[00:01] bloody good example. [00:01] terli: "patented software needs a copyright." wth? [00:01] what has copyright to do with patents? [00:02] hear hear [00:02] what does db_get do [00:03] soc : copyright, license, you name it [00:03] legal shmega [00:03] terli, they're all very different things [00:03] patented meaning proprietary [00:03] terli, terms aren't interchangable [00:04] right, right, but my brain is of the sort that can replace tomatoe with table in a second [00:04] what does db_get do [00:06] doesn't that get stuff from the template? [00:06] so if I have a file named templates [00:06] with two entries in it marked parallels/accept_licence and parallels/license [00:06] right [00:06] accept_ being the boolean and license being the text itself [00:07] yep [00:07] db_get parallels gets both? [00:07] no. can't be. [00:08] well, look at the example [00:08] db_get debconfdemo/accepted [00:08] if [ "$RET" = "true" ]; then === _neversfelde is now known as neversfelde [00:08] exit 0 [00:08] fi [00:08] so it grabs debconfdemo/accepted and then tests if it returns true or false [00:09] be careful with forcing the user to accept the agreement prior to proceeding. you're in a position to design it properly now. the major point of contention will be blocking distribution upgrade progress. [00:10] i.e., i install foobar, update && full-upgrade, walk away for a week, and find the upgrade is blocking on my input [00:10] crimsun, your a god, but this package will never, ever, ever make it into the repo. [00:10] it's proprietary and binary. [00:10] so? [00:11] so the only upgrade possible is manual. [00:11] sure, that's fine, and there are plenty of examples that did in fact use a hard-stop query, like the sun-java[56] in multiverse [00:12] terli: I'm saying prorietary and binary packages are allowed [00:12] Nobody maintains this package, and I'm not capable of complying with the universe rules. [00:12] , now, if you want to upload it for me, thats a different matter [00:13] not I, been there, done that, still reaping the "benefits" :-) [00:13] the source package I edited was sudo extracting a tar file in /usr/lib/programdir [00:13] made removing all of it a headache [00:17] terli: what's the name of the software? [00:17] parallels. [00:17] I'm doing some work to get rid of my howto [00:17] and stop breaking dash/bash [00:18] http://revu.ubuntuwire.com/details.py?package=metakit <-- comments would be appreciated [00:19] subprocess pre-installation script returned error exit status 255 [00:24] http://pastebin.com/m238dc41a [00:25] er, umm [00:25] http://pastebin.com/m699a61f3 [00:26] there [00:41] could someone review/advocate my package? ... http://revu.ubuntuwire.com/details.py?package=ttf-droid [00:42] it should be totally clean and nice now ... [00:42] Template parse error near `.', in stanza #2 of ./templates [00:47] The requestsync script in intrepid misbehaves a bit if you're a member of the launchpad betatesters group :) [00:48] Specifically, it submits the bug report 5-6 times, then crashes. [00:49] use it to submit a bug against launchpad? [00:49] :P [00:51] /c/c [00:56] Ok, what should I do with the duplicates? Is there a way to delete them, or do I need to mark them as duplicates? [00:56] It seems a bit silly now with 5 identical reports in a row. [00:57] in what? [00:57] bug 315291 [00:57] Launchpad bug 315291 in mumble "Please sync mumble 1.1.6-5 (universe) from Debian unstable (main)." [Undecided,New] https://launchpad.net/bugs/315291 === emma_ is now known as emma [00:57] bug 315292 [00:57] Launchpad bug 315292 in mumble "Please sync mumble 1.1.6-5 (universe) from Debian unstable (main)." [Undecided,New] https://launchpad.net/bugs/315292 [00:57] etc etc up to 315295 [00:59] I'll mark them duplicates for now, that should at least remove them from the trackers. [01:00] is there a db_display variable? [01:01] I need someone to tell me three db_ commands [01:11] I'm looking for a little help, OK everyone need help:) I'm looking for someone who works with the init scripts. [01:12] which init scripts? [01:13] Let me look. [01:15] somewhere before /etc/initramfs-tools/scripts/init-bottom/ in the init process [01:18] how about someone working on the netbook remix [01:24] nschembr: You are not being very specific about what you need help on - specific questions are better because they allow people to see whether they can actually help or not, quickly [01:27] hey , could someone show me how to put my license in a debconf tag? [01:28] terli, a EULA-style you-mist-agree license? [01:29] terli, an old sun-java6 package should provide something to copy [01:29] yes [01:29] no [01:30] the non-free virtualbox has a license agreement if I recall correctly [01:31] i just remember java needing it. the sun distributor license or somesuch [01:31] I have it on my screen [01:31] it's not working [01:31] it doesn't have a job? for shame, my taxes are paying for it to lounge around at home! [01:32] it's no use to me [01:32] and why is that? [01:33] its all nice and all that [01:34] hmm [01:34] I ran the script from a terminal [01:35] it just ran through, didn't prompt me or anything [01:35] but its not crashing [01:35] had you already accepted the licence once, and are now trying to accept it a second time? [01:35] no, I'm building the bloody package [01:35] oh, right. [01:35] ran which script from a terminal? [01:36] quiet. [01:36] ok, now it failed to install, but that's my fault [01:37] I can feel it [01:37] the preinst is configured to db_go parallels [01:37] the parallels file is configured to: [01:38] db_input critical license; db_input critical license_accepted;db_go; db_get license_accepted;if [ "$RET" = "true" ]; then [01:38] exit 0 [01:38] fi [01:38] the preinst is configured to db_go parallels <--- are you saying you actually have a line "db_go parallels" in the preinst? [01:39] yes? [01:39] because that would be meaningless [01:39] db_go doesn't take an argument [01:39] how do I get preinst to pass control to my script? [01:39] you don't [01:39] and does db_stop belong in there? [01:40] You've misunderstood how debconf works. You shouldn't have a 'parallels' file. [01:40] I have preinst , a file called parallels and one marked parallels.templates [01:41] preinst is supposed to use debconf to run parallels , which is a script that uses the parallels.templates file to show license details [01:41] uhm, no [01:41] then explain how it's supposed to work. [01:41] You don't use debconf to run things [01:42] You use debconf to interact with some abstract UI [01:42] I don't think you understand. [01:42] you are supposed to help me, not contradict me., [01:42] Part of helping is pointing out misunderstandings [01:42] what should i do for preinst to get my script implemented [01:43] it's just a file called parallels that starts with a line marked #!/bin/sh, think of it as any other script file. [01:43] you don't write a script, you write a debconf template. preinst calls debconf using that template, and bails if your haven't accepted it [01:44] no, I've relocated the details of what debconf does to a different file. [01:44] Well, don't do that [01:44] Doing that goes against all standards of how to use debconf [01:44] to what end? now you're talking about making that different file available outside the package, which is enormously messy [01:44] well, then, I suggest you inform me about the correct way. [01:44] he tried, you told him off for it [01:44] the different file is IN the package [01:44] it's IN /debian [01:44] The correct way was the way I did it in the prewritten example I gave you [01:45] Which is the same way sun-java does it [01:46] please look at debian/dlj.templates and debian/JB-jdk.preinst in sun-java6 [01:46] you can't use preinst to call things in data.tar.gz [01:46] so I have a line marked db_get parallels/license which corresponds to a template marked parallels/license in the file parallels.templates [01:46] direct hex, everything is in the folder marked debian in the root of the archive. [01:48] the semantics are more complex than that though. a binary package is split into multiple parts. the actual data section, the files to which you'd be agreeing to a license, are not the same as the metadata - that way you can't install it if you don't agree with the license [01:48] the license is in the metadata, and in the data. [01:48] two copies. [01:48] you agree to the copy stored within templates [01:49] it's identical to the other copy [01:49] but you don't have to explicitly agree to the other copy, you already did that with the first one during installation. [01:49] but you don't want to use the template, you want to run a script. and the script does not form part of the control.tar.gz section of the package, which contains metadata like templates and preinst [01:50] the script isn't in the control.tar.gz [01:50] it's in debian! [01:50] preinst isn't in control.tar.gz either [01:50] yes, it is. [01:50] after you compile the source package, anyway [01:50] which is the point [01:50] I don't have a source package. [01:51] however, it's 1:50am, and you clearly already know best, so i'm going to bed. [01:51] I'm using DEBIAN folder and usr folder. [01:51] oh, argh, do you literally mean DEBIAN, uppercase? [01:51] literally. [01:52] Are you seriously attempting to MANUALLY assemble a debian binary package without a source package being involved? [01:52] There is no source. [01:52] it's all binary blobs to begin with. [01:52] a source package doesn't mean source. you think sun-java6 includes source? [01:52] or nvidia-glx? [01:53] directhex@mortos:/tmp$ ls sun-java6-6-10/ [01:53] debian jdk-6u10-dlj-linux-amd64.bin jdk-6u10-dlj-linux-i586.bin [01:53] that's a source package [01:53] well , this one doesn't. [01:53] wheeeeeeee [01:53] bed. [01:54] terli: You are seriously making things hard for yourself. [01:54] I don't know what you mean by easy. [01:54] I don't know the correct way to build the package, because I don't have the requisite files. [01:55] Well, you seem to be ignoring the minimal example I gave you, and the example of sun-java, and trying to forge your own way by a different path [01:55] I didn't ignore your example [01:55] I couldn't use it [01:55] waitr [01:55] why not? [01:55] you didn't give me a different example from my own! [01:55] my list of files IS identical to yours. [01:55] I put the script in preinst [01:55] I never had a DEBIAN [01:55] and now I just have templates and preinst [01:56] well I don't know where you put it then [01:56] THIS package's source had a debian. [01:56] debian/ and DEBIAN/ are *different* [01:56] I mean DEBIAN. [01:56] You shouldn't be playing in DEBIAN/ [01:56] I don't know the difference. [01:57] it's a *debian* package, anyway. [01:58] Well, either read up about it in the Debian Policy Manual, or just trust us that if you are messing with a DEBIAN/ directory you're doing it wrong [01:58] you want to generate a SOURCE PACKAGE. christ, we could be done by now. a SOURCE PACKAGE does not require source for parallels, before you start, it just means a rational sane way of building all the required files in the right places using a plethora of helper tools. a source package consists of whatever you have from upstream, plus a debian/ folder. that's it. [01:58] well, I've gotten far enough that the installer is saying, template parse error near `text` in stanza #1 [01:58] well, directhex, I don't know how that works. [01:58] can I just rename DEBIAN to debian? [01:58] terli, if you wanted to zip a file, would you use a zip tool, or a hex editor? [01:58] Nope, you can't [01:59] if I wanted to zip it I would use zip [01:59] or tar [01:59] terli, right now you're using vi. [01:59] I used dpkg -b to generate my package from the folders specified in the wiki... [02:00] I tried to use data.tar.gz and control~, really I did, but I couldn't get it to make the . folders. [02:01] if you ever see data.tar.gz or control.tar.gz, ur doin it wrong. those files are generated as part of package creation, by dpkg-buildpackage [02:01] terli: What wiki are you reading? You've apparently found your way into some deep dark internals documentation [02:01] I did. [02:01] * Hobbsee reads the backscroll, and raises an eyebrow [02:01] What, only one eyebrow? :-) [02:01] I had to open the archive in archive manager, extract it, then edit the contents. [02:01] * directhex lowers Hobbsee's eyebrow with his poking stick [02:01] then re-compress it to a archive. [02:01] * directhex falls off his desk [02:01] UR DOIN IT WRONG [02:01] * Hobbsee raises the other eyebrow at that [02:02] well, no wonder it isn't working. [02:02] I don't know what other way there is, I wasn't taught anything about any of this. [02:02] (you silly people. i would have used Hobbsee's stick to zip.) [02:02] and it works fine, I just can't get debconf to parse my note/eula thingy [02:03] did you know, before I was using their archive(the official one) , and having to manually replace sh with bash because the maintainer screwed up and made his script call /sh instead of bash [02:03] terli: You are poking about in internals that developers don't bother to deal with directly. I (and probably most others here) have no idea how the templates are supposed to be packed into the binary deb, because we use tools that do it for us [02:04] Oh [02:04] maxb: I have no idea where the tools are. [02:04] but if you don't know how [02:04] then something has gone seriously wrong [02:04] Finally it becomes clear *why* you're approaching it this way [02:05] It was not clear to me until just now that you were actually trying to repackage an existing .deb file [02:05] exactly! [02:05] dpkg-repack - puts an unpacked .deb file back together [02:05] * maxb headdesks [02:05] I wrote my own one of those too [02:05] sorry, I've already deleted it. [02:06] * maxb wonders just how much of his ~/bin/ has packaged equivalents, if you know where to look :-) [02:06] anyway, it's clear to me that in one way, I clearly am superior to you(I'm fooling around with sith bathrooms in the debian temple), and in another way, I havn't the vaguest idea what I'm doing. [02:07] db_input critical license [02:07] db_input critical license_accepted [02:07] db_go [02:07] db_get license_accepted [02:07] terli: dude, take the attitude elsewhere. [02:07] * maxb awards terli the ultra-bizarre-metaphor prize [02:07] Hobsee; I don't have the vaugest idea how I'm supposed to package this without your help. [02:07] terli: there is no reason anyone *has* to help you, and the more you show your attitude, the more likely it is that you'll be ignored. [02:08] terli: well then, you should clearly try to be polite, stop telling people to be quiet like they're your dog, listen to what they say, whether you agree with it or not, and stop saying you're superior, no? [02:08] yesno, yes, ok. [02:11] I'm waiting. [02:12] Well, it's past time for me to be asleep, so don't wait for me. [02:12] oh, and maybe you already wore out your welcome with your comment on being clearly superior, so you may find that no one's going to deal with you anyway [02:13] so, wait away. [02:13] Well, you did see me admit in one and the same sentance that I don't know what I'm doing. [02:13] Not knowing what I'm working with is humiliating and frustrating. [02:32] I've worked on a project to use aufs as the root file system. I have posted the howto, https://help.ubuntu.com/community/aufsRootFileSystemOnUsbFlash [02:32] At this point I'm stuck. I think it should be integrated into the main init script. I think this would help the people building netbook distro's. [02:36] Is anyone working on netbooks or the init process. [03:00] Hi all [03:05] nschembr: I realize I'm being incredibly obsessive, but: There's no such thing as Ubuntu 8.4 [03:05] Sorry, personal irritation of mine :-) [03:32] maxb: Attention to detail noted and corrected. :) [03:35] I'm not sure how this fits with netbooks. Surely having your documents vanish when you reboot is a bit of a pain [03:40] Actually, none of the Ubuntu Mobile flavours use stacked filesystems post-install anymore. [03:40] The live session uses a stacked filesystem. === asac_ is now known as asac [04:35] cody-somerville: ping, I updated bug 297019 [04:35] Launchpad bug 297019 in mythexport "mythexport relies on ffmpeg from medibuntu" [Undecided,Fix released] https://launchpad.net/bugs/297019 === bluesmoke is now known as Amaranth === asac_ is now known as asac [06:30] good morning [06:37] Godd morning dholbach! :-) [06:37] hiya fabrice_sp [06:42] Morning dholbach, fabrice_sp. [06:42] hiya iulian [06:44] 'morning iulian ! [06:57] morning dholbach, fabrice_sp, iulian o/ [06:57] hiya didrocks :) [06:58] dholbach: just to keep you inform all is going very well for localized UDW preparation on your last day proposal :) [06:59] didrocks: perfect - can you follow up on the mail thread so the others know as well? [06:59] didrocks: I'll start writing up the announce in a sec [06:59] I wasn't feeling very well yesterday evening, so I do it today [07:00] dholbach: I will do it later in the week-end (we are fixing the precise schedule) [07:00] dholbach: oh :( do you feel better today? [07:00] yeah, it's all good today :) [07:01] didrocks: I was just thinking that the others might want to plan something similar so let them know about the plan early [07:01] dholbach: ok, will do it today :) [07:01] super [07:01] thanks muchly [07:02] you're welcome :) [07:12] Hi didrocks \o/ [07:31] hey guys. where can i see the queue / status for the build servers building jaunty stuff? [07:35] stefanlsd: https://launchpad.net/ubuntu/jaunty/+builds [07:35] iulian: thanks! === warp10_ is now known as warp10 === hyperair1 is now known as hyperair === hyperair1 is now known as hyperair [08:40] greetings, all! [08:48] persia: there? need some guidance. [08:48] Here. [08:49] Again, I like *contentful* pings :p [08:49] * directhex pings persia with a large textbook === hyperair1 is now known as hyperair === thekorn_ is now known as thekorn [09:00] Heh === warp10_ is now known as warp10 [09:01] persia: Btw, just out of curiosity. Will you have some spare time to vote on my application this weekend or even today? [09:03] iulian, I'm nearly done with review: I was hoping to finish yesterday, but didn't quite. Should be in the next few hours. [09:04] persia: Oh, that's great! [09:04] Thanks. [09:05] Sorry it's been so long: there's really no excuse on my part. [09:05] No worries. [09:30] persia: The problem is related to batik. The current version in Ubuntu is 1.7.dfsg-0ubuntu1. Where as Debian version in experimental is 1.7-1. According to dpkg 1.7.dfsg-0ubuntu1 > 1.7-1 so I will not be able to upload a merge from Debian. :-( [09:31] Do the orig.tar.gz files differ? [09:32] persia: yes they do. [09:33] persia: our version has some code from fop as well (inline with what was done for 1.6). But that code is not needed any more. [09:34] The main problem is how the buildd will treat the version from Debian. If it think that it is less than what we currently have then it will never get published. [09:34] OK. Then create 1.7.dfsg1-1ubuntu1. [09:34] It's not ideal, but it's probably what we need until 1.8 comes out. [09:35] persia: right. That is what just came to my mind. This problem will go away when we have 1.8. [09:35] Right. We just need to do some fakery to force use of the Debian orig.tar.gz, as I presume it's better. [09:35] If nothing else, it makes it easier to merge later. [09:38] by the way do we have any set policy for/against sync/merge from experimental? Debian experimental has fop 0.95. And it is unlikely to enter unstable until lenny is released. [09:39] slytherin, not just permitted, but more or less required during the lenny freeze [09:39] slytherin, most of the jaunty mono stack is smerged from experimental [09:40] directhex: good then. I am wondering when will be lenny released. [09:40] slytherin, it'll come bundled with copies of duke nukem forever [09:41] :-) [09:43] hehe [09:45] Best way to help lenny release is to help close the remaining RC bugs: they may well be bugs in Ubuntu as well, so it's not wasted effort. [09:45] i think i proposed a debian BSP a few months ago, but nobody seemed interested [09:46] I thought we did one a few months ago: I remember a number of Debian bugs being discussed. [09:46] Let's do another one this weekend. [09:46] directhex: Break off from the u-uk bug party in feb to do some debianing [09:47] :> [09:47] there's a u-uk? [09:47] but of course [09:48] we are planning on participating in the global bug jam [09:48] I expect to see you there! [09:48] is there a secret handshake involved? [09:48] or masks? [09:48] The secret incantation is /join #ubuntu-uk [09:49] if you can manage that then.... [09:49] onto the initiation rituals [09:49] mmm, sounds hard. never mind then :/ [09:49] boo [09:52] is it a sign i'm getting old if a major factor in which brands i buy for a new pc is the warranty? [09:53] No, it's a sign that you've reached the threshold of your performance requirement: you no longer expect it will be cheaper or better to get a new machine when that one breaks. [09:54] i get dell with 3 year next business day warranties. i cant be bother to go fix get parts to fix and replace hardware anymore [09:54] stefanlsd, the price diff between a dell with 1year rtb, and a self-build with a few lifetime warranties, on the spec i want, is already a few hundred quid [09:56] mm. sounds alot. but if u think of the time and effort lost with self built. if u have the time, i guess self built will be easier [09:56] well, cheaper anyways [09:58] it's a hobby! [09:58] * directhex looks at power supply warranties [10:02] In addition to being a hobby, once you have a base set of components, you only replace some bits. Frequently if you get a replacement for a failed component after a couple years, they send you a better one. [10:03] * directhex sets a 3 year psu warranty as a minimum [10:03] it's always the psu that pops [10:04] is there a way to make bzr diff -> quilt diff patches? [10:04] Sure. [10:05] stefanlsd: quilt has `quilt import' [10:05] Just extract the bzr patch, start the quilt patch creation process against the pristine source, apply the bzr patch, and save the new quilt patch. [10:05] Oh, that's even better :) [10:05] oh cool. thanks. will check out quilt import [10:06] i do all my merge stuff in bzr. makes life soo nice [10:06] * Laney is old school [10:06] I await james_w's session with baited breath [10:07] quilt import ffmpeg-headers.diff [10:07] Importing patch ffmpeg-headers.diff (stored as ffmpeg-headers.diff) [10:07] wow. and thats it. fixed the paths and everything [10:07] quilt is magical [10:08] hm, 5 year warranty from corsair. that's good. but they're uncheap [10:15] hm. /me reduces selection to choice of 8 [10:18] directhex: by the way, how much space is saved on CD now with the mono 2 transition? [10:18] slytherin, not enough yet, as not all libs have been done [10:19] a few meg afaik [10:19] directhex: what is your estimate? [10:19] the estimate was ~18 meg [10:19] current figures... dunno, slangasek might though [10:40] hello. could you tell me, please, if ./configure shows " not found but required" message, then should be added in Depends, or in Build-depends or in both fields in debian/control file? [10:42] ia, Typically, Build-Depends. It ought go to Depends from the ${shlibs:Depends} line. [10:42] persia: ok === james_w` is now known as james_w [10:43] sebner: are you on beautifulsoup? [10:44] james_w: I wanted to do it next but have also other things to do if you want to take it [10:44] sure, I can do that [10:44] you were quicker than me on all the others, so I wanted to check [10:44] heh [10:45] james_w: I'm sorry :) [10:45] sebner: you better be :-p [10:46] sebner: I smiled at you requesting the contributor give reasons for the sync at this stage in the release cycle :-) [10:48] james_w: well, people learn and grow up [10:48] james_w: link please! :) [10:48] james_w: but most times I checked upstream myself and since there were no heavy changes I ACKed. Besides FF if far away :P [10:48] DktrKranz: nah :P [10:49] sebner: yeah, there's no problem at this point [10:50] james_w: btw, we should add a note to the wiki -> Future MOTUs beware! There will always be wars and brutal fights for ACKing sync requests :P [10:50] heh :-) [11:06] I'm packaging my own program, how should I handle translations of the package description? [11:07] debian/control? don't. [11:07] hi room! I'm trying to do a deb package. do you know where can I find a good tutorial to do it? I read some tutorials but all show diferents way and for smal applications too. [11:08] directhex: how will it get translated then? [11:08] sll: which tutorials did you read? [11:09] sll: what are you trying to package? [11:09] Flimm: why do you need to translate them? [11:09] Flimm, i wasn't aware debian/control supported translation [11:10] Well the Debian Description Translation Project seems to translate the descriptions somehow. http://ddtp.debian.net/ [11:12] But it seems like the translations are only in the repo, not in the packages themselves [11:13] seems that way. they ARE apparently supported in jaunty though -> http://archive.ubuntu.com/ubuntu/dists/jaunty/main/i18n/ [11:16] Ok, another question, what's the best way to package po translations? [11:20] vorian: hi, I'd like to clarify what license your manpages for ncpfs are under [11:22] ok, I read spanish tutorials, and I think the problem is that I don't know about rules file. I'm trying to package this sources http://hangar.org/wikis/lab/pd/pdvjtools/src/ [11:23] they are externals/objects for puredata [11:41] can you recomendme any tutorial about deb packaging? [11:41] !packagingguide [11:41] The packaging guide is at http://wiki.ubuntu.com/PackagingGuide - See https://wiki.ubuntu.com/UbuntuDevelopment/NewPackages for information on getting a package integrated into Ubuntu - Other developer resources are at https://wiki.ubuntu.com/UbuntuDevelopment - See also !backports [11:42] . o O ( how is backports related to the packaging guide? ) [11:43] ubottu: thanks :P [11:43] Sorry, I don't know anything about thanks :P [12:32] james_w: heys. i think we can try sync imview again [12:32] is it built everywhere? [12:32] where would i check that exactly? [12:33] i've checked here - https://edge.launchpad.net/ubuntu/jaunty/+builds [12:33] for Debian? buildd.debian.org [12:33] (I guess it's for Debian as you're talking about a sync) [12:34] https://edge.launchpad.net/ubuntu/+source/imagemagick/7:6.4.5.4.dfsg1-1ubuntu2 [12:34] so...yes [12:34] pochu: it was a dep we were waiting for. [12:34] cool [12:40] stefanlsd: still fails to build for me [12:41]