[00:00] ups [00:00] yes :-) [00:00] A couple of other things - the version should have been 1ubuntu1 [00:00] im not too familiar with this [00:00] and you should put the .xpm inside debian/ [00:00] yes... [00:00] ... [00:00] o [00:00] ok [00:00] how i can do it [00:00] create a new patch? [00:02] anakron: I wonder if this patch is really necessary anyway? The bug is really minor in any event and introducing the change in Ubuntu means we have to maintain the patch until Debian or upstream take it [00:02] yes i know it [00:02] and this one? [00:03] https://bugs.launchpad.net/ubuntu/+source/firestarter/+bug/301603 [00:03] Ubuntu bug 301603 in firestarter "Impossible to launch firestarter" [Undecided,Confirmed] [00:03] i edit a patch that changes su-to-root -X -c to gksu -X -C, but gksu does'nt have these options [00:04] shouldn't su-to-root work fine? [00:04] yes It will work [00:04] but the package have this patch [00:04] that change su-to-root to gksu [00:05] i don't know why [00:05] but add gksu with -X -c [00:05] what about kde users who don't have gksu? [00:05] Laney [00:06] You need to add a dependency on it for it to be available [00:06] i know that su-to-root its better for upstream than gksu [00:06] but, the maintainer set it [00:06] i could change it [00:06] or set it as dependence [00:06] aha, I misunderstood you [00:07] I thought you removed su-to-root ;) [00:07] :) [00:07] the problem is that the patch set gksu with -X -c [00:07] so, i remove -X -c [00:07] yeah, I get it [00:08] I don't understand the original patch though - drop it and it seems to launch fine [00:08] but i can change it and add gksu in dependences [00:08] i can't too [00:10] anakron: I think whoever did the last merge was wrong [00:10] ill see debian package [00:11] From reading bug 5269 it looks like the package originally used gksudo [00:11] Launchpad bug 5269 in firestarter "Menu location" [Medium,Fix released] https://launchpad.net/bugs/5269 [00:11] but Debian changed to use su-to-root with -7, so we should have gone with this change [00:12] anakron: If you do a debdiff which drops the patch (and confirm it works) then I will be happy to sponsor that [00:12] ok [00:13] Laney, monodevelop debuggers, tuppence a bag! [00:14] Laney, got X going again then? [00:14] directhex: metacity compositor was the culprit here [00:15] although scrolling is still pretty slow especially in FF [00:15] silly metacity [00:15] evilwm! [00:15] I want my freedom-hating fglrx back :( [00:25] Laney, how i can take this patch away? it use dpatch [00:25] im trying [00:27] anakron: edit debian/patches/00list [00:27] but [00:27] that is all? [00:27] i thought in it [00:28] Well, you can delete the actual patch file too, once you are sure you want to get rid of it. [00:28] yep [00:28] delete it from 00list and then delete the actual file [00:29] actual file, patch file? [00:29] the last and first patch says that desktop file should use gksu [00:30] uh, confusing [00:30] yes [00:30] its weird [00:30] because first [00:30] desktop file says that Exec=firestarter, then apply 1st patch to > gksu firestarter [00:31] then i dont know which one but it changes to su-to-root [00:31] and the last patch changes it to gksu [00:31] xD [00:31] I'll take away the last patch [00:31] :q [00:31] you aren't vim! [00:32] ¿? [00:32] anakron: grep -r su-to-root . in the package directory will find it [00:32] ok [00:32] It seems like there were three patches touching the same line in Ubuntu [00:32] * Laney spanks everyone concerned [00:33] hi mruiz [00:33] anakron, hi [00:34] anakron: please write in the changelog that no Ubuntu changes remain [00:35] why? [00:35] anakron, sync? [00:35] yes it seems [00:36] so that the next uploader knows this [00:36] I'll look at debian package [00:36] anakron, if Ubuntu changes are dropped they must be included by upstream or Debian [00:36] or more correctly, the person who merges it next [00:37] ok [00:37] grep -r su-to-root still looking for, but nothing appears [00:38] laney@chicken> grep -r su-to-root . ~/dev/ubuntu/packaging/firestarter/firestarter-1.0.3 [00:38] ./debian/patches/11_desktop_file.dpatch:+Exec=su-to-root -X -c /usr/sbin/firestarter [00:39] mm here nothing happens [00:39] xD [00:40] it seems that the first patch must be deleted and this one must be edited [00:41] and the last patch must be deleted too [00:41] what happens if you just delete 25? [00:42] but the first patch is useless [00:42] you should report a debian bug about it [00:42] it's more easy to just delete 25 [00:43] ScottK: I found a patch that might fix the bmpx build for ppc/powerpc: http://paste.ubuntu.com/122614/ - http://www.mail-archive.com/pld-cvs-commit@lists.pld-linux.org/msg149860.html [00:43] we don't want to make additional changes over Debian if we can help it [00:43] i think that for this bug I only delete 25 [00:43] and then I'll do changes and report it to debian directly [00:43] yes, that's a good way to do it [00:43] ScottK: I'm trying the patch in PPA, to see if it works for normal architectures as well: http://launchpad.net/~medigeek/+archive/ppa/+builds?build_text=&build_state=pending [01:04] savvas: Great. Let me know how it goes. [01:05] ScottK: with gems drama and all, we're back in sync with debian, right? [01:05] If the sync from experimental gets done, yes. [01:05] ScottK: I guess I meant policy wise / culture wise. Cool. [01:06] The one gems diversion did not stand, so I don't think it ended up being a significant diversion. [01:06] It was mostly a question of enough people knowing. [01:12] gah [01:12] can we start uploading notification fixes? [01:12] I am about to rip Banshee's face off [01:14] Laney: Got FFe? [01:14] ScottK: Is it a feature? [01:15] Certainly. Changing the system to integrate with a different notification system is definitely a feature. [01:15] Not complying with something that isn't even a spec is not a bug. [01:19] Well the feature was introduced in the new notify-osd package. Is not what we have now a series of bugs in upstreams (not checking for capabilities)? [01:19] I guess you could argue that making notifications-with-actions into alerts is a misfeature [01:20] Laney: Where is there any standard that requires checking for features? [01:21] One can't take a draft, unapproved standard that is still under negotation, implement something new from it and then declare all non-supporters of this new feature buggy. [01:21] Well one can, but it's really not cricket IMO. [01:22] Right; it has been done, and now we must implement the fixes. [01:22] I don't see the point in requiring paperwork to do so, since there is really no way we are going to leave the behaviour as it is now [01:23] https://bugs.launchpad.net/ubuntu/+source/firestarter/+bug/301603 [01:23] Ubuntu bug 301603 in firestarter "Impossible to launch firestarter" [Undecided,Confirmed] [01:24] Laney, here it is [01:24] thanks [01:29] anakron: It doesn't build [01:29] applying patch 01_use_gksu to /build/laney-firestarter_1.0.3-7ubuntu2-amd64-hUSBIB/firestarter-1.0.3-7ubuntu2/debian/patches ... failed. [01:29] ¿? [01:30] i built it and then i create a debfile [01:30] and runs ok [01:30] mmm ill see it [01:34] http://pastebin.ubuntu.com/122624/ [01:34] i make debuild -S [01:34] and then dpkg-buildpackage [01:37] anakron: You should use pbuilder or sbuild to build packages [01:37] ups [01:37] often problems can be missed by using debuild/dpkg-buildpackage alone [01:37] Would someone be willing to explain gdb breakpoints to me? I'm trying to examine the values of a variable at various points in the execution of a program (finch), and when I try setting one it tells me the source file isn't found. [01:37] Laney, good point [01:38] mm [01:38] ok [01:38] tonyyarusso: You need to instruct gdb where to find the source [01:39] StevenK: aaaah. How do I go about that? [01:40] tonyyarusso: Step 1) Get the source [01:40] Got that. [01:40] Is there a way to check from the command line what the version of a package is in a given release? [01:40] xoox: rmadison [01:40] However, that will only cover supported releases [01:41] It requires Launchpad and a shovel to find out for unsupported releases [01:41] Is it just cd? (http://www.chemie.fu-berlin.de/chemnet/use/info/gdb/gdb_5.html#SEC20) [01:41] Thanks StevenK [01:41] tonyyarusso: Sounds plausible [01:42] anakron: I fixed a problem with a missing build-dep and now it's OK [01:42] ok [01:43] :) i do it when i try to do dpkg-buildpackage [01:44] StevenK: That gets me to a different error at least, so progress. Now it says "No symbol table is loaded. Use the "file" command." (just typing 'file' says "No executable file now, No symbol file now") I passed -g and -gstabs in CFLAGS when I compiled, but I'm entirely certain what that does / how to make use of it. [01:45] tonyyarusso: Why did you feed into gdb to get that error? [01:45] uh, "Why" or "What"? I had just said 'break save_pounce_cb' (that being a function I want to stop at) [01:49] ScottK: good news, lpia built! https://launchpad.net/~medigeek/+archive/ppa/+build/882034 - still waiting for i386 and amd64 though [01:50] Sounds like I have to tell it how to load symbols? [01:52] oh, found that. 'file /usr/bin/finch' is what it wanted. [01:58] StevenK: I was able to get a bit further now. http://paste.ubuntu.com/122638/ a) I'm not sure why I'm getting that message about No such file or directory, b) How do I make sense of 0x8242158? (dialog->on_status is what I'm concerned with) [02:05] anakron: I just uploaded it, thanks for your contribution [02:05] :) [02:05] thanks to you [02:07] now to see if it also ftbfs on sid [02:07] then I should go to sleep [02:09] when is UDS? [02:11] anakron: https://wiki.ubuntu.com/UDS [02:11] ok [02:12] I just applied for sponsorship \o [02:12] shot in the dark, but might as well try [02:12] :) [02:12] hey laney [02:12] one question, what can i do for this desktop file http://pastebin.ubuntu.com/122641/ [02:12] yo [02:13] what's wrong with it? [02:13] I'm guessing 0x8242158 is a memory register address or something, but I need to be able to see the actual data (a pointer that actually just contains an integer, I want to know the integer) stored there. [02:13] do you know about desktop-file-validate? [02:13] because desktop-file-validate says that is not valid character [02:13] yes [02:13] right [02:13] savvas: If you think you have it figured out, we should ask TheMuso if he'll do a powerpc test build. [02:13] but i can't validate it [02:13] you can replace it with ö [02:13] the utf-8 character [02:13] what's a good practice for changelog entries when Debian hasn't released yet? [02:13] :O ok [02:14] how i do it [02:14] i.e. when you're uploading something based on a Debian VCS version [02:14] anakron: But this is the kind of change we don't need to make [02:14] ok [02:14] LaserJock: I like -1~ubuntu1, but personal preference [02:14] -x~ubuntu1, generally [02:14] Laney: and do you keep the Debian entry even if it hasn't been uploaded yet? [02:15] LaserJock: I keep it, but add a comment at the top under my name: "uploading Debian SVN revision" [02:15] then the reason [02:16] ~ubuntu1? I thought it was -0ubuntu1. :S [02:16] Some people like that too (for new upstream versions). I think that ~ubuntu1 makes the heritage of an upload clearer [02:17] the problem is that Debian has a -1 entry, and then I'm going to add on top of that a -0ubuntu1, a lower version [02:17] that seems weird to me [02:17] I don't think LP would even let you upload [02:17] I would think it would [02:18] Oh, *debian* has it [02:18] yeah, that would work [02:18] you could either rename -1 to -0ubuntu1 or -1~ubuntu1. Doesn't matter really [02:18] Debian has an un-uploaded -1 entry [02:18] well, I guess it doesn't matter a bunch, but I am going to make it UNRELEASED [02:18] ScottK: I'll try and test it locally in about 10-15 minutes [02:19] just so people know Debian hasn't uploaded it yet [02:19] OK. Great. [02:19] bed [02:19] night all [02:19] Laney: -0ubuntu1 is the standard way to do it. [02:24] if say. i wanted ubuntu to come with a non horribly broken x264, ffmpeg and mplayer, could i volenteer to motu and make updated packages for them? [02:26] theholyduck, what's wrong with those packages? [02:26] superm1, ffmpeg is built from a horribly old svn snapshot. x264 is built from a horribly old git snapshot. and mplayer is built from a release from 2007 instead of a svn snapshot [02:27] all 3 packages are unsported and unstable for anything OTHER than latest svn [02:27] but ubuntu still packages bad versions of them [02:27] theholyduck, well for mplayer, that's the latest stable release [02:27] 1.0~rc2 [02:27] superm1, well yes :P [02:28] but svn is WAY stabler [02:28] ScottK: it works on amd64 :) [02:28] faster, and more usable [02:28] than that release [02:28] mplayer doesnt DO releases [02:28] great. [02:28] wich is why its their "Latest stable" [02:28] theholyduck, well you can certainly do the work to file a group of FFe's for these and have the motu-release time evaluate them if you are interested [02:28] there would be no guarantees such things would be accepted, but that would be the process [02:29] theholyduck, you might want to talk to siretart before you do so though [02:29] superm1, well i currently resorted to writing a script for dling all deps and building and making a .deb for local system use [02:30] so that all the ubuntu users who clog up #Mplayer #ffmpeg and #x264 could get up to date pacakges [02:30] it doesnt do actual .deb packaging just checkinstall [02:30] but i cant see why ffmpeg and x264 gets developement versions but mplayer doesnt. [02:30] and why the developement versions dont get updated more frequently [02:31] x264 and ffmpeg both enjoy very rapid developement [02:31] theholyduck, i would imagine they generally have implications on other packages [02:31] such as the gstreamer stack (for ffmpeg) [02:31] doesnt gstreamer use its own package with its own ffmpeg [02:31] gstreamer-ffmpeg or whatever? [02:32] no reason you couldnt have a ffmpeg-svn x264-git and mplayer-svn package that fills the depends for their respective "stable" packages [02:32] but for people who want a usable system [02:33] theholyduck, no, it uses libavcodec from the system ffmpeg [02:33] arghh. why do everyone persist in dynamicly linking everything :P [02:33] saves space, encourages stable ABI & API [02:34] superm1, in a perfect world yes :P [02:34] but all its doing in this case is stagnating your ffmpeg [02:34] since people depends on how that stagnated ffmpeg work [02:34] theholyduck, so you will have more likelyhood of getting FFes on the parts that don't have potential to hurt other packages (such as mplayer) [02:34] ping Laney; hey, how i can see which dependecies are needed to add [02:34] the world isnt perfect thus dynamic linking doesnt work. [02:35] theholyduck, but feel free to file the set of FFes and let the release teams evaluate them and talk to siretart about ffmpeg [02:35] ups sorry [02:35] i might just stick to my supply a script that compiles x264, ffmpeg and mplayer statically. then integrates it into the packagemanager plan [02:36] more ideal since ffmpeg should be updated every day [02:36] theholyduck: Use rvm's PPA. http://ppa.launchpad.net/rvm/ubuntu [02:37] still no ffmpeg :) [02:37] someone here [02:37] but atleast mplayer is alteast sorta usably built [02:38] if when i use pbuilder and it says that "E: pbuilder-satisfydepends failed" [02:38] theholyduck, you might consider setting up a PPA for daily ffmpeg builds then [02:38] what can i do? [02:38] x264 is still horribly outdated on that one xoox [02:38] theholyduck, it would probably be useful to these same people who use your script [02:38] TheMuso: Can you test a powerpc package of bmpx 0.40.14-1ubuntu1 on jaunty using this debdiff: http://paste.ubuntu.com/122614/ [02:38] 20080917 is so anicent its not really worth mentioning in x264 terms [02:39] theholyduck: I would try searching people's PPA's before trying to build it yourself. [02:39] i must look at debian/control and see which things are needed to add as build-depends? [02:39] savvas: I will try it later, still don't have a build environment up yet sorry, will get to it as soon as I do. [02:39] theholyduck: I agree with you about the poor support for tracking rapidly developing packages. [02:39] TheMuso: thanks :) [02:40] anakron, look at the output of the pbuilder run to find what was missing [02:40] anakron, and add it in debian/control [02:40] i try to see it doing a dpkg-buildpackage [02:40] and it says that are some deps that are needed to install [02:41] there you go [02:41] but all of thm are used by pbuilder [02:44] superm1, do i need to setup a launchpad team or project or something to get my own ppa? [02:46] google to the rescue === Andre_Gondim is now known as Andre_Gondim-afk === nhandler_ is now known as nhandler [04:20] gah, I don't understand gdb at all. I've managed to get output for 'print variable', but it looks nothing like what I expect to see in that variable. :( [04:20] You mean, it's a pointer? [04:21] Well, at least originally I was working with a pointer, but then I (thought I) created another dummy variable to hold the value of just the integer (gotten from GPOINTER_TO_INT()), and that doesn't help either, so I'm not really sure what's going on. [04:22] (Keep in mind though that I just started trying to have the concept of pointers explained to me yesterday.) [04:22] Oh. [04:22] I'm guessing the variable you're interested in is a struct? [04:23] I don't know? [04:24] tonyyarusso: Well, what is the type of the variable you're interested in? [04:24] http://paste.ubuntu.com/122663/ - I'm trying to find out the value of dialog->on_status, so I can figure out why things aren't behaving like I expect around lines 200-207 and/or 468-473. [04:25] StevenK: I believe it's an integer that's been converted to a pointer (to match what the function is expecting), and then I try converting it back. [04:26] tonyyarusso: And when you break at line 200, what is on_status_as_int ? [04:27] $1 = -1210495972 [04:28] actually, wait - that's a few lines earlier I think. [04:29] still get the same thing. === fabrice_sp__ is now known as fabrice_sp === fabrice_sp__ is now known as fabrice_sp [04:41] StevenK: any idea why it's not 1, 2, or 3? [04:41] Perhaps it's not an int ... [04:43] Time to hit the documentation for GPOINTER_TO_INT? [04:43] Sounds like a good first step [04:46] sadly, this stuff seems to be quite poorly documented :( [04:47] This is the clearest thing I've found so far, but it seems to say I had it right - http://mail.gnome.org/archives/gtk-app-devel-list/2001-July/msg00246.html [04:47] Also, http://library.gnome.org/devel/glib/unstable/glib-Type-Conversion-Macros.html, but that page is gibberish to me. [04:50] oh, wait a second [04:51] I'm not entirely clear about how things are stored in the combo box I think. Looking for examples of that too... [05:02] I found a file that does similar things - http://paste.ubuntu.com/122670/. Looking through that - hopefully it will help. [05:42] good morning [05:42] hi fabrice_sp [05:43] Hi dholbach :-) [06:04] dholbach, I've just update the debdiff of Bug #334035. I forgot to close the bug in the changelog :-/ [06:04] Launchpad bug 334035 in transcalc "Transcalc FTBFS with error "call to '__open_missing_mode' declared with attribute error: open with O_CREAT in second argument needs 3 arguments"" [Undecided,Confirmed] https://launchpad.net/bugs/334035 [06:04] and I saw too late your comment :-) Do you want me to update the patch? [06:05] fabrice_sp: no, I uploaded it already - just close it manually [06:06] ok. Thanks! [06:14] A small question: what would be the Ubuntu's version for 0.99.5+cvs20070914-2.1~lenny2 ? [06:14] dch propose me 0.99.5+cvs20070914-2.1~lenny2ubuntu1 [06:15] sounds good [06:16] just add ubuntu1 [06:18] ok. Just seemed to me a too long version number! :-) [06:18] not our problem, we just add 7 characters to it ;-) [06:18] theholyduck: I wouldn't call the ffmpeg snapshot from a few weeks ago 'horribly outdated'... [06:19] well im talking hardy currently :P [06:19] i only deal with what i have to support [06:19] and the one in hardy is from sometime in 2008 [06:19] theholyduck: it should even provide VDPAU, but I couldn't find someone to actually test it [06:19] siretart, well ffmpeg doesnt REALLY provide vdpau :P [06:19] atleast not fully :P [06:19] fabrice_sp: are you going to forward the transcalc fix upstream? [06:19] you need to get the nvidas own build for that [06:19] well, ffmpeg in hardy is indeed nearly a year old, that's right. [06:20] we could update it in -backports, but this would break nearly all media players in ubuntu [06:20] siretart, yes. and there is no reaon you cant keep updating it:P [06:20] dholbach, I didn't find a up-to-date upstream page (2007, it seems), so I was opening a bug report in Debian [06:20] fabrice_sp: sounds good [06:20] siretart, well a ffmpeg older than 1 weeks is way outdated [06:20] :-) [06:20] by ffmpeg standards [06:20] fabrice_sp: do you use "submittodebian" for that? [06:20] theholyduck: breaking all media players *is* a very good reason [06:20] a x264 outdated by 3 days is outdated by x264 standards [06:21] says who? [06:21] theholyduck: An ffmpeg outdated by one *commit* is ancient, for crying out load [06:21] *loud [06:21] the guys CREATING the damn software [06:21] dholbach, I generally use reportbug, with manual email :-/ I'll check submittodebian :-) [06:21] StevenK, well yes. but when it hits a week its horrible :P [06:21] theholyduck: Hardy is released. I can't see us updating it. [06:22] theholyduck: well, then try to provide a PPA with updated versions of x264 and ffmpeg. you don't even need to be a developer [06:22] siretart, well if players just built in the version of ffmpeg they used. [06:22] fabrice_sp: the nice thing about it is that it makes use of the stuff in https://wiki.ubuntu.com/Debian/Usertagging for you [06:22] theholyduck: I documented the update procedure pretty cleary in debian/README.upstream-upgrade [06:22] instead of insisting on the idiocy of dynamic linking [06:22] this wouldnt be a problem now would it? [06:22] ffmpeg isnt really THAT large a binary anyway [06:22] IDIOCY!? [06:23] then maintain all media players with static copies of ffmpeg in your PPA [06:23] We're pretty much in favor of dynamic linking around here. [06:23] ScottK, im not though :P [06:23] it encourages not updating packages [06:23] because other packages would break [06:23] ffmpeg is just about 8 MB of code (on i386). I would call that rather large... [06:23] ? [06:23] Not if they're sanely designed. [06:23] ScottK, well ffmpeg isnt sanely designed or developed [06:23] and never will be. [06:23] Agreed. [06:23] most apps wont be [06:24] theholyduck: No, it involves updating one package in the case of a security flaw, rather than 15. [06:24] There I disagree. [06:24] dynamic linking is perfect in a perfect world :P [06:24] dholbach, I see. I'll check all that before submitting to debian. Tahnks for the pointers! [06:24] but the world is far from perfect. thus it doesnt work [06:24] fabrice_sp: anytime [06:24] * ScottK declines to really get into it and goes to bed instead. [06:24] dynamic linking is okay for ubuntu & debian. I'd say we do a fairly good job so far, of course there is always room for improvements [06:24] ScottK: sleep tight [06:24] ScottK: good night! [06:25] Good night all. [06:25] and then you got softwares like x264. [06:25] with THIIIINY binaries. [06:25] good night ScottK [06:25] that makes even LESS sense to dynamically link [06:25] theholyduck: are you willing to actually help to improve the situation or are you only trying to make people work for your pleasing? [06:26] siretart, well i'll start working on trying to package usable mediaplayers for ubuntu. but it sorta saddens me since i wouldnt ordinarily use it :P [06:26] ubuntu already has rather useable media players: ffplay, gxine, vlc, gstreamer0.10-ffmpeg using packages like totem, phonon + applications... [06:27] the only reason im willing to do it is because there are more and more people using the horrible abomination of mediaplayers known as vlc instead of mplayer because the mplayer in ubuntu is outdated [06:27] siretart, gstreamer and vlc are abominations. [06:27] ffplay isnt MENT to play files [06:27] it cant even demux propperly [06:27] what media players are missing? [06:28] siretart, well the problem with linux is that all the playback solutions are horribly flawed in some way [06:28] mplayer from svn is currently the only close to usable one [06:28] but no distros package it :D [06:29] well, then try to update the mplayer package [06:29] preferably based on this git branch: http://git.debian.org/?p=pkg-multimedia/mplayer.git;a=summary [06:29] I could then consider uploading it to debian [06:30]