[00:00] imbrandon: just do it ;) [00:00] azeem, :P [00:00] imbrandon: I think the policy should be "don't". They're probably on mentors for a reason. [00:00] bah, mentors can't be any worse than apt-get.org [00:00] or REVU ;-) [00:00] :) [00:00] I have a couple on mentors that I'll probably upload [00:00] Updates to pybliographer and valknut since the maintainers aren't responding :-( [00:01] bddebian, yea i was just looking over awn on mentors, looks solid [00:01] crimsun: We fortunately don't do apt-get.org any more, do we? [00:02] We just have to live with the hundreds of packages we collected from there. [00:02] Fujitsu, on request we'll sycn from any apt repo iirc :P [00:02] i dont think any are auto anymore though other than maybe debian-multimedia.org possibly [00:02] debian-multimedia isn't automatic. [00:02] Fujitsu: thankfully not. [00:03] although I haven't followed the getdeb thread(s) closely... [00:03] Only Debian main seems to be. [00:03] crimsun: Yes! Let's sync from there. [00:03] mm horkage [00:04] we cant , they dont use a repo, all http + apturl / gdebi :) [00:05] ok let me put it this way ..... if i grab a package from mentors, look it over localy and am satisfied with it, upload it to REVU for a bit of peer revu does anyone see a problem with me uploading prior to debian ( with proper versioning so when it does hit debian its taken in etc ) [00:05] wow, to achieve the things that Fedora 8 has done with PulseAudio, we'll have to revamp debian/*. [00:06] crimsun: Why? [00:06] i know thats not the NORMAL way but i dont want to wait on a long debian processes at this stage [00:06] sweet, get to work ;-P [00:06] imbrandon: personally, no, though the tension between waiting for the Debian upload so that you can sync and simply uploading to Ubuntu would be considered... [00:07] Fujitsu: quite a lot of the "nice" things (session management, stream control for devices, etc.) are spread across main and universe packages, and the build-dependencies themselves have the same issue [00:07] hrm maybe i'll just get in contact with a DD and see how quickly i can get it pushed in [00:07] Yeah because Debian is a PITA in many cases :-( [00:07] atleaste first [00:08] crimsun: Ah. Sounds like a mess. [00:08] bddebian: when? :P [00:08] just a small, conquerable mess. [00:08] imbrandon: apply for NM and become the "fast-sponsoring-DD" ;) [00:08] imbrandon: or ask Fujitsu once he gets through :) [00:08] white, or me :) lol [00:09] imbrandon: you are DD? lol [00:09] i have 2 and a backup i normaly chat up first, one is super busy atm i learned earlier, maybe i'll have some luck with the others :) [00:09] joejaxx, someday [00:09] oh ok [00:09] lol [00:09] imbrandon: poke some in this channel, mainly siretart :) [00:10] white, he's the backup :) [00:10] hehe [00:10] i always said i would probably get through the debian developer process than ubuntu motu [00:10] lol [00:10] faster that is [00:10] lol [00:10] That could only be if you WANTED to do that [00:10] since the DD process is slower [00:10] heh i doubt that, i became a MOTU + core-dev rather fast IMHO [00:11] I tend to agree with Colin (kamion) in that regard; it's likely more a "people" thing than a "project" thing. [00:11] with reguard to DD's ? [00:11] so atleast with DD i would know what to do in order to get it [00:11] well, the rate at which one becomes NM, DD, etc. [00:12] motu seems to be quite different [00:12] lol [00:12] crimsun: Where there's a whole lot of "people" in that "project" ;-) [00:12] it's pretty unfair to say that Ubuntu/Debian is faster than Debian/Ubuntu [00:12] Fujitsu: what takes you so long btw? [00:12] what about DM. This is almost in place [00:13] white: Erm, studying. I haven't done all of T&S, and really don't have the time at the moment. [00:13] crimsun: i almost rather do the second [00:13] In a couple of weeks exams will be over, and I should be able to finish it quickly. [00:13] Fujitsu: go offline and study, i told you ;) [00:13] crimsun heh well yes it is quite unfair , but when you see NM on planet.d.o complaingin about getting approved and still waiting 2 years for accounts its kinda hard not to make that distinction in your mind [00:13] DM seems like a Really Good Thing. [00:13] err, I just thought of Baldur's Gate. [00:13] imbrandon: That sounds about right [00:13] crimsun: haha :P [00:13] hehe DM [00:14] imbrandon: patience is a virtue ;) [00:14] I hate to think how long something like that would take for us to implement... it would mean a change in Soyuz, which could take years [00:14] And years. [00:15] white, i have all the patients i need imho, but a process that takes 2 years imho is a tad long, hell i've only been arround ubuntu 2.75 years and only been a MOTU + core-dev for 1.5+ [00:15] or so ( without looking up dates ) [00:15] :) [00:15] anyone know if Russell Coker was at uds [00:15] ? [00:15] if he was i missed a grand opportunity :\ [00:15] porbably [00:15] I don't believe so, but I was only there one day [00:16] imbrandon: Well you might have found a sponsor by then.. [00:17] bddebian, hehe [00:17] sponsorship i have no issues with NORMALY in debian, its the rest of the stuff [00:17] imbrandon: you want to vote? [00:17] white, on? [00:17] oh don't tell me that you wanna read -private, it is not worth it [00:18] imbrandon: you mentioned the "rest" [00:18] white, nah, honestly the main reason i would become a DD would be 2 fold, access to porter machines and direct uploads, the politics dont intrest me much [00:18] well it comes with it :) [00:19] oh definately, just wouldent be a "reason" [00:19] and the porter machine thing ... well [00:19] Yeah, don't you want to feel like you're better than everyone else? That's a great motivator! :-) [00:19] it can take quite some time, until a package is installed in a particular chroot, if you do not have the right uplink to DSA [00:20] bddebian: unfortunately, people might start feeling that, when i look at some bug report communication :/ [00:20] Or hang out on #d-d [00:21] that and personaly it just feels right to contribute farther and farther upstream and almost every other project i'm involved with debian is the base , *Ubuntu , gnusolaris / indiana , Debian Live , pkg-kde etc [00:21] white, ^ [00:22] bddebian: is that an open channel? [00:22] joejaxx: Depends on your definition I guess [00:22] i remember i got flack for joining one of the debian channels [00:22] That's probably it :-) [00:22] and they did not bother putting a passkey onit [00:22] joejaxx: everybody can join there [00:22] on it* [00:23] joejaxx, its open as long as you dont interupt kinda like #u-d [00:23] unless you're a DD or an NM, talking nonsense without a realname might not be tolerated [00:23] imbrandon: nope i just joined [00:23] debian-private is a closed channel, the rest should be open [00:23] no interrupting or anything [00:23] if you want a channel no public put a key on it [00:23] lol [00:23] joejaxx: maybe you tried here and not on oftc [00:24] non-public* [00:24] azeem: nope it was on oftc [00:24] *shrug* [00:24] i only join debian channels on iftc [00:24] oftc* [00:24] someone probably just had bad wheaties, because i'm not saying there arent DD's with a vendeta against Ubuntu but 99% i have talked to are normal humans too and even some do DD work on ubuntu [00:24] :P [00:25] :P [00:25] right, that's the "people" thing that Colin mentioned at fosscamp [00:25] I'm not bagging all DDs by any stretch. We have several in Ubuntu [00:25] I'm pretty sure nobody got kicked off #d-d just for being a ubuntu guy without saying anything [00:25] just ike there are bad apples in ubuntu too, every community has their share [00:25] "normal humans" rotl :) [00:25] rotfl [00:25] haha [00:25] white, hahaha yea i laughed at that after i typed it [00:26] you know my point though :) [00:26] azeem: no i think it was an infrastructure channel or something [00:26] :) [00:26] imbrandon: lol [00:26] joejaxx: that's something else then [00:27] reminds me i gues i took #ubuntu-debian / #debian-ubuntu from oftc off my autojoin , hrm [00:27] There is such a channel? [00:27] bddebian, yea [00:27] bddebian: i was just about to ask that [00:27] lol [00:27] i forget what way arround it is, but yea [00:27] It's very small. [00:28] I'm so surprised ;-) [00:28] ajmitch is there too iirc alot, its a small ~10 people group [00:28] * joejaxx tries to join [00:30] * bddebian is bitter atm :-( [00:30] bddebian ? [00:30] too much Halloween candy, I think. [00:30] bddebian, i got my NES controller to work on a PS/2 port last night, finaly i have a use for those arn things [00:30] darn* [00:30] crimsun hahaha [00:31] imbrandon: Nice. No, just struggling getting things done over there. [00:31] only problem is i mapped the [b] button to capslock, lol so when i'm done playing i have to check the state of my capslock key [00:32] Oh, I got Witcher today so I may disappear for any of Hardy ;-P [00:32] heh [00:32] no NWN2 for you [00:32] Already have that, thought I want to get Mask of the Betrayer [00:32] -t [00:33] having to use an old 200 mhz box stops most game playing cept for NES roms [00:33] yeah, that would do it. I bought a whole new Dell XPS for NWN2 ;-) [00:33] heh [00:33] StevenK: are you going to do the flwm merge? :) [00:34] i started testbuilding big apps on my webserver its gotten so slow [00:34] lol [00:34] lol [00:34] xen! [00:34] :) [00:34] or some other virtualization platform [00:34] joejaxx, hell my webserver is only a 1.5 ghz Via C7 :P [00:34] :) [00:34] oh lol [00:35] http://www.imbrandon.com/phpsysinfo/ [00:35] not a workhorse by anymeans [00:36] Processors 1 [00:36] Model VIA Esther processor 1500MHz @ [00:36] CPU Speed 1.5 GHz [00:36] how can I do something as root in debian/rules? I need to untar the linux-source and therefore I want to add a "cd /usr/src/ && sudo tar -xjf linux-source.tar.bz2" (yet I do not know the sudo password for a pbuilder chroot [00:36] imbrandon: second hard drive a Seagate ? [00:36] My computer is a 333mhz [00:36] SWAT: Ewwww. [00:36] joejaxx, yea [00:36] SWAT: :\ [00:36] SWAT: you don't want that [00:36] imbrandon: :D [00:36] joejaxx, 500GB segate [00:37] :D nice [00:37] ok, what _do_ I want? [00:37] SWAT: you want to do stuff only in your source [00:37] * Fujitsu notes that his trusty PIII 800MHz router doesn't like running lintian over everything all that much, but sorta copes. [00:37] SWAT: you can depend on other packages or include the source [00:38] `cd /usr/src/ && sudo tar -xjf linux-source.tar.bz2` sounds ominous [00:38] joejaxx, that 500GB is where my local mirror is ( among some ISO's and such ) [00:38] imbrandon: nice [00:38] :) [00:38] little background: I want to create my own ALSA package, since I want to do this the 'nice' way, since I have sound-issues with the repo-ALSA. afaik, ALSA requires the source-tree [00:39] SWAT: err, /which/ ALSA package? [00:39] kernelspace [driver]? userspace? [00:39] the driver should suffice. I just want to check if it's a downstream issue, or at least, what's going on. [00:39] gutsy already ships 1.0.15rc3 in linux-backports-modules-$(uname -r) [00:40] SWAT, you might look into l-u-m and how it does it ( or l-b-m ) if your messing with kernel drivers [00:40] yea [00:40] Is the driver in LUM yet? [00:40] snd-hda-intel is in lum for gutsy [00:40] If so, it should be easier and quicker. [00:40] I think I saw a spec about moving the whole lot over there. [00:40] for hardy, alsa-kernel is going into lum wholescale [00:40] Aha. [00:40] yeah, I spoke with Tim and Ben about it at UDS [00:41] SWAT: if you wish to track hg, you can look at the alsa-driver source package [00:41] I have heavy background noisy on my audigy2 and I tried all the tips of the ALSA people. I want to try if compile/installing ALSA from source will help, but ./configure&&make&&make install ALSA is a bit 'nasty' imho [00:41] SWAT: it's fairly straightforward to modify the generated alsa-source binary package to contain the current hg tip [00:42] you can then use alsa-source along with module-assistant to create a 'legit' alsa-drivers binary package [00:43] I've already taken a look at the source packages from the repo. I'll take a look at the hg tip then [00:43] you can't use the source package unmodified [00:43] you'll have to checkout hg tip, copy the bits into the source package, regenerate the source package, and build the new alsa-source [00:45] Does "background noise" refer to bus noise (presuming you mean a PCI version)? [00:45] I really can't use an unmodified source? I'm trying to see if it's a downstream issue [00:45] well, you have a choice if you want to do it the Debian/Ubuntu way [00:46] PCI, yes, onboard sound disabled, no interrupts, even changed the PCI slots etc. (I did my homework) [00:46] if you simply want hg tip, you don't need it [00:47] When I attempt to build catfish in sid pbuilder, I get: [00:47] Copying source file [00:47] -> copying [catfish_0.3-1_source.changes] [00:47] -> copying [./utils] [00:47] cp: cannot stat `./utils': No such file or directory [00:47] are you attempting to pbuild the .changes instead of the .dsc ? [00:47] ok, thanks for your tips. I'll enough to do tomorrow then [00:47] yes [00:47] lol [00:47] SWAT: also, check in your bios if you have pci masking options. [00:47] somerville32: that would do it. [00:50] heh [00:50] does ubuntu only allow source only uploads or also source+binary uploads? [00:51] source only [00:51] As it should be ;-) [00:52] yea source only , makes a cleaner archive imho [00:53] since I started trying to build 'nice' packages (very recently), I noticed how hard some packages are to create (from source, like ALSA, firebird etc.). I've gained a deep respect for the MOTU's (you are doing a great job) [00:53] harder to bootstrap things, but thats only a corner case [00:54] SWAT, heh thanks, you should try and join the ranks someday seems you have the drive to doso [00:54] (alsa's fun!) [00:54] :] [00:54] right, coffee shop's closing. 'night, folks. [00:54] gnight crimsun [00:54] :( [00:55] Night [00:55] Gnight crimsun_ [00:55] Doh [00:55] imbrandon, maybe, small steps for now. I will create some (private) packages from scratch first and then I'll look how that turns out. [00:55] :P [00:56] hello minghua [00:56] Hi joejaxx. [00:57] :) [00:57] Woot [00:58] anyone here use epiphany ? [01:00] Do I need to subscribe to debian-mentors to send my RFS? [01:00] imbrandon: i do/used\ to [01:00] imbrandon, sometimes, why? [01:00] on debian unstable [01:01] how the hell can i stop it from keeping aq history in the addr bar [01:01] i looked everywhere [01:02] I guess you've just found a new feature that you want [01:02] lol [01:03] ugh [01:05] http://mentors.debian.net/cgi-bin/sponsor-pkglist?action=details;package=catfish :) [01:05] somerville32: No. But please state you are not subscribed in the email if you aren't. [01:05] Oops. too late. :/ [01:06] I misspelled the domain so I get a second chance. [01:06] imbrandon: I've used Epiphany forever... but aren't you a KDE person? [01:06] Fujitsu, yea, i'm just tinkering with some stuff [01:06] sides i still use FF in KDE [01:06] :P [01:06] hi [01:07] * Fujitsu must try epiphany-webkit again soon. [01:07] imbrandon: http://bmk789.blogspot.com/ i borrowed your title/subtitle if you dont mind :P [01:08] bmk789 haha no i dont mind, i change them every few months === bmk789_ is now known as imbrandontoo [01:08] imbrandontoo: lol [01:08] hahahaha [01:09] You might just give a link to that long e-mail instead of reposting the entire thing. [01:09] that might be a bit confusing [01:09] imbrandontoo, ^^ === imbrandontoo is now known as bmk789 [01:10] good idea, ill make it link [01:10] bmk789, actualy thats kinda a cool nick, i wish i would have registed it now LOL [01:10] lol [01:11] bmk789, just do me a favor and dont try to impoersonate me , if it becomes a problem i might not be so happy bout it :P [01:11] not that i think it wlll , just fyi [01:11] * joejaxx did the flwm merge and needs to see if stevenk is going to do it [01:11] o no problem [01:11] s/the/s/g [01:11] skdflksdjflk [01:11] s/the/a/g [01:12] joejaxx, one word, PowerPC, kthxbi [01:12] :) [01:12] imbrandon: :P [01:13] in all serousness , need a machine to build the iso on ? its not powerfull by anymeans ( 333 w/96 mb ram ) but it will work in a pinch [01:13] nope :P bakery can build mutliarch :) [01:13] i really really want KDrive for it though [01:13] joejaxx, kk [01:14] joejaxx, arent you working on getting kdrive bins in the archive ? [01:14] Morning, persia. [01:15] imbrandon: yeah i already have permission to fo it :) [01:15] Good morning Fujitsu [01:15] imbrandon: i just have not had the time [01:15] i have tried but the X packages are trickier than the kernel ones [01:15] yeah [01:15] joejaxx, if there is anything i can help with the kdrive stuff lemme know, i've been contemplating that for 2 releases now [01:15] ok :) [01:15] What's so good about KDrive? [01:16] Fujitsu, it runs a full blown X server in under 700KB memory and no config files [01:16] great for older systems [01:16] Ah. [01:16] even conposit and such [01:16] I didn't realise it was *that* light. [01:16] yea its TINY [01:17] my microwave probably takes more than that ;-) [01:17] heh [01:17] LaserJock: ! :D [01:18] LaserJock: how are you doing ? :D [01:18] Nafallo: bug 159659. :) [01:18] Launchpad bug 159659 in irssi "Merge with Debian unstable - new upstream release 0.8.12" [Low,Confirmed] https://launchpad.net/bugs/159659 [01:18] LaserJock: You seem to be here rather frequently nowadays. [01:18] honestly i dont see why they dident use kdrive for bulletproof X [01:19] but thats neither here nor there [01:19] imbrandon: yeah [01:19] that would be interesting [01:19] pochu: what should I do with it? :-) [01:19] Any core-dev for sponsoring that? bug 159659. It's an easy one! [01:19] Nafallo: I'm afraid nothing, since you aren't a core-dev... are you? [01:19] Just to let you know :) [01:19] pochu: no, StevenK sponsored me :-) [01:19] pochu: ah. cheers :-) [01:20] StevenK: around? :) [01:20] so you need a sponsor or not for irssi ? [01:20] * Fujitsu wonders if he will be murdered for attempting to add a maintainer-is-not-quite-motu check to Lintian. [01:20] i'll look if StevenK hasent already [01:20] imbrandon: yes, I do :) [01:20] i need a sponsor! [01:20] Fujitsu, hehe [01:20] Fujitsu: Just so it [01:20] imbrandon: thanks. And he hasn't. I've reported the bug a minute ago :) [01:20] Fujitsu: doesn't that go with the whole oh-god-jdong-touched-the-package check? [01:20] jdong: ! :( [01:20] pochu / Nafallo , ok i'll look at it now [01:20] jdong: you disappointed us [01:21] :( [01:21] joejaxx: aaahhh [01:21] imbrandon: cheers mate :-) [01:21] joejaxx: sorry, it's been a hellish day [01:21] kdubois: Have you subscribed a sponsorship team to your patch? [01:21] Thank you imbrandon. [01:21] joejaxx: been sleeping for the whole day until now [01:21] jdong: lol [01:21] imbrandon: just fyi, I'm using it atm... works fine :) [01:21] joejaxx: grossly underestimated how long a 5-page homework assignment should've taken [01:21] * Nafallo looks at the diff as well fwiw [01:22] jdong: oh wow [01:22] persia: no, i mean i'm just trying to get started with motu and such, and need general guidance. I can code or bughunt.... [01:22] kdubois: Ah. In that case, you don't need a sponsor yet, you just need some work :) [01:23] Fujitsu: shhhh [01:24] lol [01:24] persia: yeah, more or less. what should i do to get started? i'd like to start on a problem thats challenging enough, but easy enough for me to do.... [01:24] kdubois: There are three kinds of work I usually recommend for new people: 1) Bug Triage, 2) Debugging Crashes, and 3) Fixing broken packaging. Do any of those interest you? [01:24] imbrandon, pochu: diff.gz looked fine to me. [01:24] 02:23 [Freenode] Nafallo [n=nafallo@ubuntu/member/nafallo] requested CTCP VERSION from pochu: [01:24] Nafallo: I said the truth! :-) [01:24] Nafallo, yea here too, just doing a quick testbuild ( no offence ) and testing [01:25] pochu, ^^ [01:25] then i'll upload it [01:25] debugging crashes sounds the best [01:25] cool :) [01:26] imbrandon: if you hadn't, you would've done wrong :-) [01:26] kdubois: OK. We have a tool called apport that helps users to file bugs when something crashes. These bugs usually have a stacktrace attached, to help dig through the code, and find the problem. [01:27] persia: They're also mostly private. [01:27] persia: where can i get these stack traces? [01:27] Fujitsu: Not all, and there's not a few people who can help make them public. [01:29] kdubois: https://launchpad.net/ubuntu/+bugs?field.tag=apport-bug is a good list with which to start. I recommend picking a package that you use, and starting with a universe package. [01:29] kdubois: Also, would you mind telling me how many bugs you see on that list? It may be that making them non-private is something that more people need to do currently. [01:29] * minghua didn't notice anyone working on making crash bugs public in the packages he's interested in. [01:30] Nafallo / pochu, ok seems good to go, bug updated [01:30] * Fujitsu used to go through recent ones and unprivatise them, but just hasn't had the time of late. [01:30] persia: i can see 504 [01:30] Heya persia [01:30] persia: Are you on Debian devel games ML ? [01:30] pochu: *hint* wait until it's built before you mail upstream :-) [01:31] kdubois: OK. If you run out, complain in #ubutntu-bugs (or here), and we'll get you some more :) [01:31] pochu: build and released actually ;-) [01:31] Nafallo: I'll do. And I'll CC you too :-) [01:31] pochu: cheers mate :-) [01:31] kdubois: Also, if you pick one, and need some help in the debugging process, please ask: I'm happy to provide a walkthrough. [01:31] damnit! [01:31] I'm getting all english quickly :-P [01:31] persia: #ubutntu-bugs? :) [01:32] bddebian: One of them. I need to sort my filtering system, and join the other. [01:32] persia: alright, will do. i think i'll start on one tomorrow.. If i were to find a bug in say, vlc or something, how do we get that fixed? do we talk to the vlc guys and have them fix it in their code base, or do we apply patches to whats in out repo's? [01:33] kdubois, somethimes both :P [01:34] kdubois: Generally, I work through the stacktrace, and make a patch which fixes the problem, after testing. It's nice to send the patch upstream to help everyone else. It's also nice to package the patch for a new revision of the package, and seek sponsorship to get Ubuntu fixed sooner. [01:35] ok i'm off for some beers with some friends, ttyl [01:36] imbrandon: have fun, and thanks again :) [01:40] superm1: :) [01:40] hey joejaxx [01:40] hello [01:40] didn't see you down for free food and such? [01:40] were you around? [01:40] yes but only briefly at the beginning [01:40] ah that would explain it [01:41] yeah [01:42] i didnt get as many people to sign my key as i should have while i was here :( [01:42] i tend to have nothing to do at those sort of things :P [01:42] oh well. [01:42] well atleast you have some :) [01:42] i have zero [01:42] :) [01:42] they weren't checking id's or anything..... so you could have joined :) [01:43] superm1: i mean socially :P [01:43] i only know a small amount of people in ubuntu [01:43] lol [01:43] and none of them were there [01:43] lol [01:44] and i rather not start a random conversation either lol it seems pointless and folly [01:44] oh i ran into several random people that i met down there [01:45] for a first time [01:45] yeah i do not do well with randon encounters lol [01:46] Please rereview pq. Everything should be ok now. http://revu.tauware.de/details.py?package=pq [01:49] Good night motuland! [02:28] i am curious, are there many cases, where there is an FTBFS in ubuntu? [02:29] one argument for debian do disable source-only uploads was that people do not test the builds beforehand [02:29] if that is true, it would of course put more workload on the buildd maintainer due to more bugs (or other people to fill FTBFS bugs) [02:30] if not, source-only uploads would of course be nice to have [02:30] script those bugs? [02:30] white: I would guess that there are fewer but we do have them unfortunately [02:30] Of course maybe they are all the syncs from Debian ;-P [02:30] :) [02:31] Nafallo: sure, but you want someone to investigate it and give some more information, then only a failed build log [02:31] white: depends on what's wrong I guess. [02:31] OK damnit, UlstraStar-ng is finding boost/thread/thread.hpp but fails to find boost/ptr_container/ptr_map.hpp. The darn file is there, wtf? [02:33] white: One thing we're trying to organise for this cycle is to have LP provide a list of all current FTBFS packages, so that developers could look at the list, and fix a few. Right now, we're a little blind as to what FTBFS today vs. what FTBFS at one point in the past. [02:34] persia: is there a overview for build logs, like debian has linked form the PS site? [02:34] LP has them [02:35] white: There's http://launchpad.net/ubuntu/hardy/+builds, and you can check logs by build status, but it's not quite there yet. [02:35] white: LP has every build log ever, linked from the package page. [02:38] ah ok [02:38] ... except for if something is given-back, then the old log is erased. [02:39] Ah ha, libboost-dev build-deps are fsck'd [02:39] bddebian: As normal. [02:39] Fujitsu: I've never dealt with them before [02:41] Damnit, 1.1a patch for The Witcher is 113Mb :-( [02:41] 113Mb??? [02:41] egads [02:41] s'ok, it's only about a 10min download :) [02:42] OMFG: boost/program_options.hpp... no [02:43] lol [02:45] boost/program_options.hpp... [02:45] * bddebian smacks himself in the head with his laptop [02:46] bddebian: Be careful: you might hurt your laptop :P [02:47] Ouch :'-( [02:47] :-) [02:47] :) [03:01] Uh oh, The Witcher starteth.. [03:02] bddebian: is that a good game [03:02] ? [03:03] You have got to watch the trailer. It is one of the coolest trailers for a video game I've ever seen [03:12] psh [03:12] i'll just wait for the Witchest [03:12] pwnguin: lol [03:12] heh [03:12] alternatively, im thinking I'll have to have someone REVU this package i've got [03:13] cartoon nethack [03:13] (remember egoboo?) [03:13] Damn, how many versions of nethack can there possibly be? [03:13] ok [03:13] cartoon diablo [03:14] lolo [03:14] bddebian: Even nethack.org is a fork of a fork of a fork of a fork of a fork of a fork (if I remember correctly). [03:15] http://youtube.com/watch?v=OxlEPsfrKJ8 [03:16] pwnguin: egoboo has been resurrected [03:16] * persia cheers [03:16] Burgundavia: well, nobody's packaged this [03:16] Burgundavia: understandably, because nobody's quite sure htf to build it [03:17] hehe [03:17] isn't soulfu non-free? [03:17] theres an itp [03:18] and the liscense claims GPL [03:18] the data i donno about [03:19] theres a sourceforge project [03:19] right, soulfu just became free [03:19] it didn't used to be [03:19] well then [03:19] huzza [03:20] a bit CPU intensive [03:20] There's an ITP or RFP? [03:20] itp [03:20] hmm [03:20] its been like 60 days [03:20] A bit long. [03:21] I've got enough broken games in the games team svn to worry about for the moment :-( [03:21] and the ITP is debain games [03:21] hah [03:21] i didnt know a team could ITP without having someone in particular [03:21] team maintenance is a good thing [03:21] sure [03:21] Somewhat [03:22] It carries it's own issue [03:22] +s [03:22] as long as the team actually maintains it [03:22] instead of assuming someone else will do the job [03:22] Hmm, I don't see it in our svn yet [03:22] * persia votes for a bddebian-special package in SVN [03:23] hah [03:23] Burgundavia: do you have an authorative source on the "becoming free" part? [03:23] the source code appears GPL'd but im not sure about the data [03:24] the soulfu forums [03:26] the source is a bit strange. i think soulfu.c #includes all the other code [03:26] instead of the more usual linking .o files [03:27] and apparently it used an odd libjpeg [03:41] I'll try to check it out if I get some time soon [03:42] * bddebian wants to get updated scorched3d in [03:43] bddebian: Thank you :) [03:47] persia: Well don't thank me yet, I'm not having much luck getting a hold of people :-( [03:48] dang it, PPA needs download statistics [03:48] LaserJock: Umm.. **Ubuntu archive systems** need download statistics [03:48] pfft [03:49] who cares about Ubuntu? I just want to know how many people have downloaded *my* packages ;-) [03:49] Shit, how do I overwrite my local svn crap with that from trunk? [03:49] LaserJock: My motivation is the same, but my dput configuration isn't :P [03:49] bddebian: svn revert [03:50] persia: even though I haven't committed? [03:50] bddebian: Yep. If you don't pass a revision, svn revert defaults to the last checkout / update [03:51] bddebian: So, if you want a newer version to overwrite local changes, use `svn update .; svn revert path/to/badfile` [03:51] I can't just wipe the local files and svn update? [03:52] bddebian: wiping the local files is considered a change. It's no different than editing them from a revision control perspective. [03:52] grr, OK, thx [03:53] Hmm, I thought rpath was a no-no ? [03:54] bddebian: It's not preferred, but there are lots of leftovers. Making it work without is better. [03:56] bddebian: If you want to wipe it out, you need to wipe out the .svn dir in the parent directory and that can get all confusing. Revert is better. [03:58] ScottK: At that point isn't it better to just `cd ..; rm -rf workspace; mkdir workspace; cd workspace; svn co svn://my.project.url/trunk` ? [03:58] persia: Aye it worked, thanks [03:58] persia: Pretty much. [03:59] persia: Except if you've got a checkout of the entire Debian Python Modules Team svn and you really don't want to do that one over.... [04:00] ScottK: Ah. Is svn not smart enough to notice that if you delete a subdirectory (e.g. trunk/myproject/example), recreate it, and svn co .../trunk/myproject/example that it all belongs in the same place? [04:02] persia: IIRC the time I had to do that, I deleted the .svn for the parent and then did an svn add for it followed by rechecking out the module I'd borked. [04:03] * persia decides that svn state management is sufficiently opaque that `svn revert` is always the right solution. [04:18] Anybody here do much work on nm-applet's openvpn interface? It exists under universe/net [04:26] Nafallo: siretart: any of you MOTU-media folk wanna talk syncing x264 from marillat? :) [04:26] a LOT of really nice performance increases since our 6+month-old sync [04:32] Ok folks. Unless I see you tomorrow some time, I'll see you online once I return to Australia. [04:37] persia: Hi! [04:38] s1024kb: Good day [04:38] persia: shall i say "Good morning" or "Good afternoon" to you? [04:39] s1024kb: Good afternoon: it's about an hour later here than there. [04:40] persia: Oh, i see. :) I just upgraded my Ubuntu to 7.10 last night. [04:40] s1024kb: Great. I hope it continues to work as well for you. [04:41] persia: shall we developers keep our Ubuntu most updated to keep up with the pace of others? [04:42] s1024kb: Actually, I believe that having different developers on each supported release is best. If no developer is running 6.06LTS, it becomes difficult to support it properly. We need a mix of installations to best find and close all the bugs. [04:43] persia: did you mean a mixed installation in the same computer? Can do? [04:45] s1024kb: Some people dual-boot. I was more speaking of a mixed set of installations for all developers. [04:47] persia: I understand. [04:48] persia: so you have been a MOTU for a long time already? [04:49] s1024kb: Not really. I only became MOTU about six months ago. Previously I was a Contributor for a long time. [04:50] persia: what a Contributor is doing? Can i be a Contributor too? [04:51] contributers are people who do work but dont have upload privledges. instead an MOTU is supposed to review the work [04:52] s1024kb: Certainly. Becoming a Contributor is easy: just start doing stuff :) [04:52] pwnguin: ok, understand. so i should do something, and then hand them to one of the MOTUs to review them? [04:53] basically, yea. there's some tools to make this easier, and a schedule that makes it more likely to get attention if you follow it [04:53] persia: Still studying... wish that i can start earlier... all the books my teacher recommended me to read are almost finished... English is difficult to me. [04:54] s1024kb: Contributors usually start with small patches, but as skills develop can implement new features, handle library transitions, coordinate teams to accomplish goals, and even change how we do things (e.g. backports was started by a Contributor) [04:54] pwnguin: thank you for explaining it to me. [04:54] he [04:54] s1024kb: pay attention to persia more. he knows more than I [04:55] pwnguin: actually i take every contributor as my teacher, i hope to learn more from you. [04:56] freeflying: hi! Nice to meet you here! [04:58] persia: recently i am looking for a simple example to follow - i believe that when i can do a whole thing, no matter how simple it will be, i can understand more. [04:58] s1024kb: hi [04:58] there is a package called hello-world [04:58] the Ubuntu Packaging guide describes it in detail [04:58] s1024kb: Have you looked through https://launchpad.net/ubuntu/+bugs?field.tag=bitesize [04:59] pwnguin: yes, i have it now. but i was stopped in the last steps... i don't know if i should create the 4 important files manually. [05:00] s1024kb: which 4? [05:01] pwnguin: for example, postinst and prerm [05:02] they dont exist any more in hello, do they? [05:02] persia: yes, i see a lot of bugs there, but don't know how to do... shall i download one of the packages, build and install it, run it, modified it and so on? [05:03] s1024kb: Yes, although first make sure that the bug is not already assigned, and assign yourself to the bug. When you get a fix, prepare a patch, and submit a debdiff. https://wiki.ubuntu.com/MOTU/Contributing has some further hints [05:03] pwnguin: that's what confused me... so i can already build the package without making any change to it? [05:04] s1024kb: look at the changelog for the finished package [05:04] its no longer needed -- the documentation is just broke i think [05:05] * persia points out that a documentation fix tends to be an easy patch to develop [05:05] pwnguin: i will read the packing guide carefully again, hope to make a big step forward this weekend [05:06] s1024kb: sure. once you feel comfortable with it, then start looking at the bitesize list persia mentioned. dont start with big stuff, cuz just going through the process itself is a bit of work to learn === _nand_ is now known as nand_ [05:08] pwnguin: okay, thanks. sometimes i found that i am using a mind of a Windows user to understand Linux... yesterday i modified so packages of Delphi in Windows, i wish that was the same concept of our bug fixing here in Ubuntu. [05:08] im not familiar with delpha [05:08] delphi [05:09] so packages == translation? [05:09] pwnguin: yes. sorry, i meant some packages [05:09] s1024kb: Perhaps it is, but you'll do better to explain the specific modification methods, so we can share thoughts on possible differences. [05:10] pwnguin: some of our colleagues need to use some controls, they are in the packages, but no longer work in Delphi 2007. I modified them, repacked them and install them again. [05:11] s1024kb: same thing happens here [05:11] Gnight folks [05:11] but its helpful to know how a package is constructed first [05:11] the difference is that instead of modifying a binary (repacked?) you can modify the source [05:12] but since Univese will be given to lots of people, some peer review and quality control is a good idea [05:13] i had been developing Delphi components (VCL) in the past, so i quite understand how to make a VCL package. Yesterday when i was modifying the package, i thought - will it be the same in Ubuntu? [05:14] very similar [05:14] you grab the source package, modify it, build and test it, then push the source package back out to the community [05:14] pwnguin: now i understand more after learning for you and persia. I am happy that i understand a little bit more today. [05:15] the open source community tends to communicate these sorts of changes in patches (diffs) [05:15] thank you my friends. gotta go now because i must switch back to WinXP for my works... sigh... thank you very much and bye. [05:30] persia: hi, i am here again, from Windows (though i don't want to...) [05:30] pwnguin: hello again... [05:30] hi [05:31] pwnguin:i am happy to see you again... [05:31] pwnguin: now i am using Xchat. [05:32] neat [05:33] pwnguin: are you a programmer too? [05:33] yes [05:34] but not delphi [05:34] pwnguin: c++? a Linux programmer? [05:34] c/c++, java, whatevre. linux, embedded systems mostly [05:35] :-) embedded systems is interesting. I write single chip programmes too. [05:36] pwnguin: i guess that most of the contributors are programmers? [05:37] mostly. though i hear a few core developers can't code [05:37] pwnguin: Oh my God, hard to believe it. But how they work as core developers? [05:38] s1024kb: They are very active in documentation, image improvements, packing adjustments, etc. [05:40] persia: haha, so i feel kind of eased now... i understand that many people are experienced in programming, i am not very top actually [05:42] no worries. many people feel MOTU is a place where people can learn through doing. [05:42] i'd personally hope that core developers would be... more capable, as we already have Membership for non development [05:43] pwnguin: the first time i joined MOTU i was kind of nervous... [05:43] i dont think that phrase means what you think it does [05:43] "joined MOTU" [05:43] pwnguin: actually i have only one goal in my life - to become a top Linux programmer [05:44] you need two goals then [05:44] which 2? [05:44] 1) to become a top linux programmer [05:44] 2) some other personal goal [05:45] pwnguin: personal goal... i had never thought of it. living a normal life as others is okay to me [05:46] s1024kb: i mostly mean software [05:46] pwnguin: ... still don't understand... [05:46] as in, you can be a great linux programmer, but its very hard to do so without proper motivation [05:47] pwnguin: Oh i see now. the motivation is - i love Ubuntu, i guess, at least at the moment [05:47] there's spectators, who wish to be great, and then theres people with specific goals that require them to be experts [05:47] for example, you need working laptop suspend [05:48] linus torvalds didn't set out to be the world's greatest linux programmer ;) [05:49] pwnguin: i wish that i can change my job next year once i learn Linux better, i wish to work under the Linux OS in my next job, not being a Win programmer here like now [05:50] s1024kb: That's an achievable goal. I'd recommend building a portfolio of work you've done to improve some specific aspect of linux, and look for a job doing the thing you've already been doing. [05:50] pwnguin: maybe i am superficial... [05:51] s1024kb: i dont think so. you just need to find a focus. what's wrong with linux today? [05:52] pwnguin: :-) a focus... [05:52] well, perhaps you'll find some motivation by spending time with MOTU ;) [05:53] persia: i guess that a person should stick to what he/she wants to do... so i decide to do the things i love, so that i can concentrate to my goal. [05:54] s1024kb: Exactly. [05:54] pwnguin: Yes, :-) [05:55] persia: yesterday i had browsed the job market from the internet, i found some jobs here in my city are about Linux programming... just as what i want. Hope to hop soon... [05:56] where do you live? [05:56] Guangzhou, China [05:56] interesting [05:56] and you? [05:56] manhattan kansas [05:56] USA [05:57] persia: and you? where do you live? [05:57] Tokyo [05:57] orly [05:58] but here we seem to be together in the same city... thanks for the internet to bring us together [05:58] * pwnguin tasks persia with finding Kenta Cho and forcing him to accept patches upstream ;) [05:58] pwnguin: haha [05:58] pwnguin: Actually, he lives in the same town as I ("Tokyo" maps well to "State" for the US). What do we need from him? [05:59] well, nothing really [05:59] * persia allows Cho-san to relax and develop more fun games while bored [05:59] he's stated that he has no interest in maintaining the games hes released, and the internet has basically routed around that already [05:59] haha my friends, happy to have you all here... [06:00] tokyo isn't a state :P [06:00] anymore than LA is a state [06:01] i guess that most of you are boys? i am not... haha [06:01] pwnguin: No, it's a "prefecture", consisting of about 23 "wards", 15 "cities", and a few "villages" (for approximate translations) [06:02] i'd imagine metropolean area is also suitable [06:02] pwnguin: Well, LA doesn't have an independent government that is only answerable to the federal government: you've skipped sacramento. There is a "Toyko Metropolitan District" within Tokyo that more closely maps to LA. [06:03] ("Tokyo Metropolitan District" includes all the "wards", but none of the "cities" or "villages") [06:05] pwnguin: It is perhaps confusing: to map to LA again, it is as if "Baldwin Park" was in "California" county, in the state of "California" [06:05] politics is wierd [06:05] s1024kb: you mean you are a girl :P [06:05] im not about to call NYC a state even though it rivals the budget of one [06:06] freeflying: haha, that's why i was kind of nervous... i was afraid that i could not do as good as what the boys are doing... [06:07] heh [06:07] pwnguin: Well, no, but that's actually a little closer, in that "New York" state belongs to "New York" city, and the Burough Mayors have significant powers. On the other hand, in Tokyo, the "state" capitol is in the "City" capitol, and there's not really a "mayor" of the metropolitan district: that is directly adminitered by the prefecture (whereas outlying cities do have "mayor"s). [06:07] you'd have a harder time on IRC convincing people you were female than intelligent, I think [06:07] s1024kb: why not [06:08] freeflying: ... because i am really not very good at programming, i am a normal programmer in a normal company - though i thought that i am almost the best here in my department... [06:08] Wow, a Chinese lady in this channel. Welcome s1024kb. [06:09] minghua: thanks. nice to meet you. [06:09] minghua: your name seems to be a Chinese name? [06:09] s1024kb: you got it :) [06:09] s1024kb: the best way to learn is to read what other people have done, and discover what they did wrong. MOTU definately comes aross what other people did, and did wrongly ^_^ [06:10] s1024kb: Yes, I'm a Chinese. Althoug I'm living in US now. [06:10] ^_^ i feel here a warm world... i like it. [06:12] i feel kind of guilty these days because i had spent a very long time to understand only a little bit thing, and the other problems is my current situation does not allow me to make faster progress... [06:22] hey, if i want to get a package into multiverse how do i go about it? the ubuntu wiki page says revu is for universe. so i presume that means no multiverse packages there? [06:22] DarkMageZ: Processes for multiverse and largely identical to those for universe. [06:22] Including REVU. [06:23] ok, but should i upload my package for multiverse to revu? or is there another place for multiverse packages? [06:23] DarkMageZ: The same place, but there'll be lots of pressure to get it free unless there is a well documented use case that requires it. [06:25] persia, it's opensource. but debian decided to strip it out case it doesn't meet their dfsg. the legal restriction on it iirc is no commerical use. [06:26] DarkMageZ: OK. You've passed the basic legal test. Now, why should it be included? Is there not a better free solution? [06:26] persia, it's a music visualization. the "better free alternative" is debateable :) [06:27] DarkMageZ: That makes it tricky. It might get in, but it's not likely to get the attention that something for universe would. [06:28] wait, correction. "intended only for non-profit and hobbyist use and explicit permission to use it otherwise is required" [06:28] Is "non-commercial" okay for multiverse? [06:28] DarkMageZ: Part of why I warn you is that I don't want to see another case of one of your patches getting delayed forever: you are skilled at what you do, but seem to have poor luck with policies and guidelines :) [06:28] hmm, i think thats too restrictive even for ubuntu... [06:28] Hmm, I suppose it is, as we don't ship multiverse. [06:29] It's perhaps a harder case for restricted, though. [06:29] minghua: Multiverse doesn't ship on DVD, and there are existing examples, but the greater plan is for multiverse to go away [06:29] noncommercial licenses to the code i can understand [06:29] but to the binaries? [06:30] persia: is multiverse dissapering before or after world domination? [06:30] pwnguin: It'd be extra hard to ship restricted-source multiverse, even with free binaries. [06:30] pwnguin: Not relevant. Debian has world domination as a goal (IIRC), whereas Ubuntu just ships software. [06:31] im just saying, where in the grand scheme of things is this plan to remove multiverse? [06:31] persia: Right. I thought whether we ship it or not is important, too. [06:31] pwnguin: When we no longer need non-free software, presumably. [06:31] pwnguin: It's more of a loose goal. When there are no use cases that require multiverse, life gets easier for MOTU [06:31] Fujitsu: ah, so world domination [06:32] persia: until a use case pops up again that does require multiverse ... [06:33] pwnguin: Maybe. I suspect if we ever purge multiverse, we'll not take it on again lightly. "Edge Case" [06:33] There are also uses for multiverse that aren't just about non-freeness - parts of mplayer are likely patent-encumbered, so need to be in multiverse. [06:34] (or we could rip out mencoder and MPEG4 support like Debian did, but people would murder us quickly) [06:34] psh [06:34] mp4 playback is in gst [06:34] or is it just encoding that patented? [06:34] Fujitsu: That's just time for now. In the future, it may be that this a sufficiency of good codecs that "I want to play videos" isn't as likely to require multiverse/restricted [06:35] pwnguin: Good question. [06:37] hmm. surely theres a place to find a list of all packages in multiverse [06:38] pwnguin: You can look through pool/multiverse on a mirror, or... look through a Packages file, probably. [06:38] Or maybe something on packages.ubuntu.com [06:38] But not LP. [06:40] pwnguin: http://archive.ubuntu.com/ubuntu/dists/hardy/multiverse/source/Sources.gz [06:42] angband [06:47] some of this i dont see getting fixed in my lifetime [06:48] pwnguin: Well, it's a goal. It might take a while :) [06:50] i mean, MAME is comparable in scope to the linux kernel [06:54] hi Hobbsee [06:56] hi ajmitch [06:57] persia: Depends how long "a while" is, I suppose. [06:57] * minghua is pessimistic. [06:58] minghua: Depends? How? While "a while" is completely unbounded, I'm not sure that it affects either the goal, or the ease of adding to the burden. [06:58] And hello ajmitch and Hobbsee. [06:59] hiya minghua, persia [06:59] Good afternoon Hobbsee [06:59] * Hobbsee wonders what this discussion is on [06:59] persia: What I meant to say is, it's going to take much more than "a while" IMHO. More than my lifetime, for example. [06:59] Hobbsee: multiverse [07:00] minghua: Nah. Your lifetime is insignificant compared to the span of time that may be encompassed by "a while". [07:01] persia: ahhh [07:02] anyone up for doing a revu? http://revu.tauware.de/details.py?upid=505 (not related to my multiverse question) [07:03] DarkMageZ: Sure. [07:03] persia: See? That's why I say it depends. I happen to think my lifespan is longer than a while. I'm not really into buddhism. [07:04] well everyone has goals, at the end of my life, i might see the full release of hurd :) [07:05] minghua: Ah. nomenclature by philosophy. Not a distinction worth arguing: I believe it's not more than a dozen years until everything there is obsolete, considered an "edge case", or has become cruft from bitrot. [07:06] Gah. I need my dictionary again. [07:06] xmame isnt going to dissappear [07:06] nothing even comes close to duplicating what it does [07:06] its already like 10 years old [07:06] white: I recently has the opportunity to hear RMS speak about that: apparently there's not currently even a goal to release [07:06] pwnguin, the scope of the project will expand to cover the ps3 tho :P [07:06] na [07:06] its not arcade hardware [07:07] yet [07:07] true [07:07] persia: it would surpise me ;) [07:07] Hi all! [07:07] pwnguin: Sure, there's no replacement currently. On the other hand, someone may write a free mutiple architecture emulator that can handle all those architectures, and has sufficient free content that it no longer matters. [07:07] persia: its free enough for almost everyone [07:07] pwnguin: I forget the MAME problem. Wasn't it a lack of free ROMs? [07:08] no [07:08] persia: Right, no point arguing about terminology. Maybe a dozen years is a good estimate. Anyway, I'd better not waste your time that you could use to review REVU packages. :-) [07:08] noncommercial [07:08] and some other semi goofy restrictions [07:08] like [07:08] "you must distribute docs" [07:09] i.e., invariant sections? [07:09] its in the license [07:10] pwnguin: Ah. licensing. I still suspect one could implement alternate hardware emulators using a e.g. qemu framework, if one wanted. This would be handy for those wanting to sell a kit-box to turn any computer into an arcade console. [07:10] http://packages.debian.org/changelogs/pool/non-free/x/xmame/xmame_0.106-2/xmame-common.copyright [07:11] - Every X-Mame executable distribution MUST include AT LEAST the doc [07:11] directory, with special mention to readme.txt file ( this file ) [07:12] pwnguin: "Usage of the _information_ contained in the source code is free for any use". Just needs someone with an itch. [07:12] a massive itch [07:12] its 12 megs of compressed source [07:13] i gather much of it is historical in nature, but due to the distributed copyright, would be nearly impossible to negotiate [07:14] DarkMageZ: lintian and linda are full of complaints. Do you need hints for each one? [07:14] persia, the lintian complains are invalid. but i noticed the linda ones and having a look @ fixing them. [07:15] DarkMageZ: why are they invalid? [07:15] DarkMageZ: "changelog-should-mention-nmu" and "source-nmu-has-incorrect-version-number" are both valid. Perhaps you wanted to use -0ubuntu1 or set the maintainer Debian-style? [07:16] Fujitsu: lintian still doesn't know about hardy :( [07:16] persia: I've got a merge waiting in the u-m-s queue. [07:16] (with that change) [07:16] * persia thinks we should patch the hardy lintian with the name of the next release as soon as it's decided [07:17] That would be a good idea. [07:17] Fujitsu: bug #? [07:17] Fujitsu: Was that the only change, or did you also do the spellcheck MOTU change? [07:17] persia, k. i'll turn this package debian style so lintian won't cry. [07:17] * persia thanks Hobbsee for dilligence and attention [07:17] persia: I added original-maintainer to the list of valid fields, but other than that, no. [07:17] Fujitsu: That's a good one. [07:17] it's useful, being able to upload to main :P [07:18] persia: are you interested in adminship of the sponsorship queue, btw? [07:18] Hobbsee: Yeah, I'll have a number for you in a sec [07:18] Bug #158667 [07:18] Launchpad bug 158667 in lintian "Please merge lintian (main) from Debian unstable (main)" [Undecided,New] https://launchpad.net/bugs/158667 [07:18] Hobbsee: I'd need someone to do it as a team (or rather, a backup for the next time I disappear for six weeks). For the next couple months, I can take it, but will need help. [07:19] persia: oh, it would be team based. [07:19] * Hobbsee is going to send a mail to the MOTU ML about it [07:19] Hobbsee: In that case, sure. I'm happy to help with that. [07:20] Fujitsu: [07:20] + Your version string suggests this package is for Ubuntu, so your [07:20] + distribution should be one of hardy, gutsy, feisty, edgy, dapper, breezy, [07:20] + hoary or warty. [07:20] Fujitsu: surely not breezy, hoary, warty [07:20] Hobbsee: They're still valid distributions. [07:20] We should probably cull them at some point, though. [07:20] Fujitsu: they dont even exist anymore, do they? [07:20] i thought launchpad culled them. [07:20] I guess we can kill of unsupported ones. [07:20] DarkMageZ: 1) it FTBFS for me, and 2) If you make it Debian-style, lintian won't complain, but it can't be uploaded from REVU. [07:21] hmm, so I was last to touch lintian [07:21] Brb. [07:21] persia, build error? [07:21] DarkMageZ: configure: error: cannot run /bin/bash ./config.sub [07:22] Fujitsu: apart form that, looks fine. [07:22] DarkMageZ: Caused by the lack of a Build-Depends on autotools-dev [07:23] DarkMageZ: you're wanting to go for MOTU, arent you... :) [07:23] s/:)/?/ [07:24] Hobbsee, nope. it'd use my powers for evil went i get annoyed. [07:24] when* [07:24] ah [07:24] oh, must have been another d [07:24] DarkMageZ: On the other hand, your contributions are valued: please keep them up (and apologies it often takes us so long to commit them) [07:24] Hobbsee: DktrKranz? [07:25] yeah, him [07:27] * Fujitsu returns. [07:27] Hobbsee: So, do I want to remove those three? [07:28] Fujitsu: i think so [07:28] Fujitsu: it's not a valid distribution anymore - and for building locally, you'd probably ignore lintian warnings anyway [07:29] Hobbsee: Yep, fixing now. [07:30] Hmm... We still have some package in the archive that will have old changelog entries like that. [07:30] *packages [07:31] Fujitsu: the packages themselves? oh sure. [07:31] Fujitsu: We should really bump them. There's no guarantee that we can build them in that state. I'd like to get warnings. [07:31] persia: True. [07:31] Fujitsu: my point is that the *latest* one going in should definetly not be an unsupported version [07:31] Hobbsee: That's important because of the automated lintian against everything in the archive that Fujitsu is running [07:31] persia: uh? piuparts? [07:32] persia: we know the stuff still builds, even if it hasnt had a recompile since hoary. [07:32] Hobbsee: We have lintian on everything updating two or three times a week. We don't have a piuparts solution. I'm tempted to go with what we have. [07:32] allthough, there's no guarentee that the binaries are building with the latest packages, this is true [07:32] persia: we will get piuparts, RSN> [07:32] * Hobbsee is reasonably sure of this. [07:32] Hobbsee: That's not at all true. I've fixed a fair number of FTBFS stuff that still builds in Dapper, but not for Feisty / Gutsy [07:32] LPRSN, or something else? [07:33] Hobbsee: Good luck. I'm behind you all the way :) [07:33] Fujitsu: do you know who liw is? [07:33] Hobbsee: True, true. [07:33] (ie. yes) [07:33] persia: weird. i thought that it all got built for the most recent version. [07:33] in piuparts [07:34] Fujitsu: seeing as he's now an employee, id' expect him to start running it off people.u.c automatically. if not, request it. [07:34] Hobbsee: Umm. "All got build in piuparts" or "all got built" and the other was a different context? [07:34] persia: ah right [07:34] * Hobbsee is tired. [07:34] Hobbsee: piuparts doesn't take a trivial amount of CPU time to run. You need quite some hardware to do it. [07:34] fricking nutty customers. [07:35] imbrandon: No, I'm nat actively working on apt-proxy. Go for your life. [07:35] Fujitsu: and there are lots of machiens in the DC. but true. [07:35] Hobbsee: I'm less sure that p.u.c has the resources to run against all of universe on a regular basis. Maybe we'll get lucky. [07:35] So, after this argument, am I dropping them or not? [07:35] persia: rookery? no. but other machiens there, that arent connected to the outside world - pitti will probably do that for us [07:35] Fujitsu: i would. [07:36] Fujitsu: Please drop them. I'd prefer to see the warnings, and think that means we should bump with a -build1, if nothing else, just to check. [07:36] Hobbsee: OK. Good luck (somehow this reminds me of a similar discussion in May). [07:37] persia: true [07:37] persia: it hasn't gotten off the ground yet - some stuff's changed since then. [07:38] Hobbsee: I know, and am very much encouraged. I believe there's a good chance it will happen this time. I'm just going to keep trying to organise a backup, just in case. [07:38] persia, k. i've done some fixing, but i've got 2 new lintian warnings that i don't know how to fix :( http://revu.tauware.de/details.py?upid=506 [07:38] persia: good idea. [07:38] DarkMageZ: likely means that you're still not build-dep'ing on autotools-dev [07:39] DarkMageZ: OK. Ignore the first one: Fujitsu is adjusting the patch, and it should go away soon. Items 2 & 3 mean you need to copy config.sub and config.guess from the autotools-dev package into the source directory before configure:: [07:39] (e.g. in makebuilddir::) [07:39] persia: ew. [07:39] persia: you mean people actually do that? [07:39] Hobbsee: ew? Why? [07:39] Hobbsee: New debdiff attached. [07:40] how architecture-specific is autohell config files? [07:40] Hobbsee: Not at all. [07:40] ah, OK [07:40] You often need the newer ones to support other archs properly, in fact. [07:40] Hobbsee: Not at all. They encode information about all the architectures to support portability. [07:40] persia: ahhh. [07:40] (That's why lintian complains when they are out of date) [07:41] * Hobbsee wonders what they do contain, and ponders going and looking for one [07:41] Hobbsee: Bascially, there are two current practices. Either copy the files at packaging time, or copy the files at build time. I like build time because it means that a give-back can help. Other people like packaging time because it means that it's less likely to FTBFS if it builds locally. [07:42] * persia further notes that copying hint files at build time is *very* different than running autogen at build time [07:42] persia: *nod* [07:43] * Hobbsee tends to just run autohell during build. [07:43] and ditch debian's buildprep patches. [07:44] Fujitsu: please give me a patch that actually applies. [07:44] Hobbsee: You don't want to run autogen during build because then your configuration flags are dependent on the versions of the libraries installed at build time, which, due to apt's support for alternate build depends (|), means that it's deuced difficult to expect a rebuild to produce a similar package. [07:45] persia: *nods* [07:45] persia: how does one end up running autogen? [07:45] or does autogen run when [07:45] Hobbsee: One runs it manually at packaging-time [07:45] Hobbsee: WFM [07:46] -------------------------- [07:46] |diff -Nru /tmp/WFlb6bAD3d/lintian-1.23.36/checks/common_data.pm /tmp/tn6lKYl8oD/lintian-1.23.36ubuntu1/checks/common_data.pm [07:46] |--- /tmp/WFlb6bAD3d/lintian-1.23.36/checks/common_data.pm 2007-10-15 09:15:44.000000000 +1000 [07:46] |+++ /tmp/tn6lKYl8oD/lintian-1.23.36ubuntu1/checks/common_data.pm 2007-11-03 18:29:51.000000000 +1100 [07:46] -------------------------- [07:46] File to patch: [07:46] (Only do this if it's really out of date, or you need to alter things significantly, or it FTBFS locally) [07:46] persia: ah right. [07:46] Hobbsee: Are you applying it over the top of the old one? [07:46] so i havent inadvertantly done it without knowing :) [07:46] Hobbsee: -p4? [07:46] Fujitsu: yeah [07:46] Hobbsee: Right, that would do it. [07:47] Fujitsu: someon [07:47] Hobbsee: If you ran autogen by accident, the package has other issues :) [07:47] persia: yeah well :) [07:47] persia, this should be good. http://revu.tauware.de/details.py?upid=507 lintian only cries about hardy and linda is quiet. [07:47] Hobbsee: Um? [07:47] Fujitsu: erm. hwo were you intending me to run it then, if not over the original source? [07:48] DarkMageZ: Did you run linda and lintian also against the binaries that result when you build it? [07:48] Hobbsee: I was intending it to be over the original source. [07:48] Fujitsu: A merge against the Debian source? Or against the last Ubuntu source? [07:48] persia: Debian. [07:48] oh, duh. [07:48] whoops [07:49] Hobbsee: You were applying it over the already applied one? [07:49] * persia reminds Hobbsee to use -v(last Ubuntu revision) [07:49] Fujitsu: i was applying it to the ubuntu one, yeah. [07:50] Hobbsee: That's what I was checking with `Are you applying it over the top of the old one?', but it was rather ambiguous. [07:50] * Hobbsee uses grabmerge, then. [07:50] persia, run against the produced .deb ? what type of things does it check for with the binary? [07:50] Fujitsu: yeah. i had no idea what you meant about that [07:50] Hobbsee: Sorry. [07:50] Fujitsu: no problem. [07:51] DarkMageZ: I usually run against _arch.changes, which iterates over each of the .deb files. [07:51] DarkMageZ: Similar to the source checks, there are binary checks for a number of things. [07:52] persia, all good. cept the hardy issue. [07:53] DarkMageZ: Great. I'll take a closer look at 507 then. [07:55] * persia wishes the output wasn't in ALL CAPS [07:56] output of? [07:56] DarkMageZ: sbuid [07:56] s/id/ild/ [07:58] DarkMageZ: linda tells me you aren't shipping the upstream changelog. You might need to set a config variable [07:59] DarkMageZ: Any particular reason to use debhelper 4 instead of debhelper 5? That might be the cause of my last comment [08:00] persia, that's greater than or = to. to stop linda from crying. it should be using whatever the current version for your distro is [08:01] DarkMageZ: Right, but you've specified 4 in debian/compat. [08:01] As a result, debhelper does everything in compatibility mode, and doesn't expose the new features. [08:01] persia, k. i'll bump that. but i don't get that warning here :( [08:02] DarkMageZ: which version are you running? [08:02] DarkMageZ: The "no upstream copyright file" warning? debian/compat is just my eyes, not automated. [08:02] * persia uses the linda and lintian flags shown on https://wiki,ubuntu.com/MOTU/Contributing [08:02] * Fujitsu would like to see a space between `debhelper' and the versioning, but that's just nitpicking. [08:04] DarkMageZ: I'm not that familiar with libvisual-plugins. Are there also plugins not contained in either libvisual-plugins-goom2k4 and libvisual-plugins? [08:04] s/and/or/ [08:06] persia, official libv-p-0.4 ships what ubuntu's libv-p-0.4 + g-force. there's also goom2k4 which is what this is which was accidentially not included in official 0.4. there's also projectm and maybe others i have not found. [08:06] Fujitsu: uploaded, thanks [08:07] Hobbsee: Thanks. [08:07] DarkMageZ: Hmm.. I'm almost more tempted to get goom2k4 as a patch to libvisual-plugins, rather than looking at an entirely new package. What is your opinion? [08:07] (I didn't think it was broken - the patch applied, built, and ran over universe and multiverse) [08:07] heh [08:08] Fujitsu: i dont know why i thought that the p<#>'s go backwards [08:09] persia, i'm not bothered either way. ubuntu's choice. [08:10] DarkMageZ: Well, I'd argue it's in large part your choice, as you're the primary Ubuntu contributor for libvisual-plugins. Do you want to maintain a huge patch which can be dropped for the next upstream, or maintain two packages, and merge them later? [08:11] * Hobbsee could have sworn that used to be in main. [08:11] persia, i'll build a giant patch. gimme a few minutes [08:12] DarkMageZ: Thanks. I think that makes it easier, especially for the archive admins :) [08:13] DarkMageZ: And, as before, I'll suggest you might want to ping David about it, but the package is so different at this point that it's not nearly as important. [08:14] persia, david doesn't appear to be interested. if i tried any harder to get any sort of responce from him he'd probably file for an avo. lol [08:14] heh [08:15] DarkMageZ: If you7re up for it, and want to also close Debian Bug#431915, I suspect someone would be happy. [08:15] Err. Not actually close the bug, but provide a fix in Ubuntu so that it doesn't apply here. [08:16] * persia misses ubotu http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=431915 [08:16] Debian bug 431915 in libvisual-plugins "Please add projectM to package" [Important,Open] [08:16] debian bug #431915 [08:16] I think it needs a space. [08:17] Perhaps, but it's not talking to you either. I think it's started a URL-only diet [08:17] persia, fix already mostly provided. i personally would not be up for merging it into libvisual-plugins. too painful. good case for separate package [08:17] I think it was probably because you'd said it so quickly before. [08:17] debian bug #431915 [08:17] Debian bug 431915 in libvisual-plugins "Please add projectM to package" [Important,Open] http://bugs.debian.org/431915 [08:17] There we go. [08:18] DarkMageZ: Ah. I see you're already completely on top of it: I should read the bug logs :) [08:18] persia, apachelogger uploaded projectm to revu. i'm hoping he fixes his issue with 1.01 so he'll bump his revu packages to the current version. [08:19] cause 1.01 works fine here. [08:19] DarkMageZ: Cool. It sounds like that will be all sorted for hardy then :) [08:21] could anyone review my package on REVU? === asac_ is now known as asac [08:22] Repsa_Jih: Which package, what URL, how many advocates do you have, and did you fix everything from the last comment? [08:23] http://revu.tauware.de/details.py?package=annchienta [08:23] I need a first review so I can start fixing [08:23] DarkMageZ: I'm archiving the libvisual-plugins-goom2k4 REVU entry based on this discussion. [08:24] Repsa_Jih: On that page, there is a bunch of output from lintian and linda. You'll want to try to fix those first, before asking for a manual review. If you have any questions, please ask them here. [08:25] Kk [08:26] Does anyone have an opinion about item #2 in http://revu.tauware.de/revu1-incoming/annchienta-0711012110/annchienta-1.0/debian/copyright ? Does that make it non-free? [08:27] (or does it require special handling?) [08:28] Repsa_Jih: Also, You'll want to make this not a native package, so that it can be included in lots of distributions, rather than just Ubuntu. [08:28] Ok [08:30] Repsa_Jih: Just idly, as my request above doesn't seem to be attracting attention, would http://en.wikipedia.org/wiki/MIT_License not meet your license goals? [08:32] * Fujitsu returns from dinner and looks at debian/copyright. [08:33] persia: Sounds like Mozilla's trademark stuff? [08:33] Also like ion3. [08:33] ie. multiverse. [08:33] Fujitsu: Thanks. [08:34] It's the zlib/libpng license, actually [08:34] Repsa_Jih: Based on that, I'd encourage you to consider altering the license. We're more than happy to match upstream, as it makes life easier from a distribution point of view, but multiverse = purgatory in many ways. [08:34] So it is. [08:35] Oh. I remember something about the zlib license... [08:35] I'm really not sure why that's OK. [08:35] http://www.opensource.org/licenses/zlib-license.php [08:35] Fujitsu: I think it's based on the argument that adding -N marks it as unofficlal somehow. [08:36] persia: It mentions that the modified source versions, not binaries. [08:36] Ah. Cool. My apologies. That's all good. [08:36] Erk, changed what I was saying half-way through. [08:41] Repsa_Jih: Other than the example files, including your full name, making it non-native, the other other thing I see offhand is extra commented-out stuff in debian/rules. Nice work. [08:44] What should I comment out? [08:44] And should I add an url to the zlib license to make it more clear? [08:45] Repsa_Jih: It's not that you should comment out more, it's that lots of commented commands make it harder to read. [08:46] Regarding copyright, I don't think you need to include the URL. The archive admins are much more familiar with the definition of free vs. non-free than most MOTUs: you should be safe. [08:46] Oh I see [08:50] persia: Hey, i had recorded all those things you guys had just chatted and will read them later at home as a lecture... haha [08:50] * persia uninstalls liquidwar in the hopes this will increase the upload count [08:51] s1024kb: That's a good way to learn about packaging. We encourage questions in this channel, and the answers are usually applicable to other packages as well. [08:52] :-). By the way, it's your holiday today and you don't need to work in the office? [09:01] persia, how should i handle this? open a new bug and attach debdiff? email you a copy of the debdiff? [09:02] DarkMageZ: Open a bug, attach a debdiff, subscribe the sponsors. I promise it won't take three months this time :) [09:02] imbrandon, hey. about bug #59534 ? why did you change it to won't fix without an explaination in the bug report? [09:02] Launchpad bug 59534 in libvisual "Libvisual autoinstall by demand" [Low,Won't fix] https://launchpad.net/bugs/59534 [09:05] DarkMageZ: The big issue is that libvisual is in main, whereas libvisual-plugins is in universe. Main shouldn't depend or recommend universe [09:07] yup i know. but he didn't explain this in the bug report when he killed it off and didn't reply to a request in the bug report on why he did it. [09:08] DarkMageZ: Yeah, well, not everything gets documented as well as it might. Are you looking to do another sweep of bugs as well as adding the new plugin? [09:09] persia, i'm not aware of any other bugs atm. so it'll just be tweaks in the debian folder and the new visualization. [09:10] Even better :) [09:13] persia, i can drop the sizes of my changes by about 600KB if someone fixed bug #152528 [09:13] Launchpad bug 152528 in xmms-goom "needs to be split into multiple packages" [Undecided,New] https://launchpad.net/bugs/152528 [09:14] DarkMageZ: That's fair game if you want to try it. [09:15] i had a shot at it, i managed to expose the development sections, but i couldn't do it without breaking the splitting of the package. lack of skill on my part. [09:16] DarkMageZ: You might try uploading a work-in-progress to some net-accessible location, and asking someone to take a look :) [09:19] I'm looking a codeine for bug #159338, and I'd like some advice re: binary dependencies. Currently codeine gets the libxine dependency from ${shlibs:Depends}. Does this get automatically overridden if I add "libxine1-x | libxine1 (<< 1.1.8-2)" manually? [09:19] Launchpad bug 159338 in xfmedia "Re: Heads-up: small xine-lib transition in hardy" [Undecided,Fix released] https://launchpad.net/bugs/159338 [09:20] persia: No. You'll get a duplicate dependency, which lintian will complain about. [09:20] Fujitsu: That's what I thought. Any suggestions? [09:21] persia: Have you had a look at how the other package(s) do it? [09:21] * persia investigates xfmedia [09:23] Fujitsu: The only package marked "Fixed" isn't. [09:24] Err... Nevermind. "Needs Building". I may be inaccurate... [09:27] Nope. I was right the first time. There is currently no example of a correct solution uploaded. [09:27] persia: Not even siretart's? [09:28] Fujitsu: Which package? It doesn't show on the bug task page. [09:28] klear. [09:29] * persia wonders why there is no assignment: doesn't LP do that automatically? [09:34] persia: No, changelog-closes-bugs only sets the status [09:34] Changing assignee is a bit risky. as multiple people may have collaborated on an upload, etc. [09:35] Right. Claims are web or email based then, I suppose. Not a bad model. [09:36] Fujitsu: Not even klear. [09:37] Alternately, perhaps the duplicate dependencies are intentional. Must depend on the library, and also either a front-end-support module or a version of the library before the package split. I think I'll go with duplicate depends. [09:39] I guess there's probably no other way around it at the moment. [09:39] persia: have you the Depends line for klear at hand? [09:39] They can be removed for Hardy+1. [09:39] geser: results of a local build: http://paste.ubuntu-nl.org/43101/ [09:40] Fujitsu: rather, next LTS, no? [09:42] OK. I haven't done a KDE package in a while. kde-config isn't in my path, but I don't find a kdebase-devel package in my hardy chroot (although I'm told that is what I need to install) [09:42] persia: Aren't the duplicate dependencies only for upgrades, or did I miss something? [09:42] I must admit that I didn't read it thoroughly. [09:42] I've found a bug reported months ago, status "new". The bug has been fixed in Gutsy. What's the most appropriate status? Fix-released?? [09:43] Fujitsu: libxine1 (>= 1.1.4), ... , libxine1-x | libxine1 (<< 1.1.8-2) (or were you asking a different question?) [09:43] warp10: I think so, but you'll get a better answer in #ubuntu-bugs [09:44] persia: good suggestion... thank you! :) [09:46] persia: Won't that old version of libxine1 only be in Gutsy and lower? [09:47] Fujitsu: Right, but Debian bug #448077 tells the story of a user who didn't have the frontend module installed (pushing the transition plan) [09:47] Debian bug 448077 in xine-ui "please depend on libxine1-x" [Normal,Open] http://bugs.debian.org/448077 [09:49] Ah. Nevermind. I'm not reading you well. Yes, this would normally only be adding a dependency on the frontend module, but the | supports upgrades and backports. [09:50] So there's no reason we can't drop the second libxine1 versioned dep in Hardy+1 or so? [09:50] I really don't see the point of it. [09:50] Fujitsu: Ah, because all Hardy users will already have it. Right. [09:51] if the app links directly against libxine1 it should be mentioned in Depends even if libxine-x depends on libxine [09:51] * persia is confused about the difference between supporting things since the last LTS and supporting things until the next LTS. Dropping all the Dapper/Gutsy transition bits for Hardy+1 makes sense. [09:52] persia: http://revu.tauware.de/details.py?package=annchienta I think the latest lintian error shouldn't be there [09:52] I'm pretty sure it's done correctly, the changelogs are identical [09:52] geser: Right. The objective was to not have two dependencies on libxine (both versioned), just to make lintian a little more quiet. It appears we need it in these cases. [09:53] Repsa_Jih: That error is because you're generating a native Debian package. You'll want to 1) separate your debian/ directory into your diff.gz, and 2) use an -0ubuntu1 version number. [09:54] How would I go about the first one? [09:56] From the package directory, `mv debian ../annchienta-debian; cd ,,; tar czf annchienta-1.0.tar.gz annchienta-1.0; mv annchienta-1.0.tar.gz annchienta-1.9.orig.tar.gz; mv annchienta-debian annchienta-1.0/debian; cd annchienta-1.0; sed -i s/1\.0/1.0-0ubuntu1/ debian/changelog; debuild -S` [09:57] Umm. Except you really always want 1.0, and never 1.9 :) [09:57] k [09:57] Err... And probably want to use debuild -S -sa [09:58] And more generally, don't want to run random shell fragments from IRC [09:58] I was checking it :) [10:07] Your version string suggests this package is for Ubuntu, so your distribution should be one of gutsy, feisty, edgy, dapper, breezy, hoary or warty. [10:07] How should I fix that, then? [10:08] Repsa_Jih: Don't worry about that one. That's a lintian bug. The patch was applied a few hours back, so it won't get distributed for a couple days. [10:09] kk [10:09] Do we normally backport lintian? === allee_ is now known as allee [10:10] Fujitsu: Only in that we usually have a special backport of lintian and linda for REVU. [10:11] Well, perhaps not "usually", but at least for Gutsy. I wasn't doing much reviewing during Feisty. [10:21] Hi! What do I have to do so that a program is included in the repository? [10:21] Which program? [10:22] Detox (http://detox.sourceforge.net/) [10:23] fosstux: It needs to be packaged by someone, submitted to developer review, and submitted for archive admin review, in that order. [10:24] hi, how do i package a shared object file? [10:25] i mean an individual .so file [10:25] pyc: http://www.netfort.gr.jp/~dancer/column/libpkg-guide/libpkg-guide.html [10:25] ok [10:25] ok. I haveinstalled it using checkinstall... [10:26] Allright, is this correct now? http://revu.tauware.de/details.py?package=annchienta [10:26] pyc: And you can't :) You need to do the source package, and then the library and development packages are built from that. [10:26] oh :( [10:26] fosstux: The current release of checkinstall has some issues that mean the packages cannot be accepted easily. Sorry about that: it will need to be repackaged. [10:27] ok. [10:27] uh, what? [10:27] since when did we accept packages when they'd been touched with checkinstall anyway? [10:28] checkinstall was best when it kept segfaulting on attempting to start ti. [10:28] * Fujitsu liked those days. [10:28] Hobbsee: I haven't know checkinstall to be free of the dependcy issues, etc. since Ubuntu began. [10:28] But then some people got annoyed when we just laughed, IIRC. [10:29] Fujitsu: oh, indeed! [10:29] persia: as in, for it itself installing, or for it building other things? [10:29] persia: i think it's built using the wrnog approaches, so... [10:29] Hobbsee: I'm not a fan (although I've patched a couple things in it), but I'm not sure there isn't value in automating some of the packaging process. [10:30] cdbs ftw! [10:30] dh_make doesnt do too bad a job, actually [10:30] * persia disagrees with dh_make about many things, and doesn't like the example files. [10:31] * Fujitsu writes everything by hand. [10:31] Fujitsu: That's wise. It is, after all, only three files. [10:31] persia: true, the example files arent terribly useful [10:32] Hobbsee: And the default rules isn't lintian clean, and generates big ugly diff.gz files, even when someone adds a patch system (at least for C/C++ packages) [10:33] urgh [10:33] * Hobbsee doesnt do new packaging, didnt you know? [10:33] Hobbsee: Neither do I, but I often find myself trying to sort of a mess when someone uses dh_make, and isn't careful (and not only REVU) [10:34] * Repsa_Jih wonders if this is packaged correctly now: http://revu.tauware.de/details.py?package=annchienta [10:34] Hobbsee: For a true horror story, ask DarkMageZ and RAOF about trying to get aotoconf working for libvisual-plugins without having conflicting patches in the diff.gz and debian/patches [10:35] i've got better horror stories, that was fairly light stuff [10:35] in comparison [10:35] Repsa_Jih: Someone will look at it as soon as they have time. If you ask too often (we request less than once a day), you'll get put on a lower priority list. If nobody else does, I'll look at it in about 20 minutes. [10:35] Oh, I'm sorry. [10:36] DarkMageZ: heh, but were they related to dh_make? [10:36] I didn't know that [10:36] Repsa_Jih: No worries. You're new here. Nobody gets swatted without a warning :) [10:36] K :) [10:37] persia: yummy. [10:37] persia, ah. i hasn't read that part. [10:37] where can I find a ubuntu Tutorial how to build a dab? [10:37] deb... ofcourse [10:38] fosstux: https://wiki.ubuntu.com/PackagingGuide [10:38] persia, hehe. one of my favorite horror stories comes from checkinstall. [10:38] thx [10:38] DarkMageZ: Yep. That's why it's not currently accepted for new packaging. [10:46] !packaging [10:46] The packaging guide is at http://doc.ubuntu.com/ubuntu/packagingguide/C/index.html - See https://wiki.ubuntu.com/MOTU/Packages/New for information on getting a package integrated into Ubuntu - Other developer resources are at https://wiki.ubuntu.com/DeveloperResources - See also !backports [10:46] :) [10:47] Kmos: Thanks for pointing that out. [10:47] ubotu: packaging is The packaging guide is at https://wiki.ubuntu.com/PackagingGuide - See https://wiki.ubuntu.com/MOTU/Packages/New for information on getting a package integrated into Ubuntu - Other developer resources are at https://wiki.ubuntu.com/DeveloperResources - See also !backports [10:47] persia: shouldn't fix the first link to the new in the wiki ? [10:48] bluekuja: ping? I'm merging rtorrent from Debian, and I'm wondering whether we still want to apply the port-range patch. Bug #110803. Since upstream thinks it's better a default starting from 6890... [10:48] Launchpad bug 110803 in rtorrent "The default for port_range should be "6881-6999", not "6890-6999"" [Low,Fix released] https://launchpad.net/bugs/110803 [10:48] pochu, let me see [10:48] bluekuja: wait! [10:48] ../src/command_network.cc: ADD_VARIABLE_STRING("port_range", "6881-6999"); [10:49] bluekuja: so it looks like it's been changed upstream :) [10:49] ^^ [10:49] seems so :) [10:49] I'll remove the patch then. Thanks! [10:49] np, is it the only Ubuntu change? [10:50] bluekuja: yeah, we can sync again! :-) [10:50] * pochu tests it and requests a sync [10:50] * persia wishes *all* torrent applications had configurable port ranges, rather than hard-coded port ranges [10:50] pochu, great, let me know when you've opened a sync bug [10:50] persia: that's the default :) [10:51] bluekuja: would you mind sponsoring an upload for me? bug 159700 [10:51] Launchpad bug 159700 in tasks "Merge tasks 0.12-1 from Debian unstable (main)" [Wishlist,Confirmed] https://launchpad.net/bugs/159700 [10:51] pochu: Goody!. I haven't looked in a while, but I remember patching python-source in /usr/share for original bittorrent. [10:51] pochu, yep, let me look at it [10:51] bluekuja: and bug 159695 needs a confirmation, if you have some time :-) [10:51] Launchpad bug 159695 in anjuta "Please sync anjuta (universe) from Debian unstable (main)" [Undecided,New] https://launchpad.net/bugs/159695 [10:51] bluekuja: ty [10:52] pochu, added to TODO as well ;) [10:52] cool :) [10:52] pochu, would you mind adding a debdiff to tasks merge? [10:52] I don't like reviewing merges that way [10:53] bluekuja: sure, from Debian or from old Ubuntu? [10:53] Or from both? :) [10:53] pochu, between new Ubuntu revision and the new debian one [10:53] where new Ubuntu is your one [10:53] of course [10:54] pochu, like every normal merge [10:54] (or is it your first one?) [10:55] bluekuja: no, but I haven't done any since Gutsy ;) [10:55] * pochu adds it. [10:55] ty [10:55] * persia thinks pochu predates workflow, but tends to focus on only a few packages, and so has been lucky with sponsoring [10:57] persia: I do, but I want to change that (focus on the entire universe ;) ) [10:57] pochu, plus when asking for a sync, would be nice to have latest unstable changelog's entry as well [10:57] so I don't have to check it myself [10:57] pochu: That's a big goal, but you're more than welcome :) [10:57] bluekuja: ah, right :| [10:57] pochu, also I would like to see a build log there [10:58] writing "build fine" [10:58] bluekuja: I have it, gimme a sec. [10:58] k, fine [10:58] * persia notes that more than just the latest, but all Debian changelog entries since the last Ubuntu pull is nice [10:58] persia, true [10:58] pochu, all debian entries since latest ubuntu [10:59] even better [10:59] ;) [10:59] bluekuja: tasks debdiff attached. [10:59] pochu, let me check it [11:00] Repsa_Jih: annchienta (510) FTBFS for me. [11:03] pochu, dont need to add the lp bug again [11:03] pochu, in the new entry [11:04] bluekuja: it's for future reference. but since it isn't 'LP: ', but 'Launchpad', it won't re-close it ;) [11:04] yeah, ok then [11:11] In what package can I find the dpms dev headers? [11:12] persia: just to be sure, for the new xine, we change the build dep to libxine-dev (>= 1.0.2) ? [11:12] knights: Maybe libxcb-dpms0-dev, but I've not checked [11:13] persia: Unfortunately I've already tried that- no look [11:13] luck, even [11:13] jpatrick: http://paste.ubuntu-nl.org/43107/ is my debdiff - sample minimal adjustment for the transition [11:14] jpatrick: Well, you might get lucky, and not need Maintainer mangling, but aside from that :) [11:14] persia: just the dep on libxine1-x | libxine1 (<< 1.1.8-2) - got it [11:14] jpatrick: Remember to use libxine1-console for non-x apps [11:15] persia: I'm doing the KDE ones :) [11:15] jpatrick: Excellent. I'll do the rest of universe, and we'll be in good shape :) [11:17] Of course, as this is a Malone-managed transition, anyone else is welcome to join: just assign yourself a task (although it's lots easier if you can upload) [11:17] * jpatrick was unlucky and got a messy merge [11:17] pochu, looks fine [11:18] * persia has been sniping those that aren't merge candidates :) [11:18] yay, conflicts everywhere [11:19] 0.6.4-4+b1 means binary NMU, right? [11:20] persia: yes [11:20] geser: Thanks. [11:21] jpatrick: depending on the size of the Ubuntu delta, it might be easier to apply the Ubuntu changes by hand on the recent Debian package [11:21] (resolving conflicts in configure is no fun) [11:21] * persia always does that: patchutils can help [11:22] geser: no worries, everythings fine [11:25] http://paste.ubuntu-nl.org/43109/ - does anyone know what that _P=)'s for? [11:28] jpatrick: It's because sed is silly about close parens. [11:28] persia: ah, right, thanks, just making sure === apachelogger is now known as leinirlogger [11:33] bluekuja: bug 159695 updated :) [11:33] Launchpad bug 159695 in anjuta "Please sync anjuta (universe) from Debian unstable (main)" [Wishlist,New] https://launchpad.net/bugs/159695 [11:33] pochu, tasks uploaded. [11:33] pochu, you should fix the bug title [11:33] pochu, to merge package-version from debian unstable [11:34] pochu, plus a build log is missing [11:34] :) [11:34] I've just attached it ;) [11:34] (refreshing) [11:34] huge build log^^ [11:35] http://paste.ubuntu-nl.org/43110/ - which is the prefered format? [11:36] jpatrick: I prefer the second set without reference, but generally prefer the Debian solution, unless there is a good reason otherwise, as it makes merging easier next time. [11:37] bluekuja: just got port 6888 in rtorrent 0.7.8, so we can safely remove the patch :) [11:37] jpatrick: More generally, one doesn't want to have to update the package to build into debian/packagename/foo [11:37] * pochu request a sync [11:37] right, now to find a core-dev [11:37] pochu, yep [11:39] bluekuja: thanks for the upload. [11:39] pochu, np === leinirlogger is now known as apachelogger [11:49] bluekuja: bug 149190 [11:49] Launchpad bug 149190 in rtorrent "Please sync rtorrent 0.7.8-1 from Debian unstable (main)" [Undecided,New] https://launchpad.net/bugs/149190 [11:49] Hope it's fine :) [11:50] * persia encourages pochu to take advantage of the fine sponsors queue === asac_ is now known as asac [11:55] persia: it's because he included the patch I've dropped :) But yes, I should use it more (not that I don't use it, but not always...). [11:56] pochu: Right. That's why it's encouragement. You get by, and generally don't bother anyone, but you could probably benefit from the queue as well (and would certainly have more people seeing your work, which could help in any future application) [12:00] Hmm... I'm confused. I'm looking at one package, that needs to be merged or synced. One of the files is marked with .UBUNTU and .DEBIAN, so there should be some changes. But the diff says they're the same. So, that should be just a bug in automatic checking? [12:01] Jazzva: it's not likely a bug, either you're missing something, or the change is invisible (non-printable characters, whitespace alterations, etc.) [12:01] persia: Ok... I'll take a look at permissions or something like that :)... Thanks [12:03] Well, the permissions, size and timestamps are the same for both files :/. [12:03] pochu, is it written somewhere in upstream website? [12:04] Jazzva: I'd argue for invisible then. diff might help you track down which line number, and with the right arguments produce something small enough to pipe into od [12:06] bluekuja: not that I know, and I can't find anything about it. [12:06] pochu, that's why I asked [12:06] :) [12:07] pochu, but anyway we have ADD_VARIABLE_STRING("port_range", "6881-6999"); [12:08] that confirms that [12:08] pochu, the patch shouldnt apply anymore as well then [12:12] bluekuja: it doesn't. And I got a 6888 port with the random port enabled. [12:12] I'm just editing my debian/control file for my 1st ever deb but I think I've messed the formatting ie the leave a space at the start of each line of the package description [12:12] howshould I edit this file? [12:12] or is there a script that will format it for me? [12:13] persia: Hmm, the diff -s explicitly says they are the same, the file identifies them as ASCII PO text. I think that the same char encoding is used... [12:13] Jazzva: Odd. Are you using MoM or DaD? Which file? [12:13] When I say 'how should I edit' I mean should I be editing it in 'so many columns' mode or something? [12:14] with a specific text editor maybe? [12:14] persia: MoM, package xsane. Files are backend-po/newbackend.po.{DEBIAN,UBUNTU} [12:14] pochu, true, also I see inside /src/core/manager.cc [12:14] that the line reported in the bug is no more there [12:15] and now we have a variable for it...maybe the moved everything to the other file you linked me before [12:15] pochu, let me test if it works here as well and I ack your request [12:15] Can anyone help me edit me control file please? [12:16] persia: I also think that the same will be with the rest of the files in backend-po marked with C* in report. [12:16] knights: every line in description should start with a space. [12:16] knights: Generally, I use vi. If it's broken in some way, and you need another eye, a pastebin might help. As far as I know, there is no lint like tool. [12:19] persia: Yes but don't I need to enforce so-many column and some kind of wrapping as I don't know whee to insert the new lines in the description- that'd be the only way I could add a space at the start of each line [12:19] If I just open vi in a terminal it could be of any column size [12:19] persia: Just checked the rest with file and diff. file reports they all the pairs are the same file type. diff reports that every pair is identical. [12:19] Jazzva: That's just odd. The timestamps match as well. I think it's the wrong time of day, but I suspect a bug report may help (and I don't know where it gets filed) [12:20] I'd like to do it in gedit, kate or a gui text ed pref. [12:20] knights: Yes, you need to enforce it, and yes, vi is flexible, but you can also find your position on each line, and manually split the lines (you should be able to do this in any text editor) [12:21] persia: and that splitting position number is? [12:21] column 80? [12:22] persia: Wrong time of day? For example one is 9am, and the other 9pm? [12:22] knights: It should be less than 80 columns. If a couple lines are exactly 80, nothing will complain, but I usually shoot for 76 or so, so that diffs and diffs of diffs are also within 80. [12:22] right! Thanks [12:22] (diffs of diffs of diffs should be avoided, if at all possible) [12:22] knights: Also, check if the text editor made backup files (they end in ~), gedit does that by default. Don't forget to clean them up (I do *sigh)... === jekil2 is now known as jekil [12:23] right. MOTU ML emailed. [12:25] who is doing what now? [12:25] zul: You can have it all, if you want :) [12:26] i already have it all [12:26] persia: I looked at DaD, those files aren't reported with C*, so I suppose it's a bug in MoM... [12:27] Jazzva: Perhaps. Hard to say. I'd definitely report it as a bug in MoM, as it should be looked at, but I can't say if it's MoM, or something MoM is using to do the work. [12:27] * cyberix will change his strategy [12:28] Please surf http://revu.tauware.de/details.py?package=pq to see that there is nothing to complain about the package, and move on to sponsoring it. Lets make the World a better palace! [12:29] imbrandon: irssi is taken a while... is it normal? [12:30] s/taken/taking/ [12:31] pochu: Taking a while? [12:31] persia: well, I suppose it's the latter. [12:32] cyberix: Better strategy. Things to include in the request: whether the new upload fixes previous comments, and whether you are seeking the first or second advocate. [12:32] persia: I'll file a bug... just to find the appropriate project. If I don't, should I file a bug in Ubuntu (as the last resort)? [12:32] Fujitsu: it hasn't been announced, isn't in the archive, isn't in the new queue... [12:33] pochu: did it actually get sponosred? [12:33] Hobbsee: imbrandon told me he uploaded it. [12:33] persia: It fixes the pervious requests and some other issues too. [12:33] persia: I'm looking for the first advocate. [12:34] pochu: ah [12:34] pochu: It hasn't been uploaded. [12:34] Or it was uploaded incorrectly and rejected due to lack of signage or so. [12:34] cyberix: OK. Did you already run linda and lintian against the binary packages? [12:35] Ok, thanks. Maybe he meant he added it to his todo or something :) [12:35] Or it was rejected, yup. [12:36] pochu: What's the bug #? [12:36] why was it rejected? [12:36] Ouch! s/gutsy/hardy/ ! [12:36] :-) [12:36] just upgrade to hardy. problem solved. [12:36] pochu: That's what I was going to check. [12:36] hehe. easy enough to miss :-) [12:37] pochu: ah well, fix it, and point me at a bug # [12:37] That's probably the most common and easiest to miss. [12:37] Hobbsee: or rather bind dch to dch --release=hardy :) [12:38] pochu: that would work [12:38] unless doing backports [12:40] Hobbsee: bug 159659 [12:40] Launchpad bug 159659 in irssi "Merge with Debian unstable - new upstream release 0.8.12" [Low,Fix committed] https://launchpad.net/bugs/159659 [12:40] I've updated the package in that link. [12:42] pochu: lately, I've been maintaining a hardy chroot for development work, with a bind-mount of /home. That way I get the hardy dev tools and still have a gutsy workstation. [12:42] I've install the devscripts from hardy on my gutsy, I just now need to add hardy to the vim syntax hilighting [12:43] persia: Yes. [12:43] cyberix: OK. I'll take a look then. [12:43] persia: And got rid of all complaints [12:43] cyberix: Even better :) [12:43] :) [12:44] could anyone review a REVU package? [12:44] jpatrick: You might want to merge the jpatrick-kubuntu account on LP with your normal one. [12:44] Fujitsu: I'm trying but my @kubuntu.org and sysadmin hasn't got round to fixing it [12:44] *.org is bust [12:45] jpatrick: Ah. [12:45] Good Mrning MOTU [12:45] Morning* [12:45] You could probably poke ~admins into merging them manually. [12:46] LucidFox: You'll want to add a fair bit more to the request [12:46] if that means the link... http://revu.tauware.de/details.py?package=qconf [12:47] is there an reason mom would produce a confluct and the part that conflicts is exactly the same? [12:47] LucidFox: Link, and status, but link is a good start :) [12:48] joejaxx: We had a case of that just before... [12:48] status: new [12:48] joejaxx: jazza encountered the same issue a little while back. MoM is perhaps confused, and needs a poke. [12:48] Keeeeeeeybuk! [12:48] For /debian/manpages, do I just have a list of the man files with their relative paths, one per line? [12:48] Fujitsu: hmm, but how? [12:48] persia: oho k because the kazehakase merge i was looking at was like that [12:49] I have another question, how do reviewers build packages from sources? Because I got a FTBFS for my package, but it builds fine on my machine. [12:49] i actually need to see who did it last so i can ask them if i can do it [12:49] And I want to reproduce the error. [12:49] jpatrick: File a support request at https://answers.launchpad.net/launchpad, requesting the accounts be merged. It should be fairly clear you own both, I think. [12:49] Hi JoeJaxx! [12:49] Any news on fluxbuntu? [12:49] Fujitsu: ok, I'll do that later, but now -> lunch [12:49] knights: hello :) how do i know you? :) [12:49] knights: oh [12:49] lool [12:50] Repsa_Jih: I used `sbuild -A -d hardy annchienta_1.0-0ubuntu1.dsc` to build that (or rather, to FTBFS) [12:50] knights: yes i am going to build when i get back home [12:50] Ok [12:50] knights: :) [12:50] joejaxx: I e-mailed you recently about the PPC version and the PC RC not booting for me [12:51] knights: oh ok, do you have a date for that? [12:51] I expect it will be FAST tho :) [12:51] for that email that is :) [12:51] yeah, hold on [12:51] persia: oh ok i was just wondering if there was something different that had to be done when that happens [12:52] persia: because it was exactly the same [12:52] joejaxx: Do your best to confirm the files are really the same, and then use the Debian version? [12:52] joejaxx: 31st Oct, allcoms at gmail [12:52] persia: ok [12:53] knights: ok i will take a look for that email [12:53] cyberix: I get lintian and linda output on the binaries with the flags in https://wiki.ubuntu.com/MOTU/Contributing [12:53] cyberix: Most importantly, /usr/games/pq is not executable, so it doesn't run :( [12:54] OK, so my man files are in /man, so in /debian/manfiles I put stuff like [12:54] ./man/manfile.1 [12:55] or do I need to put the full path? [12:55] sorry ../man/manfile.1 [12:55] knights: You probably want "man/manfile.N", without the leading "./" [12:56] Thanks persia! [12:59] pochu, I'm getting only high port numbers for now [12:59] pochu, using randomly port chooser [12:59] is Medibuntu legitimate? (or is it something like Automatix?) [13:00] it's not official, no [13:00] pochu, [Port: 6882] [13:00] pochu, got fixed then...acking your request [13:01] elkbuntu, hmmm, ok, thanks. [13:01] bluekuja: cheers [13:02] in that case, it's strange it's promoted on the restricted formats page (due to libdvdcss) etc. I can imagine that a lot of new users will use it [13:02] pochu: done [13:03] Thanks :) [13:03] persia: Interresting. Will examine. === _nand_ is now known as nand_ [13:04] Does anyone know of a good example of a get-orig-source rule? [13:07] rexbron: Did the examples I gave you yesterday not include one that worked for you? [13:07] persia: the wiki page explains how to use uscan, were their others? === TheMuso_1oston is now known as TheMuso_Boston [13:08] rexbron: There was one in the recipies section, and the other in the section about changing the original tarball. [13:08] I will review motu/recipes [13:09] rexbron: You might also want to check the section in the packaging guide about changing the original tarball. [13:10] rexbron: LucidFox's looks pretty good, for a uscan based repack (http://revu.tauware.de/revu1-incoming/qconf-0711030750/qconf-1.4/debian/rules) [13:13] persia: except there is a typo in .PHONY :) [13:13] grt-orig-source [13:13] rexbron: Thanks. I'll add that as a comment :) [13:15] persia: my main issue was I do not know how to use sed or awk to get the version number, but I will modify his example to suit genpo. [13:16] rexbron: So the issue with building a regular expression? [13:16] rexbron: `man -S 7 regex` might help. [13:16] basically, ya. I found sed and awk a bit daunting to read [13:16] read/understand [13:17] rexbron: Sorry. It's my preferred regex format, so I've been encouraging people :) I'd be happy to walk you through a line, or troubleshoot something, if you'd like. [13:18] persia: I have a regex list from a python book, would that be applicable? [13:18] rexbron: I think python uses a slightly different format, but I'm not sure (both perl & ruby use the same different format) [13:20] LucidFox: upid 503 commented. [13:20] * persia ceases REVUing until REVU day (starts in ~21 hours) [13:21] persia: if you could walk me through the regex in the link you sent, I would appreciate it. [13:22] The qconf example? [13:22] yes [13:22] OK. We have s/.*\(.*\)<\/upstream-version>.*/\1/p [13:22] persia: nah, keep going. then you can revu the fixed packages again :P [13:23] Hobbsee: No, I'm trying to split oxine, and I want to finish before I fall asleep, while still answering questions, etc. [13:24] persia: ahhh [13:25] So, it looks for the string between the first two '/' characters, and replaces with the string between the second two '/' characters, and prints the results (s/this/that/p) [13:26] ok [13:28] Not fair! [13:28] Why is debuild not working now? [13:28] The first string consists of '.*' which means "as many characters as possible while still meeting other conditions", "" which is just those characters, '\(' which is really just '(': the \ is for a shell escapem '.*' which is as much as possible again, '\)', which is ')' escaped, "", which is another literal, and another '.*' [13:28] I started it using 'debuild -us -uc' ans it starte to compile [13:29] but then I CTRL-C it, thought I might change something, changed my mind, then ran debuild -us -uc again but now it doesn't start compiling [13:30] dpkg-source: unrepresentable changes to source [13:30] The "(foo)" construction is special: it indicates that "foo" should be saved in a buffer, for later use. As this is the first parentheses, this is buffer #1 [13:30] what happened? [13:31] knights: Your source dir got corrupted. Try `debian/rules clean`. If that doesn't work, your package suffers from the "Cannot be built twice" bug, and should be fixed. You can usually get the old state by deleting the working directory, and using `dpkg-source -x ` in the parent directory. [13:32] So, back to sed: when looking at the replacement string, it consists of only "\1", which is a special notation to mean buffer #1, or the data that was between the parentheses in the match string. [13:33] persia: it depends on how regex is implemented: ( could be a verbatim ( and \( the start of a match buffer and vice versa [13:33] ..... [13:33] persia: is it acceptable to call a python script that does the same thing? [13:34] geser: Not for uquoted regexes fed to sed in a makefile, but in some cases, yes (we're specifically looking at http://revu.tauware.de/revu1-incoming/qconf-0711030750/qconf-1.4/debian/rules) [13:34] rexbron: You'd have to build-depend on python, which isn't so nice. Otherwise, sure. [13:35] persia: first there seems to be ' missing at the start of the sed command [13:36] geser: My, you're right. It was intended to be quoted, and it's not. [13:37] rexbron: My apologies: that was perhaps not an ideal example. [13:37] how does the output from "uscan --force-download --dehs" usually looks like? [13:37] rexbron: The Python regex syntax is unique. [13:38] geser: It can look like http://paste.ubuntu-nl.org/43116/ [13:39] persia: see the dehs output makes the regex make sense [13:39] *seeing [13:39] * cyberix feels tasked [13:39] dh_md5sums doesn't seem to do anything [13:39] and it reports no errors [13:40] persia: correct me if I am wrong, the sed line places what ever is between the two tags into buffer #1 [13:40] in this case, the upstream version number [13:41] persia: \( \) are the matching ones, ( ) are verbatim ones [13:41] rexbron: That's the idea. [13:41] cyberix: what do you expect from dh_md5sums to do? [13:41] persia: .* <= not sure what that means, everything before? [13:42] geser: Hrm. I thought I usually did it the other way. [13:42] rexbron: '.' means something (any character), '*' means any number (including 0) of the preceeding character [13:43] geser: I expect it to create a file with md5sums for some other files [13:44] persia: it's the other way with sed -r [13:44] geser: Ah. That's it. Thanks. [13:46] persia: ok, think I understand now. Thanks for your time [13:46] rexbron: Sure. Thanks for working on the package. I'm looking forward to using it. [13:47] :) [13:54] Could anyone point me at a good guide for splitting packages with debhelper? http://wiki.debian.org/PkgSplit is frustrating me, in part because the sample package no longer exists. [14:03] How does dh_md5sums decide what to do? [14:04] It gives an error, if control file is missing. [14:04] Heya gang [14:04] I'm not sure why it needs it [14:04] But maybe that is not my problem [14:05] cyberix: debian/control ? [14:06] yep [14:06] It would be really nice to know what the script currently is/isn't doing [14:06] cyberix: man dh_md5sums and less /usr/bin/dh_md5sums might help... [14:08] Not to mention that debian/control is probably the most important file in a package :-) [14:09] hi [14:09] RainC1: Hi! [14:09] Heya RainC1 [14:09] bddebian: I wasn't questioning that. [14:09] Question regarding re-rolled tarballs, should the version number of the orig.tar.gz be changed? === RainC1 is now known as RainCT [14:10] bddebian: But the information might help in reasoning [14:10] ie, .dfsg1.orig.tar.gz [14:10] rexbron: You probably want something like pacakge-X.Y+dsfg.1- [14:10] rexbron: ... [14:10] I'm looking for something to merge, any suggestion? :) [14:10] RainCT: Feel free to grab any of mine [14:10] bddebian: I may have asked this question like 6 months ago.. :p [14:10] RainCT: You can have bacula, if you like [14:11] rexbron: No, persia beat me to the answer as always :-) [14:11] bddebian: OK. You get the next question ... [14:11] * persia requests a question to be posed... [14:12] hehe [14:12] persia, bddebian: that is in the changelog, is it not? Should the .orig.tar.gz be vesioned aswell? [14:13] bddebian: Don't forget to explain why :) [14:13] persia: ok, I'll try with bacula. seems a pretty big package.. [14:14] rexbron: http://www.debian.org/doc/developers-reference/ch-best-pkging-practices.en.html#s-bpp-origtargz [14:15] RainCT: It is indeed big, and a bit messy, but the patches should be fairly well organised, and both Debian and Upstream are pretty quick to merge things. [14:25] (do u read me?) [14:27] RainCT: Sorry: you left about 30 seconds after your last broadcast... [14:27] persia, bddebian: RFC, http://pastebin.ca/759768 [14:28] persia: were you saying something? [14:28] (I just did a little test, the router doesn't like my PC :P) [14:29] RainCT: No, just apologising for not responding to your ping quickly enough to support your test. [14:30] ah, I was still in here? [14:30] well, isn't that importat. thanks anyways :) [14:33] bddebian: Sorry to be a pain, but did you see the pastebin link I sent you? [14:34] rexbron: He's being beaten over someone else's mistake in a secret room. I'll take a look. [14:34] rexbron: Are you asking if that is correct? [14:34] persia: :-) [14:34] rexbron: You don't need line 8, but other than that, the important question is, does it work for you? [14:35] Hi all [14:35] persia: I'm now setting executing permission in rules file. I wonder, if it works. [14:35] Heya DarkSun88 [14:36] persia: I've fixed other issues. Except that lintian complains about hardy not existing. [14:36] cyberix: You might also look into dh_fixperms [14:36] cyberix: dh_fixperms would be preferred [14:36] * bddebian curses persia again :-) [14:36] bddebian: Just for that, I'm stealing wsjt. === asac_ is now known as asac [14:38] bddebian: Hi :) [14:38] I've just ran lintian on my new deb and its complained about a few things. Can I make my adjustments then just re-run 'debuild -us -uc' or do I have to run another, clean-up command first? [14:39] "It makes all files in bin/ directories, /usr/games/ and etc/init.d executable (v4 only)." [14:39] I'm using debhelper 5 [14:39] cyberix: Hrm. Perhaps I'm out of date. There should either be a replacement, or that really means (v4+ only) [14:40] I'll try that then [14:40] or do I just manually deleted the debs, .dsc files etc? [14:41] delete [14:41] knights: If your clean rule does everything it is supposed to do, you can just edit and rerun. If not, delete the working directory, unpack the .dsc, and make your changes, and rebuild. [14:43] Have I got to use my Launchpad name as my mainter name? [14:44] maintainer name [14:45] knights: You must use the name on your GPG key. Ideally this will match the name on your government issued identification (or getting your GPG key into the web of trust will be difficult) [14:45] Ack! So I NEED to get a GPG key now too? Yuk! [14:47] I don't know anything about this 'web-of-trust' either [14:49] knights: http://en.wikipedia.org/wiki/Web_of_trust [14:49] Thats the page I'm on :) [14:50] knights: Essentially, it's a mechanism to prove you are you, so that as long as people trust you, they don't need to check the package as carefully. [14:50] He! I can't prove and nor do I fully trust I'm me, so that's buggered that one eh? :D [14:51] knights: Well, if a government believes you're you, that's usually enough. You can pretend, and nobody has to know :) [14:51] I don't trust the govt to believe I'm me [14:51] ;) [14:53] * persia grumbles at seahorse failing to have enough threading in the code [14:59] persia: This is fun, bash does not like multiline (even when continued with \ ) command substitution [14:59] rexbron: debian/rules is not a shell script (or am I confused?) [15:04] persia: doesn't make just exec the commands in a shell? [15:05] as I am getting bash errors... [15:05] rexbron: No. It execs each line individually in the shell [15:06] what is the appropreiate way to test it then, calling rules get-orig-source? [15:09] persia: also from testing it myself, the get-orig source line will not work in the example you gave me [15:09] it needs a single $ and forgot to include the single quote at the start of the sed expression [15:09] rexbron: From qconf? Geser pointed out it didn't have enough quotes. Could you pastebin your error? [15:10] oh did he, ok then [15:10] not I got it to work === bluekuja_ is now known as bluekuja [15:10] But I will paste bin it when I know it works in the rules file [15:10] rexbron: Cool. My apologies: I have lots of information, but the details aren't always 100% correct. [15:11] rexbron: If you know it works, and you've tested it, no need to pastebin: just upload it :) [15:21] hmm [15:23] My rules file has the following line [15:23] binary: binary-indep [15:23] I copied it from somewhere [15:23] Does it make any sense? [15:23] The package doesn't actually have anything to compile [15:24] persia: http://pastebin.ca/759812 [15:26] cyberix: http://en.wikipedia.org/wiki/Cargo_cult_programming, it makes sense if binary-indep does something, which it should if you plan to create a binary independent package. [15:28] Each line like "foo: bar" represents a dependency. In order for "foo" to execute, "bar" much first execute. In this case, that line indicates that in order to build all the binary packages, one must first build the binary-independent packages (whereas the line might also be "binary: binary-arch" or "binary: binary-indep binary-arch") [15:28] You'll want to be sure the "binary-indep" rule creates the arch: all .deb file, and then everything is golden. [15:33] hmm why is make shorting this line version=$(uscan --force-download --dehs | sed -n 's/.*\(.*\)<\/upstream-version>.*/\1/p') to version=<\/upstream-version>.*/\1/p') && \ [15:34] as having it show up as shorted explains the bash error [15:34] You need to escape the $ [15:34] Replace it with $$ [15:35] could someone look if this is ok for multiverse? http://revu.tauware.de/revu1-incoming/avidemux-0711031810/ -- its linda and lintian-clean [15:35] ion_: Excellent. Thank you. [15:37] My rules file contains a 'dh_shlibdeps -a' but lintian is telling me I'm missing a 'dpkg-shlibdeps' can I just add this anywhere (at the end) or does it need to go in a certain place/order? [15:39] anywhere in the rules file, of course [15:40] persia? [15:41] knights: I think that belongs in binary-arch:, but I'm not sure (and not thinking well at this time of day) [15:42] OK, at the end of the binary-arch section then? [15:43] persia: * Apply 2.2.4-lost-block.patch from upstream bug#964 to address data loss [15:43] RainCT: That's already in Debian. Drop it :) [15:43] where's that patch? there isn't any file called like that [15:43] ok [15:44] ... [15:45] RainCT: It looks like I made a mistake and modified src/stored/append.c directly. Sorry about that. [15:45] when I echo the variable, instead of giving me the correct upstream version, I am getting a list of them [15:45] 1.3.5 1.3.1 0.9.6 0.4.0 0.4.0 0.4.0 [15:45] * rexbron is confused [15:45] persia: oh, ok [15:46] rexbron: Interesting. You might play with other uscan output formats to get the right thing. [15:46] So, sync requests must have the "sync" tag? [15:46] LucidFox: No, but it makes them easier to find. [15:47] Ah. [15:48] persia: the really odd thing is if I run it by itself, i get the correct version [15:48] if I run it in the rules file, I get a list [15:49] rexbron: That is unexpected. [15:49] (and most of them are not even the correct numbers [15:49] I've got a SETUID file in my package- how do I add a 'Lintian override' for it? [15:50] knights: First, why do you have a setuid file? [15:50] Its a configuration tool [15:50] knights> what package? setuid files are highly unwelcome because of security risks [15:51] knights: OK. Why does it need to be setuid? Can it not require that only root run it (or a suitably privileged user through sudo and the like)? [15:51] xdtv contains xdtv_v4l-conf [15:51] persia: That is what I'm doing. I was wondering why binary: gets executed at all. :-) [15:52] To be honest I've never had to run xdtv_v4l-conf - xdtv has always worked without running it for me [15:52] cyberix: Actually, in most cases it only gets called during manual local test builds (as I understand it), but it should really be there. [15:55] That was my best guess too. And that is why I asked instead of just removing it. [15:55] OK- so we'll make it so only root can run it. Thats no prob. What chmod command shall I use on it? [15:56] knights: 0755 is probably correct [15:56] Hi jussi01! [15:57] Thanks persia! [15:57] hi knights [15:57] heya persia [15:57] Hi jussi01 [15:58] * knights normally danboid [15:58] jussi01: Do you want another easy patch possible SRU? [15:58] persia: I fixed the issues and made various other improvements. Pour some more problems on me. http://revu.tauware.de/details.py?package=pq [15:58] s/problems/improvement ideas/ [15:58] ;-) [15:58] no motu that has time to check my package for multiverse? [15:58] persia: LOL...throw it at me... havent got time right now, but if its not urgent [15:59] cyberix: I'm not REVUing anymore until REVU day. You might catch someone else. I did recently notice the existence of libpq though... [15:59] jussi01: bug #159734 [15:59] Launchpad bug 159734 in wsjt "wsjt does not install the correct dependencies" [Medium,Confirmed] https://launchpad.net/bugs/159734 [15:59] persia: that shuld be pretty basic :) nice familiaristation one. :) [16:00] jussi01: Still needs the test cycle and extra care though :) [16:00] persia: yes. but at least i know pretty much what to do. Ill give it a go and then you can critique/answer question. alright? [16:01] jussi01: OK. See how far you can get [16:01] persia: Atleast commanding "pq" in an Ubuntu console doesn't suggest installing any packages, if that is what you're talking about. [16:02] cyberix: postgresql-8.1 produces libpq4. Not likely to be too confusing, but wanted to make sure you knew. [16:04] persia: Ok. Thanks. [16:05] OK- so in the xdtv Makefile I've got these two lines: [16:05] install-exec-local: install-binPROGRAMS [16:06] chmod u+s $(DESTDIR)$(prefix)/bin/xdtv_v4l-conf [16:06] Thishould I just comment out both these lines? [16:06] or just that 2nd line? [16:06] or something else? [16:08] LucidFox still here? [16:09] I would imagine I just comment the second line [16:09] because the preceding line might do somrthing else [16:11] what does "install-exec-local: install-binPROGRAMS" do? [16:12] would nothing get installed if I commented that line? [16:12] E: bitlbee source: debian-rules-missing-required-target binary-indep [16:12] I get that from lintian and I'm not familiar with what it means. [16:16] hi all [16:16] I want to package my favourite application (anki) for ubuntu gutsy. [16:17] what steps should i follow to create a package with the right dependencies, etc? [16:18] freakabcd: have you done packaging before? [16:18] nope. i have made an odd rpm in the old days (of rh6) [16:19] !packagingguide | freakabcd [16:19] freakabcd: The packaging guide is at http://doc.ubuntu.com/ubuntu/packagingguide/C/index.html - See https://wiki.ubuntu.com/MOTU/Packages/New for information on getting a package integrated into Ubuntu - Other developer resources are at https://wiki.ubuntu.com/DeveloperResources - See also !backports [16:19] pretty new to deb packaging i must admit [16:19] freakabcd: yeah, the packaging guide is a good place to start. [16:19] freakabcd: Also finding an existing package for a program similar to yours can be very helpful. [16:20] I'm working on my first package now.. === allee_ is now known as allee [16:20] ScottK, well, the program (anki) is a very small one. has dependencies py-qt4 (>= 4.2) and python-sip (>=4) [16:21] ScottK: I finally decided to take a more standard package to start with vs my app I talked to you about [16:22] ok, I'm going to follow that guide and make my package. I hope I won;t encounter too many problems [16:23] well you know where to find us if you do.. (granted I'm not much help yet) [16:23] heh, thats ok. atleast you're 'older' than me in the packaging area :) [16:24] any of the head-motu realize that x is b0rken in hardy? [16:25] hmm? [16:26] hardy is going to be quite unusable for some time [16:26] crimsun_: ohh I know and I signed the virtual waiver :) [16:26] crimsun_: just did an update this morning and no more X since [16:27] complains about a broken package in xserver-xorg-core [16:31] StevenK: are you going to do the flwm merge? :) [16:36] I don't understand what this dpkg-shlibdeps in rules is all about. If I've added it to the right part of rules I'm missing some commands off the end as debuild fails. Does it need to be run on every binary produced by the package? 'dpkg-shlibdeps xdtv' at the end of my rules causes debuild to fail [16:37] but it looks out of place really, all the others being dh_somewthing [16:38] If I don't have it there debuild works ok but lintian complains it is missing [16:38] would it still get accepted into universe even if lintian complains about this? [16:39] What exactly is the error? [16:40] unable to open 'xdtv' for test at /usr/bin/dpkg-shlibdeps line 117 [16:41] I'm just guessing that 'dpkg-shlibdeps xdtv' is the kinda command I'm after? [16:41] its obviously wromg tho [16:41] or in the wrong place [16:42] Have you looked at dh_shlibdeps ? [16:43] knights, ^^ [16:43] somerville32: 'dh_shlibdeps -a' is already in my rules file [16:44] and its uncommented [16:44] It is merely a wrapper around dpkg-shlibdeps [16:45] So I don't think you need to call dpkg-shlipdeps again [16:45] so what could lintian be moaning about? [16:46] It was something like 'xdtv is an ELF app with shared libs so you need to run dpkg-shlibdeps' or something [16:46] Get rid of your call to dpkg-shlibdeps and just let the debian helper script handle it for you. [16:48] Suppose that your source package produces libfoo1, libfoo-dev, and libfoo-bin binary pack‐ages. libfoo-bin links against libfoo1, and should depend on it. In your rules file, first run dh_makeshlibs, then dh_shlibdeps: [16:48] dh_makeshlibs [16:48] dh_shlibdeps -L libfoo1 -l debian/libfoo1/usr/lib [16:51] somerville32: already set it compiling before you said that but I'll tell you what its complainig about soon [16:56] ok, which is the updated guide? [16:56] https://help.ubuntu.com/ubuntu/packagingguide/C/index.html [16:56] or http://doc.ubuntu.com/ubuntu/packagingguide/C/index.html [16:56] the copyright notice is repeated twice in one version. [16:56] so i assume they are atleast (a little bit) different [16:58] freakabcd: actually, the best might be to use the wiki version [16:58] freakabcd: https://wiki.ubuntu.com/PackagingGuide [16:59] This packaging could be more work than I anticipated- must I get no complaints off lintian before its accepted into universe or are some warnings (like missing manpages) OK? [17:03] knights: where there as some lintian complaints that aren't a big deal [17:03] *well [17:03] you should have a manpage though [17:04] I do have SOME manpages, but then those bins that do have them have lots of ' hyphen-used-as-minus-sign' errors- I take it these don't mayyer much? [17:05] matter [17:06] are pdfs not available of the packaging guide? [17:06] freakabcd: not currently no [17:06] i can;t find the pdf for it. i want to read it in evince or other app. i just hate reading stuff in the web browser :( [17:07] freakabcd: https://help.ubuntu.com/6.10/pdf/ubuntu/C/packagingguide.pdf [17:08] ah for edgy, not many changes from it i guess [17:08] some, bt it should work [17:08] *but === apachelogger_ is now known as apachelogger [17:13] Is there an MOTU who would be prepared to look at my package tesseract? I've uploaded it to REVU (http://revu.tauware.de/details.py?package=tesseract) [17:14] JeffreyRatcliffe: how big are these language packs? [17:16] Depends on the language. Typically 1-2Mb. [17:19] I've packages all 8 languages currently available, but only uploaded English. When it is clear that the packaging for English is OK, I'll upload the others. [17:24] do you mean they are separate source packages? [17:25] Yes. Each language pack is available separately, and I have packaged them separately. [17:28] does the new version handle layout analysis better? [17:29] are the GPL and apache licenses compatible? [17:31] Hardly anything is compatible w/ GPL :/ [17:32] knights: replace - with \- in the manpages where appropriate (e.g. it makes it easier to search for --help (hyphen and minus aren't the same in UTF-8)) [17:34] JeffreyRatcliffe: you can remove "g++ (>=4)" from build-depends as all Ubuntu versions have a g++ greater 4 [17:35] hmm seems GPLv3 is compatible but not GPLv2 [17:35] JeffreyRatcliffe: so you might want to make your packaging GPLv3 [17:36] JeffreyRatcliffe: and move Homepage into it's new control field inside debian/control [17:36] err.. i have a small question: is pbuilder going to install an entire (although minimal) distro again? [17:36] freakabcd: essentially yes [17:36] can;t i just use fakeroot (and alike) so it avoids having to make another system? [17:37] well, as long as you have a decent connection it's no problem [17:37] JeffreyRatcliffe: does one really needs all the languages for tesseract in Recommends? Recommends should be installed by default but can be removed (week depends) [17:37] pbuilder is very advisable [17:37] LaserJock, for what do i need a decent connection? [17:38] to download the packages for the minimal install that pbuilder is going to create [17:38] LaserJock, i don;t want to have *another* install just to build packages. the one i have should suffice [17:38] well, it's not exactly *another* install [17:38] How do I go about constructing the right dh_shlibdeps command for my package? AFAIK my packages doesn't create any new libs [17:38] LaserJock, i want to _not_ use pbuilder. the guide covers this method, right? [17:39] I haven't looked at the Apache in detail, but v1 was the same. I'll change it to GPLv3 anyway. g++ is still needed to build, so shouldn't I leave that line? [17:39] freakabcd: well, basically pbuilder is required, although sbuild is also used [17:39] JeffreyRatcliffe: change what now? [17:39] eh.. ok, how much space is pbuilder going to eat up? [17:39] on my hdd [17:39] freakabcd: ~100MB [17:40] woah, quite a bit. [17:40] plus the temp space for build depends [17:40] 100MB is not a lot. [17:40] yeah, depending on the package you can chew up a lot of space [17:40] I've seen some apps go ~2GB [17:40] pbuilder will unpack that, and then grab build depends from the net and install them into a temp dir while it builds [17:40] Thats a lot [17:41] pwnguin: the packaging licence [17:42] is g++ part of build-essential? [17:42] freakabcd: do you not have enough disk space? [17:42] yes. [17:42] LaserJock, let us assume that at the moment [17:42] JeffreyRatcliffe: g++ is part of build depends... [17:43] JeffreyRatcliffe: err, part of "build-essential" [17:43] freakabcd: hmm, that's tough. you can build without pbuilder but it's not encouraged [17:43] freakabcd: have you enough space to install all build-depends? [17:44] and depending on the package you'll still have to install build deps [17:44] geser, my system is essentially a damned dev machine. most of the dev packages are installed [17:44] * pwnguin notes that while not very efficient, you can use a PPA to host builds [17:45] freakabcd: but the point is to ensure that you've properly specified the build depenencies to avoid "works for me" scenarios [17:45] if you're just being OCD about disk space, I suggest you get over it. if you're truly low on disk space, that's something you should want to work around [17:45] pwnguin, i get the idea. but my package is small enough and the dep. list is also small enough. [17:45] you guys can prolly make a package for this app in seconds! [17:46] and i am *not* joking [17:48] freakabcd: and you are running hardy? [17:48] LaserJock, nope. [17:48] gutsy [17:48] freakabcd: you should build your packages on hardy [17:48] eh.. but i just installed gutsy [17:49] right, that's why we use pbuilder ;-) [17:49] freakabcd: no. only pbuilder environment should be gutsy [17:49] what [17:49] hellboy195, what? [17:49] you mean hardy [17:49] freakabcd: and new packages should be able to build in hardy (if you want them in the archive) [17:49] freakabcd: but you could use Launchpads PPA maybe if you really don't want to use pbuilder [17:49] i don't comprehend the sentence hellboy195 wroe [17:50] wrote [17:50] grr.. please stop it! [17:50] freakabcd: that might be for the best... [17:50] heh [17:50] don't flood all the possible options and why each option might be better [17:50] this is what scares off people from contributing casually [17:51] freakabcd: well, I'm just trying to help you out :( [17:51] freakabcd: äh sry. I meant hardy -.- [17:51] heh, i know. i'm just saying that providing a lot of options to inexperienced people is a very bad thing to do [17:51] just as they are learning. [17:51] LaserJock: I think PPA is the easiest for him!? [17:52] hellboy195: well, not really [17:52] I've got an old deb for edgy for my package- can I not extract the rules file from this somehow? [17:52] knights: no [17:52] freakabcd: well, you're the one who said pbuilder took too much space.... [17:52] dang! [17:52] freakabcd: pbuilder is *the* tool for build testing [17:52] if you dont want solutions, dont make up problems! [17:53] pwnguin: easy now :-) [17:53] freakabcd: when doing packaging there isn't always one "true" solution for a problem but often several [17:54] for example: lets say you are just learning about electrostatic interaction between charged electrodes in electrolyte. I am *not* going to tell you: (a)you can assume planar symmetry and use the Debye-Huckel approximation (b)solve nonlinear poisson-boltzmann equation numerically for the right answer or (c) use symmetry arguments for spherical -> cylindrical -> planar approximations [17:54] Well I'm stuck now it seems. I got most of the way- I just don't understand the man page for dh_shlibdeps and so I'm stuck on that [17:54] good news. packaging isnt quantum physics [17:54] knights: only if you still have the .diff.gz [17:54] see that? it might make you think: ok, this is a bad thing to learn [17:54] freakabcd: ok, so we gave you the one to use, pbuilder [17:55] one solution [17:55] you said that you didn't want to do that [17:55] so we're trying to figure out something here [17:55] hehe, lemme take some baby steps & go through the 'build deb from scratch' in the doco [17:55] freakabcd: actually, maybe you shouldn't do that [17:56] heh [17:56] really? [17:56] you might get overwhelmed with that [17:56] since it walks you through 3 different ways to package [17:56] yes [17:56] what? i browsed through it. seemed not too complicated [17:56] and starts witht he hardest [17:57] Can I not compile my app, then do a ldd on the binary, then that should tell me most of what I need to know for dh_shlibdeps? [17:57] freakabcd: ok, if it didn't seem complicated go for it [17:57] you mean from scratch -> debhelper -> cdbs ? [17:57] yes [17:57] freakabcd: yes [17:57] well, scratch will make me understand the process better [17:57] freakabcd: right, that's why I wrote it that way [17:57] indeed. and i *want* to learn :D [17:57] * pwnguin is tempted to make gender comparisons at the moment [17:57] ok, great [17:57] pwnguin, ? [17:58] freakabcd: ok, go for it! [17:59] freakabcd: and if you have questions feel free to ask [17:59] yeah, i'm reading. [17:59] thanks. [18:00] Don't suppose anybody here wants to pick up where I left off packaging xdtv? Its just a case of setting dh_shlibdeps correct and getting rid of the SETUID binary now- package builds fine apart from that [18:00] i can use linda wherever lintian is used, right? [18:00] I bet an experienced packager could sort this in a couple of minutes [18:01] i just read the desc in synaptic for linda and it said: 'better' than lintian [18:02] freakabcd: it's actually good to use both [18:03] ok, i'll install lintian as well then [18:03] freakabcd: a great tip for packaging is to run lintian -i and linda -i on the .dsc (source package) and .deb (binary package) [18:03] they can save you a lot of grief [18:04] ok, i'll remember this. i am writing all the concepts/tips and stuff i will be typing out in the term onto a txt file [18:04] going through the hello example now. [18:04] so it makes it easier for me to understand the entire process later on [18:05] freakabcd: please share any feedback on package learning with us too [18:05] Whats Debian's packaging channel called? I'll take my probs over there [18:05] we're well aware that the documentation is nowhere near perfect [18:05] knights: you're giving up here? [18:05] LaserJock: weren't you involed in edubuntu? [18:05] #debian-mentors on oftc [18:05] jpatrick: yes sir [18:05] LaserJock: No-one has been able to answer my question [18:06] knights: what was your question? [18:06] LaserJock, sure, i will. i can provide whatever doco i generate for myself, so the doco on the site can be edited if necessary [18:06] LaserJock: just so you know edubuntu-es.org is going up in about an hour [18:06] freakabcd: thanks! [18:06] jpatrick: oh? [18:06] jpatrick: who's doing it? [18:06] LaserJock: I've read the man page for dh_shlibdeps but it didn't make blip sense to me [18:06] LaserJock: some guy from Spain [18:06] I need to set it up in my rules file [18:06] knights: what are you trying to do with dh_shlibdeps? [18:07] I'm 'missinga depends line' [18:07] knights: so it's not adding enough? [18:07] So I need to know how to correctly construct said line [18:09] knights: are you getting a particular error? [18:09] Not really. The short version of the error is 'W: xdtv: missing-depends-line' [18:10] I could print the long error too but it doesn't add anything useful [18:10] that's from lintian? [18:10] or at build time [18:10] No, the lintian error is a bit longer but equally unhelpful [18:10] this is from lintian after its been built [18:10] knights: can you pastebin it for me? [18:11] Hi, I am trying to start packaging, but I got a problem, How can I set owner of a file in package, I saw some packages that used dpkg-statoverride to change file ownerships but is it correct? [18:11] LaserJock: Yeah [18:11] ssh_rdp: you can do it in a postinst script [18:12] LaserJock: with chown ? [18:12] ssh_rdp: quite possibly :-) [18:12] LaserJock: http://pastebin.ca/760006 [18:13] thats the full nasty [18:13] knights: running as root? :-) [18:14] Ommm! Yeah [18:14] Thats not the cause of the prob though is it? [18:15] knights: not particularly I don't think, but you shouldn't do that [18:15] packaging is done as a user [18:15] knights: can you pastebin your control file for me? [18:16] the only thing that needs admin privileges is pbuilder [18:17] LaserJock: http://pastebin.ca/760011 [18:17] LaserJock: OK [18:17] pwnguin: g++ is needed to compile the package, and therefore is part of build-essential. Is that not OK? [18:17] knights: ready for the solution? :-) [18:18] Dunno If i can handle this much excitement but, go on... [18:18] JeffreyRatcliffe: packages that are essential don't need to be added to the Build-Depends, they are assumed to be available [18:18] JeffreyRatcliffe: build-essential is a package on its own, installed for every build [18:19] knights: in debian/control you have no Depends: line for xdtv [18:19] LaserJock: Ahhhhh! [18:20] LaserJock: Thanks! I'll mess a bit more then [18:20] knights: you should have at least have Depends: ${shlibs:Depends} [18:21] the shlibs:Depends variable pulls in the shared deps from dpkg-shlibdeps [18:21] What good is Build-Depends, then? [18:21] JeffreyRatcliffe: for stuff that isn't essential [18:21] JeffreyRatcliffe: build-essential is only a few packages. it doesnt install debhelper, for example [18:21] JeffreyRatcliffe: there's only a small set of packages that are deemed "essential" [18:22] build essential is a package, Build-Depends is a field [18:23] I don't see why in this case autotools-dev should be in Build-Depends but g++ not [18:24] because g++ is "essential" and autotools-dev isn't [18:24] "essential" is a package priority [18:24] JeffreyRatcliffe: g++ is already brought in by default via the build-essential package [18:25] JeffreyRatcliffe: this is from the build-essential package description: [18:25] If you have this package installed, you only need to install whatever a package specifies as its build-time dependencies to build the package. Conversely, if you are determining what your package needs to build-depend on, you can always leave out the packages this package depends on. [18:28] JeffreyRatcliffe: is that making sense? [18:28] OK. So you are saying that I should have build-essential in Build-Depends, rather than g++? I hadn't seen build-essential. Yes. got it now [18:29] no [18:29] build-essential is *assumed* [18:29] packages in build-essential are assumed to be already installed [18:29] so you *only* need to add in the additional dependencies [18:30] OK. So neither build-essential, nor any of its dependencies. [18:31] right [18:31] builders will install build-essential automatically for you [18:32] well, not exactly [18:32] LaserJock: Am I likely to need to add ${misc:Depends} too? [18:32] to my control file [18:32] they already have the essential packages installed [18:33] knights: can you pastebin your rules file for me? [18:33] :-) [18:33] Indeed [18:35] LaserJock: http://pastebin.ca/760039 [18:39] knights: you most likely don't need it, but it doesn't really hurt [18:40] LaserJock: OK, thanks [18:40] knights: misc:depends pulls in any additional dependencies from dh_* [18:47] So if I change the packaging licence to GPL3 and remove g++ from Build-Depends, then tesseract is OK? [18:48] JeffreyRatcliffe: well ... I don't know [18:52] lintian is still giving me the error 'W: xdtv: setuid-binary usr/bin/xdtv_v4l-conf 4755 root/root' [18:52] even though I'm commented that chmod bit out of the makefile and [18:53] a small doubt: if app exists in debian and is changed in ubuntu, then version will be 2.2-1ubuntu1 [18:53] if app doesn;t exist in debian, revision is 2.2-1ubuntu0 ? [18:53] if the debian version is 2.2-1 yes [18:53] no [18:53] it'll be 2.2-0ubuntu1 [18:53] The line 'dh_fixperms -X xdtv_v4l-conf' is commented out in my rules file [18:54] sorry, yeah i meant to type 2.2-0ubuntu1 [18:54] it's -ubuntu [18:54] freakabcd: yep, that's right [18:54] JeffreyRatcliffe: I think there's some stuff you can clean up in debian/rules [18:55] LaserJock, so *all* ubuntu packages have the version as {version}-{deb_rev}{ubuntu}{ubuntu_rev} ? [18:55] not all. Most. [18:55] freakabcd: packages changed by Ubuntu yeah, well except some native packages [18:55] but the hello src package for gutsy debian/changelog has only 2.2-1 as the revision [18:55] freakabcd: because Ubuntu hasn't changed it [18:56] so if i am making a package for an app that doesn;t have an ubuntu package and doesn;t have a debian package, what would my version be? [18:56] 0.36-1 ? [18:56] or 0.36-0ubuntu1 [18:56] ? [18:56] well, 0.36-1 would mean it's for Debian [18:57] 0.36-0ubuntu1 would mean it's for Ubuntu [18:57] LaserJock, but if *most* packages like you guys say have the ubuntuX postfix, how come most of the packages don;t show the ubuntuX postfix in synaptic? [18:58] it just means they haven;t been changed from the debian package? [18:58] freakabcd: no, I didn't say that [18:58] I said most package that Ubuntu has changed have ubuntuX versions [18:59] ok, cool. so i will be using 0.36-0ubuntu1 [18:59] OK- so I need to get rid of this SETUID file now [18:59] and my unix is very poor [18:59] I suppose I need to search all the source files for chmod right? [19:00] source files? [19:00] something like 'find * | grep chmod' or something, maybe? [19:01] yeah- ubuntu doesn't lie setuid files [19:01] like [19:01] knights, err just use grep [19:01] grep -irn 'chmod' * [19:01] JeffreyRatcliffe: you can probably clean up the ./configure line [19:01] Thanks freakabcd! [19:01] well, actually there's probably not a chmod anywhere [19:01] 'i' just in case some script is using a var like: CHMOD=chmod and using $CHMOD everywhere it needs it [19:01] So how do I get rid of this pesky file? [19:02] LaserJock: How do I attack tit? [19:02] he! [19:02] sorry! bad typo that! :) [19:02] lol, i'll pretend i didn't read that sentence wrong :) [19:03] JeffreyRatcliffe: you also don't need specificy patch/unpatch rules [19:03] As I've said already- I've commented that chmod bit out of the makefile and The line 'dh_fixperms -X xdtv_v4l-conf' is commented out in my rules file [19:04] LaserJock: I have a debian/patch, so why no patch/unpatch? [19:04] These were the two lines I thought might be responsible for the setuid file [19:05] LaserJock: dh_link I can remove, I think [19:05] The line in the makefile looks like: 'chmod u+s $(DESTDIR)$(prefix)/bin/xdtv_v4l-conf' and thats commented [19:06] JeffreyRatcliffe: one sec [19:06] This is my last prob- honest! :) [19:08] <_16aR_> Hello [19:08] the control file for my app would be saying 'extra' for priority, right? [19:08] <_16aR_> Anyone has used scons with cdbs ? [19:09] the package (anki) will only be useful to those that know what the package is for. [19:09] atleast fits the description of 'extra' from the doco. is this right? [19:09] JeffreyRatcliffe: read man dpatch.make [19:09] JeffreyRatcliffe: it'll clean up the patching stuff [19:10] freakabcd: actually, probably optional [19:10] LaserJock, ok. i will do that [19:11] knights: well, if it needs suid it needs suid [19:11] Yesss! ;) [19:12] knights, got rid of it? [19:13] well, we don't like them, but if you either need to patch the package to not need of it or you gotta install it, that is *if* it needs it [19:13] crimsun_: you got any suggestions? I haven't dealt with suid binaries [19:14] freakabcd: nope, its still haunting me [19:14] freakabcd: "extra" is for packages that conflict with other packages [19:14] freakabcd: "optional" is sort of the generic priority [19:14] knights, what package is this? and what is the suid binary exactly? [19:14] LaserJock: I haven't been reading backscroll. [19:14] crimsun_: knights has a package that installs a suid binary and lintian complains [19:14] LaserJock, the language is ambiguous then. that sentence might need to be edited [19:15] crimsun_: I know we don't like suid binaries but what does a person do about it [19:15] freakabcd: xdtv- its a video/TV capture, viewing and streaming tool. Very powerful [19:15] freakabcd: make sure to write that down in your notes ;-) [19:15] I *am* writing it down :) [19:16] i just wrote down that the language might be ambiguous for describing the package priority === bmk789_ is now known as bmk789 [19:16] I thought I just took that from the Debian Policy but perhaps I wasn't clear enough [19:17] for the wiki we should just link to the Debian Policy section on it [19:17] maybe, i read it wrong. anyway, its in my notes regarding that sentence. [19:17] LaserJock: chmod it. [19:17] crimsun_: will that be sufficient? I wasn't sure [19:17] yes. [19:18] knights: listen to crimsun_ , he's a packaging god! ;-) [19:18] LaserJock, I think we could make a wonderful change: 'add example(s) for each package priority! [19:18] LaserJock: Thanks for the tip with dpatch.make. I'll finish cleaning up debian/rules tomorrow and upload a new attempt then... [19:18] (that's a marked exaggeration) [19:18] crimsun_: well, I was going to say *the* but I didn't ;-) [19:19] freakabcd: hmm, it's rare to see stuff other than optional [19:20] crimsun_,: OK, so do you think its just a case of me changing the 'chmod u+s $(DESTDIR)$(prefix)/bin/xdtv_v4l-conf' line in my progs makefile? [19:21] knights: no, just explicitly remove the setuid bit in debian/rules [19:21] crimsun_: That didn't work [19:22] knights: at what point did you apply the chmod? [19:22] I commented out the line that said 'dh_fixperms -X xdtv_v4l-conf' in my rules [19:23] knights: err, no. That doesn't do what you intend. [19:23] knights: you need that line in addition to chmod 755 $(DESTDIR)$(prefix)/bin/xdtv_v4l-conf [19:23] (assuming you wanted it 755) [19:24] I have the previous line uncommented (included) as well as that new line? [19:24] yes [19:24] OK, Thanks crimsun_ - I 'll try that now [19:25] does that chmod line go right at the end of my rules file? [19:26] depends. [19:26] it must appear prior to dh_installdeb [19:27] sorry, s/install/build/ [19:27] but within the binary-arch section? [19:27] if that's where it appears, yes [19:28] alright folks, I gotta run and get some real work done [19:29] OK- thanks for all your help LaserJock! [19:29] gotta get this dissertation done at some point ;-) [19:29] good luck with that! Whats it on? [19:30] knights: well, shooting little molecular machinery with lasers to see how the work [19:30] *they [19:30] that's not the official title ;-) [19:30] Cor! Sounds a bit techy that! [19:31] I'm a physical chemist [19:31] Like, nanotech? [19:31] yes [19:31] a psycho chemist is more like it [19:31] nanosensors mostly [19:31] :p [19:31] that means a lot coming from a Vista abuser. [19:31] nixternal: pffft, at least I'm no lamo MBA [19:31] scary stuff! I await the grey swarm yet :( [19:31] hey, the world needs more dumb people like me [19:32] oh man, I swear, crimsun_ says nothing forever, and then everytime I speak, I swear he says something [19:32] I am starting to think he is a bot :) [19:32] of course, this bot has a trigger for nixternal. [19:32] haha [19:32] I wouldn't put it past you [19:33] nixternal: read http://www.phdcomics.com/comics/archive.php?comicid=914 [19:33] nixternal: I think of you every time I read that [19:33] * StevenK appears back from Boston downtown [19:33] wb [19:33] hardee har har LaserJock [19:34] I am starting to think that MBAs are useless [19:34] of course they are [19:34] that's why everybody gets MBA+something useful [19:34] but, the richest people in America have one...and I want to be just like them :) [19:34] I want to pay off the government so lawsuits don't effect me and such :) [19:35] Ha. "Boston: Light rain" -- it isn't so light [19:35] I'd rather not get sued in the first place [19:35] StevenK: that is good though, because Boston really needs a bath [19:35] I would go buy some detergent and throw it all over the streets [19:35] "Temperature: 7C" "Feels like: 2C" [19:35] anyway, bbl [19:36] nixternal: Boston isn't that bad. [19:36] the only thing that sucks about all of this business school crap, is it really kills my engineering potential if I decide to continue this education stuff [19:36] I looked at my options and they are bleak [19:36] hmm? Why would it? [19:36] StevenK: if you say so...head about 30 minutes south into their burbs [19:36] I know people who get MBAs and MS xEs [19:36] crimsun_: because I don't have all of the high tech science and math courses behind me [19:37] nixternal: I'd rather not, I just spent a few hours in downtown Boston getting soaked. [19:37] heh [19:37] nixternal: that's normally not an issue; many committees simply require you to take prereqs [19:37] conditional admittance [19:37] crimsun_: ya, but the prereqs would be about the same as getting a BS almost [19:38] and seeing that I am more than halfway to a BS, I thought about getting it for the heck of it...the one benefit I have is free schooling [19:39] * nixternal imagines the people in the UK going "free schooling? we thought it was free everywhere" :p [19:40] university isn't free in the UK [19:40] far from it [19:40] it is crazy watching people drop like flies because not all education is free here [19:40] nixternal: you're just in the wrong field, I get paid to go to grad school ;-) [19:40] me too :) [19:40] by the US Government [19:40] tax free too [19:40] you can't say that much when you live in the US [19:40] so what's your sentence afterward? : ) [19:41] I already did my sentence and am still doing it [19:41] nixternal: paid paid? or just tuition paid? [19:41] LaserJock: paid paid paid [19:41] nice [19:41] UK/US - independence day my arse! [19:41] only thing out of my pocket is lab fees and books...so it only costs me about $300 a semester for 20 credit hours [19:41] same people running the show behind the scenes [19:42] hehe === asac_ is now known as asac [19:42] nixternal: I get $20k, health insurance, and tuition [19:42] but it's still rough [19:42] wow, that is almost exactly what I get [19:42] yeah, well mine's comming from tax payers too [19:43] but my health insurance has no out of pocket costs, everything is free there [19:43] so is mine :) [19:43] well, you guys are paying me, so I guess everyone wins [19:43] I have to pay out the nose for my wife [19:43] haha [19:44] crimsun_: tax payers are paying me as well...the good ol' Department of Defense employment [19:44] LaserJock: you go to a public school right? [19:44] ah, the behemoth that is the DoD. [19:44] nixternal: University of Nevada [19:44] go wolfpack! [19:44] well LaserJock is getting paid by tax payers as well [19:45] right, we're all contributing to each other's demise. [19:45] * LaserJock hugs DoD and DoE [19:45] heh, my reserve job is working along side the DoE actually [19:45] I was looking at government/military jobs the other day [19:45] for some reason they're mostly by D.C. :( [19:46] NavSeaSysCom OC-J8 and DoE re-certification on nuclear subs [19:46] I don't know why they wouldn't want a national lab in Montana... :-) [19:46] yup, DoE is right on Independence in D.C. [19:46] I will be there in the next couple of months [19:46] So who has the highest security clearance here then eh? ;) [19:46] we are supposed to go to New Mexico for training [19:46] nixternal: don't be a stranger. [19:46] crimsun_: of course not [19:47] I am moving out by you in about 6 months I think [19:47] anywhere from DC to southern MD to be exact [19:47] md or dc? [19:47] crimsun_: can people visit you at work or is pretty tight? [19:47] so I could be you neighbor or within 30 minutes [19:47] LaserJock: the latter [19:47] I know where I will be working, the only way for you to visit is to the have the Civilian or Military sticker on your front windshield [19:48] You could link yr PCs and get some seriously networked packaging done! [19:48] * nixternal has a Red sticker on his vehicle, so no saluting [19:48] serious :) [19:48] I've only been to a DoE lab before and it wasn't horrible, just had to go through some hoops, but nothing heroic [19:48] knights: we can't do anything with out base machines [19:48] I know my old machine at my last job had the puttied usb ports [19:48] we badge with rfids, so no, no visitors unless there's a legit reason. [19:48] Thats right [19:48] crimsun_: who do you work for? or will they kill you for telling? [19:49] ;p [19:49] err, s/out/our up a few lines there [19:49] do you allow unidentified msgs? [19:49] nixternal: "My name is Chen, Daniel Chen" ;-) [19:49] hahah [19:49] Daniel T. Chen damnit! get it right :p [19:49] opps, sorry [19:50] Department of Education? [19:50] energy [19:50] Energy [19:50] lol [19:50] haha [19:50] that makes me laugh [19:51] * LaserJock gets a security clearance to go to a top secret national lab run by the Department of Education ... [19:51] hahahaha [19:52] Farmy, not Fermi [19:52] lol [19:52] I was just at Fermi the other day [19:52] somerville32: dude, I can't stop laughing [19:52] you made my day [19:52] seems they are looking for some Linux folks in the area [19:52] \o/ [19:52] zooom! [19:53] Sorry, don't let me interrupt. I'm fascinated! :) [19:54] my job is boring though, and thankfully is only 2 weeks a year and 1 weekend a month long :) [19:54] although, I don't get to do my 2 weeks a year any place fun anymore though [19:54] Yes, but tell us about the tech please! [19:54] I so need to get back with a combat unit so I can get back to Puerto Rico [19:54] The ENERGY devices [19:54] the energy devices are nothing more than nuclear bombs :p [19:54] What can you tell us? [19:55] I mess with big guns and that's it unfortunately [19:55] Tesla's dad apparently had access to the Vatican library [19:55] that might explain a few things [19:56] you dont need a combat team to go to puerto rico... [19:56] haha, no but when you are with a combat team in the navy reserves, you get to go to cool places [19:56] I rememberwhen this room was all compiling and scripting :) [19:56] a combat engineering team, not like grunts and what not [19:57] combat engineering -- we can rig up a trubchet from local resources in 48 hours! [19:57] haha, that is more like it [19:57] actually, my scout troop built a catapult in about 6 hours [19:58] So what percentage of the people on this channel have military connections? Count me in obviously [19:58] well, our group does more along the lines of taking apart sea and land mines [19:58] I do [19:58] 3,4 [19:59] * pwnguin has a fantastic idea: instead of euthanizing stray cats... [19:59] my only 2 tours in Iraq were doing just that...and man, I have to say, Iraqis need help with brewing their alcohol....for flavor that is, because the alcohol is obviously there [19:59] let them clear minefields [19:59] lol [19:59] hahahahhahahahahahaha [19:59] nixternal: well afaik, they're not allowed to actually have any alcohol [20:00] pwnguin: really? because you either drink the tea or the brew [20:00] their tea will probably make you fail a pee test, but oh well [20:00] i think alcohol is against iraqi law [20:00] I think now that the taliban aren't in control, they allow you to drink [20:00] being a muslim taboo and what not [20:01] I know the qu'ran forbids alcohol, but I have no idea about the law [20:01] now my only military tours are at Great Lakes Naval Base..and boy are they boring [20:01] i recall a guy getting nearly shot up for speeding and running away from a patrol, turns out he just had alcohol in the trunk and thought the patrol would turn him in or something for it [20:01] nixternal: but if canada ever invades, hoo boy! [20:02] I know nothing much about the religion, but they do drink, and I noticed quite a few listen to rap music [20:02] people drank during the prohibition too [20:02] pwnguin: hahahahaha, I was with a Canadian outfit actually [20:02] I learned 3 things about Canada and their military that I never in my life never had any clue about [20:03] 1) they have a pretty bad arse special forces unit [20:03] 2) they are pretty elite at explosive ordnance disposal [20:03] 3) they will drink just about anything [20:03] apparently the alcohol laws were lifted a few years ago. im guessing that story was from before then [20:03] http://www.aliraqnews.com/23-5-16.htm [20:04] We're also pretty elite snipers [20:04] you don't get to see the snipers, literally [20:04] heh [20:05] which is funny...nobody there knows who the snipers are, but you know who the special forces units are [20:05] you would think it was the other way around [20:05] I've met several snipers [20:05] thats just the special forces they want you to know about [20:05] But I've never met a special force guy [20:05] Although I suspect I did meet one once [20:06] i knew a guy who did the rangers test [20:06] pwnguin: true [20:06] wound up leaving my school for westpoint [20:06] who apparently doesnt accept transfer credit [20:06] Delta < Special Task [20:06] I almost got booted from the military for trying to be...I killed my left knee and they wanted to give me a medical discharge...but I was able to pass the PT tests with flying colors [20:07] PT test is pretty easy :/ [20:07] somerville32: that would be hard to believe, since Delta Force is made up from the most elite people from all around the world and different branches of the military [20:09] I don't think thats true. [20:09] it is true [20:09] that is about the only info they will publish about the df [20:10] JTS is pretty hawt stuff. [20:10] JTS is equiv to Seals not Delta anyhow [20:25] does any MOTU have time to check if my package is ready for multiverse? norsetto had a look at it some time ago [20:33] which source package? [20:39] lol [20:39] Debian is a lot stricter than Ubuntu when it comes to QA :] [20:40] crimsun_: avidemux [20:40] that's probably unfortunate. [20:40] nixternal: A Canadian sniper currently holds the world distance record. He got it in Afghanistan for some mind bogglingly long shot (I don't recall exactly). [20:41] somerville32: It depends a lot on the DD. Some are, some aren't. [20:41] crimsun_: was that for me? [20:41] Hi all! [20:42] Kopfgeldjaeger: yes [20:42] crimsun_: because of the codec stuff and so? [20:43] Kopfgeldjaeger: oh, you mean the "unfortunate" bit? No, that was to somerville32. [20:43] I thought you meant my "which source package" [20:43] crimsun_, For ubuntu or debian? [20:43] somerville32: both, really. [20:44] lose-lose? [20:44] crimsun_: ah, ok :-) as i said, the source package is avidemux (the version in feisty and gutsy, which are the same, are very old and not very useful). a link to the daily svn: http://revu.tauware.de/revu1-incoming/avidemux-0711031810/ [20:45] how do i apply a debdiff? with patch? [20:46] nxvl: yes [20:46] crimsun_: thnx === asac__ is now known as asac === bmk789__ is now known as bmk789 === sacater is now known as wraund [21:26] mm, LTS strikes again. [21:30] :) [21:34] is there any suggestions how to test LTS upgrade. not yet of course. === warp10 is now known as warp10|gashofa [21:34] we run automated attempts at work [21:44] joumetal: for something more concrete in the unclassified world, try svn co svn://svn.debian.org/vlosuts/vlosuts === neversfelde_ is now known as neversfelde [21:45] * StevenK grumbles about people filing syncs for his merges [21:45] I did the hard part, merging three packages from Debian, one of them base-files so that anjuta could actually install its Build-Depends, and then someone else files the sync. Bah! [21:46] hey, what's a suitable DFSG license for game content? [21:46] like music, models, etc [21:49] pwnguin: e.g., CC-SA v3.0 [21:49] take a look at http://wiki.debian.org/DFSGLicenses#head-48e946e55f9c5d8595e5387a3546ba11a91dee03 [22:00] crimsun_: would an email to some debian list granting a recursive license be sufficent, or do you think it needs to be done at a website === neversfelde_ is now known as neversfelde [22:01] pwnguin: the source is definitive. [22:02] the current and possibly last release is ambiguous [22:02] pwnguin: if one feels that e-mail to a public [& archived] list is also necessary, feel free. [22:02] I'm writing the author asking for clarification, and it would help if i could tell him what to do [22:03] he simply needs to consent to license clarification (or relicensing). You can then distribute the relevant portion of the response in the copyright file. [22:03] (the latter part with his permission, which I presume shouldn't be a problem) [22:04] ok [22:04] StevenK: have you already seen the NBS file for libglib1.2? [22:05] err.. which package provides dh_make ? [22:05] weird, i don't have dh_make in my path [22:06] anyway, thanks for listening :) [22:06] StevenK: you can still ACK the anjuta sync request [22:06] freakabcd: dh-make [22:07] quick question, dholbrach did the packaging class and there was a specific type of bug that was supposed to be small/easy to get into [22:07] why did some DD are so zealots? [22:07] anyone have a hint on what it was that identified the bug type? [22:08] nxvl: because the confirmation process encourages it [22:08] rick_h__: I'd like to know as well [22:08] rick_h__: you mean the bitesize tag? [22:08] that might be it [22:09] heh. the bitesize thing mostly means the bug itself is easily fixed. it might still take you a while to figure the whole process out. but at least fixing the bug will be simple ^_^ [22:10] if you're new, those are the things to hit up [22:10] pwnguin: heh [22:10] pwnguin: good one [22:10] !bitesize [22:10] Sorry, I don't know anything about bitesize - try searching on http://ubotu.ubuntu-nl.org/factoids.cgi [22:10] !bitesized [22:10] Sorry, I don't know anything about bitesized - try searching on http://ubotu.ubuntu-nl.org/factoids.cgi [22:11] that is the one, thanks [22:12] ubotu: bitesize is https://bugs.edge.launchpad.net/ubuntu/+bugs?field.tag=bitesize [22:13] pwnguin: remove the "edge" from the url [22:14] :D [22:14] :] [22:14] and perhaps make the factoid a bit more explanatory [22:14] ubotu: bitesize is https://bugs.launchpad.net/ubuntu/+bugs?field.tag=bitesize [22:14] !bitsize [22:14] Sorry, I don't know anything about bitsize - try searching on http://ubotu.ubuntu-nl.org/factoids.cgi [22:14] !bitesize [22:14] Sorry, I don't know anything about bitesize - try searching on http://ubotu.ubuntu-nl.org/factoids.cgi [22:14] You need special accesses [22:15] somerville32: yes, i was trying to see it pwnguin add it or not [22:15] :P [22:15] and i'm not going to add the factoid until it actually explains something - clicking on that URL explains nothing to me tbh [22:15] You tell 'em girl! [22:15] doh [22:17] ubotu: bitesize is Bugs tagged trivially easy to fix: https://bugs.launchpad.net/ubuntu/+bugs?field.tag=bitesize [22:17] You might want to change that again [22:18] Make it it whatever. [22:18] im not very familiar with the bot [22:18] you can fix it :P [22:18] !bitesize is Bugs tagged trivially easy to fix: https://bugs.launchpad.net/ubuntu/+bugs?field.tag=bitesize [22:18] But bitesize already means something else! [22:18] !bitesize [22:18] Bugs tagged trivially easy to fix can be found here: https://bugs.launchpad.net/ubuntu/+bugs?field.tag=bitesize [22:18] LjL: too slow :) [22:18] /kb pici [22:18] Bugs tagged trivially easy? [22:19] yes [22:19] Is that.. legal use of the English language? [22:19] things like a misleading string, or an incorrect .desktop === bddebian2 is now known as bddebian [22:19] well [22:19] trivial can mean other things [22:19] like inconsequential [22:20] in fact, that's probably the actual definition [22:20] I might try: A list of bugs that are considered easy to fix and good for beginners to attempt can be found at: [22:21] run with it, it's yours now [22:21] I don't have magic accesses powers anymores. [22:21] !no bitesize is A list of bugs that are considered easy to fix and good for beginners to attempt can be found at: https://bugs.launchpad.net/ubuntu/+bugs?field.tag=bitesize [22:21] I'll remember that LjL [22:22] !whoami [22:22] Sorry, I don't know anything about whoami - try searching on http://ubotu.ubuntu-nl.org/factoids.cgi [22:22] does ubotu remember factoids across channels? [22:22] err [22:22] pwnguin, yes [22:22] or on a per channel basis too [22:22] it can do both [22:22] which is default? [22:22] accross channels [22:22] %whoami [22:22] I don't recognize you. [22:22] : ( [22:22] %whoami [22:22] imbrandon [22:23] good [22:23] back to DFSG campaigning [22:24] pwnguin, instead of like !blah is blah you can do !blah#ubuntu-motu is blah or similar, been a long time since i have messed with editing the factoids [22:24] there is a wiki somewhere that explains the commands, but then you still need editor access [22:24] This is true. [22:27] ok, i dunno if i just had a hallucination.. [22:28] how do i build binary packages? [22:28] is pbuilder the only option? [22:28] freakabcd, no but its the cleanest option [22:28] freakabcd: no, but it's the simplest "good" option [22:29] freakabcd: pbuilder and sbuild are the preferred tools to test-build before uploading. debuild, dpkg-buildpackage, and several more manual methods may also be used locally. [22:29] ok, but i want to build without using pbuilder. is this documented in the packaging guide doco? [22:29] freakabcd: you can use the methods persia outlined (debuild / dpkg-buildpackage) [22:29] probably not because its not a good idea most of the time [22:29] imbrandon: cleanest? Wouldn't wanna-build feeding sbuild be "cleaner"? [22:29] pbuilder is pretty easy to setup [22:30] freakabcd: however if you are targeting to get these packages into Ubuntu it's not frowned upon to test without pbuilder [22:30] rather, it is frowned upon [22:30] persia, hehe maybe but thats a pita for most to setup [22:30] well, it is a small package and i have no hopes of getting it into ubuntu at this moment in time [22:30] hell i still dident even get it working 100% right [22:30] imbrandon: Oh, I completely agree :) [22:30] imbrandon: https://wiki.ubuntu.com/UbuntuBots that wiki page? There is also a page with all factoids http://ubotu.ubuntu-nl.org/factoids.cgi [22:31] freakabcd: if you just need to build it for personal reasons, you can use debuild or dpkg-buildpackage inside the source's directory [22:31] that is the fastest way of getting the job done [22:31] geser, yea, its just been a very long time since i used the editing commands :) [22:32] * persia points freakabcd at https://wiki.ubuntu.com/PbuilderHowto and https://help.ubuntu.com/community/SbuildLVMHowto [22:33] is sbuild similar to pbuilder? i.e. sets up a minimal _distro_ ? [22:34] freakabcd, yes, all build options will use a minimal env to build to get ensure the build deps are correct [22:34] freakabcd: Sortof. sbuild relies on there being a different chroot system already in place (schroot is recommended). pbuilder bundles it. As jdong said, pbuilder is a little easier. The buildds use sbuild. [22:35] freakabcd: sbuild is for if you are setting up a build server or otherwise doing a LOT of builds where pbuilder is more for lighter usage [22:35] freakabcd: sbuild takes longer to set up but is a bit more efficient than unpacking and destroying a a minimal Ubuntu install between every build [22:36] pbuilder honestly is like 3 commands to set up... it's well worth the effort if you intend on building more than just a single package or two [22:37] say i use debhelper to setup the initial src package. how much work is usually required to eventually get to the binary debs ? [22:38] You just type: pbuilder build [22:38] freakabcd: For a clean upstream, it's a quick review and edit of three files. If it's less perfect, it can take weeks. [22:38] ok fun problem: upstream tarball doesn't have a root folder [22:38] i mean, from what i see it is just editing the copyright/control/changelog minimally and a little bit more work in rules [22:38] tar xvzf dumps a lot of files in the working dir [22:38] pwnguin, lol [22:39] persia, my intention is not to get it into ubuntu repos at the moment [22:39] imbrandon: funny, sure. so what should I do to solve it "correctly" [22:39] freakabcd: Ah. In that case, it's a quick edit of three files. A few minutes. [22:40] pwnguin, ummm solve what, dpkg extracts the orig.tar.gz where it wishes anyhow no matter what the upstream tar says [22:40] imbrandon: Well, it can be confused, but it handles 90% of the cases :) [22:40] Hi, I am looking for a MOTU to review my Package on REVU : I currently have 0 Advocates : http://revu.tauware.de/details.py?upid=462 [22:40] ok, thats what i wanted, sort of. I could have just built this src.tar.gz file myself (./configure && make &&まけいんsたll　部t椀〒dと負けあ羽c影疎染め尾根絵l背canインs多ll井tとお [22:41] oh crap! sorry [22:41] i went into the scim mode without noticing :( [22:41] lol thats a command I wouldnt want to type [22:41] Is it ok if I just cut n paste it :p [22:42] heh, that is english. i said ...make && make install, but wanted to have some sort of package that other people (without any experience building any sort of package) could install and use the app [22:42] StevenHarperUK, lol, sure try it out. [22:42] it will show stars in your console! [22:42] don;t believe me? try it out :p [22:42] Yeh I want tar ball first [22:43] freakabcd, That sounds like one of those scams to get their password :P [22:43] somerville32, haha [22:43] i will put up a notice though: Install at your own risk :D [22:43] freakabcd, sure then if you are gonna distribute the deb you really need to use pbuilder [22:44] this will ensure the package is consistant [22:44] ok, for now. i still haven;t managed to build a binary deb for myself :( [22:44] pbuilder build .dsc , then look in the results dir, it will be there [22:45] freakabcd, You could upload it to PPA [22:45] Did anyone hear about that new Trjon based as a Stripping game, it makes you (on windows) type the obfuscated text hidden in images to make it work, what its actually doing is getting the image security checkfrom sites like facebook and google and using you PC in the background to create new accounts and spam : made me chuckle. [22:45] Thats cleaver. [22:46] wow, that is indeed clever [22:46] Yeh its a smart way to get peopel to do stuff, porn is always a great motive.... [22:46] Not that I condone it [22:46] i am surprised they didn;t make it a full fledged TypingTutor (tm) with *excellent* graphics [22:46] :p [22:47] Are there any MOTU's reviewing today or are they all waiting for the 5th when its teh big REVU push? [22:47] StevenHarperUK, I could review it but I'm not a MOTU [22:48] ok, I'm gonna be slightly peeved if after 30 minutes of applying these patches to Azureus they cause it to crash. [22:48] somerville32: feel free to do a review and email me anything you find [22:48] link? [22:48] its MYNAME@gmail.com [22:48] http://revu.tauware.de/details.py?upid=462 [22:49] StevenHarper: Are you going to run for PM? :P [22:49] ah canadian PM jk Yeh? [22:49] ok, grr.. a debian package already exists [22:49] for the software i wanted to make a deb for [22:50] I was rather pleased to discover I share the same name as with the canadian PM [22:50] I wouldn't be [22:50] Plus he spells it differently [22:50] bah, should i just install the debian deb? or compile the tar.gz or build the deb and then install? or build a deb and send to review? [22:50] ul Freakabcd, im sure t here are procedures to repackage stuff from debain [22:50] grr.. [22:50] StevenHarperUK, It seems like you've been at this package for awhile :P [22:51] O_O how come this patch's indentation is completely different from upstream? [22:51] somerville32: its my first ever package : ther are lots of changes [22:51] somervill32: its also All my code [22:51] somerville32: you see I made it ans wanted to share, so I have been learning Packaging [22:52] somerville32: I do genuinely feel its ready now... [22:52] jdong: What are you doing to it? [22:52] somerville32: I must admit it was in a right state on my first submit [22:53] ok, it appears i have to try to make the deb myself. [22:53] The maintainer can be you, btw. [22:53] somerville32: a comment above says it cant [22:53] StevenHarperUK, what? you can;t be maintainer for your own code [22:53] ? [22:54] If you put yourself as the maintainer, maintain it. [22:54] I quote : "You don’t need to list yourself both in Maintainer and XSBC-Original-Maintainer. I’m not sure if you should set yourself as Maintainer (and no XSBC-O-M) or set Ubuntu MOTU as Maintainer and yourself as XSBC-O-M. " [22:54] StevenHarperUK: Maintainer must be @ubuntu.com, but it could also be you, if you have such an address. [22:54] Fujitsu: well doko made it pretty clear last time that he really wants those Fedora patches to Azureus on, even though there are clear indications from our bug report comments and from upstream that they are in part responsible for introducing the crash-at-splash-screen effect [22:54] or @kubuntu.org :P [22:55] I dont have an @ubuntu.com address [22:55] Fujitsu: so I figured I would give it a shot just so that I can say that I tried it but it caused a regression.... [22:55] jdong: What does he have to do with anything? [22:55] imbrandon: Right. *ubuntu* [22:55] Fujitsu: Important Java person? [22:55] Fujitsu: he has scolded me twice about refusing to apply those patches [22:55] and I'm scared [22:55] can I easily get an @ubuntu addess? [22:55] persia: Not sure how that affects it. [22:55] StevenHarperUK, becomes a ubuntu member [22:55] and you get one [22:56] It's just written in Java, and the patches don't affect the Javaness. [22:56] If you need anyone to check JAVA code I can help I am a Java come as a job. [22:56] imbrandon: How do I do that [22:56] StevenHarperUK: We've quite a few Java packages, and some have bugs [22:56] !membership [22:56] Want to become an Ubuntu member? Look at http://www.ubuntu.com/community/processes/newmember [22:56] !membership | StevenHarperUK [22:56] StevenHarperUK: please see above [22:56] persia: I code http://search.orange.co.uk in RL [22:56] ok, i'm going to make my first package now. dunno if it is going to be ok. will someone be able to check it for me? [22:56] persia: that takes 1-2Million hits a night [22:57] StevenHarperUK, Verbose should be commented out [22:57] or do i have to do ...some stuff.. to check it myself? [22:57] in your debian/rules [22:57] StevenHarperUK: You're too AJAX-happy. Epiphany complains when I try to close the page. [22:57] persia: Blame the companies webtrends solution ; i hate it [22:58] persia : I have no choice [22:58] Fujitsu: well, then I'll stick to the packaging the way it is now (i.e., working and not crashing) :) [22:58] jdong: If someone really wants patches, they could upload them... [22:58] persia : try the video search its a YOUTUBE 2.0 solution: the only people to use it n the world [22:58] * Fujitsu finds that that feature of Epiphany can be a tad annoying at times with some sites. [22:58] persia: That's a good point. [22:58] And then they're in the changed-by, and... their problem. [22:58] * persia likes it when there is actually data that might be lost [22:59] persia: Right, that's why I don't disable it. [22:59] persia: agreed [22:59] persia: try it in Firefox : we fully support that [22:59] jdong: I find azureus to actually work and be a bit usable now! Much better than it used to be. [22:59] !membership | StevenHarperU [22:59] StevenHarperU: Want to become an Ubuntu member? Look at http://www.ubuntu.com/community/processes/newmember [23:00] Fujitsu: yeah, totally, I'm really happy that I can apt-get install azureus and it works :) [23:01] StevenHarperUK, In debian/copyright you need to have a copyright statement for the actual package. [23:01] StevenHarperUK, ie. The Debian packaging is (C) 2007, Cody A.W. Somerville and is licensed under the GPL, see above. [23:01] GPL V2 :p [23:02] jdong: It used to work after apt-getting... just crashy and slow and generally useless. [23:02] I'm looking for first sponsor for my pq package. I have fixed all problems that have been brought up. http://revu.tauware.de/details.py?package=pq [23:02] Fujitsu: in Gutsy it totally stopped working after apt-getting altogether; someone must've breathed on the GCJ stack ;-) [23:03] Ah. [23:04] Fujitsu: yeah, I remembered in Edgy it worked near-perfectly out of the box with GCJ (i.e. slow but didn't crash) and I was pretty impressed... [23:04] but it just went downhill from there [23:04] Also, you should use dh_installdirs to create your directories not dh_install [23:04] StevenHarperUK, ^^ [23:04] I probably haven't used it since around then. [23:05] somerville32: I dont use dh_install [23:05] Hey, looks like azureus -0ubuntu2 finally built. [23:05] somerville: are you checking the latest submit [23:05] Fujitsu: yeah, I saw :) [23:05] backport is already approved and awaiting archive guys to come back from UDS :) [23:06] we're moving forward [23:06] jdong: Where coming back from UDS == going to AllHands? [23:06] (which is in the same place, isn't it?) [23:06] imbrandon, If you place just a directory in debian/install will it create it? And if so, should you place them there or in debian/dirs ? [23:06] Fujitsu: way to burst my bubble of optimism :) [23:07] StevenHarperUK, disregard my comment [23:08] somerville32: do you mind if after you have finished I submit my package again and you confirm that its all correct? [23:09] jdong: That's my job. [23:10] StevenHarperUK, I would be happy to. [23:12] StevenHarperUK, Also, your changelog has no final newline. [23:12] ta [23:13] somerville32: tell me when your done and want another build n upload [23:13] I'm building right now so that I can evaluate the binary package. === asac_ is now known as asac [23:21] somerville32: great, btw I obviously can make myself Maintainer until I get an @ubuntu address I have however humped through all the hoops ( see https://wiki.ubuntu.com/CommunityCouncilAgenda ) [23:21] StevenHarperUK, E: easycrypt: menu-item-creates-new-root-section Applications /usr/share/menu/easycrypt:3 [23:22] somerville32: Thats the NEW debail Menu format [23:22] somerville32: its correct to their new spec : all hardy stuff has to meet it [23:22] I wonder if the lintian and linda packages in Hardy have that [23:23] Make it easier to evaluate hardy packages [23:23] somerville32: http://www.debian.org/doc/packaging-manuals/menu.html/ch3.html [23:24] somerville32: yeh its confirmed that the gutsy ones throw nice red herrings :p [23:24] somerville32: keeps you on your toes & tests your memory [23:25] somerville32: My lintian merge (uploaded and built yesterday) adds that check. [23:25] Erm, fixes that check. [23:25] chilling [23:28] So, who has the magic powers to install the new lintian on REVU? [23:28] Fujitsu: great is that in the Hardy version then? [23:28] StevenHarperUK: It is. [23:28] persia, i can [23:28] is there a new one ready ? [23:28] imbrandon: Would you? That'd be great. [23:28] It would be nice to have that backported, probably. [23:28] imbrandon: Yeah, Hobbsee uploaded it for me yesterday. [23:29] Fujitsu: How much "backport" does it need? [23:29] Particularly with the new menu stuff. [23:29] somerville32: how you getting on? ready for a new upload yet? [23:29] Fujitsu, cool, can you give me a dsc url, i'll build and install it on sparky [23:29] StevenHarperUK, I'm installing new version of lintian and linda :) [23:29] persia: Shouldn't need any source changes. [23:29] imbrandon: Sure, sec. [23:29] Somerville32: great I haven't done that yet [23:30] imbrandon: http://archive.ubuntu.com/ubuntu/pool/main/l/lintian/lintian_1.23.36ubuntu1.dsc [23:30] That's the one. [23:30] are we backporting lintian? [23:30] Don't forget linda! :] [23:30] jdong: Why not? [23:30] Fujitsu: sounds like fun, should I queue up a test build before heading out? [23:30] jdong: At least for REVU. If you want to push it for -backports, that might make some people happy. [23:31] jdong, atm i'm just backporting it localy on REVU [23:31] somerville32: Is there a new linda in hardy yet? [23:31] There are changes in the menu system, so it would be nice for Gutsy people to easily get it. [23:31] imbrandon: ok, it might be worth pushing it out via backports anyway if all goes well [23:31] persia: There have been no hardy linda uploads. [23:31] persia, no :( [23:32] * jdong issues prevu lp:lintian and heads out [23:32] Fujitsu: That's what I thought. [23:32] jdong: Thanks. [23:32] sure thing [23:32] well REVU is still feisty so i need to rebuild it [23:33] imbrandon: I don't think you do, actually, but best to anyway. [23:33] StevenHarperUK, All good :) [23:33] Fujitsu, well its a sparc too soo best to anyway [23:33] imbrandon: It's arch: all. [23:33] err true [23:33] heh [23:33] still [23:33] Yeah, best to. [23:34] imbrandon: It's the right thing to do, even though it might not need to be done. [23:34] ah crap I think I just found a prevu bug too :D [23:34] jdong: Yay! [23:35] ok its building now [23:36] * Fujitsu notes that sbuild for each distroseries with deb-src lines for new releases would function well as prevu. [23:36] Because you can sbuild package_version and it will look for it in its sources... hm.. [23:36] Fujitsu: prevu is a lot more trivial to set up though; and the Hardy version now doesn't rely on any deb-src lines if invoked with lp: [23:37] jdong: Aha. [23:38] somerville32: ok i sec [23:38] * persia likes diversity, as a means to demonstrate bugs in the default toolset [23:38] Fujitsu: I'm using a simple LP scrape so that I can do things like "prevu lp:lintian" and it'll grab the Hardy lintian from LP [23:38] Fujitsu: that way I don't have to apt-get update so frequently and wonder if I am up to date or not [23:38] jdong: Yep, and that'll be much easier soon. [23:39] Fujitsu: yeah, when the real LP API comes out for doing this, I'll get rid of the nasty scrapejob :) [23:39] There'll be a +dsc or something that redirects to the .dsc, so it'll be trivial. [23:40] Fujitsu: Redirects to the a.u.c .dsc in the pool? [23:40] persia: The one in librarian, I think. [23:40] LP avoids referencing external stuff as much as possible. [23:40] Fujitsu: yeah, right now the main problem is that the dsc is not in the same dir as the rest of the sources [23:40] * persia hopes it is completely dget friendly [23:40] persia: That's the point of it, yes. [23:41] jdong: That's what I filed the bug about. [23:41] I've make a quick lpget hackjob that is a UI clone of dget based on the tidbit from prevu [23:41] * persia thanks Fujitsu for diligently watching LP progress [23:41] ok, now I'm starving. foodtime [23:42] * Fujitsu files too many Launchpad bugs. [23:42] Fujitsu: If you're not filing all of them, it's not too many :) [23:42] persia: Heh. [23:43] I was rather surprised when a number of my last batch were marked High and a couple cherrypicked. [23:43] I thought they were trivial things that would get Low at best, but apparently not. [23:44] Fujitsu: Why? One presumes you're getting good at filing clear bugs by now, and are a representative of a reasonable portion of the user community [23:44] True. [23:45] * persia points at http://ubuntu.linuxdc.it/ftbfs/ as a target for people looking for somthing to do, with props to DktrKranz [23:45] somerville32: its there http://revu.tauware.de/details.py?upid=521 [23:46] ok persia / Fujitsu , new lintian installed on REVU [23:46] imbrandon@sparky:~$ dpkg -l|grep lintian [23:46] ii linda 0.3.24 Debian package checker, not unlike lintian [23:46] ii lintian 1.23.36ubuntu1 Debian package checker [23:46] imbrandon@sparky:~$ [23:46] imbrandon: Thank you. [23:46] np [23:47] is there a new linda too ? [23:47] imbrandon: Not yet, [23:47] k [23:48] i have a doubt about LG bug #137993 [23:48] Launchpad bug 137993 in mplayerplug-in "mozilla-mplayer unnecessarily depends on gecko browsers" [Low,Incomplete] https://launchpad.net/bugs/137993 [23:48] Anyone with a package pending on REVU: lintian has been updated. Please grab the hardy lintian (backport in progress) to check locally, as there will likely be adjustments necessary for REVU day (starting in ~11 hours) [23:48] nxvl: What is the nature of the doubt? [23:48] persia, i steped away for a moment, did jdong start the backport ? if not i'll do it [23:49] StevenHarperUK, Looks good to me. I'd get imbrandon to look at it now [23:49] imbrandon: He said he was, and found a bug in prevu. I don't know the current status. [23:49] i have talk to the debian maintainer and he/she has told me that it will be resolved on the next upstream release which will be released any day from now on, should i fix or wait? [23:49] btw, where is the roadmap for hardy? is it already up there? [23:50] persia, k [23:50] somerville32: thanks a million : having people who know about packaging review it is the way I have got this far [23:50] nxvl: At this point in the cycle (archive open - debian import freeze) adding a note in the bug about the conversation with Debian is best, and wait. After Debian Import Freeze, it7s often easier to just fix it. [23:50] StevenHarperUK, I'm not expert :) It appears you've done a good job though. [23:50] s/not/no [23:50] !scheudle | nxvl [23:50] Sorry, I don't know anything about scheudle - try searching on http://ubotu.ubuntu-nl.org/factoids.cgi [23:50] !schedule | nxvl [23:50] nxvl: Ubuntu releases a new version every 6 months. Each version is supported for 18 months to 5 years. More info at http://www.ubuntu.com/ubuntu/releases & http://wiki.ubuntu.com/TimeBasedReleases [23:51] ubotu: thnx [23:51] Sorry, I don't know anything about thnx - try searching on http://ubotu.ubuntu-nl.org/factoids.cgi [23:51] :D [23:51] What! That should point to the development schedule in -motu [23:51] * nxvl hugs persia [23:51] persia, i can make a adjustment for the factoid for just in here [23:52] heh, there is information until dapper, not feisty or gutsy [23:52] :P [23:52] give me a few though, its dinner time [23:52] * persia further grumbles that http://www.ubuntu.com/products/whatIsubuntu/releases doesn't talk about Feisty, Gutsy, or Hardy: where do I file a bug? [23:52] persia, against the ubuntu-website product in LP [23:52] so the website team can fix it [23:52] imbrandon: Cool. Could you point it at https://wiki.ubuntu.com/HardyReleaseSchedule ? [23:52] imbrandon: Thanks. [23:54] persia, https://edge.launchpad.net/ubuntu-website [23:54] bug #159852 [23:54] Launchpad bug 159852 in ubuntu-website "http://www.ubuntu.com/products/whatIsubuntu/releases should mention Feisty, Gutsy, and Hoary" [Undecided,New] https://launchpad.net/bugs/159852 [23:54] persia, btw or you could just poke one of ..... https://edge.launchpad.net/~ubuntu-website/+members [23:54] :) [23:55] what is this Developer Sprint thing? [23:55] imbrandon: No point poking people. They are doing stuff. They should look at the bugs :) [23:56] somerville32: Thanks now I'm off I appreciate your help - cya [23:56] StevenHarperUK, Have a good one. [23:56] nxvl: Some developers get together to improve communication in an effort to get everything fixed (or at least well planned) as the deaadline for features draws near. [23:56] persia: like UDS? [23:57] UDS is more about planning [23:57] nxvl: Not really. Fewer people. More hacking, less meetings. Less of a schedule. It's just that putting people in a room can be more efficient than IRC. [23:59] and how can i go? paying the trip?