/srv/irclogs.ubuntu.com/2014/12/29/#ubuntu-devel.txt

=== sz0 is now known as sz0`
al_exquemelinwhat's the versioning rule for PPA packages that are brand new? (I mean, not present in Ubuntu's nor in Debian's repositories)01:15
al_exquemelinshould some -XubuntuY suffix be added to the version number?01:15
al_exquemelinI guess it should be -0ubuntu1, as it is the first release for Ubuntu, right?01:25
tewardal_exquemelin: i don't think there's PPA-specific packaging schemes, but that would work, appending ~ubuntu14.04 or similar based on the security team's versioning scheme01:27
tewardI use a slightly different one on my ppas, except when they're build tests of Ubuntu-destined packages...01:27
al_exquemelinteward: I've got Ubuntu 14.10, and without the -XubuntuY part, bzr builddeb gives an error stating inconsistency between source format and version01:34
al_exquemelinso I think -0ubuntu1 should work01:35
tewardal_exquemelin: well, you'll need -* so either -0 or -0ubuntu101:36
al_exquemelinteward: ok, I'll stick with -0ubuntu1, thanks01:39
al_exquemelinand yet another question. when entering the build directory, bzr builddeb tries to clean it, as far as I can see01:53
al_exquemelinbut there are no object files in it yet, so dh_auto_clean exits with an error01:54
al_exquemelinwhat should I do to proceed with building?01:54
al_exquemelinfor now I changed rm *.o in my Makefile to -rm *.o, this should force the error to be ignored02:04
=== sz0` is now known as sz0
=== sz0 is now known as sz0`
xnoxNoskcaj: tah.09:34
* xnox is not sure why i'm not getting notifications about highlights in this hexchat =(09:35
xnoxLogan_: let me see.09:35
Logan_I have no idea what you're replying to :P09:35
xnoxLogan_: ari appears to have merged it already09:36
Logan_what package?09:36
xnoxLogan_> xnox: can you please sync or merge guayadeque? you're the TIL, and it's blocking a transition ;)09:36
Logan_ah09:37
Logan_I prefer notifying the TIL before taking a merge ;)09:37
xnoxLogan_: it is still stuck in -proposed, not sure if it needs another rebuild or not.09:37
Logan_looks like codelite would break if wxsqlite3 migrates09:40
Logan_but that doesn't make sense09:40
Logan_oh, it's because it loses ppc64el09:41
Logan_(I think)09:41
* Logan_ retries that build09:42
Logan_xnox: hmm, looks like codelite needs liblldb-3.5-dev on ppc64el09:49
Logan_is there any progress on that? I see you were involved in https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=75638009:49
ubottuDebian bug 756380 in llvm-toolchain-3.5 "llvm-toolchain-3.5: Add support for ppc64el" [Normal,Fixed]09:49
Logan_although the title is misleading, as it just "supports" it by disabling lldb support09:49
xnoxLogan_: hm?! we have llvm-3.5+ on ppc64el everywhere....09:50
xnoxwhat's the problem?09:50
Logan_liblldb-3.5-dev isn't built on ppc64el09:50
Logan_it's not in the architecture list09:50
xnoxLogan_: it appears that codelite might be able to build without lldb component.09:53
Logan_I was just about to test the same thing :)09:53
Logan_you built it locally without that build dep? and it was successful?09:53
Logan_yeesh, it takes 20 min on the buildds09:53
Logan_xnox: I'm going to sleep now, but feel free to upload another codelite with [!arm64 !ppc64el] for lldb if you find that it's fine locally09:57
Logan_I wonder what features would be lost09:57
* Logan_ shrugs09:57
Logan_ah, it looks like it's just a plugin09:58
Logan_we should be fine :)09:58
Logan_g'night09:58
xnoxLogan_: night.09:59
xnoxdoko: are you going to do an archive rebuild with gcc-5 this cycle?10:48
* xnox is excited about -std=gnu11 by default10:48
mitya57What happened to Launchpad? It says it can't resolve two of pyqt5 build-dependencies, but they are clearly present (libenginio1-dev and libqt5websockets5-dev).15:04
mitya57Oh I see, someone demoted them to universe.15:05
mitya57Hmm, they were always in universe, but the previous version of pyqt5 was synced from a landing PPA where all builds are against universe... :-/15:11
mitya57ScottK: do you think we should MIR qtwebsockets/qtenginio or disable that for Ubuntu?15:12
=== Noskcaj_ is now known as Noskcaj
tewardquestion: where do i upload to for the ubuntu repositories now that I have approved upload access for the nginx package?  There's a merge in the sponsorship queue, but since I have upload rights I can upload it myself, I just need a little bit of information to do so20:15
xnoxteward: $ dput *_source.changes20:16
Laneyhttps://wiki.ubuntu.com/MOTU/New20:16
xnoxLaney: oh, nice one =)20:16
xnoxLaney: next time i'm asking "how to upload a package into ubuntu" on the dmb meeting.20:16
tewardahh, okay, so it's just dput.  nice.20:16
Laneygood page20:16
Laneyxnox: "What is xnox's favourite drink?"20:17
tewardxnox: alternatively: put it on the FAQ list for what new uploaders frequently ask questions on :)20:17
tewardLaney: yes, that one is important.  :)20:17
xnoxLaney: .... and if one responds beer they get declined to apply for another 12 months.20:20
Laneya lovely pint of guinness20:20
xnoxLaney: i have never tried guinness =) i'm saving that for dublin.20:20
bekksWrite down your name and your address onto some note, before entiering the pub :D20:22
xnoxbekks: that's one reason why i have bussiness cards with me, when i'm going out.20:25
bekksxnox: Well, it's kinda hard for the barkeeper to get you on a plane/train/bus to your home/business address ;)20:26
bekksBut I guess you got the point ;)20:26
tewardis it atypical for web server binaries to FTBFS on powerpc, and do I need to panic with this failing to parse ddeb data?21:26
nxvlhi, i'm having a pretty weird issue with a package, when i try to build a package it's FTBFS with "undefined reference to" a function, but when i compile using make and the exact same compiler flags it works fine22:59
nxvlany pointers on what could sbuild be adding that i'm missing?23:00
nxvlhttp://paste.ubuntu.com/9642820/23:01
mdeslaur_nxvl: you probably need libimobiledevice-dev as a build-depends23:15
mdeslaur_just guessing23:15
mdeslaur_nxvl: also, go buy an android :)23:16
mdeslaur_nxvl: is libimobiledevice what you're actually trying to build?23:17
nxvlmdeslaur_: i'm building libimobiledevice23:18
nxvlmdeslaur_: what i've found is that on plain make configure status i get "dependency style of gcc... gcc3" and on sbuild i get "dependency style of gcc... none"23:19
nxvlwhich seems to be the one bit producing the issue23:19
mdeslaur_are you rebuilding the package from the archive? what release? what version of package?23:20
nxvlmdeslaur_: new upstream version, on trusty23:21
mdeslaur_and you used the trusty package as a base, or the utopic/vivid package?23:22
nxvlvivid23:22
nxvlbut that one builds just fine23:22
nxvlso, i first updated to 1.1.7+git20140505.1be575f823:23
nxvlthat worked fine23:23
nxvlnow that i'm updating to the final 1.1.7, i'm having the issue23:23
mdeslaur_hrm23:23
nxvlwhat i don't understand is why on sbuild the gcc dep method is none23:24
nxvland not gcc3 too23:24
mdeslaur_where did you get 1.1.7+git20140505? is that your own tarball?23:24
nxvlyup23:24
nxvlgit pull and re-tarball23:24
mdeslaur_hrm...would have to diff the two tarballs to see what changed23:26
mdeslaur_my guess would be it's because of this: http://cgit.sukimashita.com/libimobiledevice.git/commit/?id=4c4bbd31f52845de70f5b828121eeea62f8b451423:33
mdeslaur_but I don't know what the fix would be, and have to go23:34
nxvlmdeslaur_: thanks!23:34
mdeslaur_it's just a quick hunch, but may point you in the right direction23:35
* mdeslaur_ goes away23:35
mdeslaur_perhaps disable Use-symbol-script-to-exclude-private-symbols.patch23:36
mdeslaur_(wild guess)23:36
* mdeslaur_ goes away for real23:36
nxvlmdeslaur_: yes, that did the trick, now you can go in piece!23:45
nxvlpeace*23:45
sladenlibimobiledevice ... that's a familiar name23:52

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!