[09:35] asac: it needs some more error handling, help messages and a manpage [09:38] asac: did chromium support pdf viewer for the time being? [09:41] e-jat: why not just try ;) [09:42] asac: i have try .. but nothing appear :) [09:42] using it right now .. [09:42] ive try it for slideshare.net [09:42] i guess not. you probably would need to enable plugins [09:42] asac: owh .. ok .. [09:42] so i need to copy pdf plugin that was use in ff ? [09:42] :) [09:43] you can enable that experimental plugins feature by setting some magic in /etc/chromium*/default [09:43] is it? [09:43] no clue about the current plugin state. wait for fta [09:43] bdrung: yeah. i wrote a test suite like thing. looks pretty good. [09:44] not yet sure which name to use for it though [09:44] moz-compare-versions ;) [09:44] or mozcmp [09:44] := [09:44] or something else? [09:48] CHROMIUM_FLAGS=""i can see only [09:48] :) [09:50] e-jat: you will figure. its probably coded in chromium-browser warpper script [09:50] otherwise wait for fta ;) [09:50] mean .. need to specify : --enable-extensions --load-extension="/some/path" [09:50] asac: ok i will :) [09:51] so far so good with the chromium .. after flash seem working fine :) [09:51] http://identi.ca/notice/6796569 [09:51] so flash works? [09:51] how did you enable it? [09:53] asac: i copy libflashplayer.so to /usr/lib/chromium-browser/plugins/ [09:53] then run it with --enable-plugins [09:54] k [09:54] r0x :) === reed is now known as Guest3224 [11:27] asac, hey [11:32] aquarius: hi [11:33] asac, your CouchDB patch from yesterday (https://issues.apache.org/jira/browse/COUCHDB-435) wasn't quite enough, it turns out; the spidermonkey API apparently changed, and another chap did a different patch: https://issues.apache.org/jira/browse/COUCHDB-288 [11:34] asac, your views on those would be interesting; I don't know enough about spidermonkey to evaluate the other guy's patch to know whether it's the Right Thing or not [11:38] aquarius: what was the problem with that patch? [11:39] asac, nothing at all, it's just that in addition to JSOPTION_NATIVE_BRANCH_CALLBACK going away, the API to call spidermonkey changed too, meaning that your patch made couch compile fine but the other guys patch makes it actually work :-) [11:39] this is by no means your fault :) [11:40] aquarius: what i asked was : what didnt work? [11:40] asac, ah, when couch tried to use a JS view function, it threw the error "expression does not evaluate to a function" [11:41] ok ... will check the patch then [11:41] i mean the one from 288 [11:42] cheers, chief :) [11:58] asac, is the guy who did the torbutton fix a new contributor? [11:58] asac, if yes I won't run dch -r [12:03] asac, contents here: http://paste.debian.net/42831/ [12:20] andv: sponsoring never involves dch -r [12:21] if you change more things you can either take over changelog or be nice and just add you to the changelog [12:21] or ask the the guy to fix the patch [12:21] asac, dch -r do the work for you [12:21] (though if its trivial its not really worth a round) [12:21] asac, it creates a path for the previous changer [12:21] and adds one for you where you can add your changes [12:21] andv: dch -r will take over changelog [12:21] thats not sponsoring [12:21] asac, no, it creates a [ name ] [12:22] path with personal changes [12:22] andv: it also takes over ownership [12:22] andv: just add that [ ... ] manually [12:22] asac, yes, it's what I did [12:22] good. then its fine [12:22] just updated timestamp [12:22] asac, look at package contents and tell me if they are fine [12:22] asac, http://paste.debian.net/42831/ [12:23] andv: looks fine. if its tested go for it [12:23] k [12:23] next ;) [12:23] ok running some errands and doing lunch [12:24] ok, I finish this and I go to prepare lunch too [12:27] asac, the diff shows some changes [12:27] asac, to torbutton.js [12:27] file [12:27] - proxy_host = '127.0.0.1'; [12:27] + proxy_host = 'localhost'; [12:27] is that normal0 [12:27] ? [12:27] e.g it gets auto-generated during build? [12:29] asac, it's not on his debdiff so I guess it's auto-generated during build [12:31] andv: you are a motu [12:31] ;) [12:32] asac, I just wanted a clean diff :) [12:32] andv: yeah. start with a clean tree [12:32] apply the reviewed patch [12:32] asac, I know, and I get that change [12:32] asac, which is not in his debdiff [12:33] asac, the only thing possible is that that file gets auto-generated during build [12:34] and just debuild -S [12:34] [12:34] andv: yes. loads of packages are bad and dont properly clea n up [12:34] thats normal but one shouldnt bother with it when sponsoring [12:34] just take care that you dont produce sources from a tree you used to build ;) [12:34] its annoying, but thats life :) [12:35] asac, yeah, I was just having a look at the diff [12:35] and found out that thing [12:35] so I asked you if that was normal [12:35] nothing more [12:36] everytime i do something to packages not done by me i dpkg-source -x ... modify, create sources .... and if i do a binary build for testing i trash the tree again and unpack the dsc [12:37] ok bad connection atm [12:37] and well spotted. you are right that diff shouldnt be in teere ;) [12:37] ok ... have to run out and connectivity is gettin worse. will be back in 30 minutes i guess [12:37] asac, I leave it as it is? [12:38] it's something really minor [12:38] asac, want me to push anyway? [12:39] asac, I looked at old package previously uploaded [12:39] and that change is still there [12:40] so I assume it's auto-generated [12:40] fine then === penny|wise is now known as pennywise [13:26] andv: a clean debdiff is what we want [13:26] andv: dont use _built_ trees to produce the sources [13:26] asac, I know [13:26] good [13:26] asac, I uploaded it more than 30 mins ago [13:26] but I received no mails [13:26] great [13:27] maybe the archive don't recognize my sha-256 key [13:27] ... [13:30] I'm asking on LP [13:48] ping fta === rickspencer3-afk is now known as rickspencer3 [14:15] hi! why not include jssh support for ubuntu build of mozilla ? [15:06] exn: not sure what jssh is ;) [15:06] exn: you are asking to have it packaged or to instal it by default? [15:07] I'm ask to compile it with this options by default mk_add_options MOZ_CO_PROJECT=browser [15:07] mk_add_options MOZ_OBJDIR=@TOPSRCDIR@/firefox-jssh [15:07] ac_add_options --enable-extensions=default,jssh,webservices [15:07] ac_add_options --enable-application=browser [15:07] even redhat has firefox compiled with jssh [15:08] asac: yes, it will be cool to have it by default [15:08] I'm tired to recompile it every time [15:09] and it just option which can be enabled anytime [15:09] like firefox -jssh [16:18] exn: since when does jssh exist? [16:18] asac ? [16:20] since when does it exist [16:20] just want to estimate how likely it is that this goes away [16:21] in next release or something [16:21] we were burned in the past by shisping stuff that upstream does not support [16:21] from 1x versions [16:22] check it http://tinyurl.com/685tac [16:22] there described how to compile it with jssh support [16:23] for 3x version [16:23] yeah. i will think about it. file a wishlist bug for now [16:24] also check if there ar other extensions we might put into a firefox-3.0-extras package or something [16:24] after compiling it must looks like this http://pastie.org/561804 [16:24] well. compiling and doing is not the problem ;) [16:25] its just that unsupported stuff shoudlnt be in a firefox build with official branding [16:25] i will check with fedora guy, why and how they ship it [16:26] so for now: file a wishlist bug ;) [16:26] against firefox-3.5 [16:30] it will be cool to have jssh immediately after apt-get install =) [16:30] I don't know will it be in extras or not [16:31] I think no, because xulrunner and libs must be compiled with this option. but I not sure [16:31] i will check it. i think its basically just chrome and components [16:32] can be shipped outside [16:32] otherwise we wont do it i guess [16:32] thanks [16:32] file bug ... tell me id and I will check when i have time [17:03] maybe somebody knows how to run mozilla without an any kind of X ? [17:04] exn: any kind of X? [17:04] asac , yes, without Xvfb, without desktop [17:05] just one process of mozilla and no more [17:05] without rendering ! === ejat is now known as e-jat [17:10] exn:what do you want to do [17:10] it's sounds bad, I know. if I will work with it through jssh interface, rendering is not needed, I mean do not need to draw pages anywhere [17:11] just disable rendering, or something like this [17:11] it's possible ? [17:12] exn: why do you want to run firefox just for jssh? [17:12] what are you doing with it? [17:14] testing javascript [17:32] hey asac [17:33] hi exn [17:33] hi eagles0513875 ;) [17:34] kde is broken lol woohooo [17:39] exn: kde networkmanager? [17:39] eagles0513875: ^ [17:39] (sorry exn) [17:40] no [17:40] the desktop as a whole [17:40] eagles0513875: but ou had problems with kde network manager right? [17:40] i need you to test something then ;) [17:41] the login is broken there have been a whole bunch of kde updated packages and updating to the latest ones seems to have issues i login then my screen goes blank then all i see is the mouse and black screen nothing shows that the desktop is loading [18:19] hi [18:26] hey fta [18:36] is it just me or is http://packages.ubuntu.com/ dead? [18:41] what you mena not loading [18:42] fta: sites down [18:42] if you wanna know what is honestly dead is something with kde login === micahg1 is now known as micahg === rickspencer3 is now known as rickspencer3-afk [20:08] asac: did you wrote a test program like http://paste.ubuntu.com/235370/ ? [20:47] A chromium-v8-13149/src/test/mjsunit/proto.js [20:47] svn: Decompression of svndiff data failed [20:47] make: *** [gos-co] Error 1 [20:48] forcing a respin of the bot.. [21:05] is the version of mozjs in xulrunner 1.9.1 supposed to be the same as spider monkey 1.8? [21:05] i'm getting some weird differences between them - including a symbol being in the headers but in the library [21:30] hey hey hey . guud evening [21:30] * BUGabundo is Happy.... enjoy while you can :D [21:32] BUGabundo: is BUGaBOTdo a better nick? [21:33] * BUGabundo ignores Mika [21:38] bdrung_: something similar ... more complete [21:39] bdrung_: but in sh ;) [21:39] and done manually. but i dont think that there is much need to extend the test much [21:39] i mean manually as yours, but i also test gt ge for instance nd le [21:40] erh: i dont know about js versioning [21:51] asac: the really weird thing is that there is a symbol in a header file js_RegExpClass that is not in libmozjs.so when i build spider monkey myself, its in the library [21:55] asac: i only tested the cases listed on the website [22:01] asac, http://paste.ubuntu.com/235432/ [22:03] identica is down!!! [22:03] who will have I to nag now ?!!? [22:03] OMG OMG OMG [22:05] BUGabundo: I thought it was distributed so it can't go down [22:06] Laconica support OMB and Federation it is [22:07] *but* identica is just ONE service [22:07] it can go and has been down [22:09] you killed it [22:10] damn it [22:10] GMail spend the entire day killing my FF [22:10] both here on 3.6 [22:11] and on debian iceweasel 3.0.9 [22:11] stupid gmail [22:12] BUGabundo, so you're using debian now? [22:13] at work [22:13] I told you that like 3 weeks ago [22:16] BUGabundo, now that i've unsubscribed from your dents, i may pay more attention to what you actually say ;) [22:16] I said it HERE [22:16] fta: you shameless ex-follower :( [22:16] no offense ;) [22:17] fta: http://www.brainbird.co.cc/bugabundo [22:17] sub to me there [22:17] much less _noise_ [22:17] i'm only on identi.ca [22:18] so? [22:18] its federated!! [22:18] duh [22:19] LOLOL [22:19] I'm subbed to ppl on at least 5 other Laconica servers [22:19] from my identica account [22:19] actually my identica and my brainbird are subbed to each other [22:19] LOL [22:19] Incorrect username or password. [22:20] how do i know they didn't collect my password in the process? [22:20] shame! [22:21] ahh? [22:21] what ppassword? [22:21] you should have been returned to your idenitca accont [22:21] you are pressing the worng button [22:21] let me sign out to give you the DIRECT link [22:21] http://www.brainbird.co.cc/main/remote?nickname=bugabundo [22:22] no enter http://identi.ca/fta in the 2nd field [22:23] No suitable nodes are available to serve your request. [22:23] GRRRR [22:23] http://mashable.com/2009/07/27/firefox-40/ <- anyone see the chromium/chrome resemblance? [22:24] fta: see it now ? [22:24] assuming Identica is up [22:25] nope, nada, i gave up [22:25] :( [22:25] try again later [22:25] when its up [22:25] pace_t_zulu, ff 4.0? ff 3.7? wtf is that? [22:25] did you undertood how to sub to a remote account fta? [22:25] pace_t_zulu: saw it and read all about tit [22:26] fta: mockups of the next looks of FF [22:26] on version 3.7 [22:26] that will lead to 4. [22:27] they borrowed liberally from google [22:27] why not, chromium borrows a lot from firefox [22:28] ahahahah [22:29] fta: +1 [22:30] bdrung_: its ok. i tested almost everything. its good. i will commit it ... thanks! [22:30] erh: how do you checkout/build spidermonkey [22:30] ? [22:34] asac, http://code.google.com/p/chromium/issues/detail?id=16800#c17 [22:34] so system v8 is a problem... [22:56] fta: please disable shred build for now [22:56] but keep v8 [22:56] it's off [22:56] i never enabled shared build [22:57] ok [22:57] fta: i dont understand the problem though [22:58] can you explain what exactly is broken now? [23:00] localtime is called by v8, but the binary being jailed (sandboxed), it doesn't have access to /etc/localtime so fails to apply the correct tz [23:01] fta: aha. how does jailing work? [23:01] which binary is being jailed? [23:01] how do they detect in which binary they are (code wise) [23:01] it's a chroot on an empty dir (deleted dir) [23:02] is there some document online? that gives an overwivew? [23:03] chromium searches for chromium-sandbox, if it finds it, it starts from there, then all children (the renderers) are jailed === rickspencer3-afk is now known as rickspencer3 [23:03] yes, most probably [23:03] http://code.google.com/p/chromium/wiki/LinuxSUIDSandbox [23:08] fta: http://www.brainbird.co.cc/main/remote?nickname=bugabundo [23:08] can you try now? [23:08] and enter http://identi.ca/fta in the 2nd field [23:12] fta: yeah. but seems its not many cases that this breaks. in general it works pretty well [23:12] i will check the links though. thanks! [23:13] asac, run javascript:alert(new Date().getTimezoneOffset()) [23:13] asac: can you push the branch? [23:15] BUGabundo, Bad Request, No authorization request! [23:15] grrr [23:15] this is supposed to be easy [23:15] I've done it 250 times before [23:16] fta: are you loggin on identica?? [23:17] asac, http://paste.ubuntu.com/235480/ [23:17] BUGabundo, of course i am [23:18] are you sure ? :p [23:19] fta: (11:17:33 PM) shiny: remote subscriptions are..... flakey [23:19] (11:18:12 PM) evanpro: shiny: I think there's a problem with the session storage [23:23] bdrung_: urgently? i am not logged onto the system. [23:23] asac, i will disable it for now, please experiment in a sandbox ppa [23:24] fta: well. what i wanted the daily builds for atm is to track how well abi/api works [23:24] i cannot do that in a sandbox [23:24] for me the current regression seems ok. but if you think its too bad go ahead [23:25] but good to see that they plant to stabilize v8 [23:26] on their own [23:26] asac, my plan was always to push upstream to do the right thing for us, even if i sometimes have to do the patch, and use that *when it's ready*, but never to enable stuff before they are accepted [23:26] and settle on 1.2 branch [23:26] fta: the discussion i read didnt sounded like they would care much about sonames for now [23:26] asac, many people use chromium with gmail, so their emails have a busted date [23:27] soname is a different topic, it's about system libs here [23:27] if upstream says they will do stable abi stuff for v8 its ok. i wont bother to put any work in this. but your paste is the first i see that they want to do that [23:28] fta: yes. but what i did i didnt do for the sake of ssytem libs it was for the sake of checking how abi works out [23:28] not for chomium, but to check if v8 can be a viable alternative for mozjs in the archive and the freesoftware that needs js [23:28] chromium wsa just a good build target to check if soname breaks [23:28] far better than the shell testcase they proposed imo as it is massive real live usage [23:28] asac, as i said many times, webkit, v8 and skia should be the last libs we should try to get in, because they are the ones that are not ready [23:30] all the rest isnt important for the distro [23:30] v8 is what i want to understand [23:30] we have a pressing need to get any stable js lib [23:30] but we have none [23:30] the v8 api looked simple enough to be trackable thats what i did here ;) [23:30] asac: it not urgently. [23:30] bdrung: just want to test or to improve the rest? [23:31] bdrung: i didnt modify your patch so you can work on top of that [23:32] asac: i only replaced the dpkg command by mozcmp and added an "a" to the comparision for maxversion. [23:32] fta: anyway today i made some political progress on the mozjs abi front so the urgency of v8 might drop again [23:32] asac, i'm all for experimenting stuff, but not when it's a confirmed regression for users [23:33] fta: sure. i wasnt aware about the regressions. for me it looks ignorable for daily build archive (not in a release archive for sure). but its your decision [23:33] asac, what else would use v8 anyway? it's not even complete [23:33] asac: s/it/it's/ [23:33] fta: what is missing its a js engine that works good enough fo r abrowser [23:34] so its probably good enough for almost everything [23:34] asac, well, the problem is that we just have those daily builds, nothing else, so all users are using the same thing, and they will eventually suffer from our experiments [23:34] seems like you can do all the basics like definining functions objects, running them [23:35] fta: i understand and its your decision because you drive this. maybe we can make the packageing smart so that it will pick up system v8 in a repository that has v8 in it [23:35] and move the daily v8 to some other ppa [23:35] eh?? "fo r abrowser" ? you mean the unbranded firefox? you want to swap the js engine?? [23:35] in that way i could regularly copy over [23:35] fta: for a browser == its used by chromium. its definitly complete enough to be used [23:36] fta: i didnt want to say that another browser should use it. what i meant was that for apps that need it now (like couchdb, google gadgedts) its definitly complete enough [23:36] yeah maybe, then try it there, chromium has obviously troubles with it [23:36] though to some degree i feel tempted to build ephy webkit against chromium webkit ;) [23:37] fta: those apps do not exist yet. i need chromium to test how good the abi is [23:37] try [23:37] so i can suggest folks to use that and promote it in the archive. [23:37] i will check what we can do. [23:38] v8 could probably already go, it has everything now [23:38] what i want i sthat i can copy chromium browser to a special ppa where it at best would pick up libv8 system [23:38] asac: good night === BUGabundo1 is now known as BUGabundo [23:38] fta: thats what i mean. and i didnt want to experiment weith chrome because of chrome, just beacuse its a great app that probably uses most symbols and so we would see abi breakage [23:39] asac, maybe you should make more use of the testsuites [23:39] asac: it looks like the whole mozcmp works. i will do more testing the day after tomorrow [23:40] mozcmp? [23:40] fta: look here: http://paste.ubuntu.com/234812/ [23:41] fta: it compares mozilla versions. [23:41] fta: it's the current working title. [23:41] fta: yes. you can do it right with testsuites too probably. but they just have the shell from what i understand and not so extensive api tests [23:41] xul already has that [23:41] and the shell most likely doesnt touch enough [23:42] ping fta [23:42] fta: which script does provide it. i dont know of any [23:42] i only know of C code [23:43] fta: can chrome use pdf plugin? [23:43] asac, i assume the v8 testsuite tests everything, and as for chromium, it should too, gaps should be filled in [23:43] e-jat, no idea, i never tried [23:44] even it wont try to download its .. [23:44] example : slideshare.net [23:46] e-jat, wfm, i mean, when i click on a pdf, it offers me to save it, i accept, then it's also in the dl bar at the bottom, if it click it there, my preferred pdf viewer opens up [23:47] owh ok .. [23:48] it's not *nice* but it works [23:50] reconnect [23:50] fta: i think he wants an embedded pdf viewer (guess he already told that though) [23:51] asac_, as i said, i don't know. i don't use any even with firefox [23:52] ok [23:54] ok [23:54] asac & fta : it not necessary to embedded it :) ... just curious why mine is not offer me to save it ... its ok .. ill keep try it [23:55] maybe you auto-accept all dls [23:55] try ~/Downloads/ [23:55] (i should change that back to our default btw) [23:55] fta: another thing is ... is it i need to manually add the SSL cert for the website that they self generate? [23:56] auto-safe or not auto-safe by default? [23:56] fta: does chromium already treat .deb files as unsafe downloads? e.g. only offers to save and not gdebi? [23:56] asac__, i think it's off by default [23:56] yeah. i think it should be on ;) ... and honour XDG ... lol ;) [23:57] not sure it's ready [23:57] same for ssl [23:57] whats not ready? unsafe downloads? [23:57] you know you can file bugs, right? [23:57] whats the problem with ssl right now? [23:58] i wouldnt file a bug to find out if a feature is already implemented ;) [23:59] e-jat, well, for self signed certs, i get something similar in ff [23:59] hmm ..