[00:02] NCommander, okay, I advocated it. [00:02] finally! [00:03] * cody-somerville rolls eyes. :P [00:03] The package isn't debian perfect (I wish I could review like some of the debian folk can) but I don't think we'll have too much trouble getting it into the archive and soon into Debian. [00:04] cody can you go over mine then? please? [00:05] coppro, ermmm... I guess. I've probably missed my bus already anyhow., [00:05] coppro, link? [00:05] http://revu.ubuntuwire.com/details.py?package=libmk4 [00:05] thanks [00:45] ScottK: ping [00:46] any MOTUs around? [00:46] I am. [00:47] oh [01:18] coppro, http://revu.tauware.de/details.py?package=libmk4 [01:22] nxvl, hey [01:25] cody-somerville: hi! [01:25] nxvl, want to do me a favour? :) [01:26] yup [01:26] shot === cprov is now known as cprov-zZz [01:30] thanks cody [01:31] cody-somerville: yes, sure [01:32] nxvl, I've been assisting NCommander over the last 24 hours to get his codeblocks package on revu in shape. [01:32] nxvl, I was wondering if maybe you could take a look to see if you'd be able to provide a second advocate so we can get it uploaded? [01:33] it takes an eternity to build [01:33] nxvl, hehe. It isn't so bad if you upload it to a PPA. [01:33] cody-somerville: i'm not a MOTU yet [01:33] nxvl, no? [01:33] nope [01:33] Sorry to bother you then. [01:33] np [01:54] okay, upstream includes an outdated config.guess, what do I do? [01:54] ask upstream to update it [01:54] in the meanwhile? [01:55] try to build it [01:56] sometime it includes some .in file or something that updates it [01:57] lintian gives an error :( [01:57] and apparently I can't just replace it [01:58] well [01:58] actually in that particular case i would: [01:59] a) ask upstream to include it and wait [01:59] done [01:59] b) patch it, and send the patch to upstream [02:03] I'm having a dependency issue building iriverter using pbuilder, can anyone help please? It builds fine under 8.0.4 (x86) but not in pbuilder. [02:04] how do I download the stuff necessary to pbuilder? [02:04] apt-get? [02:05] i don't understand the question [02:05] copro... stuff like?? sudo apt-get install pbuilder installs it for me. [02:05] not that, one moment [02:05] coppro, just copy over the config stuff over at build time and delete it in your clean statement. [02:05] ok ty [02:06] oh, it's just pbuilder completion being stupid [02:06] nvm [02:07] hmm... error is in satisfying the build dependencies [02:07] don't have any problems with hardy... [02:11] someone name a random universe package [02:11] exfalso [02:12] Very random. [02:13] coppro: Have you enabled universe in your pbuilder? [02:13] no [02:13] which is likely the case [02:13] *issue [02:17] jmarsden: something about depending on a virtual package I think [02:18] coppro: Yes, there's supposedly a table specifying how to satisfy such virtua dependencies... but man pbuilder doesn't seem to tell me where that table is... ? [02:19] it's a problem with the package source I think then [02:22] coppro: Well, iriverter is already packaged and so at some point I presume it must have built OK with pbuilder? I'm fixing a minor bug... bug fix is fine, but pbuilder won't build it, which irks me. [02:23] I wonder if the new openjdk stuff in Hardy changed something about Java package build depends? [02:23] try the --extra-packages option [02:23] OK, will do. [02:23] i don't know if it will really help but [02:23] its worth a try [02:29] coppro, I followed up [02:29] ok ty [02:30] * NCommander inhales [02:31] * NCommander begins hacking apart revu [02:31] thx for the feedback [02:36] can someone help me set up CDBS please? [02:37] coppro, what do you need to setup? [02:37] the whole thing [02:37] Run dh_make, and select CDBS [02:37] If its something that works via make/make install, it's pretty much done ;-) [02:38] hmm [02:39] damn it, brb [02:44] Would anyone like to be the second advocate of codeblocks? [02:45] I would like to be, but I'm not a MOTU :( [02:45] heh, its appericated [02:45] * NCommander watches keys get imported from launchpad [02:46] NCommander: You might want to try a smaller team. [02:46] what's the format for 00list? [02:46] wgrant, yeah, I discovered that the hard way [02:46] coppro, its just the patch names [02:46] ok [02:47] * NCommander has it pulls the keys from my personal launchpad team [02:47] it appears fetch-launchpad-keys ignores the config file [02:47] Quite possibly. [02:47] SOmewhat anonying [02:47] I don't like the hardcoded paths either :-/ [02:48] coppro: I think dpatch or whatever patch creation tool you are using should update 00list automagically? [02:49] I was converting from diff patches [02:49] OK. [02:50] * NCommander makes a test team [02:50] ahah [02:51] Cool, I can just use my username and get that one specific key :-) [02:52] yeah, it appears most/all of the scripts ignore the config file [02:52] (aside from the webpage) [02:56] great [02:56] I now have a random failure [02:57] oh, I know why [02:57] insidious [03:02] is there any way to change the name my packages are signed with by default? [03:02] coppro, define the env variables DEBNAME and DEBEMAIL [03:02] ok ty [03:10] * NCommander has a fully functional revu at 127.0.0.1 ^_^ [03:10] debuild should always be passed --lintian-opts -i [03:10] it should be a law or something [03:11] NCommander: Make sure you have it turn off email, though. [03:11] NCommander: I've spammed a few people with it once when I forgot to disable things. [03:11] Is there a handy disable switch? [03:11] Since my changes last week, I belive that altering the email address in the config will get everything to respect it, but I'd grep to make sure. [03:17] yay lintian-complaint-free! [03:23] Good evening! Is there an f-spot expert around? I have a little compilation/packaging problem to sort out with f-spot 0.4.4 for hardy... [03:24] RoAkSoAx: ping [03:25] just ask anyway [03:26] worst-case-scenario: no one can help [03:27] foxmike: ping [03:27] Here is the thing: f-spot seems to try to install upon compilation some data-hooks under /usr/lib, and that makes the compilation crash since it is done via debuild, with user permissions. I've located where in the Makefile it happens, but I don't know how to fix such thing... [03:27] ok [03:28] one moment, installing build-deps [03:28] ok thanks [03:29] odds are the makefile just needs to be changed to install to $(PREFIX)/usr/lib [03:31] ok, check I'll try this... I guess $(PREFIX) will be something like (put working directory here)/f-spot-0.4.4/debian/tmp or something? (sorry for the noob question, I'm fairly new to all this...) [03:32] it should be substituted in by the build system, but I'm not sure [03:32] so you're trying to compile 0.4.4 right? [03:33] where did you get the source from? [03:35] foxmike: I need a copy of the 0.4.4 source, may I ask where you got it? [03:35] Well, I got it from f-spot website I've copied the debian directory from 0.4.3... [03:35] ok [03:35] ty [03:35] thanks to you!:) [03:42] coppro: have you got it? If not, I can send you a copy... [03:42] yeah I've got it, and I think I have a better solution [03:43] go ahead [03:43] 0.4.4 is already in intrepid [03:44] http://mirrors.kernel.org/ubuntu/pool/main/f/f-spot/f-spot_0.4.2-1ubuntu3_i386.deb [03:44] geez, I'm looking almost every day if it is there (I'm working on the package to learn, mostly...) and I didn't check today... anyway, I try to compile like this if it doesn't, well, I'll backport from intrepid!;) [03:44] thanks for the link [03:44] ok [03:49] It's still listed as f-spot0.4.2 on packages.ubuntu.com/f-spot ...:~ [03:52] thanks again coppro, I got to go! [03:56] nxvl, here! [03:56] Is this the wrong place for asking a question about packaging even if it's not a MOTU thing, or is there a better place for that? [03:56] nxvl, had to run for a while [03:57] (it is an Ubuntu thing, but not a MOTU thing.) [03:57] you can probably ask anyway [03:59] okay. What if I have a .deb I didn't make and I want to unpack it so I can make changes? I have lots of guides from various places telling me how to build them, but I haven't found how to unbuild them. [03:59] RoAkSoAx: ok [03:59] RoAkSoAx: stay here [03:59] vorian: ping [04:00] nxvl, k ;) [04:00] where did you get it from? [04:00] ideally, you want the source package [04:01] yo! [04:01] ok [04:01] vorian: this is RoAkSoAx, you mentee [04:01] RoAkSoAx: this is vorian, your mentor [04:01] :D [04:01] i hope you 2 the best on the mentoring program [04:01] howdy RoAkSoAx [04:01] make me proud! [04:01] vorian, master!! :D [04:01] hahaha [04:02] vorian, i'm ready to role xD [04:02] excellent [04:03] coppro: ah, that makes sense. I'll get that, then. [04:03] * marnanel used to know how to do this (at least to some extent), but it's been years... [04:03] well, the source package isn't necessarily the source code, it's the .diff.gz, .orig.gz, and .dsc [04:04] right [04:04] * marnanel has it now. yay [04:06] thanks. I will try this and see how it goes... [04:14] nxvl: Pong [04:16] ScottK: i was looking for you to mention my application, but you already replied, so nm [04:16] ScottK: and Thanks for the support [04:22] ScottK: you up for a REVU? [04:25] nxvl: You're welcome, you deserve it. [04:34] !seen [04:34] Sorry, I don't know anything about seen [04:34] any MOTUs willing to do a REVU? === asac_ is now known as asac [04:39] coppro> What package? [04:39] http://revu.ubuntuwire.com/details.py?package=libmk4 [04:39] thanks [04:42] * LucidFox looks [04:51] coppro> commented [04:51] ty [04:51] one question [04:51] how do I get dh_install to install do a different name? [04:52] coppro, you can't [04:52] that's why I used regular install for lintian-overrides [04:52] Ah. [04:52] Then it's fine. [04:53] coppro, all the debhelper scripts have pretty comprehensive man pages. [04:53] yeah, that's what I thought [04:53] which is why I was pretty sure I was doing the right thing (as I understand it, intrepid has dh_lintian) [04:54] Isn't dh_lintian part of debhelper 7? [04:54] and you use 5 [04:58] I should use binary-indep for Python stuff, right? [05:03] binary-indep is for arch:all packages [05:04] oh, ok [05:04] wait, but Python is arch:all [05:06] cody-somerville, would you like the honor of sponsoring codeblocks, or should I do it? :) [05:07] ok, dumb question [05:07] how do you put a paragraph break in a long package description [05:07] . [05:07] LucidFox, what would you do differently? [05:07] a space, followed by a dot [05:08] ok ty [05:08] for example, I'd use install instead of cp [05:08] like coppro just did :) [05:09] great, now dpkg-source is complaining [05:09] and I'd remove the upstream debian from the orig.tar.gz rather than diffing against it [05:09] apparently the +x mode on my patchfiles won't get through to the diff... that's a problem [05:09] coppro> Python is arch:all, but your _own_ package should be arch:all to use -indep [05:10] yeah, my source package has a python component. I just haven't gotten around to packing it [05:10] along with the docs and the tcl [05:10] ah [05:10] I was making sure that went into -indep, which is what I figured [05:10] well, if it's a binding for a C library, then the C library will typically be arch:any and the binding will be arch:all [05:10] how do I deal with 'warning: executable mode 0755 of 'debian/patches/disable-test-cleanup.dpatch' will not be represented in diff' [05:11] Ignore it. [05:11] ok [05:14] ok, new version on REVU [05:14] and now that it's gotta be perfect i can get to the docs :P [05:16] should -dev recommend or suggest -doc? [05:20] jcastro: around? [05:22] NCommander, congratz. [05:23] yeah, congrats [05:23] on what? cody-somerville? [05:23] (I hope to enjoy your new package once 8.10 comes out) [05:23] NCommander, I just uploaded codeblocks. [05:23] * NCommander faints from shock [05:23] cody-somerville, why'd you feel the package didn't mean Debian expectations? [05:24] pretty good time from "decides to make package" to package [05:24] coppro, not really ... [05:24] NCommander, I don't think I said that. [05:24] Er, I thought someone did [05:24] But my memory is about as reliable as a Windows Vista workstation [05:25] can anyone point me in right direction to fix this issue in my intrepid chroot? next post will be error [05:25] debconf: delaying package configuration, since apt-utils is not installed [05:25] Can not write log, openpty() failed (/dev/pts not mounted?) [05:25] dpkg: syntax error: unknown group `Debian-exim' in statoverride file [05:25] E: Sub-process /usr/bin/dpkg returned an error code (2) [05:25] E: Failed to process build dependencies [05:25] oops way too many lines [05:26] gnomefreak, in your chroot, type: mount -t devpts none /dev/pts [05:26] And you probably need to manually add the Debian-exim group [05:26] (groupadd -s Debian-exim) [05:27] there is not option -s [05:27] what compression does .deb use? I thought it was .tar.gz, but that doesn't seem to be the case [05:28] isnt it zip? [05:28] cody-somerville, who was the other advocate? [05:28] LucidFoz [05:28] s/z/x/ [05:28] sweeeet [05:28] he advocated it when I asked him to check my package [05:28] I think [05:28] So he did it by mistake? [05:28] ... [05:28] no [05:29] Eh, if that what it takes ;-) [05:29] * NCommander watches Ubuntu suddenly explode in an amazing display of colors and Debian swirls [05:30] anyone know what compression format a .deb is? [05:30] ar archive [05:30] use ar x to extract [05:30] and the man page to determine how to compress [05:30] oh ok ty [05:30] ar r isn't it? [05:31] what is [--gid ID] [05:31] gnomefreak, the GID of the group you are creating [05:31] assuming group id [05:31] coppro, don't know off hand [05:31] how would i know what it is :( [05:32] make one up that isn't taken I guess [05:33] * NCommander loves looking at codeblocks in the archive [05:33] It's like [05:33] wow [05:33] I did that [05:34] is there a way to list all gids? [05:34] cat /etc/group [05:34] THe first number is the gid [05:34] Just pick an unusued one [05:34] thanks [05:34] cody-somerville, so now my package is in the NEW queue, right? [05:35] NCommander, correct. [05:35] So in seven days, I'll get the email notification about REJECT(ed) ;-) [05:36] NCommander, :) [05:36] it seems i cant use a number that isnt used already. error message addgroup: The user `125' does not exist. [05:37] gnomefreak, just do "groupadd -s Debian-exim" [05:37] Or loose the -s if it doesn't work [05:37] NCommander: cant -s isnt an option [05:37] then just groupadd Debian-exim [05:38] ok i did and got no output im gonna say this is good and i will try dpkg again thanks [05:38] Will version 1.7.dfsg be considered bigger than 1.7? [05:39] NCommander: it worked ;) [05:40] LucidFox, Well, if I remember how dpkg works, the string gets turned into numbers (its a char I think), and the greater sum is the higher package version. THat's why x86suse > ubuntu [05:40] Sadly :-/ [05:40] *char cast [05:46] http://thedailywtf.com/Articles/Which-Internet.aspx - lol [05:48] ok, new confusion [05:48] upstream puts all the documentation in doc/ [05:48] * NCommander apt-get gets your confusion [05:49] all this needs to be put in /usr/share/doc/package/html [05:49] can I specify that somehow in package.docs? [05:49] it gets seriously confused if I just try doc/* [05:51] Can't you just list the path to the docs folder and it will recursively do the rest? [05:51] * NCommander admits its been awhile since he had such fuck [05:51] ... fun [05:51] damn it [05:51] Two thoughts crashed at extactly the same time; my pager going off, and that statement [05:52] it puts them all into the /usr/share/doc/package/ [05:52] how do I get it to go another level down [05:52] ,,? [05:52] er .. [05:52] I need it to be in /usr/share/doc/package/html/ [05:53] I'd just manually create the doc package with some fancy dh_install rules then ... [05:53] or I could just cp -a doc debian/tmp/html [05:53] and put debian/tmp/html into the .docs file [05:53] go for it [05:56] ok ill say you told me to do it lol [05:56] or you know what [05:56] 'doc' is an okay folder name [05:58] ack [05:59] what? [06:00] nm [06:00] * NCommander sends a rst packet into the channel [06:01] * coppro /clears [06:02] do we know status of ruby1.8? [06:03] * NCommander throws more rst packets and an x-mas tree too [06:04] * coppro clears multiple times [06:04] ow [06:09] I just came up with a new warning lintian should check [06:10] if there exists a debian/patches/*.dpatch and you don't build-depend dpatch, something is probably wrong [06:16] gaah! what options are there for Section: in a doc-base file? [06:17] found it [06:17] but the examples in the doc-base docs are invalid :( [06:19] wait, no [06:19] stupid lintian [06:22] its funny how chroot doesnt install apt [06:23] it doesn't? [06:23] nope [06:23] you used debootstrap right? [06:23] yes [06:24] well, my chroot has apt [06:24] i have 5 chroots set up intrepid is only one that doesnt install apt [06:24] not really much else, but it definitely has apt [06:24] root@Development:/# apt-get update [06:24] bash: apt-get: command not found [06:25] O.o [06:25] try reinstalling your chroot then? [06:25] coppro: 2 times [06:25] sudo chroot /chroot/intrepid [06:25] apt-get update [06:25] [06:25] ending up downloading it from p.u.c moving it to /var/chroot/intrepid/home and installing it with dpkg [06:27] this breakage is recent [06:27] last 7 days or so [06:28] I made my chroot yesterday [06:28] or the day before, I forget [06:31] i had one set up last week but i had broken it so im setting it up again. last week apt was installible [06:31] -i +a [06:32] I just made another one: [06:32] I: Retrieving Release [06:32] I: Retrieving Packages [06:32] I: Validating Packages [06:32] I: Resolving dependencies of required packages... [06:32] I: Resolving dependencies of base packages... [06:32] I: Found additional base dependencies: libgnutls26 [06:32] I: Checking component main on http://archive.ubuntu.com/ubuntu... [06:32] I: Retrieving adduser [06:32] I: Validating adduser [06:32] I: Retrieving apt [06:32] so there [06:32] * coppro ducks from the spam police [06:33] is /usr/bin in your $PATH? [06:36] coppro: it is [06:36] wow something is messed up on your end then [06:37] yep fixing it as we speak [06:39] how in gods name am i gonna get a root/sudo dhell to install sudo [06:40] from a chroot? [06:40] or does your system not have sudo installed? [06:40] if the latter, and if root has no password (the default on ubuntu), you have to boot into single-user-mode [06:41] from a chroot but sudo never got installed in the chroot [06:41] you start with a root shell in a chroot [06:41] su doesnt work [06:41] doesn't need to [06:41] you are root [06:41] just apt-get install sudo [06:42] (intrepid)gnomefreak@Development:~/downloads$ apt-get install sudo [06:42] E: Could not open lock file /var/lib/dpkg/lock - open (13 Permission denied) [06:42] E: Unable to lock the administration directory (/var/lib/dpkg/), are you root? [06:42] i downloaded it but cant use sudo dpkg [06:42] you have chrooted in? [06:42] yes (intrepid)gnomefreak@Development:~/downloads$ [06:42] that's not right [06:42] sure it is [06:42] why isn't it a root shell [06:43] chroot can only be run as root and gives a root shell [06:43] none of mine are [06:43] coppro: nope it runs fine without root shell [06:43] I have never heard of chroot not giving a root shell [06:43] shares ~/ [06:44] ok wait close out of everything [06:44] go back to your normal shell [06:44] coppro: https://wiki.ubuntu.com/DebootstrapChroot?highlight=(chroot) [06:44] and 'sudo chroot /var/chroot/intrepid' [06:44] and that gives me the same tht i was in [06:45] that should be a root shell [06:45] the chroot is set up already but somehow the packages didnt install [06:45] coppro: no its not Setting up a dchroot (non-root) environment [06:46] oh [06:46] on the link i gave you [06:46] you are supposed to install sudo first, before you mess with the other stuff [06:47] coppro: yeah hence the screw up [06:47] can't you still chroot in normally? [06:47] coppro: not to root [06:48] why not? [06:48] wait there might be a way [06:48] try 'sudo chroot /var/chroot/intrepid apt-get install sudo' [06:48] coppro: because i set it up with nonroot [06:49] coppro: i think that worked but give me a minute and ill let you know [06:49] coppro: it worked thank you [06:50] manpages are your friend :) [06:51] they were already downloaded but didnt get configured due to apt not being installed so i could have just dpkg --configure -a :( [06:51] thats what i get for working at 2am [06:52] lol === coppro is now known as Halph === Halph is now known as coppro [06:53] someone broke ruby1.8 so i see if i can fix it locally most likely just a respin [07:09] does ruby build like python or normal package using dpkg-buildpackage [08:05] gnomefreak: It is a package, so yes. [08:06] wgrant: thanks now i have to figure out the error its 3:05 am atm so im not sure if im gonna continue or not i have bad feeling its *-buildpackage [08:06] Ruby is rather temperamental. It will sometimes build. [08:16] 2 [08:17] it may be ruby since it runs -b after i run dpkg-buildpackage keystuff -i.bzr [08:43] LucidFox: Hi [08:44] slytherin> Are you sure that the TIFF and JPEG transcoders don't build with openjdk? [08:44] LucidFox: Let me check. [08:45] as I mentioned in the bug, they did build for me [08:45] same for xmlgraphics-commons [08:46] LucidFox: Ihaven't looked much into oenjdk but do you if com.sun.image.codec.jpeg.JPEGCodec is available in openjdk? [08:46] Yes, Ubuntu's openjdk-6 the contains com.sun.image.* APIs [08:47] LucidFox: meanwhile check line 164 in build.xml [08:47] yes [08:47] and for openjdk, that check succeeds [08:48] javac does warn that these APIs are deprecated, but the dependent classes still build [08:49] LucidFox: That is cool then. [08:50] Now what we can do is add debain/ant.properties file and add properties ant.build.javac.source=1.5 and ant.build.javac.target=1.5 to the file. Also include ant-vars.mk in rules file. [08:58] LucidFox: and whom do I bug now to clear batik from 'NEW'? [08:59] slytherin: batik got uploaded? [08:59] geser: LucidFox did. [09:00] geser: and as he is saying it is built with openjdk without loss of cuntionality. [09:00] even better then [09:00] s/cunctionality/functionality [09:01] Now I have to apply for motu-contributors membership. :-D [09:02] slytherin: the NEW queue gets usually processed on the archive days [09:03] geser: yes I know. I am just too eager to look into other depwait/ftbfs that depended on batik [09:03] :) [09:14] can someone point me to wher ei can figure out what the following error is. [09:14] error is: dpkg-buildpackage: failure: dpkg-source -b ruby1.8-1.8.7.22 gave error exit status 255 [09:16] gnomefreak: It would generally indicate that the build failed somewhere. You must look for the source of failure yourself. [09:16] wgrant: its ruby that failed? [09:17] wasnt sure if it was ruby or dpkg [09:17] gnomefreak: read the complete build output not just the last error. You will surely find the cause. [09:18] slytherin: i have read it and doing it again atm [09:19] just about everything in output has to do with cdbs [09:19] gnomefreak: can you paste the log somewhere? [09:19] but as for the failure im not sure since it deapplied patches than cdbs starts [09:19] slytherin: yeah [09:21] http://gnomefreak.pastebin.ca/1076441 this maybe why ruby is broken(atleast i heard it was thats why im working on it i need it for an app [09:22] was gonna say make file but it was ablet o continue [09:22] gnomefreak: line 64 in your pastebin [09:23] that wont make it fail would it? [09:23] * gnomefreak thought ive had apps sontinue but ill try it [09:23] gnomefreak: I don't know. But you have to fix it anyway. [09:24] It says 'dpkg-source: error'. That looks like an error. Errors make things fail. [09:25] :-) [09:27] what do i change the maintainer to (the one that is in there now so i can add me to maintainer [09:28] xbdc or something like that? [09:28] gnomefreak: XSBC-Original-Maintainer. I suggest you use update-maintainer script [09:29] didnt know there was one [09:29] And don't add yourself as maintainer, it is either ubuntu-devel-discuss@u.c or ubuntu-motu@u.c [09:29] gnomefreak: the script is in ubuntu-dev-tools ackage [09:34] slytherin: wgrant thanks i havenever seen build fail due to no ubuntu maintianer because of the ubuntu version [09:35] all this for alexandria book app [09:41] hey guys [09:42] Im having an issue with dh_shlibdeps finding a library in a different binary package built by the same source package.. debian/rules: http://pastebin.com/f5ee3112a debian/control: http://pastebin.com/f7068ed9c buildlog: http://pastebin.com/f6581008e Any ideas why dh_shlibdeps cant find the lib it needs from the same source package even though dh_makeshlibs is run before dh_shlibdeps? [09:54] awesome, the buildlog doesn't load here [09:54] same here [09:55] i'd be really interested because i might be suffering from the same problem [10:03] cody-somerville: Hi, I just uploaded a new debdiff for salasaga, hope it's ok now. [10:11] hi all [10:12] someone can review my package, please? http://revu.tauware.de/details.py?package=umit [10:37] Has someone seen a pywebkitgtk package? [10:39] k0p: Call dh_desktop in debian/rules. It's better just to say "Please report bugs at http://trac.umitproject.org" in your manpage and don't forget to mention the authors of the package. [10:41] Iulian, about dh_desktop is it necessary? yesterday a guy in the channel told me that it is useless. do you know something about that? [10:43] Iulian, where I can set authors of the package? [10:43] laga, oh didnt notice, one moment ill psate again [10:45] laga, http://pastebin.com/f3427fb96 [10:46] Iulian, now no warnings and I fix the bug text on manpage. [10:46] lack the authors of package. [10:46] Blimey, slytherin left [10:46] k0p: It seems that lintian complains about dh_desktop, not sure if it's necessary. [10:46] Iulian, yeap [10:46] and about authors [10:47] are you talk about manpage? [10:47] k0p: Well, if someone from here told you that it's not necessary, don't add it. [10:47] k0p, as far as I know, only if you have a .desktop file in your package, then you should call dh_desktop. [10:47] Jazzva, yeah I have [10:47] so I'll call. [10:47] :) [10:48] dh_desktop debian/umit.desktop [10:48] something like that? [10:48] k0p: Add the authors of the package in AUTHORS section in your man page. [10:48] Iulian, so.. replace AUTHOR by AUTHORS? [10:49] Exactly [10:49] k0p, manpage for dh_desktop says it can't install .desktop files. It takes care of adding maintainer script fragments to call update-desktop-database (copied from manpage). [10:49] k0p: If it has more than one author, yes. [10:49] Jazzva, really nice. :) [10:50] k0p, and if .desktop file does not contain MimeType field it will be ignored. So, if your desktop doesn't have that field, than maybe you don't need to call it :) [10:50] Iulian, nop. don't have.. [10:50] Iulian, is it authors of package or of project? [10:50] Jazzva, sure :) [10:51] k0p: Yes, upstream authors. [10:51] ok [10:52] Iulian, and can I mantain the author of manpage? [10:53] Yup [10:54] done [10:55] Iulian, what I can do to my package be added on ubuntu repositorys? only wait? [10:58] k0p: You'll need two advocates from two MOTUs. [10:59] advocates? is it like a approvation? [11:01] * directhex is still waiting for sponsorship on a couple of syncs to make some packages actually work [11:02] directhex, syncs? [11:03] syncs. as in "take debian package, place in ubuntu" [11:03] hmm sure [11:04] considering i updated the debian packages, specifically so they could easily be synced to ubuntu, it's kinda annoying [11:04] directhex, only debian package can be added on ubuntu? [11:04] k0p, no, but it's the best way [11:05] :S [11:05] it's designed straight into the ubuntu package numbering process, after all [11:05] I don't added my package in debian yet [11:05] I need to prepare debian package and send to debian-mentors [11:05] non-debian packages need to have "0ubuntu" in the version, e.g. 1.0-0ubuntu1 [11:06] sure [11:06] the idea being that if/when debian uses the package, they start at 1.0-1, which is higher-versioned [11:07] yeah [11:07] I understand [11:08] directhex, I would like that my package will be available on 8.10. But I don't have debian package in repository. Do you know that is possible? [11:08] sure. add to ubuntu, make sure you follow the correct numbering shemes & use 0ubuntuX for package versions. [11:09] yeah sure. [11:09] I would like how I can two advocates from two MOTUs [11:10] I don't know motus [11:22] https://wiki.ubuntu.com/No-Mono-by-Default - pushing at the boundaries of how much bullshit can go into one page [11:27] That guy must be nuts. [11:30] How about no stupid Evolution by default? :D [11:39] directhex> Yes, I saw that page. Some people even mistook it for official policy. [11:40] i'm busy adding my $0.02 === bastiao_ is now known as k0p [13:10] I'm looking at the changelog of a package and it is written with a mixture of english and german, is this acceptable? [13:25] elmargol, but evolution is awesome. [13:47] porthose: for Debian/Ubuntu, no [13:48] RainCT: they should be in english correct? [13:50] porthose: Yes. Imagine what would happend if everyone would write changelog entries in their native language... [13:50] RainCT: that's what I was thinking thx:) [13:51] genius idea: put debian/changelog under the control of pofiles! put out a request for translations with every package update, to keep the changelog translated! [13:53] Haha [15:03] anyone in here available to help with (what looks to me to be) a bug? [15:04] I can't seem to start audacious (and play a playlist) by using crontab [15:04] spent like 3 hours on it in #ubuntu last night [15:05] there seems to be some block on crontab using the display [15:05] even if you stipulate what it is [15:06] and I tried stipulating it 3 different ways. New line variable, inline variable, and audacious switch [15:09] the command is embedded in a script, and the script is being run [15:09] echos to a log file lemme know start and finish of the script [15:10] and 2>/blah/logfile is appended to the end of the line that calls audacious [15:11] output from that is "audacious: cannot open display" [15:11] better join #ubuntu [15:11] and I think you have to specify the display [15:12] tried stipulating it 3 different ways. New line variable, inline variable, and audacious switch [15:12] which is why this feels like a bug [15:12] try: [15:12] none of those ways work [15:12] DISPLAY=:0.0 yourapp [15:12] as command in crontab [15:12] IN crontab? [15:12] so [15:13] DISPLAY=:0.0 /path/to/script [15:13] hm.. yes. or in your script [15:13] tried in the script [15:13] will try in crontab [15:13] thanks for the suggestion [15:13] this is for the wife [15:14] wake to music [15:14] originally used mpg321 but she couldn't easly kill it [15:15] hm - i would use "mplayer /media/music/wake_up.mp3" and make a big button that executes "killall mplayer" [15:16] mplayer doesn't rely on display? [15:16] no [15:16] does it handle .pls playlists? [15:16] probably [15:16] k [15:16] thanks [15:16] I tried using xmms instead and that wouldn't work either [15:17] mplayer isnt graphical (at least as long as you dont watch a video in non-aalib mode) [15:17] someone suggested this last night right when I was giving up. I couldn't keep my eyes open any longer [15:18] Hi, I wonder how I can get debuild to ignore the debian/.bzr directory "pdebuild -i.bzr" does not work. Any ideas? [15:24] DISPLAY=:0.0 /path/to/script doesn't work [15:24] got output [15:24] No protocol specified [15:24] audacious: cannot open display: [15:24] attempting with mplayer [15:24] then specify DISPLAY=:0.0 audacious in your script (alternatively) [15:25] i mean, start audac* with DISPLAY=:0.0 before it [15:25] looks like "pdebuild --debbuildopts -i.bzr" is what I was searching for [15:27] mplayer didn't work, but there's more output [15:28] oO(i think he's doing something wrong) [15:29] http://pastebin.com/d23874861 [15:30] No such file or directory [15:30] what do you specify in your crontab? [15:30] hm [15:30] XOpenDisplay() failed [15:30] ok, show your crontab [15:30] yeah [15:30] just a sec [15:32] 27 9 * * * DISPLAY=:0.0 /home/fayte/MorningPlayer/morningmusic.sh [15:32] and show morningmusic.sh ? [15:32] s/show/your/ [15:33] hang on a sec... will jump in from that machine, it keeps droppin my ssh session [15:33] ehm [15:33] a x server is running? [15:34] +n [15:34] k [15:34] just a sec [15:34] mplayer -playlist /home/fayte/Desktop/Playlists/MorningMusic.pls 2> /home/fayte/MorningPlayer/MP.log [15:35] thats the line from the script [15:35] everything else has been commented out [15:35] ok [15:35] mplayer -playlist /home/fayte/Desktop/Playlists/MorningMusic.pls [15:35] this works when you start it by hand? [15:36] http://pastebin.com/m5d3fa1b1 [15:36] works just fine from nautilus and from CLI [15:36] ok [15:36] wait [15:36] no [15:36] haven't tried it in either since changing to mplayer [15:37] but it worked fine from CLI and nautilus when it was audacious [15:37] gimme a sec, will check it agian [15:37] remove the -playlist argument [15:37] try it without it from CLI first I mean [15:38] its working from CLI [15:38] with -playlist option [15:39] ok [15:40] http://pastebin.com/m50ed90 [15:40] does this work in cli? [15:40] http://pastebin.com/d8cc086d [15:40] with my command? [15:41] yep [15:41] try the command that works for you and mine, and then paste the _WHOLE_ prompt [15:41] no CLI output [15:41] but its playing [15:42] try your command and mine, and then paste the _WHOLE_ prompt [15:42] not sure what you mean [15:43] youve opened a terminal, right? [15:43] yep [15:44] then open a new one and enter and cancel both commands. then copy everything you see (everything!) and paste it [15:46] http://pastebin.com/d22cb5347 [15:46] thats from your command [15:46] output from the log [15:46] there was no output in the terminal [15:47] Ctrl-C just went to the next line [15:47] ok [15:47] do the same without the > and 2> stuff at the end [15:47] k [15:47] so that the messages appear in the terminal [15:48] http://pastebin.com/d1502845c [15:49] its playing [15:49] also remove the > /home/blabla stuff at the end [15:49] k [15:50] http://pastebin.com/m5151e133 [15:51] it played? [15:51] yep [15:51] the "no such file or directory" bit must be referring to something internal [15:52] ok, fine. now put DISPLAY=:0.0 mplayer -playlist /home/fayte/Desktop/Playlists/MorningMusic.pls in your crontab [15:52] not the script [15:52] ehm [15:52] but add the log stuff [15:53] DISPLAY=:0.0 mplayer -playlist /home/fayte/Desktop/Playlists/MorningMusic.pls > /home/fayte/mpl.log 2>&1 [15:53] You know cron has a different environment to the logged in user. [15:53] Which means you probably can't run X programs from cron [15:53] well, but audio should work without x? [15:54] and there's the DISPLAY var [15:54] tried? xhost +localhost from inside X? [15:54] xhost +localhost is a horrid hack [15:54] As long as the user has permissions to open the sound device, it should be fine. [15:54] StevenK: only to know if X is the problem :] [15:54] if it runs as root he shouldnt need xhost [15:54] k [15:54] IIRC [15:54] 1 min [15:54] hmmmm [15:55] Kopfgeldjaeger: correct [15:55] been trying as root crontab [15:55] ls -lah /dev/dsp [15:55] where do I download the nightly iso of intrepid ? [15:55] fayte@faytal:~/MorningPlayer$ ls -lah /dev/dsp [15:55] crw-rw----+ 1 root audio 14, 3 2008-07-15 22:28 /dev/dsp [15:56] time has past [15:56] hmmm [15:56] will check the log [15:56] syslog? [15:57] http://pastebin.com/m18e4efcf [15:57] no, a log output from the command [15:58] maybe try it with the user-crontab or use sth. like sudo -u fayte [15:58] is it supposed to be this hard? [15:59] have any of you tried it on your systems? is it just my install? [15:59] WOP: try put into crontab: echo $UID > /tmp/test [15:59] is this a bug? is there another way to do cron jobs? [16:00] http://www.getdeb.net/app/AlarmClockApp [16:01] I'd have to set 10 of those to play the album :-/ [16:01] but thanks, I'll use that app on my machine I'm sure [16:02] echo $UID > /tmp/test whats that do? [16:04] I'm just not sure what the $UID is [16:04] CyberCo1: may be cron set the uid environment [16:04] so can you see the user crontab uses [16:04] ah [16:05] I get you [16:05] k [16:05] will try it [16:06] 1 min [16:07] no output [16:08] hmmmmm [16:08] - /tmp/test was created, but it was blank [16:08] CyberCo1: [16:09] cat /proc/self/status > /tmp/test [16:11] you want that in a crontab? [16:11] or run from term [16:11] yep :] [16:11] on cron [16:11] k [16:11] i think this is empty-environment proof :] [16:13] http://pastebin.com/m6512d10d [16:15] I understand the need for the empty environment, but there's an obvious need for a scheduler with the ability to output to display [16:16] Uid: 1001 1001 1001 1001 [16:16] Gid: 1001 1001 1001 1001 [16:16] o.0 [16:16] suid?!? [16:16] lost me [16:16] I dont know what any of that means [16:16] well [16:17] if you ran that line from crontab [16:17] nah [16:17] it was a user's crontab [16:17] dont make sense [16:17] hmm [16:17] no makes sense :] [16:17] *now [16:18] well, it uses you own user to run crontab [16:18] should I do it again from root crontab? [16:19] you need to test this script from crontab you get the problem :] [16:20] CyberCo1: try create a script i some place [16:20] /home/cyberco/WakeupMachine.sh [16:21] and within the script ? [16:21] and inside tuse #!/bin/bash (newline) mplayer somethingyouwant [16:21] *use [16:21] that is how I've been doing it [16:21] still get display errors [16:21] even if I specify display [16:22] try [16:22] mplayer -vo null [16:22] k [16:22] * leleobhz dont see another way [16:24] set 1 min [16:25] http://pastebin.com/m137ba1f4 [16:26] thats what was > log [16:28] whats crazy is that this is a new problem... worked fine in feisty [16:28] hmmm [16:29] try add -v -v -v [16:29] verbose? [16:29] yep [16:29] keep the -vo null ? [16:29] keep [16:30] mplayer -v -v -v -vo null -playlist /home/fayte/Desktop/Playlists/MorningMusic.pls 2> /home/fayte/MorningPlayer/MP_.log [16:30] thats the line in the script [16:30] any changes before I cron it? [16:32] nop [16:32] 1.5 min [16:34] http://pastebin.com/m693eb8b6 [16:34] o.0 [16:34] yeah [16:35] module unknown [16:35] wtf? [16:35] I think this is bug worthy [16:35] shouldn't be this hard to get cron to open something to display [16:36] well [16:36] another [16:36] remove -vo and -v's [16:37] add --novideo [16:37] k [16:37] or [16:37] (if dont work) [16:37] like manpage say [16:37] -novideo [16:37] Do not play/encode video. In many cases this will not work, use -vc null -vo null [16:37] instead. [16:39] 1 min [16:41] http://pastebin.com/m4dc2fcee [16:41] I think I'm going to have to go deeper down the rabbit hole [16:42] need to find out why cron is blocked from using display and how it is done [16:42] but right now I have to cook some breakfast [16:42] :) [16:42] thanks for trying folks [16:42] i think the problem isnt X [16:42] CyberCo1: ok [16:42] or just use that applet from getdeb [16:42] Kopfgeldjaeger: if it will play a playlist, I'm all over it [16:42] but this still bugs me [16:42] and maybe write a script that shuffles the song in the configuration of the applet [16:42] and I'm stubborn [16:43] I will get to the bottom of this [16:43] eventually [16:43] I just figured it'd be a no-brainer for the Motu's [16:43] CyberCo1: may you try #mplayer [16:44] its not just mplayer though [16:44] its also audacious [16:44] and xmms [16:44] I'm beginning to think cron can't run any graphical apps at all, and I'll test that theory after breakfast [16:44] CyberCo1: where are you from? [16:49] kentucky [16:50] is anyone else being bit by ruby1.8: symbol lookup error: /usr/lib/ruby/1.8/i486-linux/gtk2.so: undefined symbol: gtk_file_system_error_quark when trying to run an app? [16:50] ^^ intrepid [16:51] CyberCo1: it should beable to but not reall sure why you would want it to. it will end up being very slow [16:52] to have the option would be nice [16:52] not by default [16:53] CyberCo1: ive never tried but i dont see why it cant [17:09] gnomefreak: I don't see why it can't either, but it's not [17:11] no matter how you stipulate the display, it won't take it [17:11] at least not in hardy [17:11] not a problem in feisty [17:11] dunno about gutsy === RoAk is now known as RoAkSoAx [17:27] how make a non native package? [17:28] ? [17:28] non native? [17:29] well.. i'm in trouble because it not create a diff file [17:29] it create only a tar.gz [17:29] k0p: Rename the .tar.gz to packageName_version.orig.tar.gz [17:30] jpds, does work .. [17:30] umit (0.9.5RC2-0ubuntu1) [17:30] is 0.9.5RC2 version? [17:31] I think you should rename it to umit (0.9.5~rc2-0ubuntu1) [17:31] And call the tarball: umit_0.9.5~rc2.orig.tar.gz [17:33] jpds, it's of course temporary [17:33] but we're working on stable right now [17:33] but we can't release yet [17:37] does anyone of you has the Debian New Maintainer's Guide in one single document? [17:42] RoAkSoAx: there is a PDF version of it [17:42] * nxvl seaarches [17:44] didn't know : [17:44] RoAkSoAx: sudo apt-get install maint-guide [17:45] comes with the html, pdf, txt, ps and smgl version of it [17:45] nxvl: ok cool, i found the pdf too [17:45] ok, I have a question [17:45] what are the allowable doc-base ssections? [17:45] as in? [17:46] on package.doc-base [17:46] there is a field called Section: [17:46] no matter what I put there, lintian complains [17:47] Section? [17:47] no [17:47] the package.extension files under debian/ directory only show the path to the files which needs to install [17:47] not any Sections [17:47] that i remember at least [17:49] not doc-base [18:20] Dear fellow motu-sru members: Do the reports in Bug 246301 amount to a regression that needs to be reverted? [18:20] Launchpad bug 246301 in linux-restricted-modules-envy-2.6.24 "Update the FGLRX and NVIDIA driver to 8.6 and 173.14.09" [Undecided,Fix released] https://launchpad.net/bugs/246301 [18:20] * ScottK is inclined to say yes, but I haven't followed it closely. [18:20] tseliot: What do you think? [18:22] ScottK: there is little we can do if a driver breaks some cards and solves problems for others [18:23] ScottK: what are valid Section entries in doc-base files? [18:23] lintian complains at me! [18:23] ScottK: other users added a comment on my blog to say that their ATI cards work well with the new driver [18:24] tseliot: This is why I'm not sure if it's truly a regression. [18:24] coppro: Run lintian with -I and see if it gives you enough detail to figure it out. [18:24] my experiences with fglrx can only be described as bad [18:24] -I? [18:25] tseliot: Would you please at least comment in the bug and at least tell him how to get back to the previous version then. [18:25] coppro: lintian -I filename.dsc or .deb [18:26] W: libmk4-doc: doc-base-unknown-section libmk4-cpp:6 Apps/Programming [18:26] coppro: That's a capital i if it's ambiguous with small l [18:26] yeah, that's what I see [18:26] ScottK: it's what I suggested in comment 23. I can post step-by-step instructions. What do you think? [18:26] With -I it should spit out additional explanaition. [18:26] lower-case i gives the complet information [18:26] tseliot: I think that's what' [18:26] what's needed. [18:27] coppro: Try -Ii [18:27] but that just says "the possible values are similar but different from the menu values" or something like that [18:27] ScottK: ok, thanks for the notification [18:27] but I can't find the list of valid doc-base sections [18:27] coppro: It should also give you a Debian standards paragraph reference. [18:27] furthermore, there are packages already installed that use this section [18:27] I've looked at that reference it says nothing [18:27] Not sure then. [18:28] :( [18:28] I'm thinking I have to look into lintian's source package to see what it'll accept [18:29] That's one way to do it. IIRC it's written in Perl. [18:29] having just downloaded it, I can confirm that [18:30] Mostly Perl. Some shell is sprinkled around it. [18:30] You might also file a bug with a patch suggesting a more useful long explanation in Debian. [18:30] Hello StevenK. [18:31] I think the lintian check is busted, actually [18:31] coppro: Make sure you have the lintian from Intrepid. [18:31] I don't, that could be it [18:32] * StevenK waves to ScottK [18:32] coppro: I've asked for it to be put in hardy-backports, but it hasn't yet. [18:32] ok, it's in [18:33] nope same error [18:33] OK. Well at least you've tried the latest so no one can complain. [18:34] ok well it seems the section I want is Programming [18:34] yet the reference guide lintian quotes and a bunch of installed packages use Apps/Programming [18:35] does Ubuntu have different menu sections than Debian? [18:35] Ubuntu doesn't use Debian Menu. [18:36] well doc-base sections then [18:36] (which are supposed to basically be Debian Menu sections) [18:36] Not AFAIK. [18:36] * ScottK looks over at StevenK and wonders what he knows about it. [18:36] * StevenK tries to figure out how to travel around Boston [18:37] I don't think doc-base is used much in Ubuntu. [18:38] ScottK: was this package removed from hardy-updates? I can find it only on launchpad: https://launchpad.net/ubuntu/hardy/i386/xorg-driver-fglrx-envy/1:7.1.0-8-4+2.6.24.501-501.30 [18:38] * ScottK looks. [18:40] tseliot: It's marked Superseded, so it was replaced by the later upload. [18:40] I'd suggest have him wget the .deb from Launchpad and then install it using dpkg. [18:40] ok, now I have an irritating issue [18:41] ScottK: yes, that's the only solution. Thanks again :-) [18:41] my pacakge needs separate -arch and -indep targets [18:41] is there any way to invoke the dh_install* scripts to only install for the appropriate packages/ [18:43] hello? [18:43] ScottK? [18:46] coppro: Sorry. I need to focus on some $WORK right now. Generally you're better off asking questions of the channel in general as others may be better positioned to answer. [18:46] ok [18:46] coppro: My short suggestion is find another similar package and use it for an example. [18:47] or I could just ingore -arch and -indep... [18:47] no wait, I was misreading policy [18:55] Hi all [18:55] hi warp10 [18:56] hey jpds [18:56] hey, does anyone know how to get debhelper to install only arch:all or arch:any packages? [18:58] coppro: what do you mean exactly? [18:58] well, I'm writing a library, so the binary packages are arch:any and the docs package is arch:all [18:58] policy requires that the binary-indep target make all the architecture-indep packages [18:59] but debhelper will try to install even the binary-arch targets if binary-indep is called [18:59] coppro: paste your rules file to pastebin [19:01] http://rafb.net/p/z0fNlW18.html [19:01] slytherin: did you try building electric? [19:02] tuxmaniac: No. Didn't find time. Will do latest by Monday. [19:04] coppro: Sorry, not able to figure out the problem. [19:04] * slytherin has to restart [19:35] morning ScottK [19:44] NCommander, I have a tricky one for you [19:44] ok [19:44] how do I tell dh_install to only install some packages' files? [19:45] same with the rest of debhelper [19:45] I need to be able to split up the -indep and -arch files, so I need to tell debhelper to only install ome [19:48] coppro, create package.install files, and list the files that need to go into each package [19:48] (take a look at codeblocks to see how its done) [19:48] I know that [19:48] ok [19:48] but how do I tell debhelper to only install files from a certain package [19:48] I don't understand the question (sorry, I'm a litlte out of it) [19:49] well, policy says that the binary-arch target builds all the architecture-specific packages [19:49] Right [19:49] and binary-indep makes the independent ones [19:49] and binary-indep builds any all arch packages [19:49] but debhelper will try to install all the packages [19:50] debhelper doesn't install the package ... your make install rule does [19:50] install isn't part of policy [19:50] Where's your rules file [19:50] my point is that if i run binary-arch, I'm only allowed to make the arch-dependent packages [19:50] I think I get what your saying, but I need some point of reference to work form [19:50] http://rafb.net/p/z0fNlW18.html [19:51] Are you splitting your package into multiple ones, one of which is arch all? [19:51] no, I'm trying to package the python component into another package [19:51] so you could say yes, I guess [19:52] (and the docs are arch all as well) [19:52] So you have one python package, and one that is arch all [19:52] currently i have 2 library packages (regular and -dev), both arch any. I also have a docs package, arch all [19:52] Oh, I get it [19:53] Does the documentation need to be built or somethign (i.e. pod2man, or something else) [19:53] no [19:53] but it needs to be packaged by binary-indep, as I understand it [19:53] then build-indep is a no-op; [19:53] Unless I'm very much mistaken [19:53] build-indep, yes [19:53] binary-indep, no [19:53] (build-indep will be used by the python component anyway) [19:54] Oh, OH! [19:54] Ok [19:54] Finally I got it [19:54] I haven't played with an arch all package in quite awhile [19:54] so my question is how do I get debhelper to not bother with the ones that aren't necessarily built? [19:56] Give me a sec to think [19:59] List the docs under the *package*.docs file, and move the dh_installdocs rule to binary-indep [20:00] doy! [20:00] hi all. [20:00] thanks, I'm stupid [20:00] coppro, you may also need to copy the dh_compress, and everything below it (expect shlibdeps) to the indep file [20:00] ; rule [20:00] yeah [20:01] You threw me through a loop on that one [20:01] Most packages that just move documentation do it as part of the binary-arch step [20:01] I've seen it even if it generates an all package [20:02] packages like umit-locale-pt-pt is used on python applications? [20:02] yeah, the problem was I thought the python was arch-indep, which I just realized it isn't [20:02] so it doesn't matter [20:02] * NCommander hits coppro [20:03] There [20:03] Now you'll remember ;-) [20:03] but if it was, it would be a headache to deal with [20:03] python kinda weird because you can get both arch-dep and indep-dep python packages [20:03] (it's a python library so it has to actually get compiled) [20:03] yeah [20:03] pyscho is another one like that [20:05] but if, say, there was a perl module in the source package, that would be tough with debhelper [20:05] coppro, why, it would just use build-indep for everything since the module would be arch all ... [20:06] yes, but what if there was an arch-dep package in the source package? How do you get debhelper not to explode if it finds that the -dep stuff isn't ready to install when you try and install the -indep stuff? [20:10] coppro, you lost me [20:10] nvm I know what to do [20:10] don't have separate build-arch and build-indep targets [20:11] coppro, they're required ;-) [20:12] But you can alias them to build [20:12] no they aren't [20:12] they are clearly listed as optional in policy [20:13] then all you need to is, rather than call dh_builddeb at the end, use dpkg-deb yourself for the packages you want [20:16] thus, I conform to policy, even though it's a bit of a cop-out [20:17] REVU help wanted getting started: I uploaded a package last night using dput, but today trying to log in to it, REVU still says I don't have an account...? [20:17] jmarsden: Which package? [20:18] iriverter. I fixed LP #91237 [20:18] Launchpad bug 91237 in iriverter "java library not found" [High,In progress] https://launchpad.net/bugs/91237 [20:18] jmarsden: Have you added yourself to https://launchpad.net/~revu-uploaders? [20:18] and asked for a key update here? [20:18] OK, neat. I added myself... didn't ask for a key update. Key udpate, please :-) [20:19] it should happen overnight, so try now [20:19] jmarsden: One moment please.- [20:19] I thought that since it let me dput it must haev already recognized my key? [20:20] try dputting again [20:20] coppro: No need. [20:20] jmarsden: I've moved your upload from the rejected packages. It should appear soon. [20:20] Thanks... what caused it to be rejected?? [20:21] jmarsden: Your key was not in the keyring. [20:21] OK. But dput doesn't tell you that at time of upload? Would be nice if it did. [20:21] jmarsden: dput can't usually check that. [20:22] jmarsden: dput just checks if the sigature is correct. [20:22] OK, thanks. Makes sense. [20:24] siretart: Is the move_uploads scripts suppose to have root as owner? I cannot do: sudo -u revu1 ./move_uploads.sh [20:28] jmarsden: Et voila, your upload has been successfully processed: http://revu.ubuntuwire.com/details.py?package=iriverter [20:29] jpds: Thanks! Now I need people to review it :-) BTW, Would it be reasonable to file a wishlist bug against REVU suggesting that when it rejects an upload for lack of key in keyring, it emails the signer? [20:31] jmarsden: That would be an idea, the place is https://bugs.edge.launchpad.net/revu [20:32] Thanks. OK... now I have another new packager question: I ran debuild and lintian on Ubuntu 8.04 and they said the latest standards version was 3.7.3, so I updated to that. Now REVU lintian says the current number is 3.8.0 -- how do I get a "more up to date" development environment to avoid this kind of thing? [20:32] jmarsden: Current Standards-Version is 3.8.0. [20:33] And I think that the issue may be fixed by installing lintian from hardy-backports [20:33] 1.24.1~hardy1 is in hardy-backports so using it should work fine [20:34] jpds: OK, will do, thanks. [20:34] another way of testing that is to use an Intrepid chroot but that's a little more work to create and maintain [20:34] (running Intrepid is another way but probably not the recommended one if you need a stable system) [20:34] stgraber: Is there a document describing hwo to set that up you can point me to? Yes, I'm not ready to run Intrepid yet! [20:36] https://wiki.ubuntu.com/DebootstrapChroot should be a good starting point [20:36] stgraber: Using http://people.ubuntu.com/~pitti/scripts/makechroot makes chroot creating easier. [20:37] jpds: oh, nice [20:37] Thanks, I'll try it out. [20:38] * jpds personally comments out the bind mounts but they may be useful. [20:40] evening norsetto [20:41] hi jpds [20:41] what debhelper script is supposed to add ldconfig to postinst and postrm scripts [20:46] coppro: dh_makeshlibs [20:46] yeah, it was my fault, I did a patch wrong and seriuosly confused it [20:47] norsetto: hey there [20:47] highvoltage! [20:47] :) [20:47] highvoltage: how is it mate? [20:48] norsetto: it's going well, I was wondering if you could answer a question I have... do merges only happen during the beginning of the development cycle, or does it sometimes happen during it as well? [20:48] I'm just wondering if I'll get a chance to learn how to do it during this cycle still, or if I'd have to wait for the next one. [20:48] highvoltage: well, we can merge up until final freeze, but its more and more stricter [20:49] norsetto: aah [20:49] highvoltage: right now its ok, the merge has to be justified though [20:49] norsetto: hmm, I see. is there a place where I can find out if there's something that needs to be merged? [20:50] highvoltage: mon and dad are the right places [20:51] highvoltage: well, obviously you won't know if it will be just a sync or a merge, you will have to check [20:52] norsetto: o/ [20:52] emgent: a noi o/! [20:53] norsetto: how are things your side? [20:53] highvoltage: ecticts, but keeps an old carcass like me young :-) [20:54] highvoltage: and today I discovered that watermelon comes from south africa, man, you have paradise there [20:55] * highvoltage thought that they make watermelons everywhere :) [20:56] highvoltage: yeah they do, but it originates from there [20:56] norsetto: -.- [20:56] heh, now I'm learning food/geography in this channel too. bonus! [20:56] yeah, i have filed three bugs in series with nobody reporting a bug between. [20:57] * norsetto hands over new spectacles to emgent [20:59] I have a package which will only work with python 2.5, but any version of 2.5 will do [20:59] how do I express this in my relationship? [21:00] nvm, didn't realize python2.5 is a package :/ [21:01] coppro: either let your package depend on "python2.5" or better on "python (>= 2.5)" [21:01] can't go with python 2.6 though (if and when it comes out [21:02] you know that it will not work with python 2.6? [21:06] yes, it's a build-dep and the makefile has 2.5 hard-coded in [21:12] ScottK: sorry for opening all the tasks for the mayavi2 backport, I really thought it was necessary. [21:13] heilá warpie [21:14] ciao norsetto, good to see you around! [21:14] * norsetto wonders how warp10 correctly guessed it was him warpie [21:15] * norsetto thinks that medical doctors know more than the devil ... [21:16] * warp10 guessed it because he didn't see any other *warp* in the channel [21:17] lol [21:18] warp10: I have a strange problem actually, whenever I drink coffee, I get a sore eye [21:19] norsetto: well, try not drinking it :) [21:21] warp10: ah, I discovered you! A real doctor would have immediately said "well, try taking the spoon out first" [21:22] * emgent "intrepiding little box" [21:23] norsetto: I just thought you take coffee without sugar, like real coffee-impassioned do :) [21:23] warp10: you right, I do, damned, you are a REAL doctor [21:25] norsetto: indeed! :D [21:29] Is anyone willing to review my first ever Ubuntu packaging attempt: http://revu.ubuntuwire.com/details.py?package=iriverter [21:29] https://wiki.ubuntu.com/PackagingGuide/Recipes/PackageUpdate <-- I'm trying to package a new edict version, could anyone help me? I'm not really sure how to apply this information to packaging a dictionary file. [21:30] The current edict version is almost a year old [21:33] Pretty please? :) [21:34] kaminix: I'm a newcomer to Ubuntu packaging, but I can try to help... what is the issue you are trying to solve exactly? [21:35] Okay, so the edict file is pretty much just a long text file (japanese<->english dictionary file) in /usr/share/edict, I'm just trying to update the package but I don't know how. [21:35] kaminix: which step of that recipe did you get 'stuck' at? [21:35] 3 [21:36] kaminix: So the issue is that the new edict file you want to package doesn't come in an upstream tarball?? [21:37] It came with a gz file, which I've unpacked. It doesn't have it's own directory which I can cd in to. [21:39] If you apt-get source edict and then untar the .orig.tar.gz that gives you you'll see how the original package was back then. [21:39] Then you need to try and find the new edict in a similar form, I would think? [21:40] I have the new edict, and it's just a text file by default when you untar it. [21:42] kaminix: So where did the packager of the current Ubuntu edict package find that orig.tar.gz file? I don't know, but finding out would help you. I suppose you could try creating such an orig.tar.gz yourself??? But I'm pretty sure that breaks the rules... [21:44] I think the old packager must've created it himself, because on the official page they only give .gz I think. [21:44] * kaminix opens up official page again [21:45] Then I'd suggest you do the same, and email the old packager asking about it. [21:50] The original packager was Maintainer: Ubuntu MOTU Developers , should I mail the list? [21:55] Strange. This file contains enamdict which is not installed by the edict package but kept in it's own enamdict package [22:06] I've sent a message to the mailing list now, will I get answers or do I need to register somehow? [22:10] kaminix: If you're doing packaging work, I'd subscribe to that list. [22:10] I'm subscribed now. I don't know if I'll be packaging a lot though. [22:11] Do I want MIME digests if I use KMail? [22:26] jpds: yes. it has do be run as root in order to fix the permissions and owner of the files it moves [23:11] I be back [23:14] cody-somerville: It won't even build if I add the upstream's clean rule. [23:24] So whats up world [23:26] Well, enjoying the week-end. === bdrung_ is now known as bdrung [23:52] is now revu slow? :S [23:52] I make dput and nothing appears yet [23:52] kirkland: Did you upload it before XX:50? [23:54] yeap [23:55] http://revu.tauware.de/details.py?package=umit what's last chance? [23:55] may be cache [23:55] but I think no [23:56] i'll upload again [23:56] Don't. [23:56] If it's not there, I'll poke the server and work out what went wrong. [23:56] July 19 12:00... that's old. [23:57] * NCommander works out how old it is [23:58] wgrant, ok.. but may be it's my mistake [23:58] It's Europe/Berlin. [23:58] yeah UTC+2 [23:58] Your upload is still sitting in the incoming dir :( [23:59] k0p: Ah, you upload the i386.changes, not the source.changes.