xml stuff. Most times I can even get my IE6 to like it by the time a project is done [13:26] With FF3.5.x rendering these pages some place(s) and not other(s) , there is a probably a hole somewhere in the rendering, or it's exploitations used to obtain the speed. [13:26] As to the validity of the XML , click the W3C checker links. If ya like I can throw a switch in the code and it turns on validation links in every page. I checked each and every XML,XSL and CSS file this very day. [13:27] As well as the XHTML too , which is well XML :) [13:30] There I turned the switch , now each and every page has "[" and "]" at the bottom/ The "[" checks the XHTML and the "]" checks the "CSS" . [13:35] Actually the XSL file maps the "[" and "]" into the pages and a CSS class toggles their display. I also have one PHP flag I set that allows the validation engines to map to a pseudo [13:35] user so they can scan pages that require a login [13:52] well ... i am not here to debug buggy webpages [13:52] make a minimal test ;) [13:52] case [13:53] that shows there is a bug [13:53] and if you dont test 3.5.5 i wont check anyway [13:58] dtchen: if i just want to record something using alsa ... do i need to kill pulseaudio? [13:58] e.g. will that lock the thing? [13:59] how do i best test if capturing through also would work ... i assume sound recorder uses pulse [14:04] asac: I have a new machine on order, just for the purposes of running bleeding edge versions of the browsers. My regular development machine here ( a debian ) is intentionally kept well behind the curve so as not to drop audience that does not update often. My recreational machine is ubuntu 9.10 with the FF 3.5.4. The windoze box is an XP with a FF 3.X (not sure, as I don't bother with that one too often) [14:05] On my other network on the other side of the state most everything is windoze boxes except for my Asterisk server which runs a centOS [14:31] how come noone cares about fixing bug 243344 with so many dupes? [14:31] Launchpad bug 243344 in scim-bridge "scim-bridge crashed with SIGSEGV in scim::IMEngineInstanceBase::get_frontend_data()" [Medium,Confirmed] https://launchpad.net/bugs/243344 [14:31] firefox triggers it ten times a day for me (on x64) [14:32] 92 dupes :P [14:35] we already talked about that [14:39] not about why it's no fixed, or maybe i'm just too tired to remember.. [14:39] i cant tell why its not fixed. didnt i susbscribe themuso? [14:40] i only think that there are three crashes [14:41] two of them are high dupe count and are just on shutdown [14:42] why themuso? [14:42] he does a11y [14:42] ah [14:42] sorry [14:42] then arnegoetje [14:42] he does bridges [14:42] scim et al [14:48] checking for cairo >= 1.8.8 freetype2 fontconfig... Requested 'cairo >= 1.8.8' but version of cairo is 1.8.6 [14:49] 1.9.1/3.5 red on jaunty and older [14:49] fixing that before the bot kicks in [14:50] thought micah already fixed that [14:50] really? [14:50] right, nm [14:51] at least i gave ack yesterday to bum pit [14:52] asac, do you have access to a lpia box somewhere? [14:53] yes. [14:53] but i am not sure we want to fix that [14:53] i heard rumors it might go away [14:55] just wanted to get a backtrace [14:55] but well, i don't mind [14:56] i'm not even sure we have a single lpia user for those builds [14:57] almost certainly none [14:57] let me check if i can log in etc. [14:58] chromium-browser 19172 1.46% 3508 8790 6870 4 [14:58] google-chrome-unstable 16012 1.22% 4288 5842 5863 19 [15:03] asac: hey. so I've tested ff35 on jaunty and karmic on amd64/i386 and it seems good. when do you want me to publish? [15:06] jdstrand: whenever you want ... karmic is tested here too ... jaunty i was setting up again as previous i accidentially busted [15:07] asac: wrt jaunty> do you mean you broke the testing environment or the package is busted? [15:08] testing env is broken [15:08] asac: shall I hold off on pushing jaunty then? [15:09] no [15:09] ok [15:09] if you tested it its fine [15:09] asac: also, I've got a merge request pending for some apparmor profile updates for the next karmic SRU [15:10] asac: I haven't looked at lucid, but it would be good to get all that uploaded there too [15:11] asac: would it make sense to give me commit access with the promise I will only touch the apparmor bits? (review can still happen via pings or prior to upload) [15:12] asac: oh, I've not tested lucid-- do you want me to push ff35 for lucid too? [15:14] yes ... just push there [15:14] thx [15:14] cool [15:34] [reed], http://paste.ubuntu.com/316039/ [16:00] fta: did you omit the sync-dir test? [16:00] or is this failure now before? [16:03] i used an objdir [16:03] cool [16:03] manually or already properly in package? [16:04] properly [16:07] asac: hi , did you happen to catch comment > https://bugs.launchpad.net/humanity/+bug/439172/comments/12 [16:07] Launchpad bug 439172 in gtk+2.0 "nm-applet icon blurry on 25+ pixel panel" [Low,Confirmed] [16:11] !test [16:11] yes, I'm alive. [16:47] asked in #nm about that [16:51] hehe , yeah , i thought something was wrong with my connection ;)... lets continue in #nm [17:30] asac, chromium is also impacted by the flash sucking ~20% cpu even when idle :P [17:30] bad flash [17:34] asac: pasuspender -- arecord -twav foo.wav [17:34] asac: but yes, Sound Recorder uses GSt's audiosink [17:37] asac: killing PA is insufficient; you'd also need to disable PA's autospawn (echo autospawn = no|tee -a ~/.pulse/client.conf && killall pulseaudio) [17:37] dtchen: thanks. its odd but after poking around three hours ... it suddently started to work [17:37] but i didnt change anything ;) [17:37] at least not in alsamixer anymore [17:38] so pasuspender does temporary pulse downing? [18:41] [reed], TEST-UNEXPECTED-FAIL http://launchpadlibrarian.net/35543641/buildlog_ubuntu-karmic-i386.xulrunner-1.9.3_1.9.3~a1~hg20091111r34772%2Bnobinonly-0ubuntu1~umd1_FULLYBUILT.txt.gz [18:58] because loading that whole file sucks, TEST-UNEXPECTED-FAIL | trace-test.py | /build/buildd/xulrunner-1.9.3-1.9.3~a1~hg20091111r34772+nobinonly/build-tree/mozilla/js/src/trace-test/tests/sunspider/check-date-format-tofte.js: /build/buildd/xulrunner-1.9.3-1.9.3~a1~hg20091111r34772+nobinonly/build-tree/mozilla/js/src/trace-test/tests/sunspider/check-date-format-tofte.js:302: TypeError:... [18:59] erk, sorry about that :( === av`__ is now known as av` [22:26] fta: can I talk to you about the songbird patches later? [22:26] yep [22:48] anyone having issues sending mail with tb3.0 from nightly? [22:49] i upgraded earlier today and now I seem to not be able to send via smtp [22:52] [reed], u there? [22:52] <[reed]> fta: yeah [22:52] [reed], did you see my TEST-UNEXPECTED-FAIL? [22:52] <[reed]> which one? [22:53] <[reed]> (no) [22:53] [reed], TEST-UNEXPECTED-FAIL http://launchpadlibrarian.net/35543641/buildlog_ubuntu-karmic-i386.xulrunner-1.9.3_1.9.3~a1~hg20091111r34772%2Bnobinonly-0ubuntu1~umd1_FULLYBUILT.txt.gz [22:53] because loading that whole file sucks, TEST-UNEXPECTED-FAIL | trace-test.py | /build/buildd/xulrunner-1.9.3-1.9.3~a1~hg20091111r34772+nobinonly/build-tree/mozilla/js/src/trace-test/tests/sunspider/check-date-format-tofte.js: /build/buildd/xulrunner-1.9.3-1.9.3~a1~hg20091111r34772+nobinonly/build-tree/mozilla/js/src/trace-test/tests/sunspider/check-date-format-tofte.js:302: TypeError:... [22:54] yes, where I learned to check how long a line is before copying :) [22:54] <[reed]> TEST-UNEXPECTED-FAIL | trace-test.py | /build/buildd/xulrunner-1.9.3-1.9.3~a1~hg20091111r34772+nobinonly/build-tree/mozilla/js/src/trace-test/tests/sunspider/check-date-format-tofte.js: /build/buildd/xulrunner-1.9.3-1.9.3~a1~hg20091111r34772+nobinonly/build-tree/mozilla/js/src/trace-test/tests/sunspider/check-date-format-tofte.js:302: TypeError: Assertion failed: got "2007-01-01Monday, January 01, 2007 1:11:11 AM2007-01-02Tuesday, Ja [22:54] <[reed]> nuary 02, 2007 0:35:37 AM2007-01-03Wednesday, January 03, 2007 0:00:04 AM2007-01-03Wednesday, January 03, 2007 11:24:31 PM2007-01-04Thursday, January 04, 2007 10:48:58 PM2007-01-05Friday, January 05, 2007 10:13:25 PM2007-01-06Saturday, January 06, 2007 9:37:52 PM2007-01-07Sunday, January 07, 2007 9:02:19 PM2007-01-08Monday, January 08, 2007 8:26:46 PM2007-01-09Tuesday, January 09, 2007 7:51:13 PM2007-01-10Wednesday, January 10, 2007 [22:54] <[reed]> 7:15:40 PM2007-01-11Thursday, January 11, 2007 6:40:07 PM2007-01-12Friday, January 12, 2007 6:04:34 PM2007-01-13Saturday, January 13, 2007 5:29:01 PM2007-01-14Sunday, January 14, 2007 4:53:28 PM2007-01-15Monday, January 15, 2007 4:17:55 PM2007-01-16Tuesday, January 16, 2007 3:42:22 PM2007-01-17Wednesday, January 17, 2007 3:06:49 PM2007-01-18Thursday, January 18, 2007 2:31:16 PM2007-01-19Friday, January 19, 2007 1:55:43 PM2007-01-20Sa [22:54] <[reed]> turday, January 20, 2007 1:20:10 PM2007-01-21Sunday, January 21, 2007 12:44:37 PM2007-01-22Monday, January 22, 2007 12:09:04 PM2007-01-23Tuesday, January 23, 2007 11:33:30 AM2007-01-24Wednesday, January 24, 2007 10:57:57 AM2007-01-25Thursday, January 25, 2007 10:22:24 AM2007-01-26Friday, January 26, 2007 9:46:51 AM2007-01-27Saturday, January 27, 2007 9:11:18 AM2007-01-28Sunday, January 28, 2007 8:35:45 AM2007-01-29Monday, January 29, [22:54] <[reed]> 2007 8:00:12 AM2007-01-30Tuesday, January 30, 2007 7:24:39 AM2007-01-31Wednesday, January 31, 2007 6:49:06 AM2007-02-01Thursday, February 01, 2007 6:13:33 AM2007-02-02Friday, February 02, 2007 5:38:00 AM2007-02-03Saturday, February 03, 2007 5:02:27 AM2007-02-04Sunday, February 04, 2007 4:26:54 AM20 [22:54] yep, that one [22:54] <[reed]> guh [22:55] * [reed] looks that test up in mxr [22:57] <[reed]> got a good char-by-char diff? [23:02] <[reed]> fta: seems like it's an hour off [23:02] <[reed]> interesting [23:02] TZ issue then [23:07] <[reed]> fta: mozilla bug 515254 [23:07]