=== korn_ is now known as c_korn [00:32] looking for someone interested in packaging Luz-- a VJ / music visualizer / graphic fun time / live-performance app. it's ruby and a couple C++ apps. 100% of dependencies are already in ubuntu repos. many posts about it here http://gnomecoder.wordpress.com [00:36] ianm_: oooh, colourful. I reckon that might be best trying to get that into Debian first [00:36] ianm_: Ubuntu people are all busy through to debug issues for the release next week [00:37] ianm_: and if it's in Debian then in a month it'll get pulled automatically into the next release of Ubuntu [00:37] sladen: all good points, thanks. is there a channel specifically for debian packagers? [00:44] ianm_: start at #debian on irc.debian.org [00:47] ianm_: I'd personally drop an email to http://lists.debian.org/debian-devel/ requesting somebody packaging it, with a couple of paragraphs introduction and easy to click links to the webpage, and source code === freeflyi1g is now known as freeflying === qiyong_ is now known as qiy [02:06] ianm_: #debian-mentors [02:46] whats the cut off date to get some packages into Karmic? === qiy is now known as qiyong [02:47] like a month ago? [02:48] probally missed that then! [02:48] https://wiki.ubuntu.com/KarmicReleaseSchedule [02:50] You're just in time to get some packages into Lucid Lynx! :-) [02:51] zooko, I will do that, at least I am early [03:02] Even better get them into Debian and they arrive into Lucid automatically [03:02] ScottK: how .. often is packages synced from Debian to Ubuntu? [03:03] or is it only 'when requested'? [03:03] runasand: For ~ the first half of the release cycle automatically every few days [03:03] ah ok :) [04:16] Does anyone know if TrueCrypt's license has changed sufficiently to be re-considered for universe or multiverse? (http://www.truecrypt.org/legal/license) [04:18] (albeit perhaps under a different name, a la firefox v. iceweasel) [04:18] it's a long license. hard to give a snap judgement [04:20] Phrase "Based on TrueCrypt, freely available at http://www.truecrypt.org/" must be displayed by Your Product (if technically feasible) and contained in its documentation. [04:20] is that considered a non-Free clause already? [04:20] (I think multiverse-wise truecrypt shoudl be fine) [04:20] though of course I can't say in 2 minutes I've read this thing! [04:21] directhex: Yeah - annoying that way [04:21] I'm still only partway through myself, after reading some commentary on debian MLs and Fedora's wiki [04:22] It seems it in the past was inadmissable, but changes frequently, progressing towards more freeness. [04:22] fedora's attitude is weird. they *want* to be super-Free-only, but they seem to have no coherent deterministic policy - seems more like gut feeling much of the time [04:22] generally, i mea [04:22] n [04:24] At the time of Fedora's last comment (in 2008), their legal team was of the opinion that the license placed risk on the distributor. I haven't found that yet reading this now though. [04:26] The only saving grace bits are that it exempts "Your product" to be just the stuff directly around truecrypt when dealing with aggregate/OS stuff (see definition 4) [04:26] Oh gross. [04:26] tonyyarusso: No, that's better than what I remember. Of course, I don't really remember much about what it was [04:26] but back in the day it was a pretty nasty license [04:27] You may allow a third party to use Your copy of This Product (or a copy that You make and distribute, or Your Product) provided that the third party overtly accepts and agrees to be bound by all terms and conditions of this License and the third party is not prohibited from using This Product (or portions thereof) by this License (see, e.g., Section VI.7) or by applicable law. However, You are not obligated to ensure that the third par [04:27] ^^ Does that mean it needs a debconf acceptance like Java? [04:27] Flannel: Yeah - I'm hoping maybe it's been fixed enough to finally be packaged - this has been a wishlist for years. [04:28] tonyyarusso: Yeah, I imagine their license issues are hurting their userbase significantly [04:28] Me, for instance. [04:29] I'd love to use it, as it's the only package I know of that offers plausible deniability through containers, but I'd want it in the repos first. [04:31] If people here (like jdong and Flannel) think it's at least close, I'll re-open the needs-packaging request so it can be looked at by people with legal expertise, but I wanted to get feedback from you guys first before I waste some lawyer's time. [04:32] tonyyarusso: my opinion is that it's gonna be close [04:32] tonyyarusso: and my gut feeling also is that the truecrypt team would be at least willing to be cooperative with a userbase as big as Ubuntu [04:32] That'd be helpful. === vorian is now known as obama === obama is now known as Obama [04:49] ianm_: I'd also file a launchpad needs-packaging bug if you haven't already [04:49] File it against Ubuntu, name it like [needs-packaging] foo, and tag it needs-packaging as well [04:50] then it'll end up in all the nice lists we link to on the wiki for people who want to contribute === Obama is now known as vorian [04:54] At a glance, I think it's fine for multiverse [04:54] Really obnoxious, but legal [04:55] ScottK: Was there something you see preventing it from universe? [04:55] The license acceptance thing [04:56] Gotcha. [05:03] btw, if anyone who's in here is interested in following up on that later, the relevant bug is https://bugs.edge.launchpad.net/ubuntu/+bug/109701 [05:03] Launchpad bug 109701 in ubuntu "[needs-packaging] Truecrypt" [Wishlist,Confirmed] === micahg1 is now known as mciahg === mciahg is now known as micahg [06:11] how [06:11] ack [06:11] sorry :) [06:14] @ping [06:14] !ping [06:14] pong! [07:20] im trying to build a pkg for 64bit and i have some very basic ques.... are --enable-64bit(in ./configure) and -m64(in gcc) doing the same thing? can i use either of them, at appropriate places or both? [07:24] is Tim spriggs here. === NCommander is now known as NC|Mobile === NC|Mobile is now known as NCommander [07:27] * maco does the I-Need-A-Sponsor dance around bug 272659 [07:27] Launchpad bug 272659 in gjiten "Gjiten could try to find kanjipad" [Wishlist,In progress] https://launchpad.net/bugs/272659 [07:39] maco: Looking [07:39] StevenK: thanks [07:40] maco: There was no patch system integrated? [07:40] StevenK: no [07:41] maco: Okay, I'll test-build and upload. [07:41] thank you [07:48] im trying to build nspr for 64bit and this is the error I got. i added the -m64 flag to CFLAGS and tried building .. can anyone please help me . Im just starting off and need help.. plese. http://pastie.org/661643 [07:49] do i need to write a separate rule for this compilation or how is it? I have absolutely no idea.. would really appreciate if anyone can help me. [07:49] thansk [07:50] maco: Uploaded. [07:51] StevenK: thanks much! [07:51] maco: No problem :-) [08:07] maco: Huzzah, gjiten was accepted === porthose is now known as porthose|afk [08:47] keyword ffmpeg [08:47] argl. morning byw [09:02] if I have a Hardy install, and I would like a newer version of a package (which is in say Intrepid, or Karmic), what's the best way to get it? download .dsc and rebuild it and hope for the best? :) maybe upload it to a private PPA ? === ApOgEE__ is now known as ghost === ghost is now known as Guest44168 === Guest44168 is now known as ApOgEE [11:27] lfaraone: hi, did you see bug 456155? [11:27] Launchpad bug 456155 in sugar-hulahop "remove from archive. requires not existing pyxpcom" [High,Triaged] https://launchpad.net/bugs/456155 === simon-o is now known as simon-o|lunch === simon-o|lunch is now known as simon-o === porthose|afk is now known as porthose [12:41] I need some help building a package. It's about https://launchpad.net/~cemc/+archive/ppa/+build/1300735. only i386 isn't building. amd64 and lpia built ok. I'm not sure what's wrong [12:45] maybe you should use 2.2.1 version from unstable? [12:57] guys can anyone help me build the 64bit version .. Im having difficulties understand as im starting off... would really like some help. [13:19] wrapster: The 64bit version of what? [13:36] Hi, Is submitting a "update" patch for a package different than a security one? [13:38] I have problem with python and yesterday it makes the system crached and not work till I reinstall python and gnome .. I wnat to know how can I find the source of this problem [13:40] soren: libnss [13:43] soren: http://pastie.org/661922 thats the enitre rules file that im trying to manipulate for 64bit.. I have no idea how to proceed.. you can see that i've written for 64bit but could you help me understand the actual procedure.. pls [13:57] Could anyone provide me some info about a "bug fix" update (Not a security one) for a universe package? [13:58] qnix: what do you need exactly? [14:00] dholbach: I know how to do a security patch, follow that page: https://wiki.ubuntu.com/SecurityTeam/UpdatePreparation [14:00] But is there something different for bug fix update ? [14:00] for a stable release or for the current one? [14:00] soren: could you help me? [14:01] for karmic, yes. [14:01] https://wiki.ubuntu.com/SponsorshipProcess maybe? [14:01] wrapster: I'm not sure wha tyou're trying to achieve. libnss already builds on 64 bit, doesn't it? [14:02] dholbach: Ok.. so I cannot send a simple patch on the BTS, as I did for a security fix ? [14:03] soren: yeah but thats the 32bit versions..i mean the libs.. that wont generate the 64bit libs [14:03] wrapster: Sorry, what's the name of the source package? [14:04] qnix: you can - that's what the document basically says :) [14:04] qnix: attach patch to bug, subscirbe ubuntu-{main,universe}-sponsors (depending on where it is) [14:04] soren: libnss3-1d [14:04] kk, I'll read that later. :) thanks [14:05] cool [14:06] wrapster: So you're saying that the amd64 packages we have for that package are 32 bit? [14:06] /usr/lib/libnss3.so: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked, stripped [14:07] * soren runs for 10 minutes [14:20] * soren returns [14:25] is anyone familiar with sdl-alsa vs sdl-pulseaudio ? some games have issues have issues with libsdl-alsa and work fine with -pulseaudio, would it be acceptable to set a depends on -pa ? [14:26] Someone who was working on eclipse ought to read ubuntu-devel-discuss. [14:27] joaopinto: Do recall that not all *buntu flavors use pulseaudio. What about recommends sdl-pulseaudio | sdl-alsa ? [14:29] that's a good option, but I am not sure it would be effective because most people will have libsdl-alsa pulled frm other package [14:29] well, maybe this shouldn't be fixed at the packaging level [14:44] The thing is, your alsa problems are quite likely hardware specific. [14:45] Just because pulse works better for you, doesn't mean it will for everyone [14:46] dtchen, ping. backing out the changes from alsa-utils ubuntu4 & ubuntu5 fix that bug I indicated, what's that mean though for the bugs that were "fixed" in ubuntu4 and ubuntu5? [14:53] admins and sponsors, please review bug-fix package bug #455744 [14:53] Launchpad bug 455744 in tinyproxy "[FFe] New upstream bug-fix release 1.6.5" [Undecided,New] https://launchpad.net/bugs/455744 [14:57] ScottK, don't think so, other people was able to reproduce this, it seems to be PA related [14:57] OK [14:59] bug 454879 [15:00] Launchpad bug 454879 in libsdl1.2 "hedgewars process uses 100% cpu with libsdl1.2debian-alsa installed" [Undecided,New] https://launchpad.net/bugs/454879 === rmcbride_ is now known as rmcbride [15:20] Heya gang === DktrKranz2 is now known as DktrKranz [15:20] Hi bddebian. [15:20] Hi iulian [15:20] huhu bddebian [15:22] Heya sebner === RainCT_ is now known as RainCT [15:41] hey MOTUs just to thank you for your great job this 5 Years a big Thank you !!! [15:48] james_w: probably not, let me take a look at it, [15:48] bug 456155 [15:48] Launchpad bug 456155 in sugar-hulahop "remove from archive. requires not existing pyxpcom" [High,Triaged] https://launchpad.net/bugs/456155 [15:49] james_w: can't we just import python-xpcom? [15:49] *sync [15:50] dunno [15:50] I'm not the one you want to be talking to about it, just wanted to make sure you had see it === jono_ is now known as jono === bdmurray_ is now known as bdmurray === jono_ is now known as jono === jono is now known as Guest86059 === Guest86059 is now known as jono [17:21] I have stumbled upon a case where network-manager fails due to ridiculously short (~300) lease times given by DHCP. What does work is using ifconfig to re-specify the received IP as static. Would this be "fixable" as a configuration in NM? [17:22] And no, DHCP is not accessible for configuration... [17:22] arand: yeah, that's pretty inconvenient. Have you tried asking NM upstream? [17:23] No not yet, so that'd be the correct channel? [17:24] lfaraone: It's explained in the bug [17:29] arand: IMO, yes === rmcbride_ is now known as rmcbride [18:07] ogra: hey [18:15] ScottK: around? [18:15] sebner: A bit. [18:15] ScottK: when is universe freeze? [18:16] sebner: Sunday. See u-d-a. [18:17] ScottK: do you see a chance for nexuiz (+ data, + update for a b-p). Syncable in 2-3 (maybe)? [18:17] sebner: Did a nexuiz update just get approved? [18:17] Maybe it was something else. [18:18] sebner: In theory it's possible as long as we have a good reason to want it. [18:18] ScottK: I wondering if "our version is thaaaaaaat old" is valid [18:19] improvements over improvements included [18:19] sebner: Depends. [18:19] I'd put together the FFe and see. [18:20] ScottK: ok thx === highvolt1ge is now known as highvoltage [18:38] soren: sorry for a very late reply.... http://pastie.org/662322 [18:38] that is what i was talking about when i said i needed to compile for 64bit.. if you see there are no /amd64/*.so files... [18:39] soren: and what im trying to do is exactly that.. so now could you help me [18:45] wrapster: so you are trying to cross-build an amd64 package on i386? why not build the package on amd64? [18:46] I am building it on amd64 [18:46] isainfo ---> amd64 i386 [18:47] can you pastebin the output of "dpkg-architecture"? [18:50] geser: http://pastie.org/662349 [18:53] even if your rules file works and it compiles for amd64 your will get a ..._solaris-i386.deb. I don't know anything about nexenta but it still looks wrong to me to have an amd64 lib inside a i386 package [18:56] geser: yeah i understand your point.. but since it is a packaging question i thought i'll ask here.. could you please help me.. lets forget the issue of it being anything else(nexenta or not).. as a general packaging question if i were to put this across.. could you please help me,,, [18:57] geser: if you can at least help me to some extent i'd take it from there... [18:57] geser: here is the rules file i manipulated... http://pastie.org/661922 [18:58] have a look at the build_64 section.. [18:58] do you try to build an amd64 package on i386 or an i386+amd64 package? [18:58] i386+amd64 pkg [18:59] i would like to know how i should start with the install_64 section(which is currently left blank) [18:59] this is again my first time trying my hands at 64bit versions... so needed help... [19:01] good luck with that. the current Debian (and Ubuntu) packages are made to install always into the same location (i.e. the i386 package installs their libs to /usr/lib/, the amd64 package also installs their libs to /usr/lib/) [19:02] so you would need to modify every package to build both for i386 and amd64 and also modify it to install the files into your wanted locations [19:03] you would also need to solve somehow the problem, that the i386 and amd64 build find their libs to link [19:04] geser: could you be a bit more descriptive about your last 2 points. pls. [19:04] there are some packages which build on amd64 also i386 code, so you might want to look how they do it [19:04] geser: ok [19:05] geser: if you could explain what needs to be done looking at this http://pastie.org/661922 [19:05] i would be able to understand it really fast as i have something to relate to... could you please do it. [19:06] wrapster: I've no idea how to do it and I haven't yet looked at packages on amd64 building also for i386 [19:07] geser: ok thanks for your help thus far... [19:07] can anyone else following our coversation please help me here... [19:08] wrapster: the problem you first need to solve: when the build target is called you want(?) to build first for i386 and then later again for amd64. and you need to make sure that the build for amd64 stores the build files in some other location as else you will lose your i386 build [19:09] it probably depends on the upstream build system if you can specify it or not [19:10] siretart: hey do you know Fabian Greffrath's irc nick? [19:10] with this package your are lucky that it doesn't use configure else you would need to figure out how to seperate the configuration for i386 and amd64 [19:11] geser: thanks a lot.. [19:11] it really gave me an idea to work with... [19:12] But have a look at the rules file... should i be aiming at producing all those *.so files as 64bit versions... [19:12] thats my only question. [19:12] wrapster: take a look at http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/karmic/zlib/karmic/annotate/head%3A/debian/rules [19:13] that the rules for zlib, it builds on amd64 a package for amd64 (libz1) and a package with i386 code (lib32z1) [19:15] geser: yeah i understood pretty well.. so i should be looking at something like mkdir -p debian//usr/lib/amd64 then putting all my libs there and for i386 ones the usual.. that part i got it.. how about the install part.. [19:15] yeah i will look at this and do the best i can.. [19:15] May i revert if i have issues? [19:16] sure [19:18] geser: thanks a lot [19:18] wrapster: something similar should work for the install target, but first you need to have the files build both for i386 and amd64. the target calls will be more like: build -> build32+build64 -> install ->install32+install64 and not build32 -> install32 -> build64 -> install64 [19:19] hmm ok [19:20] so you need to take care that the amd64 build doesn't overwrite files from the ia32 build, e.g. by creating a copy of the source files, so you can safely build for i386 and amd64 [19:24] geser: ok thanks [19:25] btw does this channel have logs by anychance? [19:25] irclogs.ubuntu.com [19:25] wrapster: yes, irclogs.ubuntu.com [19:25] i can use everything youve said just now as reference [19:26] wrapster: do you have strong reasons to build an i386+amd64 package instead of a seperate one for i386 and one for amd64? that would save you much work [19:27] geser: there is no strong reason. [19:27] geser: and how would i build the dedicated amd64 pkg if otherswise? [19:27] i should start with the sources only right!? [19:33] can the solaris kernel switch personalities? once returning it's an i386 (e.g. uname -m) and once being amd64? [19:35] not very sure of it.. but i think it can. [19:35] you would need to "re-do" the boostraping similar to solaris-i386 for solaris-amd64 and to modify dpkg to know also about solaris-amd64 [19:35] wow... [19:35] not an easy task for a newbie... [19:36] this bootstraping you would need to do anyway as you can't link amd64 code to an i386 libc [19:36] you would need to also have an amd64 libc [19:36] and every other lib you would need for your libnss-1d build [19:41] geser: yeah looking at it. === asac_ is now known as asac [20:04] mhh, weird [20:04] did someone disable the avc codec of vlc? [20:05] avc1 and xvid don't work anymore ... [20:29] emm.. if a package have 2 xml file, 1 .xsd + 1 .xsl. Where am I supposed to put those files ? /usr/share/xml ? [20:29] (there is nothing else related to the xml with that package) [20:30] where does the app expect them? [20:30] that's the problem... the app don't need them at all, and even don't know they exist. [20:31] It's simply a schema and a xslt that can be used or not by the user. [20:31] hum [20:31] if you don't need them, don't install them :) [20:32] but the user may need them. It would be "bad" to tell them.. "go on svn and download them manually", ^^ [20:33] qnix, maybe, install them as examples [20:33] Hey dholbach :-) [20:34] hi fabrice_sp [20:36] in fact, there are 2 xml files, + 3 test/example files. I think I'll put that in app-doc package. [20:36] somewhere in /usr/share/... === menesis1 is now known as menesis [21:36] pochu: he doesn't irc. try mailing him or the mailing list. he is usually pretty responsive per mail [21:36] soc: did you by chance add some PPA to that updates ffmpeg? [21:40] ooops [21:40] yes, no it works! [21:40] sorry [21:59] are new additions universe/multiverse frozen for karmic? [21:59] LLStarks: https://wiki.ubuntu.com/FreezeExceptionProcess [22:00] can you give me the short version? [22:00] LLStarks: we're very very frozen already [22:00] okay. [22:00] when does it open for lynx? [22:01] you need very good reasons to get an exception [22:01] https://wiki.ubuntu.com/ReleaseSchedule [22:01] then i'm not going to bother. [22:03] i was just going to suggest a new upstream tree for onscripter. i'd have to go through the debian system first, wouldn't i? [22:04] LLStarks: getting it updated in Debian first would ease things [22:05] debian roolz ko [22:06] if it's impossible now, so we can get packages by SRU later, right? [22:08] SRU are only for important bug fixes, not new upstream versions or similar [22:09] but what about if new upstream version is bug-fix only? [22:10] I'm trying to get a aptitude bug fixed in Karmic [22:10] hope I make it :( [22:15] ari-tczew: only as backports [22:15] OK thnx === dholbach_ is now known as dholbach [23:36] So I want to take the adobe 64 bit alpha binary and package it into my PPA. What is the best way to go about this? I have never packaged before. Is there a website that explains it... simpley? [23:41] AtomicSpark, https://wiki.ubuntu.com/PackagingGuide/Complete [23:42] Thank you. I'm seeing how the current flashplugin-nonfree installer script works. [23:50] AtomicSpark, you may not be allowed to incude the binary itself, and use the download stragegy like the nonfree package [23:53] joaopinto: Yes. Looking at the source for flashplugin-nonfree now. Looks like it might just be a simple link change, removeal of the ndispluginwrapper stuff and a few text edits. :)