[00:03] Laney: implemented (will be on production in a few minutes) [00:03] Hi [00:03] hi Drk_Guy [00:03] Can anyone help out making a desktop entry for Gambas2? [00:03] I'm not so experienced ;P [00:03] Drk_Guy: Doesn't it already have one? [00:04] hey jono [00:04] RainCT, I mean, i'm packaing it [00:04] ;) [00:04] It won't install on 64 Bit [00:05] Drk_Guy: but it does build there? [00:05] RainCT, Yup [00:05] hey [00:05] Drk_Guy: why don't you fix the existing package then? [00:10] RainCT, ... Any ideas? [00:10] RainCT, sorry, my modem suspended traffic for some odd reason [00:10] You got anything? [00:13] Ummm... [00:14] Need to make a menu entry for gambas [00:14] Anyone willing to help? [00:14] I'm packaging it [00:14] 01:05:31 Drk_Guy: why don't you fix the existing package then? [00:15] RainCT, um... [00:15] RainCT, It is completely broken, dependancies aren't compiled [00:15] they should be fixed then :) [00:16] RainCT, I'm a noob packager, i'm only gaining experience [00:16] Should i download that package's sources to get it's menu entry? [00:16] Any sane MOTU will reject a repackaging for a reason like that. [00:17] * ajmitch doesn't see a bug in LP about the 'completely broken' part [00:18] ajmitch, At least it won't install here [00:18] Drk_Guy: and the .desktop file should be something like that: http://paste.ubuntu.com/32773/plain/, but if you really want it, better get that from the existing package as it is probably better [00:18] Drk_Guy: because it's only being build for i386 [00:19] (don't ask me why :P) [00:19] RainCT, Thus i'm putting it on my PPA [00:19] :) [00:19] -.- [00:20] Architecture: i386 [00:20] that may just have some small thing to do with it [00:20] ajmitch, It built fine on AMD64 (My arch) [00:20] revu.tauware.de redirect now to revu.ubuntuwire.com - is there some other domain under which REVU is accessible? [00:21] Drk_Guy: just because it built fine doesn't mean it's useful [00:21] RainCT, I've never been able to access REVU [00:21] gambas2 (2.1.0-1) unstable; urgency=low [00:21] * New upstream release. [00:21] * Upload 64 bits compatible version to unstable (Closes: #464403) [00:21] ajmitch, I USED it [00:21] ajmitch, It runs real fine [00:22] it wasn't until 2.1.0 that it was changed to be properly 64-bit compatible [00:22] such issues can just show up randomly [00:22] ajmitch, Anyways, ubuntu likes to stick with stable stuff [00:22] Drk_Guy: there's hardy-backports [00:22] * Drk_Guy thanks ... it's not as much as Debian does *shrugs* [00:23] RainCT, Already enabled ;) [00:23] Drk_Guy: how about requesting a backport if there's not one already? [00:23] Drk_Guy: https://help.ubuntu.com/community/UbuntuBackports#How%20to%20request%20new%20packages [00:23] ajmitch, why backporting if i can package it natively? [00:24] RainCT, already read that [00:24] why reinvent the wheel, many many times over? [00:24] so? request a backport and everyone will be able to use it [00:25] *sighs* [00:26] * RainCT too [00:28] tacone: Yes. [00:31] uhm.. isn't the tag "needs-devrelease-testing" used anymore or is it just that the testers are great and have already tested all bugs which had it? :P [00:32] ajmitch, backports are a lil bit unstable [00:32] Something tells me i should just build it for harddy [00:32] *hardy [00:32] oh dear :) [00:32] (ah nevermind, I'm just stupid) [00:33] anyway, good night [00:33] night [00:36] ajmitch: Do you know what Drk_Guy was talking about? [00:37] ScottK: since the hardy package of gambas2 isn't compiled for amd64, he wants to package it himself & stick it in a ppa [00:37] I see. [00:37] I don't believe that to be the best course of action [00:38] Personally I think PPAs are a particularly bad idea right now until the DNS cache poisoning thing gets settled out a bit. [00:39] It seems like an odd list of archs to bulid it on in Intrepid. [00:39] * ajmitch hasn't looked at the intrepid source [00:40] the sid version just has Arch: any [00:40] sparc, powerpc, lpia, i386, and hppa. [00:40] ScottK: too late, I posted a 100k mail on the ml :) [00:40] OK. [01:07] tacone: You've got mail (on the list). [01:07] thanks ScottK [01:08] hope nothing too bad :). waiting for the mail to show up [01:09] No. Not to bad. [01:15] ScottK: I guess you got my point. I am not trying to say people don't get the point. But obviously the real measure of (the importance of) such aknowledgement is with the stuff got done until know and the in-progress efforts. I won't carry on the conversation in this channel of course. I'll make some point in the ml tomorrow. I think in someway we feel similar about it. thanks for now [01:16] tacone: You're welcome. I agree it's a problem, just not one that I find particularly enticing for my free time. [01:16] sure :) [01:30] hey guys has anyone of you set up an active/passive config using heartbeat in HH ? [01:31] * ion_ blinks === emgent_ is now known as emgent === foka_ is now known as foka [02:47] nxvl: o/ [02:52] can anyone help me to troubleshoot this error message: [02:52] package control info rmdir of `usr' didn't say not a dir: Directory not empty [02:55] gastoni: You'd need to post much more info. Such as: what is it that you're trying to do, and the full output of whatever it is that ends with that error. [02:56] thats all the error [02:56] I have posted it in ubuntu forums [02:56] http://ubuntuforums.org/showthread.php?t=876537 [02:57] im trying to package a java application [02:57] gastoni: No, I didn't want just the error, I wanted the full output of everything up to and including the error. [02:58] yeah i figured it will be easier to read in a forum post, than copy pasting here [02:58] gastoni: Ah. Your tree is all wrong. [02:58] oh I though it resembled a ubuntu installation [02:59] Control information goes in DEBIAN/, actual files which need to end up in the filesystem tree go in the directory containing the DEBIAN directory. [02:59] Also, you almost certainly want to build from source rather than try to package up an existing binary install in that way. [02:59] s/almost // [03:00] everything is inside DEBIAN [03:01] gastoni: Yes. Which is wrong. [03:02] aaah [03:02] i get it [03:03] gastoni: You should thank dpkg that it doesn't allow you to do utterly stupid things. If it did, installing that package would have removed /usr [03:03] Actually, that's a lie :) [03:03] yeah i know, i tried to install it :-S [03:03] and /usr stills there [03:04] You should use dh-make to begin the packaging. [03:04] dpkg should have complained before even making the .deb file tho [03:05] gastoni: No, why? [03:05] no its fine actually, second thought [03:06] Maybe you _wanted_ a deb with no files in it; there are perfectly reasonable use-cases for that. [03:10] Help guys [03:10] I'm trying to package flv2mpeg4 [03:10] but it doesn't have makefiles, it just needs a single line for compilation [03:11] But the thing is, they don't have a version number, they just have SVN [03:11] :( [03:11] Can anyone help? [03:13] Drk_Guy: Why do we want flv2mpeg4? [03:13] RAOF, To convert Youtube videos, lol [03:14] Anyway, at least i would like a DEB for it [03:14] Or, rather, what does it do that the hundreds of other transcoding apps we do have don't. [03:14] ;) [03:14] RAOF, I didn't see any flv transcoding app in synaptic [03:14] Maybe someone will need it someday [03:14] lawl [03:14] That's because _all_ the transcoding apps in synaptic will handle flv. [03:14] RAOF, huh? [03:14] Well, many of them. [03:15] Drk_Guy: FLV is just another video codec; there's nothing special in playing it. [03:15] RAOF, still, it's just flv2mpeg4 input.flv output.avi/mp4 [03:15] RAOF, Most cellphones won't play it [03:15] XD [03:15] mencoder --in input.flv --out output.mp4 [03:16] That won't actually work, because that's not correct mencoder syntax, but the principle is the same :) [03:16] Rats [03:16] I could do a similar one with gst-launch if you like :) [03:16] Anyway, flv2mpeg4 uses ffmpeg, so it's heaps faster [03:16] lol [03:16] No, it isn't. mencoder uses ffmpeg. [03:17] Rats [03:17] I think i lose, you win, i got pwnt [03:17] :( [03:17] In fact, ffmpeg is not very separate from mplayer/mencoder; that's the primary target of ffmpeg :) [03:17] XD [03:17] RAOF, Still, i want to make a DEB out of it [03:17] The windows version is 0.6 [03:18] Ok. Then it's the same process as for any other deb. [03:18] There’s also the ffmpeg binary that does transcoding. [03:18] An alternative to mencoder [03:18] lol [03:18] I totally got pummeled then [03:19] Eh. This is not a competition of any kind. [03:19] XD [03:20] Ok, how should i edit the debian/rules? [03:20] The thing is, you just got to cd to the "src" dir, and then run a single gcc line [03:20] Then move the output to /usr/bin [03:21] Same way as always; you need to provide the various targets. [03:21] RAOF, I'm a lil bit noob at editing debian/rules, hhehehe [03:21] The build: to do all the configuration/building part. That's easy for you; there's no configuration, and a single line build. [03:22] clean: to get back to a clean tarball extract. [03:23] RAOF, I'll pastebin my output [03:23] And binary:, binary-arch:, binary-indep: to actually build the debian package. [03:24] RAOF, http://paste.ubuntu.com/32833/ [03:24] Look at it, and modify accordinglt please [03:25] That looks broadly correct, although you won't need sudo and you're probably moving the file to the wrong place. [03:26] RAOF, /usr/bin is right [03:27] So it's on the $PATH [03:27] No, it probably isn't. That'll install it on the system that the package is built on, not on the system the package is installed on :) [03:27] lol, so, where should i move it? [03:27] debian/$packagename/usr/bin [03:28] Ok [03:28] echo 'flv2mpeg4 usr/bin' >>debian/flv2mpeg4.install [03:28] And make sure dh_install is called in debian/rules [03:28] Of course, ion_'s method is better. [03:29] I'm confused [03:29] Can anyone explain better? [03:29] Btw, the line ‘cd src’ in debian/rules basically does nothing. It doesn’t affect the following lines. Do this instead: cd src && gcc ... [03:30] Ok ion_ [03:30] Remove the installation line from the install rule, make sure dh_install is called and make debian/flv2mpeg4.install as instructed. [03:30] Done ion_ [03:31] ion_, Umm, it's confusing, can you repaste the modified rules file please? [03:32] Remove the sudo mv line and remove the # from the dh_install line. [03:33] ion_, Um..., uncommented dh_install [03:33] Now, if i remove the mv line, how is the program going to be finally installed? [03:34] dh_install installs stuff based on debian/packagename.install [03:34] ooh *puts on evil hat* [03:34] ion_, there is no file named like that? [03:34] Oh, since the build happens under src, the correct contents: src/flv2mpeg4 usr/bin [03:34] Create it. [03:34] http://www.iphonelinux.org/index.php/IBooter has essentially the iPhuc command set without the iTunes requirement [03:34] superm1: poke poke [03:35] ion_, So, create what? [03:35] ion_, the src dir is already done [03:35] debian/flv2mpeg4.install [03:35] Please pardon my noobness [03:35] superm1: I think we can write a fuse wrapper around iBooter for a native Linux USB iPhone/iPodTouch syncing agent :) [03:35] * ajmitch pries the crack pipe out of jdong's hands [03:36] ajmitch: Hey, smoke your own. [03:36] Nice, linux on iPhone [03:36] ajmitch: *MY* crack! [03:37] I don't know if this method would be any less ugly than the jailbreak+sshfs method everyone uses now [03:37] ion_, Done [03:37] ion_, Ok, should i edit control as normal? [03:37] drk_guy: Well, yes [03:37] ion_, Thanks [03:38] <[GuS]> Hi guys.. could anyone give me a hand packaging qt4.4 (from the repos)? I am just modifying the debian rules file since i need to compile Qt4 (and package it) wityh libphonon from Qt itselft, so I've changed -no-phohon for -phonon in the configure part. All the source is being compiled sucessfuly, but fails at the end saying this: [03:38] <[GuS]> http://pastebin.com/m4c954c9d [03:39] ion_, Umm, going to put it under misxc [03:39] <[GuS]> Maybe, and i think for sure i am missing something to add in debian rules, since i have base of how to package in debian/ubuntu... but i dont want to becomre expert.. just i need this... [03:42] ion_, One last question, is there any tool to edit the man page easily? [03:43] drkguy: I’ve used asciidoc to create man pages. And you might want to use the graphics section. [03:44] ion_, graphics? [03:44] That’s where ffmpeg and mencoder are, for instance. [03:44] Ok [03:49] ion_, packaging failed, still, i'll continue tomorrow [03:49] Thanks [03:50] It's late [03:58] jdong, iBooter???? [03:58] ooh have been intrigued [03:59] jdong, it's not clear, does it need to be induced on the phone? [03:59] superm1: yeah, if you've used iPhuc before, it's the same thing but without the need for iTunes libs [04:00] or is it activated solely from the PC? [04:00] superm1: err, no induction needed [04:00] superm1: it talks to the emergency bootloader thingie [04:00] DFU mode, I think it's called [04:00] the home-and-power button thingie [04:00] ah yes. [04:00] but it seems like it passes through plenty to enable FS access [04:00] well if it can do it from the computer, that's entirely a solution [04:00] yeah I just downloaded the code and am looking around [04:01] jdong, ever written a fuse filesystem? [04:01] superm1: nope :D [04:01] well i can't imagine it's "that" bad. [04:01] model it off of some easy fuse ones [04:01] like mythtvfs [04:01] yeah I can't believe it'd be that hard [04:01] <[GuS]> could someone help me? thanks... [04:01] so far this summer I've been programming space-vector control on DSPs [04:01] so I think a bit of fuse would be a welcome break [04:02] indeed [04:02] I hate it when TI tech support tells me to read a register 3 times and take the majority vote [04:02] *grumbles about buggy silicon* [04:02] lol [04:02] seriously? [04:02] no joke [04:02] the peripherals will block CPU memory access on races, no warning [04:03] and worse, in 32-bit registers you can read back the first 16 bits properly and get 0's for the remaining bits [04:03] Score! [04:03] and using operator |= could lead to trashing all 32 bits :) [04:03] Sounds like more fun than a barrel of chainsaw-wielding monkeys. [04:03] i'm assuming its too late to switch to a new DSP..... [04:03] lol after this experience, I will NEVER EVER complain about Java. Again. [04:04] superm1: haha it's soldered onto a 600V/1000A power stage [04:04] so I'm guessing that's a NO for futzing around the insides [04:04] well not safely at least [04:04] 1000A? Toasty warm! [04:04] if you're really feeling lucky though.... [04:05] just leave a note that could be interpreted as a suicide note in case [04:05] RAOF: oh yeah. and you don't want to know about when its overcurrent shutoff takes effect. [04:05] RAOF: supplying that much power probably will roast a small country [04:05] jdong: When it's power output is comparable to the sun? :) [04:06] personally I'd rather just install Linux on a 600MHz ARM and call it a day [04:06] sure beats the current solution [04:06] I can't say much about it, but it costs $40000, runs VxWorks, and requires the upstream power controller to implement a power cycle over UDP/fiber command. [04:07] I might not be around here much this summer, but TRUST ME I am NOT having fun! [04:08] can anyone help me with this: http://ubuntuforums.org/showthread.php?p=5500185#post5500185 [04:08] I corrected the file structure of my .deb package, still is not working [04:09] pe050449 < ion_> You should use dh-make to begin the packaging. [04:13] * jdong wonders if there's a way to separate the Eclipse editor from the IDE.... the editor is pretty slick but the rest of the IDE is excessive [04:14] Huh. I bet i’d find the editor the thing i hate most about Eclipse if i tried it. :-) [04:15] ion_: really? [04:15] it's one of the more friendly coder's editors I've used, particularly for mass-reformatting type tasks [04:16] plus code completion is pretty slick for digging around foreign code bases [04:16] jdong: If the editor were in fact more effective than vim, i’m sure i’d have heard a lot of talk about it by now. [04:16] ion_: it has vim keybindings :D [04:16] and emacs ones for that matter [04:16] the only department it totally fails to complete is in its ability to boot quickly [04:17] * Hobbsee likes the eclipse editor [04:17] its paste reformatting detection is great too [04:18] I wish I had it when I was doing massive Python editing [04:18] jdong, personally geany for python is the way to go in my opinion [04:18] hmm never tried it before [04:18] but that might just be because i'm very used to it now [04:19] I'll give it a shot right now [04:19] it's quick to open and most importantly has options for showing what kind of whitespace you have [04:19] so it will detect if you are using spaces or tabs [04:19] and then automatically adjust for when you hit your tab key [04:22] superm1: it is pretty good [04:22] jdong, yeah mess with the options some more too. you'll find more stuff that you like [04:22] the defaults are pretty meh [04:23] it lacks a fully automatic code re-styler though [04:23] but that's a very minor complaint [04:23] FWIW it's an extremely lightweight editor [04:23] I'll probably replace gedit with this :) [04:24] yeah i already have [04:24] gedit just takes ages to load compared to this [04:25] thanks :) [04:25] so in exchange, i expect a fully functional ipod touch sync over usb asap ;) [04:25] rofl === davroman1ak is now known as davromaniak === gaurdro_ is now known as gaurdro [06:05] good morning [06:08] Morning dholbach. [06:09] I might even be prevailed upon to prepend a "good" to the morning :) [06:09] hi RAOF [06:09] hi cassidy [06:12] hi dholbach! [06:13] hi Hobbsee [06:34] dholbach: namaskar! how was India? [06:35] hi tuxmaniac - it was absolutely fantastic [06:35] once I'm through most of the n thousand emails I got in the meantime, I'll probably write a long blog post with lots of pictures :) [06:35] dholbach: sure sure. and did you meet the delhi folks? [06:36] I met Vivek one night in Delhi - for some reason text messages some Delhi guys sent me didn't make it to my mobile phone [06:36] and I didn't check my email during that time, so .... :-/ [06:37] I'm sure we'll come back to India :-) === didrocks_ is now known as didrocks [07:06] I'm using a package from universe (ejabberd) and was wondering if it [07:07] kinema, if it...what? [07:08] I'm using a package from universe (ejabberd) and was wondering if it's unusually that both the conf dir (/etc/ejabberd/) and log dir (/var/log/ejabberd/) are masked such that non-privileged users can't ls the files in them. [07:09] It's not really a problem it's just that it screws with the autocomplete in zsh when I'm running as a non-root user. [07:16] kinema: Not usually, although sometimes when the specific application has security concerns. [07:19] persia: I understand not being able to read the rc files (plaintext passwords) but ejabberd keeps all its files in subdirectories and non-root users are able to list the contents of those directories. It just seems strange. [07:20] persia: I missed yesterday's meeting. Too busy with work :-( [07:20] persia: I mean Java Team meeting [07:21] slytherin: Lots of people missed it :( [07:22] kinema: generally Ubuntu tries to restrict things that require root to the absolute minimum. Anything not a security issues is generally permitted. [07:35] It would be great if a motu can sponsor a NEW package sync reported in bug 253324 [07:35] Launchpad bug 253324 in ubuntu "[sync request] Avagadro Sync from debian unstable 0.8.1-4 version." [Undecided,New] https://launchpad.net/bugs/253324 [07:36] the package build fine under interpid pbuilder. === thekorn_ is now known as thekorn [08:11] Ping whoever maintains the ubuntuwire rcbugs list, it needs updating [08:12] Laney: How out of date is it? [08:13] persia: For darcs for example, the version it thinks is current is was superseded on May 22 [08:14] ajmitch: ^^ [08:20] if a file created by one app won't open in another (when it should), which app should bugs be filed against? [08:21] directhex: The one which is at fault. [08:21] THere's no way to tell from that question. [08:59] never mind, tracked it down. [09:00] Hi all! [09:06] G'morning. [09:10] txwikinger: Hey. Do you have a bug for the pending upload of ichthux-meta? === rraphink is now known as raphink [10:52] I'm building a deb package and I want to ask something... When I edit the debian/control file I should separate the dependencies with a comma right? something like Depends: ${shlibs:Depends}, ${misc:Depends}, libasound (>1.0.14), libc6 (>=2.6.1-1) .............. [10:52] cherva: Yes. [10:53] persia: thanks [10:53] Also, Unless something is very odd, you ought get both libasound and libc6 from shlibs:Depends [11:03] Would someone like to ack two sync requests? [11:06] ... or more. [11:07] what about conflicts I should be below Depends: .......... ? [11:10] Iulian> I can't right now, but you can email them to me to sikon@ubuntu.com so I don't forget [11:11] Sikon: Cool, e-mailing now. [11:14] Hm. Does anyone know if it's possible to mix Automake and make conditionals in the same file. They annoyingly share nearly the same syntax, and one appears to confuses the other. [11:19] Sikon: did you take a look later at electric? [11:21] dholbach: welcome back! [11:22] re. bug 252037 I resubscribed ums, I only wish it would be made clearer in our docs who can deal with backports and how [11:22] Launchpad bug 252037 in sauerbraten-data "sauerbraten cannot upgrade" [Undecided,Confirmed] https://launchpad.net/bugs/252037 [11:24] slytherin> No, sorry [11:25] Sikon: Mail sent, thanks. [11:25] thanks, I'll look [11:26] Wait, why on Earth does Ubuntu have yum? [11:27] Sikon: Why oughtn't it? [11:27] Despite it being odd, some people like to use RPMs anyway. [11:27] In a deb-based system? [11:28] Sikon: 1. It is free software. 2. Someone packaged it for Debian/Ubuntu. :-) [11:28] Well, with alien, who can be sure :) [11:28] heh [11:28] * Sikon shakes fist at LaCie only providing its LightScribe writer in rpm [11:29] See, that's why there's yum :p [11:30] If i was creating a watch file for mp3wrap - would http://downloads.sourceforge.net/mp3wrap/mp3wrap-(.*)-src\.tar\.gz|bz2 be a sufficient url? do we need to add debian uupdate to each entry? [11:30] stefanlsd: I actually prefer that debian uupdate is not present. [11:31] Also, you might want to read the uscan manpage again: there is a shortcut form for the sourceforge mirror network (thank you Debian) [11:32] Lastly, .gz|bz2 is awkward: we ought use whichever upstream has, and prefer .gz if upstream has that. [11:34] stefanlsd: alternatively, read https://wiki.ubuntu.com/PackagingGuide/Howtos/DebianWatch and there is a simple url for sourceforge downloads [11:35] thanks guys. do i understand debian uupdate correctly. That will attempt to download and build automatically if an update exists? [11:36] stefanlsd: thats not correct. Remember that almost every executable has a man page, in this case man uupdate gives all the info you need about it [11:36] stefanlsd: Roughly. It makes a number of assumptions which aren't always valid. === RainCT is now known as RainCT_ [11:48] norsetto: Note that the package names have tooltips now, so you don't actually have to navigate to another page to see the component and version on a bug page - just wait a while. [11:48] * norsetto checks the tooltips [11:49] wgrant: thats correct, its still annoying that if you have to use that data you have to write it down somehow [11:50] wgrant: and the tooltip fades after few secs ... [11:50] norsetto, about the amoebax revu, you commented that I should contact an archive admin to understand if the Free Art License will be accepted, where do I find an archive admin ? [11:50] joaopinto: in ubuntu-devel [11:50] norsetto: Of course, it would be impossible for Launchpad to have a solution that worked for us. [11:51] * persia looks at the license to see if there are any obvious gotchas [11:51] the license seems fine except for the fact that is listed by GNU as GPL incompatible [11:52] norsetto: your REVU comment style is one to learn from :) [11:52] joaopinto: exactly, its incompatible with the gnu gpl, this doesn't mean that its not acceptable for us, thats why I'd rather you check with an archive-admin [11:52] persia: yes, I'm mad :-) [11:52] joaopinto: The issue is that the code is GPL and the art is GPL-incompatible. That's a little awkward. [11:53] joaopinto: also, its not very clear what all the data is licensed with, there is an xml with the FAL, so, that means music, effects, etc. [11:54] While I'm not an expert, nor an archive-admin, I'd think that putting parts of each together would be considered a single work, which makes it undistributable as being part of each license. [11:54] persia, well, that is according to GNU, not to the author, otherwise it would not be distributed as it is [11:55] joaopinto: There's lots of upstreams who aren't deeply familiar with law. Doesn't make what they do legal. I can download copyrighted stuff with no license within a couple days of release, typically. [11:55] norsetto, the README is clear regarding that, "The sound effects, music, and graphics are licensed under the Free Art [11:55] License (LAL)." [11:55] I know of at least one upstream that has a license constructed in such a way that they can distribute the code, but nobody else can. [11:55] joaopinto: ok, thats what I assumed as well [11:57] Sorry - watch question again! - what process uses the watch file? Is it a manual thing, is it only something like qa.ubuntuwire.com? [11:57] stefanlsd: uscan [11:57] stefanlsd: can be both [11:57] (which is used by qa.ubuntuwire.com) [11:58] norsetto: ok. so wouldnt qa.ubuntuwire.com want the debian uupdate option? or would it add it to the watch file itself? [11:58] stefanlsd: I do it manually (actually with a script) when I sponsor new updates or new packages, its also done automatically by things like qa.ubuntuwire.com [11:59] stefanlsd: I think uupdate does more harm than good, but thats my personal opinion [11:59] hum, this brings me an interesting question, is the author violating the GPL with it's own software distribution ? [11:59] I'm having troubles making a deb file for recordmydesktop that uses oss instead of alsa here is what i get when I run debuild http://pastebin.com/d7707d0a1 what is wrong ? [11:59] stefanlsd: UEHS isn't going to use the uupdate option, as it's just checking. People doing updates may find it convenient, but as norsetto says, manually is often safer. [12:00] norsetto: i understand that i can download a package with a watch file, run uscan and be informed that there is an update. if there was a debian uupdate command - it would try download and build the package (i agree, i would prefer to do this step myself manually). [12:00] Didn't we leave OSS to die a horrible painful death years and years ago? [12:00] joaopinto: Very likely if FSF says FAL is GPL incompatible, although it depends on where the authors distribute it from. [12:00] persia: ok. thanks. that makes sense. i think i would prefer to run uupdate manually also. alright, im not gonna put it in :) [12:00] wgrant: No. We didn't port everything. Then, ALSA didn't fix everything. [12:01] joaopinto: he is also infringing the gpl since he is having binaries without source in its tarball [12:01] norsetto: Can the copyright owner be infringing? [12:01] it seems to be a common myth that you need to ship the source code with your GPL'ed application. [12:01] you need to provide it on request, AFAIK. [12:01] wgrant: of course he can, the terms of the license are clear [12:02] laga: yes, and how do we do that if we don't have it? [12:04] i was under the impression that the source must be made available [12:04] so this would be putting it up for upload... [12:04] *download [12:04] norsetto: that's a different issue. you need to make sure you have the source - eg by asking the author for it. he's compliant with the GPL if he provides it on requests [12:04] hi folks [12:05] laga: that was my point [12:05] norsetto: why are you claiming the's infringing the GPL then? [12:05] s/the/he/ [12:05] laga: because we cannot redistribute it, we would be infringing the gpl [12:06] well, that's a different issue ;) [12:06] laga: thats the issue for me ... [12:06] would it be solved if we made a request, and we put the source code in the deb src? [12:07] its kinda silly anyways that there is no source provided... [12:08] stefanlsd: if that source its not used to build the binaires, how can we maintain the package? How can we apply security fixes? [12:08] norsetto: yeah, agreed. [12:08] norsetto, are you referring to those .dlls which were probably included for the windows build ? [12:09] joaopinto: yes, and the pdf doc too [12:10] well, I am just going to rm those [12:10] Can't you get upstream to release a legal tarball? [12:11] joaopinto: Yes, we need to address these with upstream. I'd rather have the source for the pdf since I don't think we have other installable docs? [12:12] sure, but that would add time to my work [12:12] It would also add a large amount of legality, and a similarly large amount to the likelihood of it not being rejected. [12:13] joaopinto: welcome to the wonderful world of distro maintenance [12:13] joaopinto: Copyright is 9/10s of the work of packaging, I find. [12:13] Just for the record, upstream can likely distribute anything they want, they may just not be providing it in a format that can be redistributed entire. [12:14] ok, so the choice is, add an unknown time task, or do not add it, and run the risk of have a determined amount of time being wasted with a rejection, nice :P [12:14] persia: thats strecthing it a bit, copyrighted material which is not redistributale is not redistributable to usptream too [12:15] norsetto: Unless they hold the sole copyright on the GPL'd stuff; in that case they can do pretty much what they like. [12:15] norsetto: Yes, but if I am a copyright holder, I can distribute my work, regardless of what license I claim it is under. I can e.g. license it so you may redistribute it only if you don't. [12:15] joaopinto: being rejected would waste more time than just yours [12:15] persia, raof: sure [12:16] persia: the "must pet a cat" license is the typical example for such clauses ;) [12:16] persia: RAOF: what I mean is not my stuff, those dll for instance IIRC, were not from upstream they were some library necessary to build his binaries [12:17] sistpoty|work: I'm thinking of the "must pet a dodo" type license, but yes. [12:17] persia: that's not impossible, at least if you own a time machine :P [12:17] sistpoty|work: Err. Right. Pet a hippogriff then :p [12:17] sistpoty|work, persia: one of my cats name is dodo :-) [12:17] hehe [12:30] If im adding a watch file to mp3wrap - do i build it as an ubuntu release file now? It looks like its just a sync from deb and it was never built for ubuntu. [12:51] stefanlsd: Yes. You'd use Nubuntu1 as the revision identifier [12:54] persia: thanks thanks [12:56] how valuable is just adding a watch file? wouldnt it be a bit of a waste of time for the maintainers to check and approve? also it changes the maintainer to us (MOTU) so we potentially become responsible when debian could be...? [12:58] stefanlsd: erm, actually we are responsible for all ubuntu packages, wether these are modified or unmodified from debian [12:59] stefanlsd: but I agree that it's not really worth to change a synced package just for a watch file [13:01] * wgrant agrees with sistpoty|work. [13:01] siretart: The removal of source package information from bug pages is very much deliberate, and AFAICT they can't be convinced to change. [13:05] stefanlsd: As long as you're working from UEHS, all the packages should be either only in Ubuntu (so maintained by us anyway) or orphaned in Debian. [13:06] Getting the watch file in helps us build the list of ubuntu-local stuff that needs updating for each release. There's a big effort for the first couple months of each release to try to get everything in sync, and for the ubuntu-local stuff we rely entirely on the watch files. [13:06] sistpoty|work: Not even for an orphaned package? Might we not want the update? [13:09] persia: I don't know about a bug, but yes, I am trying to get to that [13:09] * txwikinger is in the middle of an intercontinental move [13:10] txwikinger: Ah. Good luck with the move. I just heard that ichthux-desktop was being dropped for not being installable in #ubuntu-devel, and knew you had an interest. [13:10] Ah ok thanks [13:11] sistpoty|work, persia: thanks. makes sense. The fact that it helps build a list for a new release makes a lot of sense if we had watch files. It probably makes more sense to do a watch file when updating a package or closing a bug, but yeah. This is my first package, so im content with just practicing by making a watch file and doing the changelog for now. [13:11] ichthux is part of kubuntu.. so I am not sure why there is a discussion on ubuntu-devel, and not on kubuntu-devel ;) [13:11] or not part of, but build on [13:13] txwikinger: Ah. I didn't know ichthux was based on kubuntu: it's just on my list of minor flavours. The discussion was in #ubuntu-devel because that's where the archive-admins chat. [13:13] Ah ok :).. I was just teasing ;) [13:15] Is there a a log report about the building errors? [13:16] any members of the Ubuntu Development Team? [13:16] Well. likely doesn't matter anyway... ichthux must be ported to kde4 [13:21] cherva: Many. [13:21] wgrant: perhaps we can work around that using leonov or some other tool using the lp API. [13:22] persia: I agree (about the source package on the bug page). It seriously makes me wonder if the excercise is just window dressing. [13:22] txwikinger: It's the NBS list: it was depending on things that are no longer there. As you say, the important bit is to port to the current stack. [13:22] persia: I need two to review my deb package [13:22] siretart: If we're going to go that route (provide our own U/I) then we ought to be focusing on comments on the API. [13:23] persia.. yes I would expect such dependency issues :D [13:23] cherva: The traditional method is to paste a quick advertisement as to why your package is cool with a link to the REVU page. [13:23] does anyone have a screenshot or something that shows how the source package information on bug pages looked like previously? [13:23] Can anyone confirm i followed the correct steps or left anything out for me please.. https://bugs.launchpad.net/ubuntu/+source/mp3wrap/+bug/253945 [13:23] Launchpad bug 253945 in mp3wrap "debian/watch file for mp3wrap" [Undecided,New] [13:23] ScottK: I see no problems in that approach, do you? [13:23] persia: ok I'll make the REVU page and come back [13:24] or should this package go into REVU as its a new name and revision? [13:24] siretart: I'm more hopeful that in a project we can control ourselves we would do better. [13:24] stefanlsd: That looks good, except why did you bump the standards-version? Is the package now compliant with the newer version? [13:24] ScottK: I take this as agreement [13:25] siretart: I have mixed feelings. [13:26] siretart: In one respect, one part of Canonical has hired another part to provide a service for Ubuntu. We ought to insist on getting reasonable service. [13:26] siretart: OTOH, I'm pretty sure that's not going to happen. [13:27] persia: ammm how to upload to REVU ? [13:27] Personally I'm pretty sick of the "Of, we don't know anything about distro development, sorry we messed that up, but no we aren't going to change anything because it's just the way it should be." reactions. [13:27] persia: mmm. good question. i guess i should first understand what changes between versions and check that it complies. I was under the impression that we always bump the standards (although i agree, it doenst make sense to bump it if you dont know what standards you are complying with...) [13:27] cherva: Log in at revu.ubuntuwire.com, and then dput. There are some instructions on the wiki. [13:27] Can I rate about 10 of the 15 options 1, as they should have been there three years ago? [13:27] !revu [13:27] REVU is a web-based tool to give people who have worked on Ubuntu packages a chance to "put their packages out there" for other people to look at and comment on in a structured manner. See https://wiki.ubuntu.com/MOTU/Packages/REVU [13:28] stefanlsd: We only bump the standards when we know the package complies with the new standards. Also, for Debian-derived packages, we try to minimise changes (unless we do an upstream ahead of Debian). [13:29] ScottK: I see. [13:29] persia: ah. k. where can i read about the standards and what they should adhere too? [13:30] siretart: One of the biggest problems I forsee in the roll our own U/I approach is latency. Do you know if the proposed API makes any quality of service requirements for latency? If not, it should. [13:30] -> copying [./games] <- any ideas why is pbuilder --build trying to copy a games directory on the copy source stage ? [13:30] ScottK: It can't be any worse than the web UI. [13:31] persia: and what should I do regarding this package now? make another debdiff leaving the debian standards in place? I did do the change of Maintainer and the XSBC-Original-Maintainer which i assume would be in the new ubuntu standards... [13:31] joaopinto: Make sure you're building the .dsc file. [13:31] stefanlsd: The debian-policy package is the official source. Debian maintains web mirror that Google finds fairly well. [13:31] ops, that's it :P [13:31] stefanlsd: You don't need to mention that you changed the maintainer in the changelog [13:31] wgrant: It can. Whatever process builds the web U/I is designed based on some internal knowledge of the system and (as slow as it is) may well be optimized. [13:32] Laney: kk. point noted. [13:32] ScottK: Hrmmm. [13:32] stefanlsd: Well, check first. Maybe you don't need to alter the debdiff. If you do need to alter it, yes upload a new one. Remember to subscribe the sponsors queue when it is ready for upload. [13:33] wgrant: Additionally, if we are throwing our hands up on the official U/I, then we need to focus on what the API needs. Latency in the API calls will have a major performance effect on things like Leonov and so we should push to get them quantified with useful requirements. === thekorn_ is now known as thekorn [13:35] ScottK: Is that going to still be true as Leonov migrates to the new API? [13:35] ScottK: TBH, I don't see that latency is a huge concern here. You can hide latency by means of caching most of the time [13:35] ScottK: If only LP didn't have a 'one size fits all' philosophy... [13:35] persia: k. will check what the new standards involve and see if i comply. i have apt-get installed debian-policy, so i will read it. Last thing - if i do need to change it back, or change whatever i need to change to make it comply to 3.8.0 - do i go through the exact same process and land up with mp3wrap_0.5-2ubuntu2 - or should it still be mp3wrap_0.5-2ubuntu1 as that build was never released. (or does it even matter with a debdiff?) [13:36] stefanlsd: ubuntu1 as it was never released. [13:37] persia: thanks. understood. will just use dch and not dch -i. [13:37] anyone willing to review pocketsphinx and sphinxbase in REVU (the first build-depends on libsphinxbase, which is provided by the second)? Uploading these packages to the archive will enable us to upgrade gnome-voice-control to 0.3. [13:38] The links: http://revu.ubuntuwire.org/details.py?package=sphinxbase , http://revu.ubuntuwire.org/details.py?package=pocketsphinx [13:38] Thanks :) [13:38] norsetto I will be addressing the PDF source question to upstream, but regarding the SDL dlls. they are only used on windows builds, they are not relevant for the linux build process, is there any real issue on shipping them on the source tarball ? [13:38] joaopinto: Still a violation. [13:39] Hm, actually I don't know which license SDL uses. [13:39] wgrant, as per GPL the source only needs to be available upon request, what violation are you referring to ? [13:39] it uses GPL IIRC [13:40] We do not distribute binaries on their own. [13:40] I hope an archive admin would reject on that kind of thing. [13:40] those binaries are part of the upstream tarball, not the package, and they do not constitute a license violation [13:41] thanks for the patience guys, gonna head home. bbl [13:42] so it would be wiser to touch the original tarball and rm the .dlls ? I am not going to ask upstream to change their build process/source to be Ubuntu/Debian specific [13:42] joaopinto: the gpl applies to both sources and binaries. espc. for the source. [13:42] joaopinto: They really shouldn't be distributing binaries in their damn source tarball! [13:42] siretart: Perhaps, but caching only helps the second time you want data. [13:42] ScottK: background fetching also helps. I like leonov in this respect [13:43] siretart, repeating myself, there is no GPL breach, the source is available upon request, you are not forced to distribute it with the binaries, the same way you don't actually do it with the binaries provided on the Ubuntu CD [13:43] joaopinto: That's one very long chain. [13:43] persia: I'm getting Illegal Files line in .changes when I try to upload the .changes file [13:43] siretart: Agreed. [13:43] cherva: Did you build a source package or a binary package? [13:44] siretart: I do think we should try to get some QoS requirements in the API. [13:44] persia: I build a source dpkg-buildpackage -S -sa -rfakeroot [13:44] wgrant, they have a different opinion, and they are probably make other's peoples life easier [13:44] siretart: I'd like to know they don't provide data more slowly via the API than to their web U/I. [13:44] cherva: Hmm. Can you pastebin your source.changes file? [13:45] joaopinto: It may not be a GPL violation for them, but it would be for us if we don't have the source. [13:45] I'd say their opinion is wrong. [13:45] persia: http://pastebin.com/d4d0864dc [13:45] Just because it's legally distributable under the GPL by them, doesn't necessarily mean it's redistributable. [13:46] joaopinto: Essentially, we can't redistribute until we have the source, or we are not in a position to extend the offer to provide source on request to those to which we distribute. [13:47] cherva: Aside from the thought that it probably shouldn't be Extra, I'm not sure why that is broken. Sorry. [13:47] Heya gang [13:48] hi bddebian [13:48] Hi sistpoty|work [13:48] ok, i will just rm the dlls, to end the question [13:48] persia: regarding orphaned packages and watch files: yes, even there I would not add such a change alone. But rather forward a patch to unstable. [13:48] persia: I made it Extra because the normal package in universe is using alsa and the binaries has the same names just my recordmydesktop is using oss and they can't be installed at the same time [13:48] persia: (was afk for coffee break in case you're wondering ;)) [13:49] sistpoty|work: Ah. I tend to do both, as I figure Debian QA doesn't necessarily score importance the same way I do in terms of making sure we have them towards getting new upstreams. [13:50] * persia had thought it was just the IRC timewarp: without timezones, one person's few minutes is another's couple days. [13:50] oh well, I am getting bored of new packages after all :P [13:51] persia: I think I know where is the problem... when I build a DEB file i get http://pastebin.com/d493e65b2 [13:52] cherva: That's the first time someone told me a good reason to have something extra. Good choice. Of course, I'm not sure what is wrong with the Files section. [13:53] persia: if you need a file to look at just tell me what to pastebin [13:54] cherva: No. The problem is with the .changes file, which you've sent. I just don't understand why REVU rejected it. [13:55] persia: any ideas how to fix this / [13:56] Unforunately not. You've created the file in the right way. It looks good. REVU didn't like it. I'm guessing NCommander might be the right person to ask. [13:57] sadly he is not online [13:58] cherva: Right. [13:58] cherva, persia: which source package are you talking about? [13:59] persia: maybe revu do not expect two words for Section ? [13:59] sistpoty|work: I'm making a recordmydesktop package compiled to use oss and I can't get the .changes file to REVU [14:01] cherva: Two words for section? Can you post your debian/control? [14:01] hm... there's s.th. entirely wrong on spooky... hrmpf [14:02] http://pastebin.com/d49b5fcc4 [14:02] cherva: Ah. Drop the " (universe)" from that. [14:03] persia: I knew it [14:03] I'm not sure that fixes it, but it doesn't belong there. [14:04] cherva: when did you actually try to upload recordmydesktop? [14:04] (as I can't find any mention of it in any logs) [14:05] I'm trying right now [14:05] ok It's there [14:06] cherva: worked like a charm :) [14:07] anyone willing to revu and sing my recordmydesktop-oss package ? It's the latest version of recordmydesktop compiled to use oss instead of alsa [14:08] cherva: why oss instead of alsa... and also if you insist on oss, why not providing two binaries from recordmydesktop instead of adding another source package? [14:11] sistpoty|work: I was having troubles with the sound using ALSA+Pulse ( strange bugs) and I didn't manage to configure it that all programs can use the sound card at the same time so I swithed to ossv4 and it worked great and i didn't provied two binaries because 1) i don't know how 2) i waned to learn how to become a MOTU and 3) if I was looking for a recordmydesktop package I would find it simle to just install the other package [14:12] 4.1 Comment is not in accordance with Gnome HIG <- could someone provide me a pointer for this ? [14:13] joaopinto: Your patch (or so) doesn't comply with the Gnome Human Interface Guidelines [14:13] joaopinto: http://developer.gnome.org/projects/gup/hig/ [14:13] this comment is related to a .desktop description [14:14] StevenK, I understood that part, the question is, is there any specific section, or do I will need to read and assimilate the entire HIG ? [14:14] the current field is: Comment=Amoebax is a free Puyo Puyo-style puzzle game [14:15] joaopinto: http://library.gnome.org/devel/hig-book/stable/desktop-application-menu.html.en [14:15] joaopinto: http://developer.gnome.org/projects/gup/hig/2.0/desktop-integration.html [14:15] persia 1 - rainct 0 :P [14:17] RainCT: Well, yours has the advantage of possibly being more forward looking :) [14:17] well, I have read it, I don't see anything wrong with my description [14:18] any suggestions ? [14:19] What is the current Comment? [14:19] Comment=Amoebax is a free Puyo Puyo-style puzzle game [14:20] Yeah, that doesn't belong in a Desktop file like that. I think you want something like "Play a Puyo-Puyo style puzzle game". The Comment ends up as the Tooltip (see 2.12) [14:20] joaopinto: a) It includes the application's name, b) it doesn't start with a verb [14:21] ok, changed, thanks [14:22] * RainCT would use "like" instead of "style" [14:25] For English at least I think you'd end up having to hyphenate -like as well at that point though, with a long dash for --like. [14:25] * GreySim could be mistaken, however. [14:25] Saw the eclipse! [14:26] can we use per package copyright files for mutiple binary packages, like copyright and -data.copyright ? [14:26] GreySim: right (at least until the first comma, then I got lost :P) [14:28] sistpoty|work: well ... is my package worthy to live in universe ? [14:28] RainCT: When hyphenating at different levels, you use an em dash for one of them, I believe. If you were really curious, I'm probably remembering it from the "A List Apart" article "Em en and other shady characters" article. Probably not really all that important though. Depends on how much detail you want to focus on. [14:29] LucidFox: WHich eclipse are you talking about? [14:29] Not the software :p [14:30] Today's solar eclipse [14:30] joaopinto: There's no facility for that. [14:31] ok, so it will be a plain description on debian/copyright [14:34] // Based on the work of Darrel Walisser <- should this be considered a copyright assignment ? [14:34] siretart: lp bug on source info reported as bug 253934. Re. the homepage there is a bug already which I believe covers it (bug 113573). This is however marked invalid in favour of bug 73116 and bug 113748 [14:34] Launchpad bug 253934 in launchpad "Please bring back the source info to the bug page" [Undecided,New] https://launchpad.net/bugs/253934 [14:34] Launchpad bug 113573 in launchpad "Link to homepage of a package in overview" [Undecided,Invalid] https://launchpad.net/bugs/113573 [14:34] Launchpad bug 73116 in soyuz "Source package pages don't describe what the package is for" [Undecided,New] https://launchpad.net/bugs/73116 [14:34] Launchpad bug 113748 in soyuz "URLs aren't auto-linked in package descriptions" [Undecided,Confirmed] https://launchpad.net/bugs/113748 [14:34] on my understanding, based on does not mean copy from [14:36] well, I will copy the statement to play safe [14:37] joaopinto: There ought be copyright assertion in each file. That's the part that is important. If Alice writes something, and Bob modifies it to make something else, if Bob says it's copyright Alice & Bob, then it is. If Bob says it's copyright Bob, it's Alice's issue, not ours. Mind you, in truly egregious or contested cases, we won't listen to Bob, but that usually means not distributing it, rather than correcting Bob's assertion of copyr [14:37] ight. [14:37] If files don't have copyright assertion, they may not be considered suitable for inclusion: I've seen 3-line files get rejected by archive-admins before. [14:43] uff, all comments reviewed, now I will be pending on the manual.pdf source :\ [14:44] joaopinto: That's often one of the trickiest ones. Note that the "preferred form for modification" must be included, but that there is precedent for not building the PDF from that source (in the rare case that the source is e.g. a MS Word file). [14:46] this is a simple game, the manual is not really crucial [14:49] joaopinto: The other option is repacking the tarball to not include the manual, although fixing it is preferred. [14:51] I would appreciate some comments for http://revu.ubuntuwire.com/details.py?package=coverfinder [14:54] joaopinto: Useless README.Debian, might want a dependency on ${misc:Depends} as you don't know what CDBS is going to do, Homepage should not be in the description, ought specify which version of the GPL in the link: source is GPL 2+, link is to GPL 3 [14:55] persia: about the link, you mean in debian/copyright? [14:56] Useless inclusion of README in debian/docs (the file is without useful content), Do you really need both dpatch/dpatch.make and cdbs/1/rules/dpatch.mk? [14:56] Yes, in debian/copyright [14:56] Anyway, that's not a proper review, because I'm going to bed, but it should give you something to get started. [14:57] persia, thanks, that's done already [14:58] persia: If it is "or later", I prefer to refer to "/usr/share/common-licenses/GPL", as it points to the latest version of the GPL which is obviously that one which the FSF recommends. (I've done that in packages and all got accepted into Debian). That's just my personal preference, though. [14:59] joaopinto: also, change "Authors(s):" to "Author:" in debian/copyright; else lintian will complain [15:00] RainCT: Understood. I like to point at the specific terms the author intended, and allow others to relicense if they wish, rather than intentionally relicensing. Both are acceptable. [15:00] ok, done === RainCT_ is now known as RainCT [15:24] cherva: well, it's not good to duplicate the source, so two binary packages built from one package would be better [15:25] sistpoty|work: well 1) I don't own the firt one and 2) the first one is outdated [15:25] cherva: in ubuntu noone "owns" one source package ;) [15:26] sistpoty|work: so what should I DO ? [15:26] cherva: probably best to update recordmydestkop instead? [15:27] I don't know how to make 2 binary files into one deb [15:28] * cherva is back [15:31] sistpoty|work: and if I make to binaries the frontends ( gtk-recordmydesktop and qt-recordmydesktop ) will not work [15:31] cherva: why? [15:33] sistpoty|work: because they I made to work with a binary named recordmydesktop and if I make a second one eg. recordmydesktop-oss if you start gtk-recordmydesktop they will start the alsa version not the oss version ... or I'm wrong [15:33] cherva: maybe I was not precise enough... I meant two binary packages from one source package [15:34] binary package = DEB files [15:35] yep [15:35] well I made one you want to build a second one for alsa ? [15:36] exactly... from the same source package [15:37] what is the best practice for that ? to copy the dir and change the rules file to make an alsa verion and then upload it to REVU too ? [15:39] cherva: you'll add a 2nd binary package to debian/control and modify the rule so that it results in two binaries [15:41] hmm I have to copy everything from Package:.......... and down agan ? [15:43] sorry, can't really give you an in depth guide right now, have to do some work still :/ [15:44] sistpoty|work, ping, i wanted to ask you about a merge for nvidia-settings you did way back when [15:45] sistpoty|work: ok I'll try and write back to you (if not today, tomorrow _ [15:45] ) [15:45] cherva: heh [15:45] superm1: just ask, don't ask to ask ;) [15:45] sistpoty|work, wasn't sure if you were around [15:45] sistpoty|work, okay anyhow. [15:45] sistpoty|work, did you submit changes back to debian at all? [15:45] because there are a few deltas in there that probably make sense for them too now [15:46] superm1: nope, as there was a blocker bug (some missing static library from somewhere I don't recall right now) filed in debian already [15:46] superm1: and later we jumped upstream versions a bit [15:46] superm1: some stuff (conflicts) don't apply for unstable [15:47] sistpoty|work, ah. well they're ahead of us now, and tjaalton has been trying to get the deltas with nvidia-glx* packages to be smaller, so this seemed logical to do now too [15:47] yes, it does [15:49] superm1: ah... now I recall. I made nvidia-settings provide the static lib... which was already filed as bug in unstable for a long time [15:49] sistpoty|work, has the maintainer for it not been responsive then still? [15:49] or is that taken care of now [15:50] superm1: doesn't look like its fixed: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=434885 [15:50] Debian bug 434885 in nvidia-settings "Missing headers in nvidia-settings from unstable" [Normal,Open] [15:51] well it's marked as fixed in 1.0+20060516-2 though? [15:52] * sistpoty|work looks at the newer debian package [15:55] superm1: indeed, the headers + static lib is there [15:55] superm1: so I guess you can drop this change (but should then also update sensors-applet to b-d on nvidia-settings instead of libxnvctrl-dev) [15:56] sistpoty|work, okay well i'll try to do a merge this weekend with it and submit some bugs for the other deltas then [15:56] thanks superm1 [16:04] persia: i've been through the debian-policy upgrading-checklist.txt from 3.7.3 to 3.8.0 and it looks ok. So, i can now bump the standards version? [16:11] * sistpoty|work heads home... cya [16:15] Would i say in LP - attachment is a patch if its a debdiff? [16:16] stefanlsd: yep [16:17] james_w: thanks [16:28] Hello [16:30] When we look at syncing a debian package - do we just sync it. or should we look at closing open bugs as well? [16:30] I have problems with uploading an updated version of my package "debcleaner". Since July 26th, 17:20, packages I uploaded via dput are not shown in the list. Any idea? [16:31] MrKanister: What command are you using to upload the package? [16:32] nhandler: I use "dput revu /path/to/package....changes" [16:32] MrKanister: Try including the -f option [16:32] nhandler: I already tried that [16:32] MrKanister: Have you signed into REVU since it got updated last week? [16:33] I think no [16:33] Try doing that. When you log into REVU, it will sync the keyring. You should also merge your Launchpad account with your old REVU account [16:35] Ok..i will do so and see what happens...thank you [16:39] I can't merge the accounts because I didn't saved the passwort for REVU...I always klicked on "forgot passwort" and it generated a new one for me [16:40] MrKanister: Maybe one of the REVU admins can merge the accounts for you. [16:42] Can I ask a MOTU? [16:42] A MOTU does not have the power to merge the accounts. It has to be a REVU admin [16:43] MrKanister: Login to REVU, there's a "Merge account" link. [16:43] jpds: He doesn't know the password to his old account [16:43] MrKanister, ask RainCT or NCommander [16:44] Ok...thank you for your help === menyort is now known as ApOgEE- [16:59] Hello, can someone help me with swt-gtk build failures: https://launchpad.net/ubuntu/+source/swt-gtk/3.4-1 [17:02] Anyone willing to tell me how to configure the debian/rules file of a multy binary package ? I need to give different parameters to ./configure at the second package [17:10] It works...the "recover password" function of "merge accounts" was defect, but rainct managed to fixed that and now it works. Thank you all for your help === ApOgEE- is now known as Sempudal === krusaf|bnc is now known as krusaf [17:23] comments? http://revu.ubuntuwire.com/details.py?package=tv-grab-dvb === Sempudal is now known as ApOgEE- [17:25] persia: are you here [17:30] doko: Hello [17:31] Hey I just made a .deb and it works, I can install it and unistall it succesfully [17:31] gastoni: congrats [17:32] The only annoyance it has is that when I want to uninstall it with apt-get [17:32] the autocomplete feature adds a slash to my package [17:32] sudo apt-get remove "package name"/ [17:33] why does that slash appears? [17:35] hi. Use unicode like ą ł ć in DEBFULLNAME or not? [17:36] DEBFULLNAME would be something inside the control file? [17:36] according to http://www.youtube.com/watch?v=VyEl3w7SFK4 I have it in bashrc [17:37] LimCore: do you run a utf-8 locale? [17:37] overall yes [17:37] you should be ok then, [17:37] does the FULLNAME have to match the name in gnupg key? or only email is checked? [17:38] that get's added in to debian/changelog, and that file must be utf-8 encoded [17:38] because, some gpg programs have problems with localized names so I dont use localized in gnupg fullname [17:38] ok [17:38] the name in debian/changelog should match the id on the gpg key you want to use byte-for-byte [17:38] name and email [17:38] Overall, I see that ubuntu fails to provide libfltk2-dev, so Im trying to provide it. (build from SVN of fltk2 into ubuntu package) [17:39] name too? ok then I go with ascii transcription of it [17:43] Anyone know a of anywhere in the wiki or documentation on how to do a sync from debian? [17:44] Learning MOTOU video - http://www.youtube.com/watch?v=VyEl3w7SFK4 - could use bigger font [17:44] like x3 bigger [17:44] but it is a nice idea :) [17:44] stefanlsd: yep there is [17:44] stefanlsd: https://wiki.ubuntu.com/SyncRequestProcess [17:45] stefanlsd: See: "man requestsync" too. [17:45] thanks jpds [17:45] jpds: i've read that. seems more about requesting a sync, rather than explaining how to do it. will check it again [17:45] stefanlsd: That script does all the fiddling for you. [17:45] jpds: kk. thanks. will have a look at it [17:47] stefanlsd: you don't need to do a sync, requesting it will get it done, they are done by the archive admins with a script [17:48] james_w: yeah, i've assigned myself to do the sync - someone else requested the sync :) [17:48] stefanlsd: what's the bug number, I'd like to look? [17:49] * dholbach hugs norsetto [17:49] dholbach! [17:49] norsetto: great work on the Mentoring Program!!! [17:49] dholbach: welcome back :) you look tanner ;-) [17:50] james_w: https://bugs.launchpad.net/ubuntu/+source/diald/+bug/253675 [17:50] Launchpad bug 253675 in diald "Please sync diald 0.99.4-9 (universe) from Debian unstable (main)" [Low,In progress] [17:50] stefanlsd: Sycns are done by archive admins. Merges are done by MOTUs. [17:50] dholbach: you have to thanks persia, he is the deus-ex-machina ;-) [17:50] jpds: oh. didnt realise that. [17:50] thanks to everybody who's doing great work in the Mentoring world right now! :) [17:51] persia has been helping me a lot with all my questions - been great [17:51] yes, thanks to all the hard-working mentors and the ass-kicking contributors [17:52] stefanlsd: Debian may be interested in the new release tho. [17:52] absolutely - I was amazed to come back and see so so many applications on the MC list :) [17:52] jpds: it wont compile.. almost compiles. but upsteam last release was 2001 and everything is depricated... [17:53] stefanlsd: Do you know if it still works? If it's that old/unmaintained prehaps a removal is a better idea. [17:54] * dholbach hugs y'all [17:54] \o/ [17:54] dholbach: We do our best. [17:54] jpds: I'm glad to hear it :-) [17:57] dholbach: I don't want to ruin your return, but you may want to look again at bug 252037? [17:57] Launchpad bug 252037 in sauerbraten-data "sauerbraten cannot upgrade" [Undecided,Confirmed] https://launchpad.net/bugs/252037 [17:57] norsetto: I got the mail about it - will take a look at it tomorrow [17:57] norsetto: I wasn't aware of that [17:57] dholbach: sure, thanks for being so active in ums, its really appreciated [17:57] no problem, lots of stuff to catch up with after the holidays [17:58] it seems I'm still on IST - I got up at 5:44 this morning [17:58] dholbach: hehe, at least you survided the food [17:58] norsetto: the food was absolutely fantastic [17:58] dholbach: glad you lik spicy food ;-) [17:58] I could have gone on and have indian food for an additional three weeks [17:59] Anyone interested is packaging NCTnus ?? === fta_ is now known as fta [17:59] dholbach: have you posted already the pictures? [18:00] norsetto: no, but we did ~1100, so I guess it'll be no problem to select a few good ones :) [18:00] once I'm on top of things again I'll write a longer blog post I think :) [18:01] dholbach: good, I'm looking forward to see them! [18:01] :-) [18:04] have a great weekend! [18:45] Anyone who's got a local source mirror for Intrepid willing to do some serious grep'ing (or pick your tool) to figure how many package use MIT (or ISC) license? [18:48] heh [18:49] RainCT: Is that a yes? [18:49] ScottK: no, I don't even have an Intrepid installation :( [18:50] I need someone with a source mirror, installation isn't relevant. [18:52] ScottK: By the way, why don't you try to get this done on the Debian side? [18:53] Because Debian just froze for Lenny so this sort of thing they are extremely unlikely to consider until after Lenny releases. [19:01] ScottK: lifeless may be able to do that for you [19:01] That'd be nice. [19:02] IIRC StevenK or wgrant have a local source mirror. I don't recall which. [19:08] I'll take whatever help I can get. [19:11] Nevermind. Shot down by slangasek. [19:13] oh, and thanks for the advocation ScottK [19:14] james_w: You're welcome. You certainly deserve it. [19:15] james_w: I wish more people at Canonical who are not employeed to develop Ubuntu were involved. I think there are a number of projects at Canonical that would benifit. [19:16] well, I kind of am paid to develop Ubuntu, so I don't think you can give me that much credit [19:16] it would be good though, not least to have the extra help in development [19:18] I just don't think there should be an expectation of them, a desire is fine though. [19:20] OK. I thought you were paid to develop bzr. [19:21] I'm reasonably confident that if a reasonable sampling of Launchpad developers where regularly involved in Ubuntu development, I'd be less grumpy about it. [19:23] no, I'm on the Ubuntu team [19:24] bzr is my hobby, though I am expected to contribute to it for some things [19:25] Interesting, I got that completely backwards. [19:44] DktrKranz: hi. gResistor upstream has replied saying he will update his tarball with the licese copies :-) [19:47] tuxmaniac, great! [19:48] DktrKranz: andanother upstream reply also I got. :-) this time a fix for bug 134617. Really a long open one shall be closed in a week's time now . I am feeling good after string of no responses and frustrations [19:48] Launchpad bug 134617 in xcircuit "xcircuit crash middle click dragging a selection" [Undecided,Confirmed] https://launchpad.net/bugs/134617 [19:49] :) [19:49] tuxmaniac, ping if you need sponsorship :) [19:50] DktrKranz: sure after the tar ball is officially released. Right now I need sponsorship for 253324 :P [19:50] bug 253324 [19:50] Launchpad bug 253324 in ubuntu "[sync request] Avagadro Sync from debian unstable 0.8.1-4 version." [Undecided,New] https://launchpad.net/bugs/253324 [19:51] tuxmaniac, I can't sponsor anything right now, I'll be off soon, but I'll have time tomorrow/next week, I'm on holiday \i/ [19:52] * NCommander forces himself to wake up [19:52] DktrKranz: no issues. :-) [20:04] * ScottK smacks NCommander with a frozen fish to help. [20:04] * NCommander smacks ScottK with a gnome [20:04] Fortunately gnome has no effect on ScottK. [20:05] Just trying to be helpful. [20:05] brb, voicemail [20:11] ScottK, ok, whats up? [20:12] Nothing, just trying to help you wake up. [20:17] ScottK: gnome gnome gnome \o/ [20:17] Right. Ne effect at all. [20:17] Ne/No [20:18] * sebner needs the magic stick of Hobbsee to convert ScottK to gnome :P [20:27] does anyone know when the next REVU day is? [20:31] sebner: I like KDE better than I like Ubuntu. No Kubuntu, no ScottK [20:31] "I like KDE better than I like Ubuntu" [20:32] now what kind of statement is that ? [20:32] ogra: very good question+ [20:33] I use Ubuntu as a base both for my KDE (Kubuntu) desktop and my servers because it works for that. [20:33] right, so how can you say you dont like it :) [20:33] you dont like gnome :) [20:33] I like the fact that it's common under the hood and a lot of other things about Ubuntu, but fundamentally, as a user it's KDE I interact with. [20:35] I like Ubuntu platform + KDE = Kubuntu and Ubuntu platform + server stuff. The Gnome part, not really. It just feels wrong to me. [20:35] * laga nods gravely [20:35] If Kubuntu were to disappear, so would I. [20:36] sebner: have you given any thought about giving a lecture? [20:36] norsetto: about? [20:36] sebner: perhaps you can cohost nhandler one on merging [20:37] sebner: or anything else you woul feel confortable to share? [20:37] ScottK, well, to me gnome feels like a professional desktop while KDE feels like a kids toy ... but thats completely a personal preference ... if ubuntu would switch to kde as default i would just do a base install and drop gnome on top ... [20:38] really nothing to wind up about :) [20:38] or to leave a community for [20:38] * norsetto sits down and open a chips bag [20:38] It'd be less work to just switch to Debian. [20:38] It's all hypothetical anyway. [20:39] yeah [20:40] * norsetto thinks that flamewars are not what they used to be [20:40] * sebner sits near norsetto and shares beer with him [20:40] norsetto, yeah, arent we boring :P [20:40] * ogra gets a beer too [20:40] Careful. Is sebner old enough? [20:41] scottk: he is austrian, they are raised on beer [20:41] * laga randomly reports beer drinkers to the community council [20:41] in EU you are allowed to have beer and wine with 16 in most places [20:42] OK. I thought it was 18. [20:42] ScottK: 18 \o/ --- though I don't drink alocohol [20:42] norsetto: well, I once tried guiness. *horrible* [20:42] Greetings, people of #ubuntu-motu! [20:42] heya zooko. [20:42] 18 for spirits [20:42] Ah. [20:43] sebner: you're not supposed to drink the whole keg [20:43] and 18 for driver licenses [20:43] Who knows python packaging here? [20:43] * ScottK is just about to head out for several hours. [20:43] * norsetto is learning perl :-P [20:43] Right, but you know Python already. [20:44] I am investigating this bug: [20:44] https://bugs.launchpad.net/setuptools/+bug/254035 [20:44] Launchpad bug 254035 in python-setuptools "easy_install will install a package that is already there" [Undecided,New] [20:44] norsetto: Would you help zooko. He's got some kind of ez_setup problem he's needing help troubleshooting. [20:44] This bug began life when debian upgraded its version of setuptools, and later migrated into Ubuntu in Hardy. [20:44] zooko: If no one else helps you with it I should be able to in 4-6 hours. [20:45] The Debian bug report (linked from the launchpad page) narrows down somewhat precisely which version of Debian package introduced the bug. [20:45] ScottK: thanks! See you around. [20:45] ... but /me cannot reproduce it on Debian sid [20:46] ScottK: I can certainly look into that, dunno if I can actually help him [20:46] ogra: well, 17 [20:46] I also can't reproduce it on Debian sid because I don't have access to Debian sid. :-{ [20:46] Maybe someone else will jump in and appear. [20:46] debian is bad -.- [20:46] sebner, 17 ? where ? [20:47] sebner: Without Debian there is no Ubuntu. [20:47] ogra: austria. costs more but possible [20:47] ah [20:47] One could write to the original poster, Brian Warner, and ask him to reproduce it on his sid box, but he also did so and posted transcripts of what he did an how it failed on the allmydata.org ticket [20:47] ScottK: I know but there are some bad boys [20:47] (also linked from the launchpad ticket) [20:47] i heard soe german states switched to 17 as well [20:47] but only some [20:47] sebner: Same here. [20:47] ScottK, but they are less rude here [20:47] CoC just requires them to be more subtle, not actually nice. [20:47] sebner, but ScottK is right [20:47] * POX_ hits sebner [20:47] no debian, no ubuntu [20:47] * POX_ now feels like DD ;) [20:48] ;-) [20:48] löl [20:48] Ironically, Debian developers have to promise to care about users. Ubuntu has no parallel requirement. [20:48] For what it is worth, the folks on the #debian-python channel have always been very polite and helpful to me. [20:48] For which I am grateful. [20:48] ScottK: ok, I hate the debian packaging beginners :P [20:49] zooko: That's a part of Debian that works very well with Ubuntu. [20:49] Now if you Ubuntu people want to get similar kudos for being helpful, then you can start by seeing if this setuptools bug exhibits on your Ubuntu system. ;-) [20:49] sebner: You're in the wrong channel then. [20:49] * zooko laughs. [20:49] So one thing that I did to investigate this problem was to uninstall python-setuptools (Ooff of Hardy) [20:49] and replace it by the same version of setuptools upstream. [20:50] This fixed the behavior on my hardy box. [20:50] ScottK: because? [20:50] So this suggests that the problem is in the packaging -- it is specific to Debian (and then to Ubuntu once those patches got into Hardy) -- [20:50] it doesn't happen on the other half-a-dozen platforms that I support. [20:51] sebner: If you don't like debian packaging beginners, this is bad place to be. [20:52] ScottK: point me to some sebner's packages, lets see how good he is [20:52] ScottK: not "debian package" but newbies packaging for debian [20:52] POX_: it's a missunderstanding [20:52] ScottK: *only* for debian. they stole me a package :P [20:53] ;-) [20:53] ScottK: and my package was better since the guy overrided a manpage warning without having a manpage xD [20:54] sebner: hint: upload your packages to Ubuntu via Debian - nobody will stole it [20:54] :) [20:55] steal* [20:55] POX_: it was on debian mentors. they guy filed the bug in BTS 2 days after me! [20:55] they = the [20:55] sebner: but don't ask POX_ to sponsor you, he always find excuses (a broken motherboard, holidays ....) [20:55] zooko: Fundamentally Debian and Ubuntu have a different vision of what "Works" means than upstream for ez_setup. We want people to use the Debian packages. [20:55] But I really have to go. [20:56] POX_: really a DD? [20:56] norsetto: uhuhuuhuhuhu! good idea :D :D :D [20:56] sebner: Yes. [20:56] norsetto: yeah, and still >100 sponsored packages, I need to find a better excuses [20:57] Bye for now ScottK. [20:57] POX_: go and teach the guy how to make it right :P and why the **** are you (debian) accepting such a package? [20:57] sebner: did you file a ITP? [20:58] POX_: wnpp [20:58] who sponsored it? [20:58] maybe he missed your ITP [20:58] did you try to contact the guy? [20:58] Okay I just reproduced this bug that is bugging my partner Brian. I just reproduced it on a current Hardy box. [20:59] POX_: just saw it yesterday or the day before in the new queue and now it's in unstable [20:59] package name? [20:59] POX_: almanah [21:00] Now I'll try installing setuptools from upstream sources... [21:00] POX_: though it disappeared on mentors (at least my package) [21:00] oh, it was uploaded by a girl, I will not send f*cs to a girl, sorry ;) [21:01] and seriously, I'm sure she just missed your package [21:01] sorry [21:02] POX_: it wouldn't be soo bad if I didn't have to fight on revu with reviewers and fixing all the complains to have a shiny package (with a manpage) ;) [21:06] sebner: what is really important is that you learned something [21:07] norsetto: sure but it would have been nice to be debian and ubuntu maintainer [21:07] sebner: you will, now that you know better how to package ;-) [21:08] since I'm sure that the other guy didn't contact upstream [21:09] !pastebin [21:09] pastebin is a service to post multiple-lined texts so you don't flood the channel. The Ubuntu pastebin is at http://paste.ubuntu.com (make sure you give us the URL for your paste - see also the channel topic) [21:09] !! What? Stop the presses -- I get the same bad behavior with upstream setuptools. [21:09] zooko: Error: I am only a bot, please don't think I'm intelligent :) [21:09] I will post again to this channel once I have investigated. [21:10] zooko: ah [21:11] Hm... It could have something to do with this: http://bugs.python.org/setuptools/issue20 [21:11] But I wonder why it started happening with a specific Debian revision... [21:11] More investigation is required. I'll be back. [21:14] POX_: "Ángel" is a man's name, actually ;P [21:15] slangasek: who-uploads almanah [21:15] Amaya Rodrigo Sastre [21:15] ah [21:15] so the sponsor, not the maintainer, ok [21:15] POX_: I have another package on debian mentors :P [21:15] sebner: Python related? [21:16] sebner: http://people.debian.org/~piotr/sponsor [21:16] POX_: yes (with fingers crossed behind the back) [21:23] joaopinto: still interested in coverfinder? for example it seems that you are missing a manpage. should I take it over? [21:26] sebner, i did some fixes on my localcopy today, but mostly on debian/copyright [21:27] but I was not planning to create a manpage, so feel free to take it over [21:27] joaopinto: ubuntu is all about manpage xD [21:27] joaopinto: unfortunately [21:28] I don't find much value on a manpage for a gui app wich supports no parameters [21:28] Anyone willing to tell me how to configure the debian/rules file of a multy binary package ? I need to give different parameters to ./configure on the second package [21:28] joaopinto: not even --version? [21:29] sebner, not sure, but does it matter for a coverfinder app :) ? [21:30] joaopinto: not really but as I said. ubuntu is all about manpages *hrhr* [21:30] cherva: you can find this useful: http://lists.debian.org/debian-mentors/2007/09/msg00487.html [21:30] manpages -> terminal, help menu -> gui :P === zooko is now known as zookonap [21:31] joaopinto: hrhr [21:34] sebner, since you are on it, debian/copyright, the ...GPL file pointer should be GPL-2, and I have missed the icon copyright (there is a text file with the license for the app icon, CC) [21:35] joaopinto: well if it really doesn't have any terminal options you may find motus to ACK it without manpage ;) [21:37] Can anyone help me? [21:37] I'm not sure how to resolve debian-changelog-file-a-symlink lintian warning [21:37] I hope I can get the amoebax package in times, is one less software ultamatix *powered* [21:37] (it appears CDBS is doing that on its own) [21:39] !ohmy | joaopinto [21:39] joaopinto: Please watch your language and topic to help keep this channel family friendly. [21:40] joaopinto: don't say anything wit *matix please :P [21:40] :P [21:43] norsetto: I can't really understand it [21:46] cherva: what is the problem? === leleobhz is now known as maedinah === maedinah is now known as leleobhz [21:48] I made two configure "things" and then I don't know what to change after that here is the rules file now http://pastebin.com/d6a2ea7aehttp://pastebin.com/d6a2ea7ae [21:48] http://pastebin.com/d6a2ea7ae [21:49] cherva: hmmm, you know a bit about makefiles? [21:50] norsetto: not really [21:50] cherva: right [21:50] I made a single binary package but I'm having troubles with the multi binary [21:51] cherva: ok, the configure, build etc. words, with the :, these are targets [21:51] cherva: after a target there are instructions which are run when that target is executed [21:52] cherva: after the : there are the pre-requisites for that target [21:52] cherva: a pre-requisitie is something that is needed for that target to be executed, it can be another target, or a file [21:52] cherva: ok so far? [21:53] norsetto: it's ok I knew a part of it... [21:53] cherva: ok, now, check your configure targets, when are they going to be executed? [21:54] norsetto: never , because I never call them [21:54] cherva: exactly [21:55] norsetto: here we come to the part I don't understand in your link ... this part http://pastebin.com/d37282fcb [21:57] cherva: yes, thats what you need to do, you need to have two build target, each with as a pre-requisite its corresponding configure target [21:58] cherva: both build-a and build-B have to be pre-requisite of build, so that when build is called all those are called [21:58] norsetto: [21:58] norsetto: I'm afraid I'll need some more info on that.... [21:59] sebner: 490535 never had ITP tag, that's why Angel and Amaya missed it (Angel should rename 490535 to ITP instead of filling 490655 though) === zookonap is now known as zooko [22:00] cherva: you need to have two build targets, build-alsa and build-oss [22:00] POX_: damn :( [22:00] norsetto: where to put them ? above ?#Architecture [22:00] cherva: then you have to have build depends on both [22:02] cherva: you don't need build-arch and build-indep unless you are building both arch dependant and an arch independant targets [22:02] norsetto: ok I'll delete that [22:03] norsetto: from line 56 to line 71 ? [22:03] cherva: I stumbled upon a package that was building two binaries with this method not long ago, this could be a good example for you to follow [22:04] cherva: unfortunately I forgot the name, but it was packed by nixternal, and if IIRC was a kde package, he surely remembers it [22:04] nixternal: do you remember the package that norsetto is tallking about ? [22:06] cherva: yes, I think he made two binaries from the same source, one specific for kde the other not [22:07] norsetto: good because I really want to upload a package to universe and the debian/rules file is my last problem [22:08] norsetto: so I'll delete lines 56 trough 71 [22:10] right? [22:10] cherva: kchmviewer [22:11] norsetto: so now apt-get source kchmviewer ? [22:11] cherva: I'd rather not tell you line by line what you should do, you won't learn anything this way, check the example I gave you, it should be enligthening [22:12] cherva: yes, use apt-get to fetch the source and check what is in debian/rules [22:12] norsetto: ok I'll be right back [22:12] wasabi? [22:13] norsetto: no time or ignoring me? [22:13] i have like 2 seconds for whatever question you have...then it is end of development cycle meeting for me [22:13] nixternal: I remembered, it was kchmviewer [22:13] nixternal: thx :-) [22:13] k [22:14] sebner: what is your problem? [22:14] norsetto: nothing but I thought we are talking in query ;) [22:14] Does anyone know what I need to include in my rules to get the CDBS flag, "DEB_CONFIGURE_EXTRA_FLAGS" to work? [22:15] sebner: don't query, it won't help anyone, I don't have time to query myself [22:15] It looked like I just needed to add autotools-vars.mk and autotools.mk -- but it doesn't seem to be doing anything [22:15] id liek to start helping in the maintenance of the nvidia-glx96 and 71 packages. there are new releases upstream and i have started installing and configuring pbuilder but iv ehad alittle tyrouble getting it to create a base file. (running intrepd) anyone familiar with this issue? [22:16] craps out regarding a tzdata necessity === zooko is now known as zookophone [22:16] michaelfavia: same here (with ltsp-build-client) [22:17] i guess intrepid just a bit broken right now? i haven't checked if there is a bug report, though [22:17] laga, soo are you trying to setup pbuiilder as well then? [22:17] or are you going another route? [22:17] norsetto: well the things we talked about doesn't help anyone but ok ^^ [22:17] it seems to be common for pbuilder not to work on dev versions. [22:17] but i cant get it to work by specifying distro=hardy either. [22:18] Majost: why do you say it doesn't work? === zookophone is now known as zooko [22:18] michaelfavia: i'm trying to run ltsp-build-client, which will also set up a chroot - just for a different applications [22:18] is there a missing include /usr/share/cdbs/1/class/autotools.mk? [22:18] because the options I am passing don't seem to be present in the output when I attempt to build the package [22:19] Majost: how are you passing this option to DEB_CONFIGURE_EXTRA_FLAGS? Can you paste your rules? [22:19] crimsun: I did include /usr/share/cdbs/1/class/autotools-vars.mk and then include /usr/share/cdbs/1/class/autotools.mk right below it [22:20] DEB_CONFIGURE_EXTRA_FLAGS := --x-includes=/usr/include/xorg --x-libraries=/usr/lib [22:20] I also tried: DEB_CONFIGURE_EXTRA_FLAGS := "--x-includes=/usr/include/xorg --x-libraries=/usr/lib" [22:21] cody-somerville, ping [22:21] norsetto: regarding lecture: there are lots of others that are a lot better in it. so why me? [22:21] sebner: I asked all other uuc not just you [22:22] I am defining the DEB_CONFIGURE_EXTRA_FLAGS option before the includes [22:22] norsetto: well that it's not really a problem to find someone better than me :) [22:22] I haven't tried after, but I hadn't thought to do so until now. [22:22] heh [22:23] norsetto: I came up with this one http://pastebin.com/d7be5b5fa [22:23] sebner: all uuc are welcome to participate actively in the motu school [22:23] norsetto: hrhr, a lot applications recently btw [22:24] ah... yeah... I needed to put it after the includes.. heh. sorry [22:25] cherva, you are using the same target install dir for both binary packages [22:26] joaopinto: I'm using what ? [22:26] $(MAKE) DESTDIR=$(CURDIR)/debian/recordmydesktop install, DESTDIR is the target installation directory [22:26] cherva: that won't work, you have to use different dirs [22:27] for the oss, it should be $(CURDIR)/debian/recordmydesktop-oss [22:27] when I see this messy makefiles I remember why I love CDBS :P [22:27] configure-alsa should be a pre-requisite for recordmydeskop [22:28] opss :) I was thinking about that but I wasn't shure [22:28] joaopinto: do you know debhelper 7? :P [22:28] configure-oss should be a pre-requistie for recordmydeskop-oss [22:28] sebner, nope, I always use CDBS [22:28] joaopinto: I already noticed [22:29] like this http://pastebin.com/d39f32cbc [22:29] http://revu.ubuntuwire.com/details.py?package=pangomm - can someone please review me [22:30] btw, anyone here running intrepid 32 bits has such a schroot ? [22:30] cherva: and you need to change directory when you make (or better, use make -C) [22:30] joaopinto, I have an intrepid 32 bit chroot jail if that's what your asking [22:30] Okay, it is definitely still present in upstream, non-Debian, non-Ubuntu setuptools. Sorry for the false alarm. I'll update the tickets. [22:31] NCommander, can you check if "gambas" is installable ? It is not on amd64, I was going to check for a bug which is present on Hardy [22:31] ops, the package is "gambas2" [22:31] joaopinto, You want me to see if its installable on a i386 chroot, or adm64 chroot? [22:31] *amd [22:31] i386 [22:32] on amd64 I am sure is not, I am using it [22:32] checking [22:32] norsetto: you mean $(MAKE) DESTDIR=$(CURDIR)/debian/recordmydesktop install -C build ? [22:32] joaopinto, you an MOTU> [22:32] no [22:33] joaopinto, its not showing up in the archive [22:33] Package not found [22:33] er wait [22:33] hold on [22:33] * NCommander enables universe in his chroot ^_^; [22:33] joaopinto, its' coming up as installable [22:34] (if I try and install all its dependencies to see if it will really install, it might take a few hours on this connection) [22:34] ok, thanks, so I will need to build a complete 32bits schroot to test it :\ [22:34] joaopinto, debootstrap is your friend [22:35] cherva: you have to call ../configure not ./configure [22:35] well, I have an helper script for that, but is failing with Intrepid, and I am not in the mood to fix it, doing it manually would be take more time [22:35] norsetto: ok I fixed that. is $(MAKE) DESTDIR=$(CURDIR)/debian/recordmydesktop install -C build-alsa ok ? [22:36] cherva: yes [22:37] norsetto: ok ... here is the updated file :) http://pastebin.com/d6cdf3bdc [22:38] cherva: the dh_install should be specified by package [22:39] norsetto, if you have a minute later, can I nab you to help review a REVU package? [22:40] norsetto: so....... dh_install -s recordmydesktop and dh_install -s recordmydesktop-oss ? [22:40] NCommander: sorry, I'm on my way to bed, just gimme the link and I'll see if I can look at it the next few days [22:40] norsetto, http://revu.ubuntuwire.com/details.py?package=pangomm [22:40] cherva: nope, if you use -s it won't work, you have to use -p [22:40] its a CDBS package so its pretty simple ;-) [22:40] NCommander: I hate cdbs ;-) [22:41] NCommander, you are the one working on the REVU improvements, right ? [22:41] joaopinto, yeah [22:41] norsetto: ok fixed that anything else ? [22:41] norsetto, normally, I'd agree, but you can't beat a ten line rules file [22:42] NCommander, a suggestion, add an "Your packages" section, filtering the packages for wich you are listed as the uploader [22:42] NCommander: is there anything you haven't patched !? :) [22:43] joaopinto, great idea, can you please file a wishlist bug again the revu project on LP? [22:43] norsetto, ?? [22:43] * NCommander is looking into a new cell phone [22:45] norsetto, I patched the Makefiles because of a bug in upstream (reported, and resolved in their svn) [22:45] NCommander, done, https://bugs.launchpad.net/revu/+bug/254078 [22:45] Launchpad bug 254078 in revu "Add a new "Your packages" section" [Undecided,New] [22:46] joaopinto, Thank you. Out of curosity, how do you like the recent REVU improvements? [22:46] * NCommander needs a new cell phone [22:47] NCommander: looks horrible xD [22:47] NCommander, I like a lot, you did a great a work [22:47] NCommander, whatcha thinking of getting? [22:48] SolarWar, I dunno, I looked at the freerunner, but it doesn't meet my requirements [22:48] I have a T-Mobile data plan with a small voice plan tacked on [22:48] (80 dollars a month for 300 voice/weekends, unlimited SMS, and unlimited data) [22:48] I had a T-Mobile MDA which finally died [22:49] NCommander, are you in the states? [22:49] (took long enough, probably the most duriable smart phone I ever owned) [22:49] SolarWar, yeah [22:49] i heard the MDA kinda sucks [22:49] It did in many respects [22:49] the freerunner's sold out from when i checked last time [22:49] But I loved it compared to my old Treo [22:49] ahh [22:49] SolarWar, The Freerunner doesn't support DUN-Bluetooth, or PAN-Bluetooth [22:49] NCommander, and I wasn't sure if it would work with tmobile [22:50] It's GSM, it would work [22:50] but its triband right? [22:50] My laptop has a Sprint EVDO card, but since a. it isn't supported under Linux and b. Sprint's data plans cost more then my entire monthly bill ;-) [22:50] hahaha [22:50] (actually, the EVDO card does show up as a COM device, I might be able to talk to it like a modem ...) [22:50] yea, i don't use bluetooth much, it would be fairly nice to have an openplatform to work on, but i hear the software stack isn't complete yet [22:51] good luck with that :) [22:51] Getting bluetooth tethering to work in Linux is a massive headache [22:51] Not to methon most smartphones need to switch into a special tethering mode to do it [22:51] (the Treo 650s was quite unstable, the MDA was ok if I used PAN) [22:52] I'm looking at the Nokia N95 [22:53] The code signing aspect is a turn off though for Symbian, but at least you don't need to hack your phone to add a new certificate [22:56] cherva, http://revu.ubuntuwire.com/details.py?package=recordmydesktop-oss , this is expected to be nuked, right ? [22:57] joaopinto: if nuked = deleted yes [22:58] please add such comment, otherwise someone may waste time reviewing it :P [22:58] ok [22:58] hello [22:59] can someone point me to documentation or a howto on how to repackage an already made deb package? I'd just like to add a few files [22:59] !packguide | TomaszD [22:59] TomaszD: The packaging guide is at http://wiki.ubuntu.com/PackagingGuide - See https://wiki.ubuntu.com/UbuntuDevelopment/NewPackages for information on getting a package integrated into Ubuntu - Other developer resources are at https://wiki.ubuntu.com/UbuntuDevelopment - See also !backports [23:00] I hope you are not asking to add files directly to a .deb file :P [23:02] joaopinto: what is wrong with this rules file ? I can't see the making of the directories and the configuring part being made ;( [23:02] http://pastebin.com/ddbafacb [23:04] joaopinto, I believe I do want that, it's just a matter of altering the control file right? [23:05] cherva, not wanting to lower your motivation but I found your rules too big for my reading, also I usually fllow a configure/build/install approach, on your case I can only see a configure and then a make install, which I presume takes care of the regular "make" first [23:06] TomaszD, do you want to repackage using the regular debian building process, or you want to directly edit an already generated .deb binary file ? [23:07] joaopinto, I'd much rather prefer the latter, I do not want to pointlessly rebuild the whole kernel just to add a few already compiled .ko files (yes, I want to add some files to the linux-image-blabla deb) [23:08] joaopinto, but there doesn't seem to be any easy way to do it [23:08] TomaszD, a .deb is just an archive, you can extract from it with ar, then you just need to manipulate the data.tar and control.tar [23:09] but that is a bit offtopic to this channel, that is not packaging, that is .deb hacking :P [23:09] joaopinto, ar --help and man ar is just amazingly complex :P [23:09] nah, that's packaging still [23:09] "ar x file.deb" is quite easy [23:10] tar xvf also :P [23:11] yes, what about the md5sum for the control file joaopinto :P [23:12] I need to add the files, bump the version number and then somehow add the md5sum [23:12] I'll figure it out [23:12] (hopefully) [23:12] it's just irritating with that "." folder, makes it hard to navigate [23:12] just add the file md5sum manually :) ? [23:13] md5sum newfile >> md5sums [23:13] I will, I just can't remember which file's md5sum to calculate, data.tar.gz? [23:13] give me a moment :] [23:13] still downloading the deb [23:13] no, the md5sums are listed on the md5sums file contained on control.tar.gz [23:15] yes, they're md5sums of the files in data.tar.gz, so I need to somehow recursively calculate them [23:15] not really, if you are going to add one file, you just need to append the md5sum for that file [23:16] ahh, true [23:16] to summarize: ar x; tar x; hack it; tar c; ar r [23:18] and you should also add an entry to the changelog, just to note that it was changed somehow [23:18] anyone wanting to help me with a rules file ? [23:18] so unpack the deb, unpack the data, unpack the control, hack both, then how do I pack it back up with the dot file... [23:18] cherva, I am leaving, sorry, anyway I would replace your rules with a CDBS based one :P [23:19] joaopinto: I'll leave it for tomorow [23:19] TomaszD, regular linux commands help, #ubuntu please [23:19] joaopinto: bue [23:19] joaopinto: bye [23:19] bye :) [23:19] ok [23:35] Should I be actively searching for sponsors for my update? -> https://bugs.launchpad.net/ubuntu/+source/malbolge/+bug/251311 [23:35] Launchpad bug 251311 in malbolge "new upstream version available" [Undecided,Confirmed] [23:48] cyberix: Did you subscribe the appropriate sponsor's team? [23:50] ScottK: I suppose so [23:50] "Ubuntu Sponsors for universe" [23:50] cyberix: That's it. That's as active as you probably need to be. [23:50] I'm just hoping to get it in before freeze [23:51] That's several weeks away, so you should be fine. [23:51] Ofcourse I wouldn't mind having time for fixing unexpected problems. [23:52] Plenty of time for bug fixing after Feature Freeze. [23:52] ok [23:59] zooko: Apply for the pythonistas team first (not pythoneers). Pythonistas covers universe and is more open to accepting new people (I can approve that).