[00:00] dget -x http://com/file.dsc [00:01] crimsun: i'll try it =) thanks for the tip [00:02] cyberix: you could just backport devscripts from edgy. [00:02] Already got it done by looking at the script [00:02] Thanks anyway [00:03] crimsun: Did you see my upload of libsdl? [00:05] StevenK: noted, yes. [00:05] I closed the associated bug yesterday. [00:06] Great. [00:06] thanks, BTW! [00:08] No problem, any time. :-) [00:09] ScottK: not sure if the key still needed syncing or not , so i'm running it now to be sure [00:09] ( just fyi ) [00:10] moins crimsun StevenK and all [00:10] 'lo imbrandon [00:11] TheMuso: i got an "broken" xbox -given- to me today , that i fixed by sodering the points and reflashing the TSOP , so soonish ( when i get time after work durring the week ) we'll have a "dev" xbox [00:11] :) [00:13] And for what? :-) [00:13] bddebian: the port of ubuntu to the xbox we;re working on :) [00:13] Hey imbrandon [00:13] port , not deriv [00:13] :) [00:14] imbrandon: Again, for what? :-) [00:14] bddebian: a $69 USD 733mhx workstation :) [00:14] or similar [00:14] oh, only to take over bddebian's house. Nothing more. [00:16] bddebian: we figured if the ps3 can get a cd rolled ( unsupported of course ) we can roll one for the xbox with only a few ( about 15 ) patches to the kernel and X ( both of witch will eventualy be upstream upstream ) [00:16] which* [00:16] :-) [00:16] crimsun: No xboxen in this house :-) [00:17] so really for minial work we can support the xbox, and the only other distro that does this is xebian ( debian deriv ) and its 2.4 kernel based [00:17] bddebian: no no, it's like X10, but you won't even see/hear it coming [00:18] crimsun: Ah :) [00:18] support == installing and questions awnserd on the forums etc etc etc when possible, you know, not canonical ( other than possible hosting the cd's on cdimage.u.c ) [00:18] hi! [00:18] have fun guys... cu [00:19] sides , theres something satisfying about installing linux/ubuntu on a MS product :) [00:19] Is this the place where to ask to re-sync the REVU uploaders keyring? :) [00:19] psicus78: i have one in progress, one moment [00:20] ahh infact it JUST finished [00:21] it's ok [00:44] . [00:48] (yes?) [00:49] imbrandon: Cool. Did you by chance see my message about apt-mirror? === tritium_ is now known as tritium [00:56] anybody around who is familiar with SbuildLVMHowto? [00:57] Sort of. [00:57] As in: I'm sort of around, and I'm sort of familiar with that howto :) [00:59] LaserJock: ^^ [01:00] do you need to have already set up LVM before running the script? [01:00] LaserJock: It's not going to create the LVM partitions for you. [01:00] That would be more than mildly dangerous. [01:01] that's what I was thinking [01:01] but there wasn't any instructions or links [01:01] So yes, you need to have a VG set up. [01:01] but down on the bottom under "Previous Instructions" there was some [01:02] 'It assumes a passing understanding of LVM, and having an available VG to work with.' [01:02] oh ... [01:02] darn ;-) [01:03] TheMuso: nope , i dont think so [01:03] was it on irc ? [01:03] and do you really need 5GB per chroot?! [01:03] LaserJock: Not really. [01:04] I use 3 normally. [01:04] OOo will want more, but you'd be silly to try to touch OOo. [01:05] # Allocate the "golden" chroot LV [01:05] sudo lvcreate -n "$CHROOT_LV" -L 5G "$VG" [01:05] I think if I run the script it's gonna give me 5GB [01:05] Modify the script, then :) [01:06] Note that there's no point having an initial chroot that's smaller than the snapshots. [01:07] imbrandon: Yes. [01:07] imbrandon: One sec, I'll see if I can dig it up. [01:07] TheMuso: then no , sorry i dident get it [01:07] ok [01:07] hmm, I just don't know how much space this is all gonna take [01:07] I have a 12GB partition [01:08] how much do I have to account for? [01:08] Oh dear. [01:08] imbrandon: I seem to be having a problem with apt-mirror. The error I get is: http://www.pastebin.ca/841988 My mirror.list file is: http://www.pastebin.ca/841990 [01:08] Where did you get a 12GB disk? [01:08] * imbrandon looks [01:08] Fujitsu: no, 12GB partition for sbuild/LVM [01:08] Ah, so your system isn't on LVM? [01:08] heck no [01:09] I dislike LVM [01:09] Why not? [01:09] Why? [01:09] it eats things [01:09] Never done so for me. [01:09] or rather my ability to control it causes me great pain ;-) [01:09] *inability [01:09] I just kept losing partitions and things [01:09] Heh. [01:10] I fell much better in general with have a nice big ext3 / [01:10] * TheMuso doesn't think he'll touch luks encryption for a while at least, after having his laptop battery die, and the whole root filesystem going caboom. [01:10] TheMuso: it seems to ( rightfully , i cant seem to get it manualy either ) getting a 404 on http://mirror.internode.on.net/pub2/ubuntu/ubuntu//dists/gutsy/multiverse/source/Sources.gz [01:11] i am getting a 404 error too [01:11] TheMuso: I have been running LVM on LUKS on my laptop for several months now. [01:11] imbrandon: change pub2 to pub [01:11] (and yes, the battery has died a couple of times, without ill effect) [01:11] pub2 is for internode customers only [01:11] so /pub/ubuntu/ubuntu works [01:11] s/works/should work/ [01:11] right that works [01:12] imbrandon: Note that this is apt-mirror from svn [01:12] Well I tried my config file with s/pub2/pub/ and got the same result. [01:13] Fujitsu: my pbuilders (5 different ones) only take up 400MB, are you telling me that'd take up 25GB with sbuild/LVM? [01:14] Fujitsu: Hrm interesting. [01:14] I'm sure there must be a way to avoid that, but I haven't found it yet. [01:14] Funny how one experience like mine burns you. [01:14] TheMuso: I had to hard-reset it a couple of times a day for weeks too, while X was being dodgy in early Hardy. [01:14] Fujitsu: Right. [01:15] Good afternoon Hobbsee. [01:15] right, svn should be stableized ( just not tarred, 1.4.6 is whats current svn ) [01:15] hum, try to change your mirror.list to s/pub2/pub/ and see if it works [01:15] ( not a full run, once it gets past the preceed indexes is fine to tell if its gonna work ) [01:15] something strange might be going on with http headers or something, but it just uses normal wget [01:15] do you use a http proxy or anything ? [01:15] O_o [01:15] imbrandon can type quickly. [01:15] also it would be nice for internode to redirect non internode customers to pub/ :) [01:15] * Hobbsee waves [01:16] imbrandon: not afaik. If so, its transparent, but I doubt it since I'm within internode's network. [01:16] heya Hobbsee [01:16] hrm [01:16] LaserJock: Ideally, one should be able to create a snapshot of an LVM volume, and grow the snapshot of the volume. But I don't think LVM can do that at the moment. [01:16] Fujitsu: ?? [01:16] heh [01:16] imbrandon: I'm running it again now. [01:17] TheMuso: ok [01:17] imbrandon: 5 of your messages appeared at the same time. [01:17] if it fails again, try to just normaly run wget [01:17] * Fujitsu heads to lunch. [01:17] Fujitsu: ahh , i must have lagged [01:17] Fujitsu: so what's the advantage of sbuild/lvm again .... :-) [01:18] TheMuso: if it fails again , try to just run wget on the file it fails on , to soo if its a possible problem with wget and that mirror [01:18] s/soo/see [01:18] imbrandon: Yeah will do. [01:18] i am getting this error http://paste.ubuntu-nl.org/51031/ [01:19] on your box, or a chroot, or pbuilder [01:20] pbuilder [01:20] sorry [01:20] run "sudo pbuilder update" ( where pbuilder == how you normaly call pbuilder , e.g. pbuilder-dist hardy or pbuilder-hardy or pbuilder [01:20] ) [01:21] alrighty [01:21] :) [01:22] :) === Nightrose_ is now known as Nightrose [01:23] imbrandon: that did the trick :) [01:23] how often should pbuilder be updated? [01:24] I do it once a day that I'm working [01:24] Yeah once a day is good. I update my sbuild chroots once a day. [01:24] Vorian: i do it once a day or so, but it really depends on how new the packages in the archive you need are, its the equiv of running "apt-get update" localy [01:24] that makes sense [01:34] c [01:34] ugh [01:34] * TheMuso thinsk this keyboard needs replacing [01:35] well, I guess since I wiped out my 12GB partition I should at least give this sbuild thing a go [01:35] What does "Depends: ${shlibs:Depends}" actually do? [01:36] cyberix, replaces Depends: with all of the depends it detects during dh_makeshlibs [01:36] replaces ${shlibs:Depends} specifically [01:37] there's also ${misc:Depends} which you should add, but it's not yet used [01:39] imbrandon: One problem with apt-mirror is that it fails if a file can't be retrieved, however the way my mirror.list file is set up, its asking apt-mirror to check for paths that may not be there now, but could be there when packages get updated in the future. Any ideas? [01:39] SO for example, gutsy-updates/multiverse/debian-installer may not exist now, but if a package that places content in that dir gets updated in gutsy-updates, that path will exist. [01:39] nenolod: it is used [01:40] nenolod: misc:Depends is used by debhelper to insert and additional dependencies that come from the dh_* [01:40] So far, I'm thinking of going back to debmirror. [01:42] LaserJock: So which one should I prefer? [01:42] cyberix: prefer? [01:43] ${misc:Depends} is ok for dh_makeshlibs too as it falls into dh_* [01:43] So should I always use ${misc:Depends} then? [01:43] no, no [01:43] instead of ${shlibs:Depends}. [01:43] they are different things [01:43] ok [01:44] let me try to find an example [01:45] LaserJock, it is now? well, that's good. [01:45] nenolod: has been as long as I've been around I think [01:46] LaserJock, dpkg-buildpackage was warning about it being an unrecognised subst a few months ago [01:46] yes, that is correct [01:47] which lead me to believe it wasn't used yet ;) [01:47] it's used for debhelper, not dpkg [01:48] I think dh_gconf may be an example for misc:Depends [01:48] Oh. It is used only by some dh_* [01:48] basically, some dh_* scripts need additional dependencies [01:49] I think there are some more [01:49] ${shlibs:Depends} is for shared library dependancies, ${misc:Depends} is used for example, by debconf [01:49] dh_installdebconf will put debconf into ${misc:Depends} [01:49] dh_installdefoma same sort of thing [01:51] is there an easy way in bash to make all files in a dir have a lowercase ext [01:51] like mass rename .TXT to .txt [01:51] s/rename/mv [01:51] imbrandon: Install renameutils [01:51] k [01:52] imbrandon: Run 'qmv -f do' which drops you into vi with a filename on each line [01:52] Other than that: [01:52] killer, thanks [01:53] for i in * ; do j=$(echo $i | tr 'A-Z' 'a-z') ; mv "$i" "$j" ; done [01:54] /win 48 [01:54] Bah. [01:54] cool, thanks StevenK [01:55] btw it must use the default editor , because it was s/vi{,m}/nano [01:55] :) [01:55] It probably calls sensible-editor [01:55] Actually, nano shouldn't be in that list. :-P [01:55] * Fujitsu agrees with StevenK [01:56] hahah [01:56] * StevenK high fives Fujitsu [01:56] that for loop will do the whole filename wont it ? [01:56] It ought to [01:57] How is nano not sensible? [01:57] If it doesn't, use qmv [01:57] minghua: In many ways [01:58] no delete line [01:58] ^K [01:58] StevenK: Okay, I'm not arguing. But the best hacker I know in real life uses nano. [02:03] I'm looking after first advocate for my package malbolge. I've fixed all problems that have been brought up. http://revu.tauware.de/details.py?package=malbolge [02:25] minghua: it's just seriously underpowered/underfeatured [02:27] is this the development team channel? [02:30] for Universe [02:30] hmm, sbuild wants to install exim4 :/ [02:32] LaserJock: You don't already have a mta? [02:40] RAOF: no [02:40] I don't like having mta's if I don't need them [02:40] Well, sbuild will want to email you the buildlogs. [02:41] but I don't want it to :-) [02:41] you don't need an mta [02:41] you can just have local delivery [02:41] seems odd that it'd be a dep and not a Recommends === cassidy_ is now known as cassidy [03:00] hmm, is there an easy way to see what provides m-t-a? [03:06] LaserJock: aptitude search "~Pm-t-a" [03:07] I just found synaptic has it :-) [03:07] * RAOF wanders through the monodevelop 0.18 dependency stack [03:11] hola everyone [03:11] I'm alive... I think :) [03:12] upgrading my macbook to Hardy at the moment for kicks [03:12] jdong: Heh have fun watching everything break. [03:12] oh my ;-) [03:12] :) [03:12] TheMuso: ooh what's scheduled to break? [03:12] other than the stuff I have scheduled ready to break :D [03:12] jdong: I don't know. [03:13] * jdong tries to play a quicktime mp4 [03:13] look at that, it still works [03:13] well... can't have that happening! [03:13] jdong: Thanks for reminding me. Wavpack playback has broken; I need to file that bug :) [03:13] lol [03:13] are we under any freezes yet or soon? [03:14] Alpha 3 freeze is either now or soon. [03:15] * jdong takes a look at that libmp4v2 stuff that he left off on... [03:26] grumble I hate circular dependencies [03:26] aye... [03:26] fuuuuuuuuuuck [03:27] would anyone cry if at this point I upload a new faac that may break AAC encoding in mencoder, avidemux, and/or gst-plugins-bad-multiverse till I upload rebuilds of those? :) [03:27] * bddebian wouldn't [03:27] nobody likes aac anyway right ;-) [03:28] I have been working on building newpki-client with wx2.6 for over a week now and there's all kinds of shit in my diff... :'-( [03:28] If you can get that all done by the alpha 3 freeze tomorrow... :) === bigon is now known as bigon` [03:29] lol well I'll wait for alpha3 to finish then, meanwhile I'll just re-test this build order locally... again. [03:30] I'm not sure I trust these tomboy notes dated on the 17th of december [03:33] I though universe/multiverse wouldn't be affected by alpha freezes? [03:33] thought* [03:34] Oh, yeah. We had this discussion before, didn't we. [03:35] * TheMuso won't be using hardy full-time till around the last alpha/beta. [03:35] Universe/multiverse will be free rain as usual. [03:37] Does anybody know if bazaar has already been proposed for backport? [03:39] persia: You about? [03:40] somerville32: Look for a bug in gutsy-backports [03:41] Anyone have a clue how I get rid of cruft from a diff that I don't even know how it got there? In almost every file I touched I'm getting the complete copyright statement re-generated [03:46] bddebian: You look for the script that is regenerating it. [03:47] Hmm, I think it's because the files were in MS-DOS format so had extra CR on them :_( [03:51] why do I get this error from soyuz ? i don't quite understand it :dpkg-genchanges: failure: cannot read files list file: No such file or directory [03:51] GRR fscking.... [03:52] bddebian, You can user filterdiff [03:53] yay lovely, that's why... libmp4v2 is removed from the archive [03:53] bddebian, or if the cruft is isolated, you can just remove it from the patch manually [03:53] Could i get a second advocation for http://revu.tauware.de/details.py?package=hardware-connected (a program that checks whether given hardware is connected to the system, nice for scripting)? Thanks [03:55] LucidFox: hey sup? === bmk789 is now known as bmk789_sleep [04:00] oooooh shiny new x264! [04:00] must upload crack! === cassidy_ is now known as cassidy [04:01] somerville32: Unfortunately I don't know what showed up :-( [04:01] Isn't there some utility to strip carraige returns from files? [04:01] bddebian, Can you pastebin your diff? [04:02] It would be HUGE :-) [04:05] Help to increase the amount of public domain in Ubuntu. Review my package malbolge, now! Right here on R-E-V-U. http://revu.tauware.de/details.py?package=malbolge [04:06] Enough syrup in my speech? [04:07] Public domain isn't always good. [04:07] Aha, dos2unix saves the day!! [04:07] jdong> I was wondering about mpeg4ip [04:07] LucidFox: uploaded 7 seconds ago [04:08] Ooh. [04:08] Awesome. [04:08] :) [04:09] now I'm wrangling faac then going to queue rebuilds of the libmp4v2-0 dep chain [04:10] Fujitsu: The software or the "license" (or lack of it) [04:14] cyberix> I think it's better to add a disclaimer for legislatures with no concept of public domain [04:14] like Wikipedia does [04:15] " [04:15] I, the copyright holder of this work, hereby release it into the public domain. This applies worldwide. [04:15] In case this is not legally possible, [04:15] I grant any entity the right to use this work for any purpose, without any conditions, unless such conditions are required by law. [04:15] " [04:16] But I'm not the copyright holder of all the stuff in the package. [04:16] Ah. Sorry then. [04:20] And using multiple licenses would be far too complicated. [04:21] Of course I might have the right to "relicense" it because it is public domain. [04:21] But then again I'd ask a US citizen to do that. [04:21] And then again the spirit is what matter most. [04:22] cyberix> You don't need to. [04:22] I'm way too technical to work with these concepts anyway. [04:22] And now we'll have to wait until mpeg4ip passes NEW... *sigh* [04:24] LucidFox: yep :) [04:24] LucidFox: then I got another mpeg4ip I need to upload after the first one and faac build. [04:25] LucidFox: had a circular uninstallable chain, the mpeg4ip I uploaded is crippled faac-less, which frankly I don't think all but 3 people on the planet care. [04:25] lol but oh well I keep everyone happy ;-) [04:25] * cyberix found speed-game (currently in revu) to be lots of fun. [04:27] On the plus side, it will finally be possible to get rid of gtkpod-aac patches. [04:28] LucidFox: yep, that's what I'm currently removing :) [04:29] In 0.99.12, I hope? ;) [04:29] LucidFox: did debian release the new one? [04:29] ep I see it [04:30] grumble well merge time :) [04:30] LucidFox: you wanna do it? :D [04:30] I feel like catching up on some sleep tonight ;-) [04:31] Well, I have already uploaded an interdiff for 0.99.12, so I might as well remove the patches === brandonperry_ is now known as bperry [04:33] LucidFox: ah, awesome, looks like you've got that under control then [04:33] jdong! [04:33] looks like just exfalso needs rebuilding then [04:33] Vorian! [04:33] welcome back [04:34] thanks, doing laundry tonight and got bored [04:34] otherwise I was ENJOYING my VACATION.... [04:34] * Fujitsu sends jdong away again. [04:34] heh [04:35] ok, looks like I've hit a dead end for the night, time to wait for FTBFS-mail :D [04:35] oh wait libx264 is asking to be broken [04:54] ok, x264 is all happy now [04:54] snoozing time [04:56] Night jdong. [04:57] night Fujitsu [05:12] I don't suppose I could interest anybody in some grub recovery help? [05:12] I'm trying to reinstall grub by chrooting off the LiveCD [05:12] but grub-install is not working [05:12] What is the problem? [05:12] !doesn't work [05:12] Doesn't work is a strong statement. Does it sit on the couch all day? Does it want more money? Is it on IRC all the time? Please be specific! Examples of what doesn't work tend to help too. [05:13] I love that factoid. [05:13] * StevenK too [05:13] root@ubuntu:/# grub-install /dev/sda [05:13] Could not find device for /boot: Not found or not a block device. [05:13] hehe, that's funny [05:13] Have you bind-mounted /proc, /dev inside the chroot? [05:13] Fujitsu: yep [05:14] I can go to /boot and I see it in /dev [05:14] What about /sys? [05:14] I didn't bindmount /sys [05:14] I haven't historically needed /sys, I don't think, but you never know. [05:14] I've never had to do that before [05:14] I'm not quite sure how grub-install enumerates devices [05:14] Neither. [05:15] no go [05:15] can I tell it manually? [05:15] strace it? [05:15] Hmmm. [05:15] You might need to ask it to regenerate its maps. [05:16] Have you moved drives/partitions around? [05:16] just tried that [05:16] Fujitsu: nope [05:16] Remove /boot/grub/device.map ? [05:16] I think that's what it's called. [05:17] nada [05:17] grr, my wife needs some files :/ [05:18] stupid sbuild [05:18] sbuild doesn't eat boot records. [05:18] Neither does mk-sbuild-lv. [05:18] Carbon unit problem? === bigon` is now known as bigon [05:23] Fujitsu: no, but the partition I wiped to do it had my previous grub install [05:23] ok, so in grub running find /boot/grub/stage1 says it can't find the file [05:26] man, I'd really hate to have to re-install this laptop :( [05:28] Er, why would you consider reinstalling? [05:28] Does /boot contain the appropriate GRUB files? [05:28] yep [05:29] Fujitsu: well, I can't boot [05:29] I'm getting off what I need for now via LiveCD [05:29] What was on the partition? Just the boot sector? [05:30] I have a separate partition for /boot [05:31] You copied all the stuff from there to /boot on your root partition before obliterating it? [05:31] no, I didn't obliterate /boot [05:32] I had an openSUSE install on another partition [05:32] and of course it installed it's grub [05:32] so today I wiped that out [05:32] I think I may have just gotten it [05:32] I started up grub itself [05:32] and ran: [05:32] root (hd0,2) [05:32] setup (hd0) [05:33] and it said it worked [05:33] Aha, good. [05:33] ok, rebooting [05:33] * Fujitsu shakes his head in dismay. [05:34] One probably shouldn't consider reinstalling because the bootloader is broken. [05:34] man-di: hello, how's the icedtea package ? [05:44] man-di: I got bad news, I just installed icedtea-java7-jre on my system, when I attemped to run JOSM, it crashed [05:47] Hello, I was building one of my packages under hardy amd64 using pbuilder, and I got this: http://pastebin.com/m461bd24 [05:51] hey persia [05:52] you mentioned on my package that "15) uscan --report-status reports no matching files, which is often an indication that the watch file doesn’t work properly." i just ran it again, it returns up-to date for me... can you try it again? [05:56] where can i find the information on the policy change in debian/menu in hardy? [05:59] jscinoz> sudo apt-get install menu [05:59] :) [05:59] the Hardy version, of course [05:59] how would i get the hardy version rather than the gutsy version (im assuming it'll ge the one for the dist im currently on) [06:00] Well, you can download the source package [06:00] https://launchpad.net/ubuntu/+source/menu/2.1.36 [06:02] thanks [06:03] Hello, I was building one of my packages under hardy amd64 using pbuilder, and I got this: http://pastebin.com/m461bd24 , can anyone help on this ? I'm pretty sure that there's nothing wrong with the package because Debian has an amd64 binary for it [06:03] sorry about all the Soyuz questions,but is there a way to find the build log of your project if it never shows up (it publishes, but never comes up with a deb package ) [06:04] actually Ubuntu too has and amd64 binary for it [06:04] http://packages.ubuntu.com/hardy/misc/acon [06:05] mattva01: Ho long ago was it uploaded? [06:05] *How [06:05] ant1: https://edge.launchpad.net/ubuntu/+source/acon/1.0.5-4 [06:05] there we go got what i needed from that, games/arcade is games/action in hardy. [06:06] * Published on 2007-10-24 [06:06] about 45 minutes ago , but it shows it as published , but with no build failed or success log [06:06] mattva01: Wait a while. The builds probably won't have been created yet. [06:07] StevenK: so why I can't build it ? [06:08] mattva01: You should be watching https://launchpad.net/people/+me/+archive/+builds?build_state=all [06:08] ant1: You probably should find the package that contains the missing file and investigate why it hasn’t been installed. [06:12] hey guys, im creating a package of the game urbanterror which contains non-gpl code but allowed to distribute, i need to make it display a prompt to allow the user to review the license and accept or decline it, i've seen this done in the sun-java packages but i cannot figure out how its done from that source package. how would i accomplish this? [06:12] jscinoz: use debconf [06:13] thakns [06:19] ant1, the short description line in debian/templates should be limited to 80 characters correct? [06:19] what does this even mean? dpkg-genchanges: failure: cannot read files list file: No such file or directory [06:24] ugh, i hate it when people abuse nominations [06:24] we're not even half way though the hardy cycle and people are already nonimating bugs to be fixed in hardy [06:24] uh, duh? [06:26] jscinoz: not sure [06:27] jscinoz: dunno about that [06:28] hmm [06:32] man-di: hmmm, it didn't work with sun java either, I think it's a problem under AMD64 arch. [06:38] argh the documentation on debconf is confusion [06:38] confusing* [06:39] you wouldnt think it so hard to simply display some text and a true/false prompt, continue installation if true and exit if false. [06:40] Can someone remember a package that does something different on say amd64? [06:40] * StevenK can't remember the magical dpkg-architecture runes in debian/rules [06:41] jscinoz: http://64.233.183.104/search?q=cache:0uKuMO_uzYIJ:www.fifi.org/doc/debconf-doc/tutorial.html+debconf+tutorial+fifi&hl=en&ct=clnk&cd=1&ie=UTF-8 [06:42] thats the one im looking at [06:42] ive got the templates file [06:42] cant figure out how im supposed to call it [06:42] in debian/config? [06:43] jscinoz: debian/config is for dh_input mostly [06:43] jscinoz: postinst is for dh_get [06:43] db_* sorry [06:43] well it has to be shown and the license accepted *before* the package is installed [06:43] so i assumed config would be a good place [06:43] jscinoz: the preinst [06:43] jscinoz: then preinst maybe ? [06:45] hmm alright.. [06:45] is there any way to test it works without compiling the package and such? [06:46] dunno [06:57] good morning [06:59] dholbach: hello [07:00] hey ant1 [07:00] how are you doing? [07:01] Hey dholbach! [07:01] dholbach: I'm fine thanks, can you review this package: http://revu.tauware.de/details.py?package=ubuntume-themes , one of the themes is based on a theme you worked on (Human theme) [07:01] heya TheMuso [07:01] dholbach: particularly I'd hope you check the copyright file [07:02] ant1: not right now - I'm just catching up with myriads of emails - if you'd drop me an email with the link I'd do it later on and not forget it [07:02] k [07:02] Hi! Is this the place where to ask to re-sync the REVU uploaders keyring? :) [07:03] thanks ant1 [07:06] np === Zic_ is now known as Zic [07:09] psicus78: It is. [07:11] thanks [07:12] You would like it synced, then? [07:13] is this script valid for calling debconf to ask a boolean question and continue package installation if true, and cancel installation if false? http://pastebin.ubuntu-nl.org/51060/ [07:17] alright... [07:17] Fujitsu: it looks like already synced, I just tried to get my password and it worked, thanks! [07:18] jscinoz: no, "exit" doesn't cause the maintainer script to return an error [07:18] if i want a debconf script to present a boolean question, and cancel package installation if the result is False, can i do this straight from debian/rules (in the install-indep section) [07:18] ugh [07:18] this is confusing [07:19] dholbach: can I ask a review for another couple of artwork packages too ? [07:19] ant1: just add the links to that email [07:19] dholbach: ok, thanks [07:22] how can I use dpkg to search for which package provides the file X ? [07:23] dpkg -S [07:24] StevenK: aha, the file posix_types_64.h is provided by linux-headers-2.6.24-2-generic, does linux-headers package get installed in pbuilder's base.tgz ? === asac_ is now known as asac [07:26] hmmm, could that be a bug that posix_types_64.h is provided by linux-headers-2.6.24-2-generic and NOT linux-libc-dev ? [07:27] wierd, it is provided by linux-libc-dev [07:30] ok, I think I found the problem [07:30] posix_types_64.h is in /usr/include/asm [07:31] slangasek, what about "exit 2" [07:35] slangasek, i've updated my script, would this work: http://pastebin.ubuntu-nl.org/51064/ [07:35] can someone confirm this with me, I think there's a problem with asm/posix_types.h that comes with hardy [07:36] in hardy it has #include "posix_types_64.h" [07:36] Why are you including asm/posix_types.h at all? It's a kernel header. [07:36] in gutsy it was #include [07:36] StevenK: I am including linux/kd.h [07:37] StevenK: which in turn includes linux/posix_types.ht [07:37] StevenK: which in turn includes linux/posix_types.h [07:37] StevenK: and the last one includes asm/posix_types.h [07:37] It changed because the kernel in Hardy is 2.6.24, which pulled i386 and x86_64 into one arch. [07:37] StevenK: here's the compile error: http://pastebin.com/m461bd24 [07:38] StevenK: yes, but should the new asm/posix_types.h read something like: [07:38] #include , instead of #include "posix_types_64.h" ? [07:39] Maybe your source should I changed to include -I/usr/include/asm [07:40] StevenK: I wonder how acon got compiled on Ubuntu's hardy ? [07:42] StevenK: isn't this -I/usr/include/asm a bad workaround ? [07:44] I'm not sure. [07:51] persia are you here? [08:01] jscinoz: yes, that looks like it should work as you describe [08:12] Aren't packages that automatically update debian/control forbidden? [08:15] LucidFox: no - the pkg-gnome team for example makes use of debian/control.in a lot [08:15] but they just update the maintainer automatically in there [08:19] congratulations pochu! :-) [08:27] * StevenK twitches, getting used to using quilt [08:32] * Fujitsu sews some patches onto StevenK. [08:33] * StevenK doesn't need patching [08:35] * Fujitsu shrugs, and continues patching. [08:35] moring dear MOTUs [08:35] Hi huats. [08:36] Hi Fujitsu [08:37] slangasek thanks [08:38] where should i call it from? [08:39] jscinoz: typically you would do that in a package preinst, if it's some kind of license that the user has to agree t [08:39] o [08:43] is anyone planning to update classpath packages? Or should I file a sync bug? [08:47] slangasek, are preisnt scripts automatically executed if they have the right name? or do they need to be executed from debian/rules? [08:50] Hi... does anyone know how I can set the default application for a mime type without using defaults.list? [08:52] jscinoz, afaik it just depends on the name of the script [08:52] you don't need to run them from debian/rules [08:53] debian/rules is run at package build time, the preinst & postinst etc at package install time [08:54] what is the preinst called by? [08:54] or is it run automatically if it has the right filename? [08:56] It's run automatically [08:56] by dpkg [08:56] It needs to be executable too, of course [09:04] Is it resonable to have echo "application/tmd.textmaker=tm.desktop" >> /usr/share/applications/defaults.list inside the postinst script? [09:05] won't that lead to a lot of cruft when you upgrade the package? [09:06] maybe you can grep defaults.list first to see if that line is already present [09:06] The trouble is I cant think of any other way to do it... one of the rm scripts has some sed magic to remove it [09:08] It seems odd since everything else I've needed to do is done with .desktop's and an .xml, which seems like a clean way to do it... then this (KDE appears to use another .desktop so avoids the issue, but i'm after Gnome as well) [09:09] Yeah, i see. It might work but it seems like there should be a better way [09:09] Thats what I was thinking (i'm new to this so may be missing something) [09:11] I have a comment on my REVU upload that the .desktop doesn't validate in Hardy, something about "Encoding=UTF-8" being wrong - where can I find out what it should be, and if at all possible, validate it myself? [09:11] I've got /usr/share/applications/tm.desktop (a menu entry), /usr/share/icons/... for the icons, /usr/share/mime/packages/tm.xml to define the mime type, /usr/share/mimelnk/applications/tmd.desktop (for KDE file assosiations)... now i'm stuck :D [09:16] Judging by the gnome documentation it seems like defaults.list must be modified though, so cant see any other way to do than in postinst [09:16] Hi all, could someone point me to a tutorial for making a debdiff? [09:18] gday all [09:21] Anyone know the correct way to set a default application for a mimetype in Gnome? [09:25] good morning [09:44] persia: may I request the merge for sympa? [10:35] warp10: Go ahead. [10:36] jussi01: https://wiki.ubuntu.com/MOTU/Contributing [10:36] chazco: look into dh_installmime [10:36] persia: Ok. I'll take care of sending that patch to Debian too [10:36] * persia thanks those who answered previous questions [10:36] warp10: Thanks. [10:37] warp10: Wait, isn't that a python 2.4 -> python 2.5 change? Debian likely doesn't want it. [10:38] persia: mmm... no. It's about verifying /var/run/sympa on startup. [10:38] warp10: Ah. Then Debian likely does want it. Thanks :) [10:38] :) [10:42] * amarillion is learning dpatch [10:43] Somehow I keep getting unwanted changes in my patches after dpatch-edit-patch [10:43] and then I have to go through a lot of trouble to get it clean again [10:43] amarillion: my personal preference is simple-patchsys [10:43] amarillion: Are you running build inside dpatch-edit-patch? [10:44] persia, yes, how else to test the changes? [10:44] jpatrick: No comments, only works for CDBS, and doesn't allow selective enablement of patches (but I agree). [10:44] amarillion: Exit dpatch-edit-patch. Do your testing. dpatch-edit-patch the patch again, etc. [10:45] ah right === _czessi is now known as Czessi [11:07] if a source release produces 5 shared libraries, which depend on each other quite heavily. Should I separate them into separate packages or just 1 package [11:08] DaveMorris: It's generally advised to separate them early, so you can avoid the pain of separating them later (in case a client only uses some of the features), but there are exceptions in the archive. [11:08] DaveMorris: can each be used without the others? [11:09] ok atm I have libs a, b, c, d, e. [11:09] a is useless on it's own. b depends on a. c/d/e all depend on a,b but are independent of each other [11:09] dholbach: welcome back =) [11:10] hi Kmos :) [11:11] dholbach: Oh lovely. Seems that LP is somewhat late in changing that wordpress bug you just closed. :p [11:11] DaveMorris: In that case, I'd suggest at least splitting into four packages (a,b), c, d, and e. [11:11] and 4 dev packages ? [11:12] DaveMorris: Err. Right. Eight. Sorry. [11:12] TheMuso: was there a LP: #xxxxx in the changelog? [11:13] thanks, whilst I'm here. My small package http://revu.tauware.de/details.py?package=libserial is looking for a revu. [11:13] dholbach: Yes. Soyuz even indicated as such with the launchpad-bugs-fixed field in changes [11:13] TheMuso: weird [11:13] Indeed. [11:25] could someone check if daloradius-0.9.5-1 uploaded successfully to revu? [11:26] ahh nevermind, I see it now [11:26] c [11:26] ugh wrong tab === ChrisGibbs_ is now known as ChrisGibbs [11:47] Hello, hope you don't mind answering a question. https://bugs.launchpad.net/ubuntu/+source/flightgear/+bug/177615 states that a new version of flightgear has been released and that no debian package exists. We are currently at 0.9.10-2ubuntu2 should the new version number be 1.0.0-0ubuntu1 ? [11:47] Launchpad bug 177615 in flightgear "New upstream version: 1.00" [Wishlist,Confirmed] [11:48] I'm new to this packaging stuff, as you might have guessed, but keen to learn. [11:50] dannioni: 1.0.0-0ubuntu1 would indeed be the correct version for a first upload to hardy. However debian has 1.0.0-1, so you might do better to test that, and consider either a sync or a 1.0.0-1ubuntu1 for a merge. [11:51] dannioni: possibly, more likely 1.00 vs 1.0.0 if the bug title is correct [11:51] e.g. 1.00-0ubuntu1 [11:51] imbrandon: It's 1.0.0-1 in Debian, so we really want 1.0.0-$revision [11:51] ahh he said it wasent in debian [11:52] * persia is paranoid, and doesn't trust anyone :) [11:52] rather the bug report in debian said there was no package [11:52] dannioni: did you look? [11:52] no, just at the bug report :P [11:53] ahh, well persia is correct on all accounts here, personaly if its 1.00 upstream then imho debian botched the version number, but its not a big deal nor something that should be fixed at this point [11:53] base the merge or sync on 1.0.0-1 :) [11:55] great, thank for the help [11:59] dannioni: When testing, please be sure to also look at simgear and fgfs-base, as these packages are all fairly tightly integrated. [12:00] Oops. fgfs-atlas too. === Toadstoo1 is now known as Toadstool [12:07] RAOF: hey! about miro ... would like to talk about moving miro to xul 1.9 ... just ping me when avail [12:07] Ah I see. games should be chown root:games with chmod g+s so they can write to /var/games. [12:07] Is there a way to do that in debian/rules, or do I need a postinst script for that? [12:09] amarillion: You need to do it in postinst [12:09] ok [12:10] asac: Hi! [12:11] asac: I've tried the PPA miro, and it seems fine to me. [12:23] hi all [12:23] hi mruiz [12:24] We don't want to preserve PPA entries in the Ubuntu changelogs, do we. [12:25] s/./?/ [12:25] persia: will do [12:25] flightgear depends on them, right? [12:25] RAOF: not unless there's a really good reason to, no [12:25] Hobbsee: Thanks, just checking. [12:26] Removing information from changelogs seems excessive [12:26] Is anyone looking into the new monodevelop packages in Debian? [12:26] persia! [12:26] perhaps it can be "zipped up", like a new upstream version pull-in from DEbian === jscinoz__ is now known as jscinoz [12:27] hey guys, ive got a script in debian/preinst am i correct to assume this is automatically called pre-installation? [12:27] sladen: We drop ~ppa revisions because as Ubuntu, we consider that all part of the preparation process. Similarly, Debian drops all Ubuntu changelogs when importing, as a part of the preparation process. [12:28] jscinoz: http://women.debian.org/wiki/English/MaintainerScripts has all the secrets of how and when for maintainer scripts. [12:28] * RAOF checks out this new link he hasn't read before [12:30] Can anyone point me in the direction of a package which creates a totally new mimetype, registers a default open application for it as well as icons etc and works on KDE and Gnome? I'm not having much luck currently writing my own deb to do this for an app... [12:36] I'll leave KDE users out for now then, at least it works for Gnome [12:38] chazco: Does dh_installmime not do it for you? [12:39] I'm doing most of it manually currently... [12:39] Oh, and you'll also need the .desktop file for the MIME type, and dh_desktop [12:39] thank you persia [12:39] But ive found very little documentation which helps with this === apachelogger__ is now known as apachelogger [12:41] Please give me an advice, I filed a bug that fixes FTBFS and attached debdiff. In the meanwhile I also forwarded bug to debian where it was fixed. So do I wait for my debdiff to be accepted in Ubuntu or file a sync request? There are no other significant changes in Debian [12:41] slytherin: sync request. [12:41] slytherin: saves you the effort of merging, and/or requesting the sync then [12:43] Hobbsee: There were no other changes in Ubuntu. Do I still file sync request? [12:44] slytherin: yes, as the autosync is off (and has been since debian import freeze) [12:46] persia, but is debian/preinst automatically called or does it need to be referenced somewhere [12:46] jscinoz: dpkg calls the scripts at the times and with the arguments explained on the Debian Women wiki. [12:48] DktrKranz: Please mention the package and the issue when mailing the BTS: there is no context in notifications. [12:48] dholbach: ping [12:49] persia, oh. right. It isn't in the subject or in the body. Thanks. [12:49] dholbach: didn't the MC violate point #1 of https://wiki.ubuntu.com/CommunityCouncil/Delegation ? [12:50] * nenolod wonders why dholbach pinged somebody who is entirely unrelated to the audacious package on the audacious merge bug and ponders idly if he meant to do it on some other bug instead [12:51] dholbach: and why is that the opposite to the bottom of https://wiki.ubuntu.com/MOTU/Council ? [12:52] persia, can i request you try uscan --report-status again for my two packages, i tried it a few minutes ago after seeing your comments, and it works fine for me. [12:53] jscinoz: I don't have time to download them before I go to sleep? I'll give it a shot when I have time again. [12:54] persia - Still cant get KDE to work (testing on a liveCD)... thanks anyway :) [12:54] chazco: Where are you installing the .desktop file? [12:54] also out of curiosity, would packages one maintains in debian which do not require ubuntu changes count as collateral reference should someone choose to join MOTU? [12:55] I've created /usr/share/applications/tm.desktop (which is a menu entry type one) and /usr/share/mimelnk/tmd.desktop which is the mimetype one [12:55] isn't it mimelink, not mimelnk ? [12:55] hmm, it is mimelnk [12:55] persia, i thought you'd already downloaded it since you've commentedo n the new ones [12:55] chazco: Put the MIME .desktop in /usr/share/applications, or dh_desktop won't do the right thing. [12:55] go figure. [12:55] jscinoz: I delete after reviewing: I don't have infinite disk space :) [12:56] >_< [12:56] So I should have two .desktops in /usr/share/applications [12:56] chazco, also, generally speaking, you can combine mime and launcher into a single .desktop [12:56] one quick thing... idk if you can answer it with out the package [12:56] chazco, so probably, you only need one [12:56] "11) The manpages say they are GPLv2, but provide a reference to GPLv3. Please either narrow the reference, or change the license." I cant find any references to GPLv3 in either man page. [12:56] hmm, ok... will have a look [12:57] Hobbsee: that's a proposal [12:57] Hobbsee: it'll be discussed in the next CC meeting [12:57] How can you combine the two .desktops? One has a type of mimetype and one has application...? [12:57] dholbach: which is? [12:57] * persia notes that one needs two .desktop files when there is one icon for the application and a different icon for the files. [12:57] nenolod: I asked somebody to review the packages [12:58] Hobbsee: what do you mean? [12:58] dholbach, oh. ok. [12:58] dholbach: when's the next CC meeting? [12:58] Hobbsee: I proposed a few meetings times to the members, but didn't get replies yet [12:58] I'm getting the impression this just isnt going to work... it sounds simple but I just cant get it to go [12:58] Hobbsee: we'll announce the new meeting time asap [12:58] chazco, what's your .desktop ? [12:58] Which one? Application or mime? [12:59] chazco: Please pastebin both of those. [12:59] chazco, you really should have all of it in one file [12:59] Oki, one moment [12:59] No idea how to merge them though (really new to this) [12:59] chazco: Also your debian/package.sharedmimeinfo file [12:59] Ive got what i have so far by looking at other packages [12:59] nenolod: Only if there is only one icon. [12:59] persia - I dont have one of those... [12:59] chazco: That might be why dh_installmime isn't working :) [13:00] persia, hmm. afaik desktops can provide more than one mimeicon. [13:00] Would I be better off trying to start the mime type stuff from scratch? (trying to pastebin desktops) [13:00] nenolod: Really? That's different than my understanding, but would be a nice feature. I hope you are correct. [13:00] persia, i believe i have seen it before. but i could be mistaken. [13:01] ,prmomg [13:02] Why is it bad to include the upstream changelog in /usr/share/docs/packagename/ [13:02] persia - http://paste.ubuntu-nl.org/51079/ [13:02] *confused* [13:03] linda gives me an error about no man page for a binary I'm installing into /usr/bin The binary is GUI only with no paramaters to be passed in, so does it need a man page? [13:04] how does one differentiate between hypens and minus in a man page. [13:04] DaveMorris: any binary in /usr/bin must have a man page [13:04] DaveMorris: Yes it does. The manpage might only explain what the command does, that it is a GUI app, and that it takes no arguments, but it should be there. [13:04] jscinoz: \- vs. \(hy [13:04] persia, there are several duplicates to our bug, for what it's worth. i'm going to list them now. ;) [13:04] hmm [13:04] nenolod: Don't list them. Just mark them dupes. [13:05] persia, that'll do. [13:05] nenolod: Anyway, unless it gets rejected upstream, it will likely be applied later. [13:06] persia, so youre saying in every argument listed in the man page i need to change -- to \-\-? [13:06] ie change --help to \-\-help [13:07] chazco: Small notes: you don't want an absolute path for the executable, shouldn't have an extension for the icon, and Encoding is deprecated (non UTF-8 isn't supported anymore). Other than that, looks relatively sane. [13:08] persia: aha, that answers one of my questions about the comments you left for my upload :) [13:08] Tried removing the icon extension, didnt help (but thanks)... app is relative for now because it clashes with something else i'm testing, but wont be in the final... didnt know about encoding, thanks :) [13:08] Ng: Ah. I thought you had an answer. Also try lintian and desktop-file-validate from hardy to check against the latest standards. [13:08] Any MOTUs here willing to review package photoml (http://revu.tauware.de/details.py?package=photoml)? It's already been advocated once (the previous upload, at least), so it should be very close to being acceptable. [13:08] This debian/package.sharedmimeinfo file... i have an xml file in usr/share/mime/packages which looks similar to what ive seen... [13:09] chazco: Removing the icon extension won't help, but it means that themes are supported (which is a good thing). Please do remove it, while also trying to sort your other issue. Also, consider debian/package.mime, which tends to be helpful. [13:10] I'm totally lost tbh now... it works on Gnome, but KDE doesnt do anythign (the menu entry doesnt show, file assosiations dont work). Its probably something simple since I built it by trial and error. [13:10] persia: you've just answered my next question with desktop-file-validate. thanks :) [13:11] chazco: You might ask for someone to comment in #kubuntu-devel (although they may be focused on actually developing kubuntu and not have time to answer). [13:11] persia, 4 duplicates ;) [13:11] For now i'll probably leave KDE out, its difficult to test without a KDE machine [13:12] nenolod: Thanks for helping triage :) [13:12] persia, well, it also demonstrates to sebastian that there are indeed duplicates [13:12] the problem is that my bug is the only one which had a clean backtrace... [13:13] i think it's related to having amd64 [13:15] nenolod: Maybe it's an SMP thing? [13:16] maybe [13:16] So... i have a .desktop in /usr/share/applications (menu entry), an xml in /usr/share/mime/packages which defines the glob patten for the mime type, icons in usr/share/icons/gnome/scalable/mimetypes for files which match, and postinst which updates defaults.list and runs the update-* commands... does that sound correct? [13:16] i finally got my memory issue worked out too [13:16] turns out the motherboard's firmware is on crack and set up the memory with a ridiculous CAS timing [13:16] :D [13:16] chazco: Sounds right. Does it work? [13:17] It does for gnome... most of it is reverse-engineered, so im not sure its the correct way to do it [13:17] persia, that's what i'm thinking [13:18] chazco: As long as you don't do anything especially GNOME-specific, if it doesn't work on KDE, and people want it, you'll get a bug, which might help towards a solution (or at least identify a willing tester). [13:19] Ok, hopefully it'll get fixed at some point [13:19] I was unaware of the debhelper stuff... where can I find more info, they could help... [13:20] chazco: debhelper is typically exceedingly helpful. There are manpages which help some, and the source is fairly readable. [13:21] Odd... mime wont show any man pages (tried it already)... will try to reinstall it [13:22] persia, this is the best backtrace of the bug i've found. it's better than mine. http://launchpadlibrarian.net/11083488/Stacktrace.txt [13:23] just notice that linda and lintian don't complain about a binary going into /usr/bin with a name of fcdActor even though all the other binaries are all lowercase. Is it something I should change? [13:23] persia: I'm not being emailed the recovery password, can you check it out please? [13:23] Just to be totally clear - i can safely remove Encoding=UTF-8 then? [13:25] liri: recovery password for what ? [13:26] REVU ? [13:26] imbrandon: yeah, sorry if I wasn't clear [13:26] liri: Have you uploaded anything? [13:26] liri: it dosent email it to you, you click recovery and follow the instructions [13:26] imbrandon: yeah I uploaded the daloradius package and saw it on the website [13:26] imbrandon: ahh, the documentation on the wiki is a bit unclear on that then [13:26] nenolod: Is there a threadedstacktrace for that bug? [13:28] maybe [13:28] persia, http://launchpadlibrarian.net/11083489/ThreadStacktrace.txt [13:29] Can somone please verify and ack for the sync request in bug 181007? [13:29] Launchpad bug 181007 in libitpp "[SYNC] Debian Unstable sync of libitpp 4.0.1-3 requested" [Undecided,New] https://launchpad.net/bugs/181007 [13:29] Bah. Only shows one thread. I was hoping to see the clearing function running ahead of the populate function. [13:29] persia, that happens [13:29] persia, because the populate function is asynchronous [13:30] tuxmaniac: subscribe the sponsors queue to ask for review [13:30] persia, so it's possible that the clearing function is already called, but there's another populate in the queue [13:30] persia, even for sync request from Debian Unstable? [13:30] nenolod: Exactly. That's why I was hoping for two threads in a threadstacktrace that would demonstrate the issue to actually be the one you fixed. [13:30] but since priv->menu and blah blah blah [13:30] tuxmaniac: Yep. You need approval from a member of ~ubuntu-dev, which is what the sponsors do. [13:30] gnome makes me want to headdesk repeatedly. [13:34] persia, the only reason why i succeeded in fixing that bug is because i'm quite familiar with the craqery that is gtk+ [13:35] things like gksudo scare me, knowing just how many bugs just like that in gtk+ haven't been fixed yet [13:35] :P [13:37] persia: could you check that the package was uploaded and signed successfully? [13:38] I'm wondering if it could be a permissions issue now... [13:50] persia - You know I said it would be something simple.... [13:50] which simple thing was it? [13:50] Log out and back in... which is harder on the live CD :D [13:51] I cant believe it after all that [13:51] I was expecting it to be like Gnome, where it updates without restarting [13:51] Anyway, thanks for your time, it was really appreciated :) [13:53] Think i'm going to have to setup a duel-boot from now one :) [13:57] How do rebuild uploads work? I upload the package sourceful with adapted build-deps and a XbuildY version number? [13:58] pkern: -XbuildY means you only changed the changelog. If you have to touch control, mangle the maintainer and use -XubuntuY [14:00] persia: Which kills autosync. Bah. [14:00] persia: So if I do build uploads, the deps already need to be built and published? [14:00] pkern: You don't want to autosync if you changed control. That's why. [14:01] persia: I do want autosync if I changed control. (: [14:01] persia: I know what I'm doing. [14:02] pkern: The guidelines assumes you aren't 100% sure, and should check again when hardy+1 opens. If you're very, very, very certain, and you break the guidelines, such is life. On the other hand, you ought set an Ubuntu maintainer for an Ubuntu variation. [14:03] persia: It's my package in Debian. \-: [14:04] persia: But ok, thanks. So XbuildY will not prevent autosyncing? [14:04] siretart: hi, the patch from bug report 104332 is missing in the gutsy package. Can I reopen the ticket? [14:04] bug 104332 [14:04] Launchpad bug 104332 in rdesktop "rdesktop crashes when logging in to windows 2000 (or pressing cancel)" [High,Fix released] https://launchpad.net/bugs/104332 [14:05] pkern: yes [14:06] Whoopie: is the patch in the hard package? [14:06] Whoopie: re the bug: sure, but please reassign to yourself === thekorn__ is now known as thekorn === cprov is now known as cprov-lunch [14:08] siretart: what's a hard package? [14:08] Whoopie: hardy [14:08] hardy, right [14:09] siretart: hardy has a CVS version of rdesktop where the bug is fixed. [14:10] siretart: we could just backport it. ;) [14:11] * persia suggests that propagating VCS snapshots can lead to user confusion and complaints, even when bugs are fixed. Is a new upstream coming soon? [14:11] persia: no idea, sorry [14:11] Whoopie: which would miss everyone not using backports. an upload to -security would be better, and requesting a backport as well for more adventureous users [14:14] I'm looking after first advocate for my package malbolge. I've fixed all problems that have been brought up. http://revu.tauware.de/details.py?package=malbolge [14:15] siretart: I don't know how to add Gutsy to this bug report. It's not "project" nor "distribution/package". [14:15] Whoopie: "Nominate..." [14:20] there is a way to enable pkgbinarymangler in pbuilder ? [14:21] Kmos: Yes; install it in the pbuilder environment. [14:21] RAOF: ah ok =) thank you [14:22] Kmos: So, you'd pbuilder login --save-changes (or whatever that switch is), then apt-get install [14:22] and don't forget to enable it [14:22] exactly [14:22] geser: how to do it ? [14:22] it's installed now [14:23] in /etc [14:23] found it =) [14:24] persia: done, but it's not the same as it's shown for the others (edgy, feisty) [14:25] Whoopie: I accepted your nomination and assigned that task to you [14:25] geser: thanks [14:25] Whoopie: Moving from Nomination to confirmed needs-fixing for a given release needs confirmation from a developer. [14:25] siretart: so what is my task now? ;) [14:25] Whoopie: fix the bug [14:26] e.g. by attaching a debdiff and getting someone to review and upload it for you [14:26] siretart: ok [14:27] grep: debian/control: No such file or directory [14:27] /usr/bin/pkgstriptranslations: Error: not in source package directory [14:27] EtienneG: pbuilder-satisfydepends failed. === n3xu|laptop is now known as nexu [14:27] not EtienneG but E: [14:29] siretart: what if I want to add the patch to a debian/patches directory? any changes needed for debian/rules. because last time, I think, it got lost because it was applied "inline". [14:30] Whoopie: please fix the package with minimal intrusion. this mean try to keep the debdiff as small as possible [14:35] geser: pbuilder gives me that error now.. [14:36] hmm [14:37] i've it installed and set enable to true [14:37] pbuilder updated [14:38] I've enabled pkgstriptranslations too but pbuilder still works for me [14:39] siretart: f*, please remove gutsy, it's applied. sorry for the inconvenience. [14:39] really strange [14:39] Whoopie: you should be able to do that yourself. please add a meaningful comment for that [14:41] I'm still looking for someone to revu my package - http://revu.tauware.de/details.py?package=libserial All previous comments have been adressed [14:52] Is glibc (>=2.2.5) a valid depend? [14:53] no, as glibc is the source package name [14:53] Ah ok... [14:54] Trying to write a Depends line for the vague requirements of: glibc 2.2.5 or higher, X Window system with any Window Manager [14:54] chazco: forget about that, dh_shlibdeps will generate depends needing newer glibc anyway === \sh_away is now known as \sh [15:05] <\sh> moins [15:05] <\sh> crimsun: what do I had to read on p.u.c. you are stepping down? === cprov-lunch is now known as cprov [15:26] hi. is anyone here from both motu and doc teams? [15:27] Legendario: I contribute to both, what's up? [15:27] hi. I would like to translate the packaging guide to portuguese [15:28] is there a way to do it using launchpad? [15:32] Legendario: yes, sorry, I was doing somethin [15:32] it's ok [15:32] so. how can i do it? [15:33] Legendario, I'm not really sure it's available for translation in LP [15:33] Legendario: the packguide is maintained at the wiki [15:33] * joejaxx wonders how much work it would be to package kde from scratch [15:35] hellboy195, around? [15:37] Heya gang [15:37] bddebian: ! [15:37] :D [15:37] Hello joejaxx [15:37] anyone know how packaged the kde packages originally? [15:37] bddebian: :) [15:38] joejaxx: The bottom of the changelo for kdelibs says Adreas Jellinghaus [15:38] oh ok [15:38] hmm [15:40] Hi bddebian [15:40] Hi geser [15:42] Legendario, the translation of the packaging guide was available with dapper: https://translations.launchpad.net/ubuntu/dapper/+source/ubuntu-docs/+pots/packagingguide/pt/+translate [15:42] broonie: Hi, have you had time to look at the aqsis build failure? Is it a problem with aqsis or scons? (the problem was that it got caught in a loop) [15:44] Hattory: but that's Dapper :p [15:46] I know... jpatrick..... just to see that it was on Launchpad [15:47] :) [15:48] Hattory: must be so outdated [15:50] do you guys know if the version listed on the doc.ubuntu.com as work in progress is to different from the dapper one? Hattory, do you have the link for this one? [15:50] dholbach: thanks a lot! [15:51] pochu: greetingz :) [15:51] Thank you Kmos :) === _nuu is now known as nuu [16:01] jpatrick, why wasn't it update anymore? [16:02] Hattory: it's moved to the wiki [16:21] pochu: Congratulations. [16:22] pochu, great news! congrats ;-) [16:23] pochu: wow congrats, though I thought you were a MOTU for like a year :D [16:23] pochu, congrats ;) [16:33] Thanks ScottK2 mruiz jdong and Hattory! :) [16:39] hi guys. now the package builds on pbuilder hardy base. request someone to please review and comment. http://revu.tauware.de/details.py?package=alliance === LucidFox is now known as LF_GroundhogDay === iceman_ is now known as Iceman [17:21] Could i get a second advocation for http://revu.tauware.de/details.py?package=hardware-connected (a program that checks whether given hardware is connected to the system, useful for scripting)? Thanks. [17:23] is there anyway to tell debbuild not to sign a package? Building a local upgrade, don't have my key on the server... [17:25] pochu: probably by passing in -us -uc [17:25] oops pfein [17:26] jdong: yup, thx === x-spec-t is now known as Spec [17:30] If a piece of code is licenced under the lgpl and you compile it with --enable-gpl such that the resulting binaries are gpl, should debian/copyright reference the LGPL, GPL or both? [17:36] I'd think both [17:38] Ok, i'll put in a "source licenced under" and "binaries" === LF_GroundhogDay is now known as LucidFox === Ubulette_ is now known as Ubulette [18:43] Hi, I am trying to build the python2.5 package via "debian/rules binary" inside the source, but its running all of the regression tests, and that takes _ages_ [18:43] any way to disable the regression tests? [18:44] unenough: export DEB_BUILD_OPTIONS=nocheck [18:44] will still run the benchmarks, though [18:45] ScottK2: were you following up with the Debian maintainer about bug #154730 as requested by seb128? [18:45] Launchpad bug 154730 in secvpn "Please remove secvpn source and binary from Hardy" [Wishlist,Incomplete] https://launchpad.net/bugs/154730 [18:46] azeem, Thanks! [18:46] azeem, any way to cancel the benchmarks too? [18:46] dunno [18:47] look inside debian/rules, comment it out if it's a seperate command [18:48] K, I'll hope for now that the benchmarks are not as long as the regression tests [18:51] slangasek: It's on my "list", but I haven't actually done it yet. [18:52] ScottK: ok === nuu is now known as nu[year] === nu[year] is now known as nuu [18:58] slangasek: newpki-client builds now, I think that just leaves ctsim with wx2.4? [18:59] I dunno, I haven't been following the progress [19:05] Well you're fired :) [19:06] I'd work on the vflib stuff if I really understood the issue :( [19:11] bddebian: When has that ever stopped you before? ;-) [19:13] ScottK: Never, but thanks for pointing that out :-) [19:14] my cdbs rules file has 1 big target which is independent (doc from doxygen) of the other targets. Is there anyway to have it run in a separate thread from the rest of cdbs, thus making use of dual core systems? [19:15] No. [19:15] shame [19:15] There is work in progress on parallel builds in Debian, but nothing production ready yet. [19:15] would save me around an hr [19:16] If you look in (I think it was) the debian-devel mail list archives for parallel build, there was a recent thread on the topic. [19:16] The package libfontbox-java in hardy shouldn't be in multiverse? it needs checkstyle package that's in multiverse [19:16] ? [19:17] the question is: why is checkstyle in universe when its in main for Debian [19:17] multiverse I mean, sorry [19:17] :) [19:18] ubuntu names... [19:19] man-di: that's also a good question [19:20] well, checkstyle was moved from contrib to main some long time ago... [19:21] And it looks like no one ever touched it in Ubuntu to move after the tarball got repacked [19:23] I'll fill a bug about this to move it to universe.. [19:24] ScottK: well, one can certainly configure one's own package build rules to support parallel building in line with the d-d discussion, and invoke the package build accordingly on one's own system [19:25] since he seems to be concerned with saving his own time, not the buildd's :) [19:25] True. [19:25] DaveMorris: ^^^ [19:26] I was talking earlier this afternoon of merging our flightgear with debians, should I do this even though we've passed DebianImportFreeze? [19:26] Are there any changes in Debian's version we don't have that are worth having? [19:27] I havn't gone through that yet, as I wasn't sure of what to do i wanted to check first [19:28] Judgement call at this point === cprov is now known as cprov-out [19:29] Go mok0! (Just saw your Debian ITP for Mustang) [19:30] We're at verion 9.10 and they're at 1.0. I went through the changelogs and there's a lot we're missing [19:30] ScottK: Heh! It's my first try [19:30] dannioni: I'd go for it then. [19:31] ScottK: ... and part of my MOTU training... [19:31] Excellent [19:31] ok, thanks for the advice, needs to be done sooner or later anyways [19:32] slangasek: I'm curious if you saw my perl-module post on ubuntu-devel and if you have an opinion? [19:32] Started using git-buildpackage today. Pretty nifty, but it has some quirks [19:33] hi astharot ;) [19:33] dpatch has the annoying habit of touching the files in debian/patches, which means git thinks they have been modified... [19:33] ScottK: yes, seen; I think dropping those modules from perl-modules and making them a dep appears to be the correct course of action, but since this is apparently a MIR I'll leave it to others like pitti to dredge up any concrete objections [19:33] K. Thanks. [19:34] re [19:35] Anyone else here has experience with git-buildpackage? [19:35] s/has/have [19:35] mok0: I’m using it. [19:35] mok0: I know lamont (on ubuntu-devel/server) uses Git. Dunno if he uses that. [19:36] ion_: How do you handle upstream updates? [19:36] Hello [19:36] Hello [19:36] ion_: Do you untar it on the upstream branch? [19:36] ion_: or create a new one? [19:37] Say I have samba-3.0.26a, which is having problems, but I want to do a quick build of 3.0.28, what's the quickest way to swap out the old .orig source for the new? [19:37] mok0: If upstream doesn’t use git in the first place, you should be able to use git-import-orig to import the source to a git repo. Then just merge the changes from the upstream branch to the ubuntu branch, update the packaging and use git-buildpackage as usual. [19:37] Hello [19:38] is Emmit Hikory here ? [19:38] bddebian: hide [19:38] AnAnt: He goes by persia [19:38] ah, so he's persia [19:38] ion_: do you have to be sitting on the upstream branch, or will git-import-orig do that for you? [19:38] persia: Hello, thanks for reviewing the ubuntume artwork packages [19:39] mok0: Look at its man page. [19:39] ion_: :-) [19:39] man-di: Hello, I ran josm using icedtea jre under i386, it didn't work so well [19:39] * mok0 feels stupid [19:39] mok0: It has defaults for the upstream and packaging branches, but you can override them with e.g. a .gbp.conf in the packaging branch root. [19:40] man I love Ubuntu! [19:40] ion_: might as well stick to the default for now... I have no better scheme anyway [19:41] For instance, http://gitweb.heh.fi/?p=ion/hardware-connected.git;a=blob;f=.gbp.conf;hb=ubuntu [19:41] LaserJock: glad you think so, consider sticking around and getting involved!! [19:41] :p [19:41] jharr: like grabbing the 3.0.28 package for hardy directly from the Ubuntu archive and rebuilding it with dpkg-buildpackage -B? [19:41] ion_: cool [19:41] ion_: I guess you can run pbuilder as well from there [19:42] AnAnt: file a bug report [19:42] man-di: ok [19:42] jcastro: I just got the uni VPN working with Network Manager [19:42] AnAnt: note that icedtea is a pre-version of what will be Java7 [19:42] mok0: Perhaps. I haven’t checked. [19:43] man-di: btw, did you tried icedtea firefox plugin on an amd64 ubuntu ? [19:43] ion_: I thought gbp.conf went in .git/ === thekorn__ is now known as thekorn [19:43] AnAnt: no [19:43] mok0: Again, look at the man page. You can put it in .git if you don’t want it to go to the repository. [19:44] AnAnt: man-di is a Debian user, not Ubuntu. [19:44] oh [19:44] well, I have Ubuntu chroots for some testing [19:44] man-di: so, you made an icedtea package for Debian this week ? [19:44] ion_: ok [19:44] yes, I work in icedtea package (which is the same as for Ubuntu) [19:45] Should I use debian-maintainer field even when the package does not exist in Debian? [19:45] man-di: thanks [19:45] ion_: Are you in Finland? [19:45] ok, I have a question, I was asked to close a bug for initial release, is there something like ITP bug in Ubuntu ? [19:45] cyberix: Yes (generally). [19:45] mok0: Yep. [19:45] AnAnt: bug tagged 'needs-packaging' [19:46] ion_: Perkele! :-) [19:46] ScottK: ok, I should file the bug against what ? [19:46] Against Ubuntu [19:46] i.e. No package [19:47] mok0: is this how you greet all Finns? [19:47] ok [19:47] slangasek: I don't think so :-) [19:48] slangasek: It's a swearword AFAIK [19:48] * ScottK starts to wonder what that meant. [19:48] yes, it is a swear word [19:48] ScottK: Is this explained somewhere? [19:48] slangasek: not rude I hope [19:48] so "Are you in Finland?" "yes" "perkele!" is kinda strange [19:48] cyberix: I'm sure it's in wiki.ubuntu.com somewhere. Dunno where as stuff was recently re-arranged. [19:49] slangasek: It's the only finnish word I know... apart from yksi kaksi kolme [19:49] ScottK: what does X licensed code mean ? it was on a comment on http://revu.tauware.de/details.py?package=usplash-theme-ubuntume [19:50] slangasek: are you finnish too? [19:51] AnAnt: From what it says, I'd say there's a thing call the "X License" and there's code in your package licensed under it and you didn't mention it in debian/copyright. [19:51] AnAnt: grep -ir copyright * is your friend. [19:51] mok0: no, just a polyglot [19:51] slangasek: you speak finnish? I'm impressed! [19:52] mok0: I only know a few key phrases in Finnish [19:52] ah, ok [19:53] can any LP user have access to LP's Vcs ? [19:53] Like "where's the beer?" [19:53] like "hyvää huomenta" and "vittu perkele" [19:54] slangasek: gonna tell us what that means? [19:55] nope :) === apachelogger_ is now known as apachelogger [19:55] Google's my friend [19:56] indeed [19:56] slangasek> Thanks for moving libhiglayout-java and freecol! [19:57] You are an archive administrator, I presume? [19:57] yep [19:57] slangasek: you are bad! [19:57] :-) [19:57] mok0: I prefer to think of myself as "authentic" [19:57] Do you manage NEW? Because I'd like to get mpeg4ip into Hardy ASAP [19:58] I should not use XSBC-Original-Maintainer? [19:58] LucidFox: I do, but you'll have to wait your turn :) === jpatrick_ is now known as jpatrick [19:58] ah [19:58] Is the NEW queue visible for the public somewhere, btw? [19:59] http://launchpad.net/ubuntu/hardy/+queue [19:59] Thanks [20:00] vittu perkele saatana! [20:00] not to be confused with the Finno-Latin fusion group, "vittu perkele santana" [20:01] cyberix: Generally for packages new to Ubuntu and not in Debian, put yourself in that field if you are packaging it. [20:02] slangasek: lol [20:06] ScottK: if you have an @ubuntu.com address [20:11] ScottK: so for packages that are not in Debian, I put myself in both Maintainer: & XSBC-Original-Maintainer: fields ? [20:16] AnAnt: no, you wouldn't need XSBC-Original-Mantainer if you're setting yourself as Maintainer [20:17] LaserJock: ok, what does that comment mean "Please use an Ubuntu maintainer. If nobody on the team is an Ubuntu member, consider applying for a mailing list for the team" on http://revu.tauware.de/details.py?package=usplash-theme-ubuntume ? [20:18] well [20:19] Maintainer requires either an @ubuntu.com address *or* an Ubuntu mailing list (ubuntu-motu for Universe) [20:21] which is to say, everything should have a *@*ubuntu.com email address as the Maintainer [20:21] ok, so what should I do ? [20:22] AnAnt: do you have an @ubuntu.com email address? [20:22] put myself in Maintainer: and ubuntu-motu list in XSBC .... ? [20:22] LaserJock: nope [20:22] reverse it then [20:22] AnAnt: the other way round [20:22] ubuntu-motu as Maintainer and yourself as XSBC.... [20:22] pochu: ok [20:22] AnAnt: have you considered uploading this to Debian too? [20:23] pochu: it's Ubuntu specific I think, it's artwork for a distro based on ubuntu, why put that in Debian ? [20:24] heh, nevermind then :-) [20:24] ok [20:24] thanks for the help guys [20:26] what's the name part of Maintainer: ? [20:26] I mean what's the name part of ubuntu-motu mailing list ? [20:26] Ubuntu MOTU Developers [20:27] https://wiki.ubuntu.com/DebianMaintainerField says Ubuntu MOTU Developers for the field [20:27] thanks [20:28] what's FTBFS stands for ? [20:28] Fails to build from Source [20:29] ok [20:30] ok, one last thing (I hope), I'm doing a usplash theme [20:30] AnAnt: https://wiki.ubuntu.com/UbuntuDevelopment/Abbreviations [20:30] AnAnt: I keep that page open when in this channel :-) [20:30] when I look at usplash-theme-ubuntu, I find that the C file is GPL licensed, yet in debian/copyright I only find a CC license [20:31] I did the same with my usplash theme, and the reviewers didn't like that [20:31] hi all! :) [20:32] AnAnt: what do you mean, "find" a CC license? it's your package, surely you're responsible for the contents of debian/copyright? [20:32] let me ask this question first, can the .c file of a usplash theme be licensed under CC ? [20:33] slangasek: I'm talking about what I saw in Ubuntu's usplash theme package, hence I did similar in the usplash theme I was packaging [20:33] legally, yes; but debian/copyright needs to reflect the actual license of the package (and CC licenses aren't as good for source code as GPL, anyway) [20:33] could anyone review sdlmame? it would be nice to see it in hardy :) [20:33] http://revu.tauware.de/details.py?package=sdlmame [20:34] slangasek: well, in usplash themes, there are image files (.png) and a C file that calls them [20:34] slangasek: so I think in that case, the C source code can be considered some artwork more than it is a software [20:37] AnAnt: debian/copyright still needs to reflect the actual license that the software has been placed under by the copyright holders [20:37] slangasek: ok [20:37] thanks for the help === Iceman is now known as iceman [20:41] Did libwxgtk-2.4 have a wxgtk-2.4-config that libwxgtk-2.6 doesn't? [20:47] Is there a Debian-Maintainer filed which is separate from XSBC-Original-Maintainer ? [20:47] field [20:48] Even apt uses XSBC-Original-Maintainer [20:54] is there some ubuntumembers administrators on launchpad? [20:54] I have a question about my membership [20:58] bddebian: how did you put together whatever list you're using of wxgtk2.4 reverse-deps? [20:59] slangasek: I've just been using what was sent on the oldlibs e-mail so far [20:59] I think gnue might still be an issue too iirc [21:00] bddebian: url? [21:00] n/m, found it [21:16] anyone to review my package, please? http://revu.tauware.de/details.py?package=sdlmame [21:29] yay! Almost got a Python2.5 package that puts .pyc's in another cache directory, rather than messing the source dir, and that treats .pyc's as purely cache, does not use them as sources when the .py's not there [21:32] unenough: don't install .pyc files in your package [21:32] Grr, I can't get ctsim to even attempt to build with wx2.6 :( [21:32] persia: Wake up man.. :) [21:32] geser, for now its just a temporary local hack for me and friends [21:33] geser, if it becomes more serious, there are some issues to handle, such as having a system-wide cache dir to search first [21:33] bddebian: that's ok, if that's really the last package and the maintainer/upstream is being obstructionist, we can just pull wxwindows2.4 out of unstable out from under him [21:34] unenough: why do you want to have the .pyc file inside the package? [21:34] slangasek: Well persia said he got it to build, he just had the sefault issue, which may have been a similar wx-config issue [21:40] And, of course, I don't know if the newpki-maintainer is going to accept my patch either [21:43] bddebian: then we would have two packages that are SOL in unstable due to daft maintainers. [21:43] especially when the code has already been written to move to wx2.6, there's no reason at all for us to keep wx2.4 around [21:43] You're the boss :-) === \sh is now known as \sh_away [21:46] come on, is revu day! no reviewers online? ;) [21:47] What does a good get-orig-source rule look like? Would something like this be ok? http://paste.ubuntu-nl.org/51135/ [21:49] btw, what does SOL mean? [21:54] amarillion: you should also rename the directory before recompressing and delete the .zip and the extracted directory [21:54] after building the tar.gz [21:55] I though the name of the directory is ignored by all utilities === neversfelde_ is now known as neversfelde [21:58] amarillion: well, to be honest, I don't know... :) the packaging guide suggest to normalize the dir name [21:58] ok, I'll do it just to be safe [21:59] I also pasted my suggestion on the pastebin... didn't know I could ;) [21:59] so I need to add "mv speed speed-game-1.00" and rm "speed.zip", probably "rm -r speed-game-1.00" afterwards as well [22:00] yes [22:00] oh I didn't know that either. Do you get a new link for that? I don't see it yet [22:02] should be http://paste.ubuntu-nl.org/51137/ [22:02] I didn't notice the link was changed [22:03] Ah yes, I could have just have incremented the number twice [22:03] thanks for the suggestions [22:04] you're welcome :) [22:21] http://revu.ubuntuwire.com/details.py?package=libfprint is ready for a review if someone has time [22:23] and so is http://revu.tauware.de/details.py?package=sdlmame, thanks :) [22:25] While we're on review requests, let me add one for http://revu.tauware.de/details.py?package=photoml It's already been advocated once (the previous upload, at least), so it should be very close to being acceptable. [22:28] I just updated my package: http://revu.tauware.de/details.py?package=speed-game. All comments made yesterday have been addressed, except the license clarification because I'm still waiting for an answer from upstream [22:28] And I learned everything there is to know about hiscore files in the process :) [22:35] bddebian: Sorry. Slow today :) There was a wx-2.4-config, but wx-config should replace it. ctsim worked for me in early gutsy, but segfaulted on execute. I haven't looked since. [22:39] persia: With no changes other than the build deps? [22:40] bddebian: By memory, yes. I appear to have removed my ctsim working directory, and as I never had a working patch, didn't generate anything for my patch archive. [22:40] persia: I would ask to nuke http://revu.tauware.de/details.py?package=shorten [22:41] as norsetto pointed out, licence is quite uncomfortable and I'd like to stop packaging it [22:41] wallyweek: I seem to be having difficulty nuking lately. Would archiving meet your needs? [22:42] persia: it doesn't matter to me... ehr... what's the difference, btw? ;) [22:42] wallyweek: Archiving hides it from the default listings. Nuking removes the package and all comments from REVU entirely. [22:43] In case of later interest, you can recover from an archive with a new upload. [22:43] wallyweek: shorten archived [22:43] wallyweek, can that package not be included? [22:43] persia: thanks! [22:44] Too bad, after all that work [22:44] amarillion: Since the original packager no longer wishes to continue, feel free to grab it, work with upstream to resolve the licensing issues, and submit a new candidate, if you have interest in the package. [22:45] persia: licence is weird... it says it's free to decode, but permission should be asked for encoding [22:45] wallyweek: Might be related to some of the encoding patents that are floating about. (not that I've investigated) [22:46] a.k.a. rar license? [22:46] persia: yes. Shorten is not a free codec, even if opensource [22:48] I could patch the source, but I feel this would not be good [22:48] That makes sense then. I suppose a decoder-only could be packaged, or a use-specific patent grant could be given (which would make it multiverse). Needs coordination with upstream . [22:49] persia: unfortunately, original upstream maintainer doesn't care about original code [22:51] new releases are carried on by another guy, which has no rights to modify original code licence [22:52] wallyweek: Right. Needs coordination with the copyright holder / patent controller / whatever, if upstream isn't that party. Upstream can maybe help with that (although you may not be the person who works with them to achieve the goal). [22:52] amarillion: if you find shorten useful, you can download package from my ppa https://launchpad.net/~c.falco/+archive [22:53] wallyweek, oh I don't really have a personal interest, I was just reading the comments page on REVU [22:53] Seeing all the comments, a fair bit of work went into it [22:54] amarillion: yes... but that was good learning anyway ;) [22:54] true [22:55] Also, it's likely that much less work for someone else, if they pick up the packaging effort. [22:55] persia: true [22:55] I was reading a few comments pages on REVU. Some packages go through dozens of revisions [22:56] This is especially true for very complex packages, with lots of upstream problems. Most reviewers stop looking carefully when reaching some threshold of comments (for me, around 10). [22:57] makes you wonder how we ever got to tens of thousands of packages that are in ubuntu now [22:57] amarillion: yes, sometimes it's quite frustrating... sdlmame has been on REVU since feb 2007 [22:57] Only about 500 came thorugh REVU. Most are imported from Debian. [22:58] persia: good answer... I wondered the same :) [22:58] true, but I understand the process for inclusion in Debian even longer [22:58] mine has had loads of revions [22:59] * DaveMorris cruses OpenSG [23:00] * DaveMorris can't spell this evening [23:01] before going to bed, let me call once again for reviews on http://revu.tauware.de/details.py?package=sdlmame [23:01] thanks :) [23:07] anybody familiar with pbuilder-dist around? [23:07] I'm trying to create a pbuilder and I just get: [23:07] E: Type '' is not known on line 1 in the source list /etc/apt/sources.list [23:17] jdong: Have you seen this aac encoder? http://teknoraver.campuslife.it/software/mp4tools/ [23:17] persia, patch accepted upstream. http://bugzilla.gnome.org/show_bug.cgi?id=507605 [23:17] Gnome bug 507605 in recent-files "[patch] gtk_recent_files_menu_populate() does not guard properly against recursion" [Critical,Unconfirmed] [23:17] jdong: specifically aacplusenc [23:17] TheMuso, that's a new one [23:17] I know [23:17] TheMuso: yet another mp4 encoder [23:18] Yeah I know. Does anybody know how it compares to faac? [23:18] TheMuso: pretty sure it uses faac? [23:18] jdong, no [23:18] jdong: doesn't appear to. [23:18] jdong, it uses it's own encoder [23:18] TheMuso, i'll play with aacplusenc tonight [23:18] lemme grab the sources before opening my mouth... [23:19] TheMuso, it'd be nice to have a decent aac encoder in hardy [23:20] g'night everyone! :) [23:21] goodnight [23:22] yeah, looks interesting [23:22] hmm [23:22] would be nice to put it too the test [23:22] the ubuntu packages are incorrect [23:22] to* [23:22] they'll have to be redone [23:22] nenolod: you don't say? ;-) [23:23] aacplusenc doesn't tag however [23:23] (they are packaged as a native archive) [23:23] TheMuso: tagging is trivial via gpac/MP4Box though [23:23] TheMuso: I'm more interested in its encoding abilities [23:23] TheMuso, that's fairly trivial [23:23] jdong: Same. [23:23] if you want to tag it, you can always open it in audacious or whatever [23:23] jdong: One thing I found out is that from reading an ubuntuforums thread, it doesn't work properly on amd64. [23:23] TheMuso, i can fix that [23:24] TheMuso: yeah I was seeing that in the makefile [23:24] http://ubuntuforums.org/showthread.php?t=628434&highlight=aac [23:24] TheMuso, also passed it onto gentoo-media team, so maybe we can collaborate on it together ;) [23:24] TheMuso: it looks immature but promising [23:24] Cool. [23:25] jdong: In the makefile? [23:25] most amd64 bugs are due to bitwidth oversight [23:25] by using stdint.h, it's not a major issue [23:25] :P [23:26] heh [23:26] * TheMuso cranks it through powerpc [23:26] compilation that is [23:26] (you can also do int foo : 4; to make a 32-bit int) [23:26] nenolod: looks like this thing uses x86 assembly a bit... afraid it won't be that easy [23:27] jdong, yuck [23:27] Great. So PowerPC is out of the question then. :) [23:27] * TheMuso grumbles./ [23:27] Userspace apps should not use asm unless they are arch specific IMO. [23:27] #ifdef __LP64__ [23:27] if (inputInfo.nChannels > 1) { [23:27] fprintf(stderr, "Stereo encoding is not supported in 64 bit builds [23:27] "); [23:28] whee! [23:28] lol [23:28] SOunds like the code is hackish. [23:28] TheMuso: well IMO media apps have the right to drop down to assembly [23:28] jdong, where's the asm? [23:28] TheMuso: often times their hand-tuned assembly turns in 30% minimum speedi mprovements [23:28] jdong: I doubt that not, but they should get help for other arches then. IMO all arches should benefit. [23:29] I know PowerPC is dying, but many people still use it, myself included/ [23:29] i don't see any x86 asm [23:30] Hrm ok. It builds on PowerPC. [23:30] yeah [23:30] nenolod: You're right, I don't think there is any asm specific stuff. [23:30] as i said [23:30] nenolod: you're right no ASM [23:30] "i don't see any x86 asm" [23:30] well [23:30] nenolod: I saw the ADD/MULT/MOVE macros and got worried [23:30] jdong, i'll import this into some bzr branch [23:30] looks like it's just some sort of hackjob profiling [23:30] eep. [23:31] jdong, that's right. i intend to fix it. [23:31] :D [23:32] nenolod: gogogogogogo [23:32] also, no autoconf = wtf [23:32] Oh, and its built statically. [23:32] nenolod: Not everything needs autotools. [23:33] sure it does, just not everything uses autotools ;) [23:33] TheMuso, autotools would be good for this ;p [23:33] I'd rather not see autotools :D [23:33] Well, it would make turning those libs into shared libs easier. [23:33] :) [23:34] nenolod: also chase through the changelog, I saw a few cases where the author references switching from his own to some GPL'ed routines [23:34] nenolod: wonder if he is bundling a bunch of unneeded routines or if they'rej ust single functions [23:35] What is the Nero aac encoder like? [23:35] it seems to encode stereo just fine on 64-bit [23:35] although mplayer is having problems with it [23:36] but mplayer has problems with SBR anyway [23:36] * jdong looks at mpeg4ip [23:36] nope still in NEW [23:37] jdong: What package is mp4box in? [23:37] TheMuso: gpac [23:37] oh ok [23:37] hmm [23:37] weird [23:38] it does indeed produce crap on amd64 [23:38] yay! [23:38] I'm surprised the author hasn't just passed -m32 into gcc [23:38] lol [23:39] (crap when done with stereo. so i think it's a bitwidth bug somewhere.) [23:39] i imagine sed -i s:unsigned int:uint32_t:g s:int\s:int32_t:g will do. [23:39] maybe he uses sizeof [23:40] LaserJock: Was it a Debian pbuilder you were making? [23:40] If so, I had the same kind of problem. [23:40] I just hacked stuff out of the script until it went away. [23:40] nenolod: unless there's hardcoded bitmasks that are different on x64 [23:42] jdong, maybe. we'll see. [23:42] :) [23:48] ScottK: no, hardy [23:48] I just rand pbuilder-dist hardy create [23:49] Did you have universe enabled? When I had the problem on Debian ones it was getting confused about extra repos and then I got that error when I removed them. [23:54] I tried it as both pbuilder-dist hardy allcomp create and pbuilder-dist hardy mainonly create I think [23:54] Ok this encoder also doesn't accept high bitrates. [23:56] anyone else having packaging issues with kde4 packages in gutsy? [23:57] i know kde4 is "beta" but packaging is packaging :P [23:58] joejaxx: where are you getting the packages? [23:58] well the ones that i did today were from backports [23:58] http://pastebin.ca/845932 [23:58] looks like fun :D [23:58] does it not? [23:58] :) [23:59] ScottK: Is Debian affected by the "needed perl module version is higher than the one in perl-modules" problem? [23:59] joejaxx: well, seems somewhat typical :/