[00:07] Looking at bug 157129 [00:07] Launchpad bug 157129 in blobwars "Please merge blobwars 1.07-2 from Debian unstable" [Wishlist,Confirmed] https://launchpad.net/bugs/157129 [00:11] IIRC (re libowfat) at least in Debian you can use .tar.bz2. [00:11] It's a recent dpkg? change [00:11] Bout bloody time IMO. [00:11] Or maybe standards version 3.7.3. [00:12] IIRC it turned up in the lintian diff when I was reviewing it for syncing. [00:12] Ah right. [00:12] * somerville32 cries as his package doesn't compile. [00:12] Which is also conveniently available in gutsy-backports for those interested in that [00:12] c [00:12] ugh [00:13] somerville32: Step away from the keyboard to avoid risk of electrical shock. [00:14] Can someone take a look at this? [00:14] http://pastebin.com/m1ba00dad [00:15] still trying to connect [00:17] Hrm. I'd have to look at the code to work that one out. [00:19] somerville32: Could you put your .dsc/.diff.gz up somewhere and give me a link to the tarball? [00:19] somerville32: And I'll see what I can work out. [00:19] I'll put it on revu [00:19] somerville32: Take a look at libapm.h, it seems like it includes both and [00:19] mok0, okay [00:20] somerville32: if that's the case, try removing the latter [00:20] somerville32: __KERNEL_STRICT_NAMES might be of interest... [00:21] somerville32: Google will explain. I haven't the time. Sorry. [00:23] somerville32: Also, in your cc statement (7) it looks like the CFLAGS are present multiple times [00:23] somerville32: e. [00:24] somerville32: f.ex you have -DPNG_NO_MMX_CODE three times [00:24] does anybody here have any interest in helping package Ubuntu artwork? [00:24] LaserJock: Yeah whats up? [00:24] "#include [00:24] #include " [00:24] Those are the only two include statements in libapm.h [00:24] somerville32: Hmm [00:25] somerville32: linux/apm_bios.h is likely including linux/types.h [00:25] TheMuso: basically, the art team is looking for somebody to handle the packaging of the artwork [00:25] * soren idly points at __KERNEL_STRICT_NAMES again.. [00:25] TheMuso: currently they have nobody that's got packaging experience [00:26] somerville32: just for fun, see what happens if you comment out the line [00:28] LaserJock: Right. [00:28] TheMuso: interested? [00:28] LaserJock: Well I'm actually looking at a better way to do the UbuntuStudio art packaging, so if and when I come up with something, I could help out with the ubuntu artwork packag. [00:28] package. [00:28] TheMuso: I think that would be stellar [00:29] <_MMA_> LaserJock: Ive been talking to kwwii about this as well. [00:29] right [00:29] I just thought I'd ask as I told him I would see [00:30] <_MMA_> Oh yeah. You were there. :) Well I thought you were gonna put a "wider call" out. Actually I think Ken did. [00:30] * _MMA_ looks. [00:30] well, I thought I'd see here first before announcing something on ubuntu-devel ;-) [00:31] it'd be nice to get somebody who is a core-dev or wants to be one [00:31] LaserJock: Yeah well I think the biggest issue with the artwork currently, is the way its built. I think for something as simple as art, this can be simplified. [00:35] mok0, more errors [00:35] ajmitch: Your rcbugs page doesn't seem to have been updated in almost a month. [00:35] sys and linux still seem to be included somewhere [00:35] I've uploaded it to revu [00:35] somerville32: hmm, something is seriously wrong with the source [00:36] somerville32: it was a wild shot :-) [00:36] somerville32: Ask google about __KERNEL_STRICT_NAMES and you'll likely be much happier. [00:37] Ok [00:38] I'll try that [00:38] Fujitsu: oh well [00:41] somerville32: it says to include features.h before linux/types.h [00:41] ajmitch: It would be nice if you could update it, so we can fix more bugs. [00:41] somerville32: features.h should define __KERNEL_STRICT_NAMES [00:42] mok0, I saw. [00:42] :) [00:42] hehe [00:43] can I possibly pick a package to work on that *doesn't* take at least 1 hr to build? >:/ [00:44] LaserJock: No. [00:44] Fujitsu: I'll consider it, if I can rsync to qa.uw.c properly [00:45] ajmitch: Why can't you rsync to there? [00:45] ok, any git-cvs or git-svn users here? [00:45] because it wanted to use my regular ssh key on LP [00:45] Oh, right. [00:45] I think imbrandon was working on that. [00:46] Not sure what came of it. [00:46] Fujitsu: have you had a look at axiom by chance? [00:46] yes, and I was waiting on that [00:46] LaserJock: I heard you were. [00:46] Fujitsu: well, I tried [00:46] Fujitsu: i was waiting on your input , i got stuck, thats why i pointed you to my copy ;) [00:46] looking at the changelog it looks to be a sync [00:46] Should be syncable, I think. [00:46] LaserJock: I use git-svn, why? [00:46] * Fujitsu checks. [00:46] but 4 people tried to build it and it gave 3 different errors [00:46] DId you point me at your copy? [00:47] Wow. [00:47] yea, one sec i'll do it again [00:47] hehe [00:47] Fujitsu: I have absolutely *no* clue why it would fail, for me it fails during a cp [00:47] imbrandon: Erm, oops, thought it was LaserJock saying he'd pointed me at his copy. [00:47] Amaranth: do you commit using it? [00:48] LaserJock: I have (once) [00:48] hmm, well that's not all that much [00:48] some guys were using it on an upstream project [00:48] and it was doing funky things [00:48] mok0, Ok, got rid of the redefinition errors [00:48] like reviving removed directories :-) [00:48] I mostly use it for tracking things [00:49] Fujitsu: my copy of what? [00:49] How can I see if something built successfully in debian? [00:49] LaserJock: I presumed axiom, but it was in fact imbrandon saying it. [00:50] oh man, we're so messed up [00:50] maybe I need more sleep [00:50] Which must be while I feel at home here. [00:50] ScottK: lol, so true ;-) [00:50] ;-) [00:50] mok0, <<< [00:52] ajmitch: Fujitsu is looking at the code i started now, hopefully he can finish the python-foo on it and have it working soon-ish/tonight/today [00:52] Oohh. [00:52] * ScottK reads the new devscripts changelog. [00:52] debdiff: Support tarball in tarball (Closes: #439667). [00:52] hmm, so I wonder when we could get this "merge Universe/Main" thing done, Hardy+1? [00:52] LaserJock: huh ? [00:52] We should do it for Hardy [00:52] seems like it'd take quite some Launchpad work [00:52] imbrandon: See the revolutionary email to ubuntu-devel. [00:52] lol [00:53] somerville32: It's not going to happen. [00:53] LaserJock: Yeah, just a bit. [00:53] imbrandon: k, I won't be looking at anything for awhile though [00:53] * imbrandon looks, subject ? [00:53] ajmitch: kk [00:53] somerville32: I don't know dude, it'll take a lot of soyuz work and governance issue resolution [00:53] imbrandon: Keybuk proposed to merge Universe and Main [00:53] imbrandon: and to manager permissions via seeds [00:53] basically [00:54] kick ass, i sugested it a few months ago, no one listened , leaste someone will listen to him maybe [00:54] yea [00:54] if a pacakges is seeded core-dev does it, if not -dev does [00:54] imbrandon: Maybe someone did listen. [00:54] it's would be sooo messy though [00:54] can you imagine the LP team web that'll create [00:54] apparently MT is gpl'd or something [00:55] LaserJock: I don't see it as particularly messy. [00:55] LaserJock: And we can't have messy because it's all so neat and clean now? [00:55] Just ubuntu-dev will be a member of a few teams. [00:55] ScottK: no, but it'll take some work [00:55] Fujitsu: starts out as a few [00:55] but people will have to be careful who they let into their team [00:55] will the TB have to approve all uploaders? [00:55] LaserJock: they should be now [00:56] imbrandon: why? [00:56] LaserJock: The emails says that will be up to the seed-managing teams, not the TB. [00:56] right [00:56] I'm just saying as we say get 30, 40, ... seed teams [00:56] this could be a bit interesting [00:56] the thing about overlap sucks [00:56] I think Keybucks followup about how to deal with it makes sense. [00:56] I'll be interested to read other Canonical dev thoughts, such as from people like Colin and pitti. [00:57] somerville32: cool, did it finish? [00:57] mok0, Look at private query [00:58] So then you'd be able to be, say, a Kubuntu developer for KDE stuff even if you weren't even a MOTU yet. It allows for a new class of specialist that I think would be useful. [00:58] See you later. Off to go Christmas shopping. Yeah!!! [00:59] ScottK: agreed, and cya [00:59] ScottK: Fun... [00:59] * TheMuso has already done his. [01:00] Keybuk's followup cleared most things up, which is good. [01:00] I think it also kinda goes to this issue that was brought up by Till [01:00] <_MMA_> ScottK: Arundel Mills FTW! ;) [01:00] it's sometimes very weird to have people go for MOTU who've never worked in Universe and never plan to [01:01] LaserJock: Right. [01:01] while, as Keybuck said, not locking into a Debian-style maintainership [01:02] man, I can just hear cprov groaning [01:02] Yep. [01:02] and as _MMA_ just mentioned to me, what happens to all the bugmail? [01:02] It loses the component information and gets seed information instead. [01:03] Hm, both would need to exist. [01:03] say ~motu is a part of a team that wants to get all it's bugmail [01:03] But is that what you meant? [01:03] Ah. [01:03] I see. [01:03] then ~motu will also get the bugmail [01:03] That requires LP to drop its one-size-fits-all idea. [01:03] *gasp* [01:03] And provide *gasp* user-configurable settings. [01:04] Damn, you beat me to the gasp. [01:07] hmm [01:08] well, I guess I better get home so I can get some work done ;-) [01:11] _MMA_: If you can find a parking place. [01:14] <_MMA_> :D [01:19] * somerville32 goes to merhe electricsheep [01:19] somerville32: How did you go with that code compilation problem? [01:20] TheMuso, Still working on it. Got rid of issue #1 [01:20] now working on issue #2 [01:20] Ah right. [01:50] slangasek: Where does the universe sponsors queue stand WRT DIF and processing requested merges/syncs? Are they still ok to go through even if DIF comes into effect? [01:51] Isn't DIF more of an advisory rather than actual freeze? [01:51] Yeah, but still. [01:51] it's when we don't get auto-sync [01:51] and if somebody bothered to do the work I don't think we should turn them down [01:52] IMO [01:52] Another one down. [01:53] TheMuso: well, the discussion on ubuntu-motu@ covers everything IMHO [01:53] ok [01:54] Which is you can continue to do merges if you can justify them? === nexu|laptop is now known as nexu [01:56] Is there some reason that seamonkey isn't in the repos? [01:57] LaserJock: Did you take care of bug 175725? [01:57] Launchpad bug 175725 in regina-normal "Please merge regina-normal (universe) from Debian unstable (main)" [Undecided,New] https://launchpad.net/bugs/175725 [01:57] I still see it on the sponsors queue list. [01:57] TheMuso: still waiting for it to build [01:57] Ummm... [01:57] I have a problem :/ [01:58] TheMuso: I know how to pick'em [01:58] LaserJock: Oh ok. In the future, could you please unsubscribe uus from the bugg, so that an indication is given as to whether someone is working on it? [01:59] ahh [01:59] I wondered if I should assign it to myself or something [01:59] * LaserJock goes to read the sponsoring wiki page again [02:02] I'm doing a native package, arch=all, using cdbs. It FTBFS at the end with "dpkg-genchanges: failure: cannot read files list file"? indeed, no debian/files are created. Any hint ? [02:02] somerville32: we know you have problems, but what's up? ;-) [02:03] LaserJock, xfce4-battery-plugin doesn't compile with 2.6.24 linux [02:04] evening [02:04] surprise surprise [02:05] hi zul [02:07] What does the S in FTBS mean? [02:07] Source [02:07] somerville32: FTBS or FTBFS? [02:07] Fails To Build From Source [02:07] Oh, ok [02:08] LaserJock: Both: unsubscribe & assign (unless it's a quick comment/rework, in which case unsubscribe is usually sufficient) [02:09] man, things have changed so much ... [02:09] LaserJock: That's not changed since it was presented at a MOTU Meeting in May :) [02:10] well, I know [02:10] but umm... it's been a while [02:10] he [02:10] h [02:10] * txwikinger gets the feeling MOTUs never sleep ;) [02:11] sleep? [02:11] what is that? [02:11] :D [02:11] More like, there is always a MOTU on hand, 24 hours a day. [02:11] txwikinger: As a group, we don't. Individually, occasionally. [02:11] persia: hehe [02:12] Sleep is for the week [02:12] weekor weak? [02:12] * persia likes "week" :) [02:12] * StevenK too :-) [02:13] * TheMuso likes sleep. [02:13] It ensures I don't make packaging mistakes. [02:15] I've had packages in the queue for like, over 24 hours now :P [02:15] boo! :P [02:15] persia: Good follow up mail. [02:16] Heya gang [02:17] Hi bddebian [02:17] Heya bddebian :) [02:17] Heya DaSkreech, somerville32 [02:18] has anyone tried to package seamonkey? [02:19] DaSkreech: I think there's something up with that [02:19] you might as the Mozilla Team [02:19] LaserJock: Something technical or legal? [02:20] most likely technical [02:21] cause it occurs to me that it's been a long time since Ubuntu came out and there is still no Seamonkeys :( [02:21] ajmitch: fyi, you can add custom ssh keys to /srv/ssh-keys/.key now too whenever your ready, the new system is in place [02:21] DaSkreech: seamonkey is in hardy: request a backport if you can compile it on gutsy (but I think you may have trouble) [02:22] persia: hooray! More reasons to jump to hardy! :) [02:22] hello [02:22] DaSkreech: There's also lots of reasons not to jump. [02:22] * imbrandon jumped a few days ago, lots of "small" things, hehe [02:23] persia: don't tell me about it :) still waiting for the not to jump reasons to come low enough for me to jump over them! [02:25] * ajmitch will probably upgrade to hardy about 2 months after release [02:27] somerville32: if you declare typedef unsigned short apm_event_t; [02:27] in libapm.h it compiles in hardy. [02:27] but I doubt that it works [02:27] mok0, It does. [02:28] I'll upgrade at RC most likely [02:28] mok0, That change is not intentionally I don't think [02:28] * TheMuso is running hardy on non-critical boxes. [02:28] mok0, Upstream (ie. Linux) has been notified [02:28] somerville32: cool [02:28] somerville32: you mean the panel app works? [02:29] mok0, we'll find out in a second :P [02:29] Ooo... [02:29] * somerville32 gets an idea. [02:30] mok0, I'm going to apply the the patch to the kernel so that I have it on my upload list :P [02:30] :-) [02:31] another bug bites the dust [02:32] TheMuso: did you do a fresh hardy install? [02:34] mok0: On the non-critical boxes, yes. [02:34] I wanna upgrade a box soon [02:37] * mok0 is going to bed [02:37] g'night all [02:37] Gnight mok0 [02:38] wow, that regina-normal build to 90 minutes [02:39] ouch [02:40] yeah, spent the first 90 verifying that our current package doesn't work [02:41] then another 90 building with the fix [02:41] heh [02:42] persia, you told me earlier that 'Uploaders' has no meaning in ubuntu and should be dropped, yet, i have to add it in a new (native) package i'm doing, in order to stop lintian from thinking i'm attempting an invalid NMU [02:42] haha [02:43] Ubulette: what? [02:43] Ubulette: on what release do you get that? [02:43] LaserJock, a new package i'm preparing [02:44] Ubulette: yeah, but I took care of NMU in lintian quite some time ago [02:44] Ubulette: Did you set the maintainer to Ubuntu MOTU? [02:44] Ubulette: are you using Ubuntu versioning? [02:44] 0.1 (native) [02:44] Ubulette: set an Ubuntu package version, and an Ubuntu maintainer, and it will go away. Anyway, why native? [02:44] Oh, LaserJock "fixed" it..?? ;-P [02:44] mozillateam as maintainer [02:45] Ubulette: yeah, 0.1 is not a valid version [02:45] native because i'm upstream [02:45] LaserJock: Yes it is, just not an Ubuntu version. [02:45] Ubulette: So? package it as upstream, and make a distro package. [02:45] persia: I don't think it is is it? [02:45] oh wait, it's hative [02:45] LaserJock: Yep. [02:45] *native [02:45] for non-native it wouldn't be valid I don't think [02:45] LaserJock: Right. [02:46] Frick I still can't build half my games packages because of this stupid sdl bug [02:46] with uploaders matching changelogs, lintian and linda are both happy, even with version 0.1 [02:47] right [02:47] because it thinks it's a Debian package [02:47] Ubulette: Yes, but that's because nobody has bothered to change lintian and linda to ignore Uploaders. It's a very minor bug in those packages. [02:47] should it ignore Uploaders? [02:47] (err, ignore "Uploaders" for Ubuntu-versioning, that is) [02:48] LaserJock: For Ubuntu, yes. For Debian, certainly not. [02:48] persia: but it's not ubuntu versioned? [02:48] LaserJock: That would be a bug for a new package in Ubuntu, no? [02:48] i wanted to do a native ubuntu only package [02:48] but he's not got Ubuntu versioning [02:48] Ubulette: you *don't* want to do a native package [02:48] why ? [02:48] because native packages are for packages that are not useful outside the distro [02:49] Ubulette: I am an opponent of native packages. Anyway, even if you did one, it should be ubuntu-versioned, or you'll need to preemptively blacklist. [02:49] LaserJock, that's the case [02:49] Ubulette: what is the package? [02:49] Ubulette: Describe the package. [02:49] no use at all outside the mozillateam. [02:50] it's a kind of cdbs-like package to create all the proper tarballs for mozilla packages [02:50] doing embedded or not, nobinonly clean-up stuff, and more [02:51] so it will be a build-dep [02:51] that's why i wanted a native package [02:52] hrm. I see the point, and yes, that type of package should be native. Any chance it would be useful to mozilla maintainers in Debian? [02:52] s/should/could/ [02:52] i doubt it. the branding thing [02:52] yeah, but that's just a patch to your package [02:53] but it's highly customizable so who knows [02:53] Ubulette: It's the "who knows" that makes me extra unsure. [02:54] * somerville32 hugs bryce [02:54] * persia grumbles at the Mozilla Foundation for forcing code duplication in multiple tarballs due to annoying attitudes about branding [02:55] bryce, It has been 15 minutes and I don't see the upload [02:56] * TheMuso should tackle the festival merge. [02:56] persia, providing debian uses git, dpatch, standard dh_ stuff, does unbranding and remove all non free files, while ubuntu uses bzr, quilt, cdbs, keeps the original branding and only remove binaries, it's a loooong way to resync [02:56] somerville32: hmm, well I've not gotten a reject yet, so maybe things are just slow? [02:57] dang, it's really been too long since I've done real packaging work [02:57] * somerville32 nods. [02:57] Ubulette: Sure, I'm just not sure about whether your add-on package would be useful there (as I have no idea what it does) [02:57] I keep forgetting that Launchpad closes bugs these days :( [02:58] Ubulette: to be honest, it seems more like a script you'd keep somewhere rather than something you'd package [02:58] persia, the idea is to use if for firefox, xulrunner, thunderbird, sunbird, seamonkey, and all their mozilla friends [02:59] we have too many copies of the clean up script, too many rules diverging, it's a pain in the *s to maintain and to review for everyone [03:01] Ubulette: Understood, but that's a description of why the package exists, not what it does, and so doesn't provide any indication of whether non-Ubuntu native versioning is acceptable. [03:01] let me past you the core of the stuff [03:01] As another workflow alternative, you could keep common scripts in a branch, and pull the branch at packaging time, with an include statement in debian/rules. [03:01] * persia suggests a pastebin [03:03] http://paste.ubuntu.com/2682/ [03:04] Ubulette: Is that essentially the entirety of it: a makefile snippet? [03:04] the core ony, a dozen of files, makefiles + patches for the multi-stages cvs fetch [03:04] anly [03:05] Ubulette: That seems awfully light for a package. Have you asked about including it as a default CDBS extension? [03:06] persia: I agree [03:06] it has no relation to cdbs [03:06] it's just cdbs-like [03:06] Ubulette: Except that it serves much the same type of purpose: a set of things that get included in debian/rules to make packaging easier. [03:07] Further, if it's only 12 files, it's not that big, and so it may be less archive space / bandwidth to just push it there. [03:08] Of course, if the answer is no, then it makes sense to go for a separate package (ubuntu-mozilla-dev or something) that might include this and more as required (and yes, non-Ubuntu native versioning would be acceptable for such a package) [03:09] well, forget it, i'm tired. the package is ready, asac welcomed the idea (and we've been using my previous version for 6 months inside the team). I'll hand it to him and he'll do whatever he want, drop, merge, keep.. [03:10] it's 4am here, sorry to be rude :( [03:10] Ubulette: Sleep on it, but come back: the code is good, but why the separate package rather than just CDBS integration. Have a good night. [03:11] Ubulette: in any case you can ignore NMU warnings in Ubuntu [03:11] LaserJock: Well, that's dangerous blanket advice: in many cases it indicates that the package isn't versioned correctly (although it's not the case for this package) [03:12] well yeah [03:13] c [03:13] ugh damn orca [03:13] Er? How does orca drop a /? [03:15] TheMuso, can you map that key that orca uses, 'c' to a different one? Perhaps a period would be less conspicuous, and easier to brush off as coming in to say hi [03:15] :) [03:15] now i'm all nervous, not good for sleep [03:15] persia: By locking the modifier key used for keyboard commands. :) [03:15] TheMuso: Ah. I see. Annoying that. [03:15] yeah [03:17] Ubulette: Nervous? Why? Relax: you've created a good thing, that should be in the archive. You've packaged it, and if the NMU error is your only automated check report, you've packaged it well. We're recommending it be further integrated with CDBS, but that's something you can look at tomorrow. [03:26] * TheMuso kicks GNOME mag. [03:36] if all goes well, this will be my most productive package day in a couple of releases [03:36] :-) [03:41] LaserJock: More sponsoring? [03:41] well, I did 3 sponsors [03:41] and I'm gonna have at least 2 uploads of my own [03:42] Right. [03:42] I'm taking a break from the sponsors queue myself as well, as there are merges I want done. [03:42] but then I gotta get some other work done :( [03:43] Yay! Finally a new festival into the archive. [03:48] persia: Are you likely to get to freqtweak any time soon? [04:00] TheMuso: No real point. I might get to it, but Debian's tarball is ugly, so I'd be doing a fakemerge at best. [04:01] TheMuso: Just for context: both represent VCS pulls from dead upstream. [04:05] persia: Ok. [04:06] I won't worry then. [04:15] ok, any mozilla gurus around? [04:15] I've got a debian package that build-deps on libxul-dev [04:15] is that ok for Ubuntu? [04:28] LaserJock: Provided by xulrunner_1.8.1.11-1ubuntu1, but you might want to test it. [04:30] persia: right, I'm wondering if that build dep is ok for a Dep on firefox [04:31] I've used it in my PPA and nobody has complained [04:31] LaserJock: Ah. That'd be different then. Most packages seem to depend on firefox-dev, but I don't know. === asac_ is now known as asac [04:31] There's the man who would :) [04:31] I just wondered if there was some policy/guidance for how we're dealing with it === calc_ is now known as calc [05:18] * ScottK butchers some more SQL. [05:20] Watch out. SQL bites back. [05:20] delete from ... [05:21] bddebian: Well the funny thing is the admin was in a hurry and I think I have the permissions to do that. [05:22] Heh. [05:23] Heh, just drop any table with the name sys in it ;-P [05:25] shesh, keescook has screwed up my fonts ;-) [05:25] LaserJock: Tut tut. What do we say about running untrusted code from the 'net? [05:26] Fujitsu: If you do it, you get what you deserve? [05:28] StevenK: Something like that. [05:28] I just wonder why my gnome-terminal looks completely different from his [05:28] Because he is Kees, and therefore rocks. Simple. [05:28] I guess so [05:29] * Fujitsu intercepted your HTTP request and modified the content on the way. [05:29] Hah [05:29]