[00:00] You can also try the mailing list: https://lists.ubuntu.com/archives/ubuntu-art/ [00:00] domo, I'll ping them~ Thanks! [00:00] Awsoonn, no problem [00:09] NCommander, so... is it needed? [00:10] cody-somerville, well, the package got rejected, so I'd say so ;-) [00:10] NCommander, yea but you could make the lintian warning go away maybe [00:10] cody-somerville, its a bug in lintian; its detecting the bash shell script parser as a shell script [00:10] (and then complaining its not executable) [00:11] NCommander, bug #? [00:11] Haven't filed it [00:11] Because its not something that can be progamatically detected [00:11] and you're sure it is a bug? [00:11] It's detecting the parser as a script [00:12] you mean the shebang? [00:12] cody-somerville, have you some coment about my package now? [00:12] yeah [00:12] and what I'm suggesting is that if the file were to be installed elsewhere than maybe it wouldn't give a warning [00:13] the file is a sample/example right? [00:13] k0p, yes, please update it to meet python policy standards [00:13] cody-somerville, it goes in share/codeblocks/lexers ... [00:13] cody-somerville, python policy standards.. is it on debian/rules? :/ [00:14] yes [00:14] k0p, and debian/control [00:14] I read the NewPolicy page [00:14] and I don't know what I can change :/ [00:15] k0p, start on step 2. It says "Ensure that there's a ${python:Depends} in the Depends field " [00:15] NCommander, lexer_bash.sample is an example, no? [00:16] I thought it was what codeblocks used to lex bash scripts [00:17] split [00:17] wow, netsplit, been common today [00:17] Could be, thats why I was trying to clarify with you. [00:18] cody-somerville, is it about "update package " section? [00:18] k0p, you need to meet the same standard [00:19] cody-somerville, right. :) [00:33] cody-somerville, ping? [00:33] NCommander, pong [00:33] I'm making the changes right now [00:33] NCommander, what is the change? [00:34] Fixing the original maintainer for one [00:36] NCommander, yea but I can do that [00:37] No need to upload to revu just for that [00:37] Oh [00:37] Let me give you what I changed in lintian-overrides [00:38] cody-somerville, http://pastebin.ca/1080124 [00:39] maybe it would better fit in /usr/share/doc/codeblocks/examples? [00:40] azeem, no, its used as a test for the bash lexer it seems [00:40] (now that I actually sit down and look at it more closely; it won't work on Ubuntu; we don't have emerge/Portage) [00:41] used by what? [00:41] azeem, that was my idea [00:41] azeem, but apparently CodeBlocks needs that file [00:41] it needs a file that runs emerge? [00:42] azeem, it doesn't actually run it, but it parses it [00:42] I dunno, this is kinda weird === nhandler is now known as nhandler_AFK [01:14] Did us MOTUs do anything this month? :P [01:14] It is team reporting time. [01:17] Iulian: You can usually convince the md5sum to work properly for conversions if you use gzip -9fn [01:18] persia, gzip -d works fine for me [01:18] cody-somerville: If you have the same .tar with the same md5sum and a different timestamp, you will get a .gz with a different md5sum. [01:19] You can test this by playing with `touch` [01:19] -9fn ignores this timestamp (amoung other things). [01:19] Okay. Neat. [01:19] The doc I read just said the .tar files needed to match [01:21] Well, nothing needs to match. If the .tar files don't match, the paranoid among us will complain that the upstream source has been modified in undocumented ways. Bonus points for making the .tar.gz match so that debuan/rules get-oprig-source; cd ..; rm -r $(package)-$(version); dpkg-source -x foo.dsc works cleanly. [01:22] * persia notes that spelling things correctly grants even more bonus points [01:26] Is there an easy way to get all the package maintainers gpg pubkeys? === mcasadevall_ is now known as NCommander [01:27] I'm just trying to rid myself of warnings about unable to verify a package when I update my mirror. [01:28] Majost: source or binary packages? [01:29] persia, what did I spell incorrectly? [01:29] cody-somerville: Not you: me. Read the previous line carefully. [01:30] oprig? [01:30] As one example :) [01:32] debuan would be the second and final one? [01:32] It would be another. I stopped counting at that point, so can't verify final. [01:33] cody-somerville, so how goes it? [01:33] still building [01:33] Ouch [01:38] persia: both [01:39] Majost: For binary packages, the ubuntu-keyring package ought have everything. I don't believe there is a correct solution for source packages for Ubuntu at this time, due to the various ways these packages are imported. [01:40] * Majost nods [01:41] the best I have come up with is to run a script against the dsc files and try to pull the keys from various keyservers [01:42] but it takes a loooong time, and isn't very accurate. heh [01:58] is there a way to make audio work inside a chroot of intrepid? [02:01] Awsoonn, try mount -o bind /dev *chroot*/dev [02:01] THat might do the trick [02:01] Awsoonn: That's an interesting question, and one to which NCommander has already given my answer :) [02:02] If your package needs ESD or firends [02:02] YOu'll have to kill them on the host and run them in the chroot [02:02] Or possibly bind-mount /tmp, too? [02:02] Alternately, also bind-mount /tmp, but that can have even more interesting side effects [02:02] Note that you need matching /etc/passwd and /etc/shadow in the chroot for either of these to work properly. [02:02] #ubuntu-motu is apparently _all about_ messing around in chroots :) [02:03] Certainly nothing else happens around here. [02:03] ;-) [02:03] * NCommander watchs ScottK-laptop's laptop explode [02:03] Pretty sparks! [02:04] w00t, I just got REVU to login using OpenID [02:04] * NCommander laughs evilly [02:12] emgent: Please make sure that packages you work on are buildable and installable before uploading. I'm just having to fix up audacious after you uploaded a new merged version that has broken dependencies. I know you didn't make those changes yourself, but if you had checked the packages installed, you would have caught this problem. [02:12] uhm? i tested it.. just a moment TheMuso [02:13] * TheMuso notes that the current state of audacious in intrepid for i386, amd64, and powerpc is uninstallable. [02:15] emgent: Don't worry about it now, I'm uploading a fix, but please check in the future. [02:15] TheMuso: no, there is a big problem. [02:15] * TheMuso cancels the upload quickly. [02:16] emgent: What problem? [02:16] hahah :) [02:16] cody-somerville, did that upload? [02:16] TheMuso: just a moment i'm re-testing it in my real intrepid box. [02:16] No [02:16] NCommander, your override didn't work [02:16] Crud [02:17] TheMuso: ok can you upload. sorry. [02:17] I'm just going to remove the override. It is just a warning and not an error. [02:17] emgent: Ok will do. [02:18] TheMuso: thanks for this. [02:18] cody-somerville, I thought you should always have them overriden ... [02:18] emgent: No problem, just make sure to check in the future ok. [02:19] sure. will do. I do not understand how I escaped. :-\ anyway thanks TheMuso [02:20] cody-somerville, hold on, I think I found why I didn't work, and uploading to PPA now [02:20] can anyone help me with this: http://ubuntu.pastebin.com/d55ae8b34 [02:21] NCommander, you forgot the binary keyword? [02:21] cody-somerville, forgot the : :-) [02:21] THat too [02:28] NCommander, I also just noticed that tips.txt is getting installed to /usr/share/codeblocks/ [02:29] That's a data file [02:29] It's what it uses to show tips at startup [02:31] error: could not create '/usr/lib/python2.5/site-packages/blueman': Permission denied [02:31] how can i overcome this error [02:31] ? [02:32] NCommander> https://lists.ubuntu.com/archives/ubuntu-archive/2008-July/019791.html [02:33] whoops [02:33] WTF, I didn't get that email [02:33] I had talked to the archive admin over IR [02:33] you aren't subscribed to that mailing list, probably [02:33] Anyway, why don't you take package descriptions from my version? [02:33] I'm suprised REJECT emails aren't CCed to the direct to the changelog email [02:33] THat's how it works in Debian [02:34] They usually are [02:34] I didn't get that reject e-mail either [02:35] * NCommander grumbles [02:35] persia, Can't LGPL 3 be substituted with GPL 3? [02:35] Yeah, it says it can be [02:35] cody-somerville: No. LGPL code may be used in a larger GPL 3 work. [02:35] Wait [02:36] (err. LGPL 3) [02:37] That must have changed in LGPL 3. LGPL 2 you could, IIRC. [02:37] If one file in GPL 3, then the whole thing becomes GPL3 [02:38] cody-somerville: I suspect the point is that the COPYING file you include should be consistent with the debian/copyright file of the package? [02:38] NCommander: No. The collection becomes GPL 3. The individual components may still be LGPL 3. [02:39] Right, I realize that [02:39] But then the copying file would be GPL 3 [02:39] And as an added WTF, the files ARE licensed until the GPL3 [02:39] The only thing I can see that are LGPL are some of the plugins [02:39] cody-somerville: No. LGPL 2.1 had the fixes to be GPL 3 compatible. LGPL 2 was GPL 2 compatible, but not by direct replacement (although relicensing was acceptable). [02:40] * persia prefers debian/copyright to specify the actual licenses of the code, rather than just some random overall license [02:40] * RAOF was always under the impression that was actual _policy_. [02:40] The overall license of the code though is GPL3 [02:40] If code from LGPL 3 and GPL 3 is mixed in a single file, the file becomes GPL 3, but if the code is in separate files, it is interesting to preserve that separation to allow other LGPL 3 projects to use the LGPL 3 portion of the code. [02:40] NCommander: As a collection, yes. [02:40] If an LGPL3 project and GPL3 project are linked together, it becomes a GPL3, its GPL3 as a collection [02:41] Yes, the individual files can kept a seperate license however [02:41] And if the individual files have separate licenses, this should be documented in debian/copyright. [02:41] NCommander, have fun going through the 1302 source files :) [02:41] *grumbles* [02:41] I'm asking the archive admin for clarification and a proper course of action [02:42] This is why I find debian/copyright invariably the most complex and annoying part of packaging. [02:44] yup [02:44] I'm asking the archive admin to clarify what he wants fixed [02:44] * persia recommends `less $(find . -print)` as a useful way to review source. [02:45] It's just the SDK that's LGPL3, the vast majority is GPL3, which is why I missed it [02:45] `less $(suspicious-source)` and `grep -ri copyright` are additionally useful. [02:45] so much for trusting the COPYING file to be right [02:46] NCommander: does that make it a copyleft file ? :-) [02:46] * NCommander whacks jmarsden for such a horrible pun [02:47] Ok, now its up to the archive admin [02:49] Better to be correct through diligence than to wait for an archive admin to report that debian/copyright must accurately represent the licensing of the code... [02:53] jmarsden, what does this mean?: " Replaced generic .desktop file with a koverartist-specific one" [02:54] Legendario: People in this channel were mainly busy at the time you posted your question; so no-one has picked it up yet. [02:54] The original had stuff related to KParts in it. [02:54] I think it was auto generated? [02:54] Legendario: Generally the way to deal with that is to wait until the current thread of conversation has ended, and give another prod. [02:54] Maybe something along those lines would be a better description for your changelog jmarsden [02:54] OK. [02:55] Legendario: As to your actual question: I believe that you're not passing the correct prefix to setup.py, so it's trying to install system-wide rather than in $(CURDIR)/debian/where-the-package-should-go. [02:55] jmarsden, and why is that patch named 01_internationalize_desktop_file.patch ? [02:56] Because it adds several languages to the file that were not in it before... [02:56] Per the bug sumbitter's request. [02:57] I thought the bug was about making it show up in Gnome [02:57] Thanks a lot RAOF, i thought i was misbehaving somehow... [02:57] I could call it 01_replace_desktop_file.patch if you prefer. [02:57] and considering the .desktop file is already internationalized, your name makes things confusing [02:58] cody-somerville: Yes, and he asks that a set of languages be "ideally" added... right? [02:59] To quote: "It is necessary in ideal that somebody translate these sections to those languages: French, German, Italian, Polish, Portugese and Spanish. Need help." [02:59] I dpn't speak Italian, Portugues or Polish, but can do French, German and (somewhat) Spanish. [02:59] So.. I did :-) [03:00] jmarsden, What I'm saying is that your patch is not to provide internationalization, that is not the *purpose*. [03:00] OK... I can change its name, no problem. [03:00] It is *a* purpose, but not *the* primary purpose. [03:01] jmarsden, the purpose of your patch is to reduce the number of languages the comment is translated into? :S [03:04] It's renamed now.\ [03:06] Any ideas on why I still see those lintian messages, which was my initial question? :-) [03:08] hello all, i want to learn the intricate magic of packaging. ive been watching the motu videos on youtube, and im wondering if I can get someone to bug so they will teach me. thanks [03:09] posingaspopular, no, sorry. : ( [03:09] :( :( [03:09] cody-somerville: now everybody hates me? [03:09] just joking! \o/ [03:10] everybody jokes at my expense, because they hate me? [03:10] posingaspopular, yes, sorry : ( [03:10] Heya gang [03:10] lol thanks [03:11] jmarsden, can you pastebin the lintian errors you get? [03:11] cody-somerville: Sure. [03:11] posingaspopular: Rather than bugging someone in particular, consider chasing some bug, and asking here for help in getting the fix uploaded once you have a working solution. After several repetitions , you'll have it down. [03:14] cody-somerville: http://pastebin.ubuntu.com/29461/ [03:14] ah i am, how do they say, 'new at this' [03:14] posingaspopular: https://wiki.ubuntu.com/MOTU/Contributing might be worth a look [03:15] thanks [03:16] * jmarsden was new at this (Ubuntu packaging) about 3 days ago... [03:16] I just realized I haven't ate today :/ [03:16] No wonder I have a headache. [03:16] jmarsden, I'm going to go fetch some food [03:16] cody-somerville: Then koverartist can wait. No problem! [03:18] where do i go to apply for a sponsor [03:19] posingaspopular: We sponsor update candidates, rather than people. [03:19] ? [03:19] Prepare a good debdiff for something, and subscribe the sponsors queue to get it uploaded. [03:19] erm... mentor [03:19] posingaspopular: https://wiki.ubuntu.com/MOTU/Mentoring [03:20] and https://wiki.ubuntu.com/MOTU/Mentoring/Contributor [03:20] thanks, sorry for asking so many simple questions... they are all in the chan. topic i assume? [03:21] posingaspopular: Some of them are, but we're generally happy to answer most questions here, as long as it's roughtly related to universe packaging. [03:21] posingaspopular: Maybe not... browsing the wiki starting at https://wiki.ubuntu.com/MOTU will help too [03:27] hey folks, after running a sudo pbuilder build *.dsc I get the following error: dpkg-genchanges: failure: cannot open upload file ../qlix_0.2.3-0ubuntu1~ppa1_i386.deb for reading: No such file or directory [03:27] what is going on/ wrong? [03:30] SolarWar: the ~ppa in the filename looks odd for something you are building locally... should it really be there? [03:30] jmarsden, why would it be odd? [03:31] Because the PPA is up on a remote server, not on yoru desktop... right? [03:31] jmarsden: Some people test-build before uploading to the PPAs :) [03:31] well there you have it :) === nhandler is now known as nhandler_AFK [03:31] my deb in launchpad is mysteriously missing debug symbols, so i decided to compile on my local machine [03:33] SolarWar: That'd be because standard procedure is to strip the binaries. [03:33] oh is that true? is there a way to avoid that behavior? this application is not stable :( [03:34] You can remove the call to dh_strip (and then lintian will complain), or you can generate foo-dbg packages. [03:34] i'm not sure what foo-dbg packages are [03:35] foo is the name your package [03:35] oh i see [03:35] dh_strip has the option of stripping the debug symbols and including them in a -dbg package rather than just throwing them away. [03:36] SolarWar, so you'd have to also add a binary package to debian/control and modify the call in debian/rules from dh_strip to sh_strip --dbg-package=qlix-dbg [03:36] erm [03:36] *dh_strip --dbg--package=qlix-dbg [03:36] What would be even nicer is for the PPAs to have pkg-create-dbgsym installed & the other magic required to build dbgsym packages. Then all PPA builds would automatically generate debugging symbols. [03:36] * cody-somerville nods. [03:37] RAOF: You could file a wishlist bug against LP asking for that :-) [03:37] I file a bug for that some months ago :) [03:37] Ah. [03:37] jmarsden: Too slow! :) [03:38] hrm, thanks guys, I will look into it [03:38] =) [03:39] hey SolarWar [03:40] heyo [03:40] kostmo, be glad *you* don't have to deal with dh_strip ;) [03:41] cdbs all the way, man [03:41] haha [03:42] SolarWar, so you can vouch for how awesome pyrocket is, right? [03:42] RAOF: The big problem is the lack of Soyuz support for storing them. [03:42] I met my wife through pyrocket, its totally awesome ;) [03:43] lol that makes no sense at all [03:43] anyone have any thoughts to the pbuilder error i was receiving earlier? [03:43] NCommander: Is the codeblocks you guys have been working on packaging the same one in https://bugs.launchpad.net/ubuntu/+bug/105428 ? [03:43] haha yea i know- pyrocket is still pretty sweet [03:43] Launchpad bug 105428 in ubuntu "[needs-packaging] codeblocks" [Wishlist,Confirmed] [03:44] looking for more MOTU advocates, of course - http://revu.ubuntuwire.com/details.py?upid=2940 [03:44] jmarsden, yes [03:44] NCommander: OK. I wonder what the packages mentioned there did regarding the copyright issue -- any help there for you?? [03:45] I emailed the package admin [03:45] er, archive [03:45] if I don'thear back from him in a day or two, I'll do what I think is right, and request a reupload [03:47] When packaging an app that uses init scripts... how I specify which runlevels get symlinked? Do I just manually add them in a links file? [03:49] Majost: look at dh_installinit ;) [03:49] ah [03:49] thanks [03:54] cody-somerville: Have you recovered from your accidental starvation? :-) [03:55] I only get 20 minutes to eat?! Sheesh. This is a shadier job than I thought. [03:55] cody-somerville, lol [03:58] cody-somerville: Hehe... I got my food when you mentioned food... and my plate is now empty, so... :-) [03:59] jmarsden, I haven't moved yet [03:59] I'm still trying to shake off everyone else [03:59] cody-somerville: Ah, OK... shake them harder :-) [04:00] * jmarsden returns to fridge, thinking... "dessert?" [04:03] hey guys, can someone re-sync the REVU uploaders keyring please? thanks :) [04:08] Incidentally, is there a (security?) reason why REVU doesn't just look for and grab a new key from keyserver.ubuntu.com when someone uploads a package to REVU and it doesn't have a matching key already?? [04:09] * jmarsden has returned with strawberries and icecream :-) [04:16] nellery, i had the same confusion when i started working on: https://bugs.launchpad.net/ubuntu/+bug/250995 :) [04:16] Launchpad bug 250995 in ubuntu "[needs-packaging] Qalculate plasmoid for KDE 4.1" [Wishlist,In progress] === nhandler_AFK is now known as away === away is now known as nhandler [04:17] jmarsden, I'm going to head to bed. I didn't reaized how late it is. [04:17] can somebody look at https://bugs.edge.launchpad.net/ubuntu/+source/inspircd/+bug/247358 [04:17] Launchpad bug 247358 in inspircd "FTBFS in Intrepid" [Undecided,In progress] [04:17] marnold, sure [04:18] cody-somerville: OK, no problem! [04:18] marnold, the patch looks good to me [04:19] anyone have any idea whats causing this problem: http://rafb.net/p/R0R1fg77.html [04:19] someone else did it but i'm one of the Debian maintainers so i kinda feel responsible [04:19] SolarWar: Generally it's considered rude to ask the same question on multiple channels at the same time. [04:20] SolarWar: Do you not have a debian/compat file? Or does that file contain a really small number? [04:22] NCommander, what happens now i haven't done anything for Ubuntu directly for awhile [04:22] ScottK-laptop, I asked it earlier and it didn't receive much attention here, thats when i turned to kubuntu-dev [04:22] all my stuff has been through Debian [04:22] SolarWar: OK. I just saw you asking in both places one minute apart. [04:23] marnold, are you an MOTU? [04:23] no [04:23] marnold, DD? [04:23] no [04:23] (or DM on inspiricd) [04:24] no its kinda hard to get keysigs out here in the sticks [04:24] It's quite possible to get stuff into Debian just by getting sponsored. No need to be any of that. [04:24] i'll see if i can find one of my sponsors [04:25] ScottK-laptop, I've never met anyone who would sponsor a NMU [04:25] I've had NMUs sponsored. [04:25] * StevenK has sponsored a few NMUs [04:26] i'm Matt Arnold so its not an NMU [04:26] In one case the package maintainer just uploaded my NMU rather than bother with his own update. [04:26] StevenK: Would you be up for a sponsorship request in Debian? [04:26] ... speaking of which ... [04:27] Oh, d'oh [04:27] marnold, just remember to change the version number on the changelog to a non-ubuntu one ;-) [04:27] RAOF, my compat file contains 4 [04:27] and it doesn't ftbfs in Debian for some odd reason [04:27] marnold, that's because Debian uses an older libc even in sid [04:27] We upgraded to 2.8 [04:27] sid is still 2.7 [04:28] * marnold slaps self [04:28] * marnold should have known [04:30] SolarWar: It's obviously an old package, then. [04:32] RAOF, the compat is five now I think [04:33] NCommander, 5 or 7 [04:33] There's a 6, too :) [04:33] 4 is not deprecated. [04:33] If it'll build with debhelper 4, there's no reason to bump it. [04:34] i forgot about those [04:34] True. I'm surprised that it's claiming the compat version is too old, though. [04:36] well the machine that built the dsc had debhelper 6 [04:36] i'm still receiving the same error at the end of the log- namely "cannot upload file ../qlix_0.2.3-0ubuntu1~ppa1_i386.deb for reading: No such file or directory" [04:36] SolarWar: Can you pastebin the full build log, then? [04:36] sure [04:53] RAOF, http://rafb.net/p/9XvEWB22.html [05:12] SolarWar: Oh. You've messed up some versioning somwhere. [05:12] dpkg-deb: building package `qlix' in `../qlix_0.2.4-0ubuntu1~ppa*2*_i386.deb' ... dpkg-genchanges: failure: cannot open upload file ../qlix_0.2.3-0ubuntu1~ppa*1*_i386.deb for reading: No such file or directory [05:13] hrm [05:13] its also looking for 0.2.3 [05:13] and in some cases its doing weird stuff with 0.2.2 [05:13] oh wait no thats an older log [05:15] okay lets try this again [05:17] now i just get: dpkg-genchanges: warning: duplicate files list entry for file qlix_0.2.4-0ubuntu1~ppa2_i386.deb (line 2) [05:20] SolarWar: That's very strange. What's in your .dsc? [05:21] RAOF i have a "files" file in my debian folder still that aparantly i don't need and was causing the initial grief, i think removing that will fix the problem [05:21] lintian complained about it as well [05:21] Ah. You'd built the source package incorrectly :) [05:23] :) yea [05:24] man, launchpad is really lenient when it comes to building source packages [05:26] hey all, im trying to follow along with packaging ed on youtube at this link http://www.youtube.com/watch?v=SwTp1YnehoI&feature=related but i get this build error http://pastebin.ubuntu.com/29500/ [05:26] can someone shed some light on this msg? [05:27] Line 13 of debian/control is strange. Can you pastebin debian/control? [05:28] RAOF, is that for my benefit ? [05:28] SolarWar: No, for posingaspopular [05:30] RAOF i have created a separate debug package for this application, does this package contain debug information only? or should it contain the whole application with debug information? [05:30] yes [05:30] SolarWar: Just the debug symbols. [05:30] If you've done it with dh_strip, at least. [05:31] RAOF i'm assuming that debian tells gdb where to look for these symbols then? [05:31] Right. [05:31] http://pastebin.ubuntu.com/29502/ [05:31] i was just trying to follow along with daniel... [05:32] Your description needs to be space-indented; each line of the description needs to start with a space. [05:33] Otherwise it thinks that each of those lines is meant to be another field, like Maintainer: me , and it gets confused :) [05:33] RAOF: and line 13, which is currently empty, needs to not be there at all, right? [05:34] A good idea, yes :) [05:34] hmmm [05:34] posingaspopular: where did you get that file from? Did you type it in yourself? [05:34] ah i see [05:35] jmarsden: no im following along on the motu chan. on youtube [05:35] posingaspopular: Yes, but you can't cut and paste frm a Youtube video... can you? [05:36] .. no but i was trying to write in what i saw daniel typing. must have messed up the syntax [05:36] Anyway, try something more like http://pastebin.ubuntu.com/29503/ [05:36] OK. [05:38] ah thank you, i apprecaite it [05:38] No problem :-) [05:40] it worked! [05:40] ;p [05:40] aright i'm off for the night [05:40] thanks guys :) [05:41] posingaspopular: Contragulations :-) [05:42] Can anyone confirm that my comment yesterday re https://bugs.launchpad.net/ubuntu/+source/openoffice.org/+bug/223476 is somewhat sane? If so, I might even try to implement it, although the build time for OOo must be looong. [05:42] Launchpad bug 223476 in openoffice.org "Failure to have openoffice.org-math intalled causes "data corruption"" [Critical,Confirmed] [06:06] jmarsden: When you say "loooong", I believe what you mean is "overnight, and I hope you've got multiple gigabytes of RAM" ;) [06:07] That bad! Ouch. Overnight is doable, but I don't have that much RAM... not in a box running Ubuntu, anyway. [06:07] Long as in way worse than Eclipse and that's horrible. [06:07] * jmarsden wonders about "borrowing" hardware from work temporarily :-) [06:08] WHat happens if you do a debbuild -S and then upload to PPA -- will it build there?? [06:08] Yes, although that may be a bit of an abuse of the PPA system. [06:10] Any way to tell PPA "build this in the background" to avoid the abuse issue? [06:10] * wgrant would do a change like that without a rebuild. [06:10] *test build [06:11] As long as it had been built recently. === beuno_ is now known as beuno [06:12] hi [06:12] i'm trying to fix bug 249452 in bzr packaging [06:12] Launchpad bug 249452 in bzr "bzr overrides the shell prompt settings" [Critical,Triaged] https://launchpad.net/bugs/249452 [06:13] we have a debian/bzr.install file [06:13] which lists some things to be installed into /etc [06:13] can someone point me towards documentation on how this works [06:14] man dh_install [06:14] wgrant: So I could just edit the control file to add the dependency, debuild -S and debdiff and attach that to the bug in LP? [06:14] jmarsden: And add a changelog entry, most probably. [06:14] thakns [06:14] If it's an SRU, it should be test-built beforehand. [06:14] As a simple rebuild might break things. [06:15] Sure :-) OK, that sounds very doable. I'l let someone with way more experience in Ubuntu packaging than me decide if it should be an SRU :-) [06:17] poolie: man dh_install should give enough info, but feel free to ask more if it's insufficient. [06:18] what controls whether they're treated as conffiles or not [06:18] and specifically whether they're removed when an upgrade is installed? [06:19]