[00:06] No worries, thanks for your help. [00:18] http://pastebin.com/fua09b5C is my lintian output so far. Whee, progress. [03:16] tumbleweed: On these DMB replacement announcements it'd be nice to know if the incumbent is running for re-election. === santiago-ve is now known as foursixnine === almaisan-away is now known as al-maisan [07:45] good morning [07:53] ScottK: no, I'm not (for re-election) [07:54] ajmitch: Hi, are you still available for getting nominated for DMB? [09:23] Morning dholbach. [09:23] hey iulian [09:26] geser: I reckon he is, otherwise we'd be very disappointed. :) [09:26] Go go ajmitch! [09:27] heh [09:27] * nigelb cheers for ajmitch as well! === kklimonda_ is now known as kklimonda [10:46] geser: yeah I may as well :P [11:09] gogo ajmitch [11:10] what's this "*still* available"? … [11:12] ajmitch mentioned in the past that he would be available for election, and I hoped he didn't changed his mind [11:12] I'm sure I trolled him about this before :P [11:24] Laney: yeah but you troll about everything [11:25] you love it [11:25] it's Our Thing [13:46] geser: Thanks. [13:47] Hiya masters! Heard REVU is still down. Can someone look at it, or has someone already? === al-maisan is now known as almaisan-away [14:07] is REVU supposed to still be up and used? [14:08] probably, but it is down === Laney changed the topic of #ubuntu-motu to: REVU is down | Precise: open for business | Want to get involved with the MOTU? https://wiki.ubuntu.com/MOTU/Contributing | Sponsor queue: http://bit.ly/fz6AyQ | http://qa.ubuntuwire.com/ftbfs | http://people.canonical.com/~ubuntu-archive/NBS | http://qa.ubuntuwire.com/bugs/rcbugs/ [14:09] Laney: is there any recourse if i wanted to get a kernel (per an approved blueprint) into the archives? [14:10] are you working with the kernel team? [14:10] they use git afaik [14:10] otherwise you could use mentors.debian.net or make a bzr branch [14:11] Laney: per the blueprint we were supposed to submit to REVU and have two MOTU approve it (themuso being one of them) [14:11] my understanding is that since this would be a community maintained kernel, the kernel team would not have direct purview over it [14:11] I don't know what it is, but they might be interested in working with you anyway [14:12] anyway one of the other places I mentioned should be fine [14:13] thanks for you help, Laney :) === almaisan-away is now known as al-maisan === al-maisan is now known as almaisan-away [15:23] hello all [15:30] Is it common practise to package a software in n-number available language separately? [15:30] how is patch applied to all of them in that case? === 20QAAK923 is now known as jbicha [17:21] udienz, about bug 913513 ... well, debian has added: [17:21] Launchpad bug 913513 in libomxil-bellagio (Ubuntu) "Please merge libomxil-bellagio 0.9.3-1 (universe) from Debian testing" [Undecided,Incomplete] https://launchpad.net/bugs/913513 [17:21] CFLAGS="${CFLAGS} -Wall -Werror -Wno-error=unused-but-set-variable -Wno-error=unused-but-set-parameter" [17:22] as a workaround of problem [17:22] we should delete that and insert the patch you mentioned ? [17:23] if the patch properly fixes the issues, you might want to ask the Debian maintainer why it wasn't applied (this might be documented in a bug already) [17:24] cjwatson, around ? you're the author of the patch :) [17:24] have you forwarded that to debian ? [17:26] l3on: i think we can use cjwatson patch, you can take liberty to forwarding to Debian [17:27] don't forget to CC cjwatson in bug reports [17:27] udienz, ok, I'm going to prepare the new debdiff [17:28] l3on: you can reference debian 625367 in your bug [17:28] Debian bug 625367 in libomxil-bellagio "libomxil-bellagio: ftbfs with gcc-4.6 -Werror" [Serious,Fixed] http://bugs.debian.org/625367 [17:28] thank you micahg [17:29] you can reopen it http://www.debian.org/Bugs/server-control [17:30] yes, but it's technically fixed, so best to open a new bug [17:31] l3on: I don't remember, sorry. But you really shouldn't need to ask me about every instance where you find a build-fixing patch of mine in a package; if it's a build fix, use your judgement to determine whether it's still needed [17:32] and if it is and I forgot to forward it then feel free to just go ahead and forward it [19:52] Hmm. I've been playing around with my pbuilder sid chroot, and I'm getting http://pastebin.com/sHc8p54f. What've I broken now? :-) [19:55] Put another way, I've got compiled python modules (cython) that are arch specific; do they belong in dist-packages? [19:58] yes [19:58] but not in usr/local [20:01] jtaylor: Ah, so I want to stash it in /usr/lib/python2.6/dist-packages instead. I presume there's a way to inform setup.py of this fact? :_) [20:02] dh_auto_build usually does it correctly [20:02] if not it probably did not find your setu.py [20:02] jtaylor: You'd think: https://github.com/KB1JWQ/salt/blob/master/debian/rules [20:03] setup.py is in the top level dir. [20:03] well thats wrong [20:03] drop the overrides [20:03] for _build and _install [20:03] jtaylor: Without them setup.py never gets called. [20:04] hm is there a makefile too? [20:04] No. [20:04] try this: dh $@ --buildsystem=python_distutils [20:04] jtaylor: That github fork is "life" [20:05] Live, even. [20:05] jtaylor: Huh, build completed successfully. [20:06] jtaylor: But I'm still not seeing the build having compiled the stuff that setup.py build compiles. [20:06] And nothing's getting stashed anywhere but in /usr/share [20:11] whats the setup.py line shown when you build with DH_VERBOSE=1 [20:15] jtaylor: Throw DH_VERBOSE=1 into rules? [20:16] export in front [20:19] jtaylor: The install is python setup.py install --force --root=/tmp/buildd/salt-0.9.5/debian/tmp --no-compile -O0 --install-layout=deb [20:19] do you copy the stuff from debian/tmp in an install file?` [20:19] jtaylor: Nope. [20:20] That... would probably explain it, wouldn't it... [20:20] than thats the reason the result is empty [20:20] This package will work with python 2.6 or 2.7; how do I reflect that in the .install file? [20:20] if you build depend on python-all-dev [20:20] usr/lib/python2* [20:23] jtaylor: And I only want to include the compiled stuff? [20:23] then make that more specific [20:24] I'm asking, the rest is just copied in. [20:24] So that stuff is already being copied in from the source itself to /usr/share, I don't need to redo that I figure. [20:25] But for the msgpack stuff, I'd want to have the install line look like: debian/tmp/usr/lib/python2.7/dist-packages/salt/msgpack /usr/share/python2*/dist-packages/salt/msgpack ? [20:25] some stuff is handled by default, e.g. copyright [20:25] everything else you have to do explicitly [20:26] usr/lib/python2*/dist-packages/salt/msgpack is enough [20:26] for compat >= 7 [20:26] jtaylor: Right, but I've explicitly had it done from the topdir-- it's static python files. [20:26] jtaylor: Ah, and that'll know where to find it? [20:27] See, what I've been doing is this: in the tarball, there's a salt directory, so I have lines like salt/utils /usr/share/salt/ [20:28] compat >= 7 will fall back to debian/tmp if not found [20:29] sry you do need a destination too [20:29] hm no, if the path in debian/tmp isok you can skip it [20:29] Yeah, it seems to have worked. [20:29] Ooh. [20:30] I threw the python-dev-all into the control file, now it looks like I'm building the socket a few times. I have a /usr/lib/pyshared/python2.7/salt/msgpack/_msgpack.so and a ./usr/lib/pyshared/python2.6/salt/msgpack/_msgpack.so as well. [20:30] you can use wildcards [20:30] I just want to verify that installing the package isn't going to deb both versions of python. :-) [20:30] dep* [21:51] I'm seeing a few python-script-but-no-python-dep errors, my control file looks like https://github.com/KB1JWQ/salt/blob/master/debian/control; what did I blow up in my deps? [21:51] ^ crossposted from #debian-packaging on OFTC, in case it looks familiar. [22:02] Ahhh! [22:02] Found it. Stupid nagging bug. [22:03] TO fix it, I have to have my packaging process patch the original source code; how do I do that properly? [22:03] I mean worst case I can have a DH_build that does it with sed, but that's inelegant. [22:14] Corey: use a patch system, like quilt or dpatch. [22:15] also, python-support is deprecated these days, see http://lists.debian.org/debian-python/2011/06/msg00136.html for info. [22:16] dpatch is deprecated these days as well :) [22:16] but you need to use it if you want to support lucid [22:17] oh damn :) [22:31] Yeah, Lucid is required. [22:31] Will DHPython2 work with Lucid? [22:34] doesn't look like it's available there [22:34] Okay, then as nice as it would be to use the latest shiny, that's not an option for me. [22:34] just dodn expect you to be targeting lucid, since you were using a sid chroot [22:35] jcfp: I'm trying to target everything reasonable in production. [22:35] This'll be submitted to Ubuntu, Debian, it'll go into an autobuilder for nightlies, etc. [22:38] jtaylor: why do you say that you need to use dpatch to support lucid? why not use quilt? [22:38] maxb: I never said that [22:38] oh, you could interpret it that way [22:39] I responded to dh_python2 [22:39] 3.0 is fine in lucid [22:39] oh right, you were talking about python-support [23:34] Okay, https://github.com/KB1JWQ/salt/blob/master/debian/control still throws a number of python-script-but-no-python-dep, even after I stripped out the version qualifier for Python. How do I sanely fix this?