/srv/irclogs.ubuntu.com/2009/06/30/#ubuntu-motu.txt

=== RoAk is now known as RoAkSoAx
=== YokoZar1 is now known as YokoZar
=== lukjad007 is now known as ShadowChild
=== keylocker is now known as leleobhz
james_wajmitch: mail won't be sent if there are no open tasks on the bug01:33
=== WelshDragon is now known as YDdraigGoch
ajmitchjames_w: I'll have to take a look at the ones I received it for then01:54
james_wplease do, if you find a case where it shouldn't have happened then let us know01:55
james_w(I'm not totally satisfied with that solution, but it was something that was quick to do and reduced mails by probably two orders of magnitude while being generally applicable)01:56
ajmitchI just noticed them all from a single package that I've subscribed to bugmail for, it wasn't a large number of mails01:56
ajmitchjames_w: it's ok, they appear to have a debian task open on them01:58
james_wyeah, the fact that a lot of bug watches haven't been updating exacerbates the problem02:04
ajmitchI just noticed that on one of the bugs I checked :)02:04
kb9vqf_ajmitch: Recompilation of everything succeeded; feel free to review it when you have time ;-)02:29
cpscottiCan anyone review this package http://revu.ubuntuwire.com/p/harpia?04:00
=== SEJeff_ is now known as SEJeff_away
=== SEJeff_away is now known as SEJeff_home
=== santiago-pgsql is now known as santiago-ve
ingenthrhey all, question on packaging: if i want to put together a package which depends on two less-common-libraries which are only available static, is it acceptable to bundle them into the package i would like to contribute?04:35
RAOF_ingenthr: For some values of "acceptible", yes.04:36
RAOF_ingenthr: Why are they only available static?  That'll almost certainly break on AMD64, for example.04:36
ingenthrdon't know for sure, i could check04:37
ingenthri believe it's because they are considered of more limited utility04:37
ingenthrit's libstrophe, an XMPP library in C04:37
RAOF_That doesn't sound like a limited-utility library.04:38
RAOF_:)04:38
ingenthrdoesn't sound like anyone's contributed it yet either :)04:39
Hobbseedirecthex: looks like your box has fallenover again :(04:39
RAOF_ingenthr: That would mean that you need to package the lib as a part of packaging the app.04:40
persiaOr rather, first package the library, and then package the application (rather than bundling them)04:41
ingenthrRAOF_: i have that done effectively, as a subdir04:41
ingenthran automake SUBDIR04:41
RAOF_ingenthr: persia said it more clearly.  Package the library separately, then make your app depend on that package.04:42
persiaOr rather, have it build-depend on the -dev binary package produced by the library package, and have done the shlibs stuff correctly so the dependency is automatic with ${shlibs:Depends}04:43
RAOF_Again with the clarity :)04:44
ingenthri just pulled the latest source to see what it does by default, and sure enough it builds a .a file04:44
* persia goes to find some oil to wipe on the screen04:44
RAOF_ingenthr: Well, that's unfortunate.  It probably needs to be fixed to build a shared object.  Also, as I said, that's likely to fail on AMD64.04:45
RAOF_Because of the wonders of PIC.04:45
persiaHasn't there been some effort underway to do PIC on i386 as well?04:46
RAOF_Yes, but it's not mandated by the buildchain; Non-PIC code will simply FTBFS on AMD64.04:47
RAOF_Well, will likely FTBFS, if it's linked anywhere near a library.04:47
ingenthrso, what's going on is AMD64 requires PIC, but i386 buildchain doesn't do PIC by default?04:48
persiaAh.  I thought that work was farther along.  I may be confused with another TLA beginning with P/04:48
RAOF_I didn't think we'd be ever mandating PIC no i386, anyway; the register it eats has a noticable (~10% for most stuff, IIRC) performance impact.04:49
RAOF_s/no/on/04:49
persiaOh my.  Yes, then it would make sense not to do that.  All the more reason for everyone to drop i386: 1) it doesn't handle very much ram, 2) it can't be made secure, etc.04:50
persia(mind you, that would have to start with HW vendors, so there's not much we can do about it directly)04:51
RAOF_Also, PIC buys you much less security on i386; there's still only a 4GiB address space, which you can apparently walk through in the order of seconds.04:51
* RAOF_ draws on hazy memories of security discussions.04:51
ingenthrso, with that digression :)   if i want to have a package considered for universe, best thing to do is fixup and contribute it's dependencies04:53
persiaingenthr, Indeed.04:54
ingenthrand even if i want someone else to run my binaries on i386 running on amd64, i'll likely have to do this anyway due to the toolchain building non-PIC04:54
ingenthrthough, if i build my binaries, i can control PIC or not04:54
ingenthror is it more complicated than that04:55
* ingenthr is familiar with position independent code, but not so much with this environment04:55
persiaingenthr, You'll want to build a source package.  The source package will be compiled for each architecture in the peculiar way best suited for that architecture.04:56
persiaIf you have an amd64 system, you can use pbuilder or sbuild to construct build environments for each of i386 and amd64.  For other architectures, you'll need to have them (or emulate, but that makes for slow compilation).04:57
ingenthrthanks for the help05:04
=== micahg1 is now known as micahg
* RAOF_ wonders whether he should respond to the rather incoherent "overgod" ubuntu-motu@ mail.05:51
* kb9vqf_ would respond and ask for his bank account # to assist with the transfer of 10 million dollars out of the country ;)05:54
kb9vqf_That is, if it's the same spam I fight all the time05:55
persiaIt's not.  It's a complaint about the terms used in a specific piece of software.05:55
RAOF_It's hard to say "I think you're making an incoherent mountain out of a sub-atomic molehill" in a respectful way.  I shall therefore not try.05:57
ScottKRAOF_: I kind of shocked him when he was here when I pointed out that it was pretty well impossible to publish anything that touched religion that didn't offend someone and predicted the outcome of his request is we'd have to remove Bible study tools from the archive.  He didn't like that idea, but it didnt' deter him.05:58
persiaIt's interesting.  I believe he just wants a s/G/g/ somewhere, but I'm not convinced it can be provided.05:59
RAOF_That doesn't surprise me.  He comes across as being supremely confident that the only action reasonable individuals could take is to remove overgod from the archives.06:00
ajmitchhe has a point, however badly he presents it :)06:00
ScottKajmitch: He has a very, very small point, but if we start playing censor on such things there is no end to it.06:01
ajmitchoh certainly06:01
RAOF_I'm not sure that's the case.  I thought, rather, that the fact that it's (possibly - the font isn't exactly the most readable set of glyphs ever) spelt "OverGod" in the title screen is evidence that the author meant to denigrate God.06:01
ajmitchfirst up against the wall is the wallpapers package from long ago :)06:01
RAOF_I wonder what he thinks of the "god" package :)06:03
persiaajmitch, That appears to be gone (or at least I can't find it in karmic)06:03
ScottKRAOF_: If I were an ardent athiest (I'm actually not) I could make a much stronger argument that the presence of material in the Ubuntu archive that supports such silly superstitions as supernatural beings of any kind is deeply offensive to me and should be removed.06:03
ajmitchpersia: oh well06:03
* ajmitch has been around too long, I even considered cleaning up a hoary chroot to get some space back :)06:05
* ScottK has been up too long and probably ought to go to bed.06:05
ajmitchheh06:05
RAOF_ScottK: I don't find that particular end of the athiest spectrum particularly convincing, either.  Particularly since there's nothing inherently evangelical about athiesm.06:05
RAOF_But yeah.06:06
ScottKDon't even get this hypothetical athiest evangelist started on the fact that there are two Ubuntu derivatives that have a specific religious focus and much of their stuff is in the official archive.06:07
persiaScottK, We've had that hypothetical individual present making that argument before.  Please don't recap by proxy :)06:08
ScottKOh.  OK.  Nevermind then.  I missed that one.06:09
* ajmitch missed it all too06:09
ScottKBTW, just backported 3 clamav packages with 18 rdepends total for dapper/hardy/intrepid and only one build failure (that on hppa, so I don't care).06:09
ajmitchI'm surprised we still get hppa build failures06:10
persiaI thought there were some Dapper and Hardy hppa users.06:10
ajmitchI thought that port was being turned off06:10
TheMusofor karmic only06:11
ajmitchso *-backports still lives on for hppa?06:11
ScottKCertainly06:11
RAOF_Thinking of ports, I need to point out that PPC kernel headers don't work properly.06:11
ScottK-security/-proposed/-updates too06:11
ajmitchwhat a shame :)06:11
TheMusoRAOF_: For what version of Ubuntu?06:11
ScottKRAOF_: Did you point it out to TheMuso?06:11
ScottKHeh.06:11
* ajmitch wouldn't mind having a PPC box to play with06:12
RAOF_TheMuso: For (at least) Jaunty onwards; nouveau-kernel-source fails to build against it, and it's due to... some code missing in the headers.06:12
persiaRAOF_, which release?  I thought that was being sorted for karmic.06:12
RAOF_Let me russle up the bug.06:12
TheMusoRAOF_: Right, I am aware of that for jaunty06:12
TheMusoRAOF_: tis fine, I know of it.06:12
RAOF_Yay.06:13
TheMusoThat reminds me, I need to fix that up for karmic, now that ports is being merged back into the main kernel source package.06:13
RAOF_Heh.06:13
RAOF_Then PPC users will get a mostly working nvidia driver!06:13
* RAOF_ hits the road.06:16
mrooneyIs there a build-dep for apt that gets the dependencies for a package, but not the package itself?06:32
mrooneyIt seems like what you'd want to run a python app from trunk06:33
Hobbseeapt-get build-dep foo?06:33
mrooneyHobbsee: no I want the depends not the build-depends06:37
mrooneyI meant a build-dep equivalent for depends06:38
mrooneybut, it should also do recommends06:39
Hobbseeoh, right06:40
Hobbseei did wonder...06:40
* Hobbsee spies dpkg-depcheck06:41
Hobbseehm, that will only give you a list06:42
dholbachgood morning07:20
ajmitchmorning dholbach07:39
dholbachhi ajmitch07:39
YokoZarhmm if dholbach's waking up that means I need to put my blog post up sooner ;)07:49
ajmitchYokoZar: another dholbach appreciation post? :)07:51
YokoZarsomething like that ;)07:52
YokoZarActually I was gonna write about branding07:52
YokoZar(apt-get install ubuntu-branding in karmic, then open solitaire)07:52
ajmitchwhen dholbach shows up I know that I've got to get back to sponsoring & uploading before I feel the whip ;)07:52
YokoZarJust put that together today07:52
ajmitchyeah, I saw the branding upload07:52
gesergood morning08:07
ajmitchhi geser08:09
=== Quintasan_ is now known as Quintasan
gaspasiretart: LOL.08:24
gaspa:)08:24
gaspasiretart: yes, it was a failed upload on my ppa. :)08:24
=== Madkiss_ is now known as Madkiss
=== krusaf|bnc is now known as krusaf
=== AnAnt_ is now known as AnAnt
gaspasiretart`: hi.09:43
siretart`hi gaspa09:46
gaspa:)09:46
=== lifeless_ is now known as lifeless
=== pochu_ is now known as pochu
Laneyanyone fancy approving the pidgin backport? bug 38932210:15
ubottuLaunchpad bug 389322 in pidgin "Yahoo server authentication changed: Pidgin =<2.5.6 will not connect to Yahoo! servers." [Undecided,In progress] https://launchpad.net/bugs/38932210:15
Laney(to hardy)10:15
Laneyscroll to the end, there's a lot of noise in that bug10:15
slytherinLaney: is it backport or SRU?10:17
Laneybackport10:17
Laneyprobably should be on another bug, but somebody already opened the task there10:18
slytherinLaney: somebody who? Was it one of the people who usually work on backports? If not then it makes sense to have separate bug.10:19
Laneyit was chriscoulson who is a developer afaik10:19
Laneyanyway the paperwork doesn't matter so much10:20
Laneyi honestly can't be bothered to open a new bug, this problem has had enough of my time already10:20
slytherinIn that case, I think you should make the backports task assigned to you. nit the SRU task. :-)10:22
Laneya backporter needs to take action on that task now10:23
geserLaney: have you seen that pidgin 2.5.8 is out? fixing CVE-2009-188910:34
Laneygeser: seb just told me10:34
Laney(bah)10:34
HewHi MOTUs. My attempt at revelation has FTBFS yet again, and this is getting quite frustrating. bug 387231. It looks like site-packages must be changed to dist-packages. How can I do this? I found references to site-packages in the configure file, can I find/replace those?11:17
ubottuLaunchpad bug 387231 in revelation "revelation 0.4.11-4ubuntu1 FTBFS" [Medium,Incomplete] https://launchpad.net/bugs/38723111:17
geserHew: have you a build log at hand?11:26
DktrKranzHew: to test it, install pkgbinarymangler, so you can simulate buildd's behaviour11:26
geserand don't forget to enable it (don't know the defaults)11:26
Hewgeser, http://launchpadlibrarian.net/28480488/buildlog_ubuntu-karmic-i386.revelation_0.4.11-4ubuntu2_FAILEDTOBUILD.txt.gz11:27
DktrKranzgeser: IIRC, it's enabled11:27
HewThanks I'll take a look at pkgbinarymangler.11:28
geserHew: you probably need to recreate the configure script so it knows about dist-packages when building for python 2.611:31
Hewgeser, how do I "recreate" the configure script? Is it sufficient to find/replace site-packages with dist-packages?11:34
mok0kees: ping11:34
geserHew: no, as it will break it for python2.5 which still uses site-packages11:37
runasandhow do you actually close a bug in launchpad?11:43
Laneyyou mean from an upload?11:43
runasandLaney: no, close it because it has been solved.11:43
Laneychange the status11:43
runasandah, found it. Thanks :)11:44
stefanlsdRainCT: when u get a sec, a revu of gears would be appreciated!  :)11:47
RainCTstefanlsd: License: MPL 1.1/GPL 2.0/LGPL 2.111:54
RainCTstefanlsd: that field is wrong, should probably be something like (check the DEP as I'm just guessing):   mpl-1.1 or gpl-2 or lgpl-2.111:55
stefanlsdRainCT: mm. yeah, i remember something like that... will check. thanks11:55
A|iis the package maintainer for openjdk around?12:19
slytherinA|i: probably not, but you can ask question.12:23
A|islytherin, i'm trying to use openjdk on opensuse build system, but installation fails:12:28
A|ihttp://pastebin.com/d1b8d468b12:28
A|islytherin, i get the same message for ubuntu 8.10/9.04 32/64 bit12:28
slytherinA|i: is your chroot updated? I believe this problem is fixed in last 2 days.12:29
slytherin*was fixed*12:29
slytherinA|i: or there is another error that the JAVA_HOME variable does not point to proper directory corresponding to openjdk12:30
A|islytherin, we cannot have access to the virtual machines, how come this problem was only fixed 2 days ago?12:30
A|ilet me try the JAVA_HOME approach..12:30
slytherinA|i: I guess I am mixing your problems. I first thought the issue was related to ca-certificate-java. But your JAVA_HOME seems wrong.12:31
slytherinA|i: What are you trying to build by the way?12:31
A|islytherin, I'm just trying to install openjdk in order to build my own java package..12:32
A|islytherin, there is no JAVA_HOME in the log?12:32
A|islytherin, should it be set to /usr/lib/jvm/java-6-openjdk ?12:33
slytherinA|i: yes12:33
slytherinA|i: if this is first time you are installing openjdk and seeing this error then you should probably log a bug.12:35
A|islytherin, yes, the virtual machines start from scratch everytime12:36
directhexthe opensuse build system is sorta like the PPA thing, but suseier. and with support for buckets more distributions12:44
directhexi don't know if its buildds do any funnies to the environment12:44
hyperairdirecthex: i don't know if they do any funnies to the environment, but they don't allow you the freedom of editing your package locally and uploading a .dsc12:45
hyperairto summarize, it's incredibly stupid for debian-based systems.12:45
hyperairi hear they don't even have support for .install files or something =\12:45
directhexo noez, my freedoms! :(12:45
hyperairthings may have changed since then12:45
directhexbasically they have little experience with debian build systems, so there are bugs12:46
cjwatsonthat sounds impossible unless they're not using the debian/rules interface, in which case most packages would entirely fail to build12:46
directhexi've filed some, and had them fixed12:46
cjwatson(.install)12:46
hyperairwell that's why i refused to move the banshee(-unstable)-team PPA to obs.12:46
directhexe.g. build-depends-indep are actually installed now12:46
cjwatsonnot that I have any reason to suggest people use the opensuse buildservice but that does sound like unsubstantiated rumour ...12:46
hyperaircjwatson: single binary packages work. and you can pass arguments to dh_install.12:46
directhexi built mono on OBS, so the mother of all splittypackages works.12:47
cjwatsonif you can't upload a .dsc, how do you interact with it?12:47
hyperairi can't remember12:47
hyperairyou upload each file individually i think12:47
hyperairin the debian/ directory12:47
directhexor upload the diff.gz12:47
A|islytherin, with setting JAVA_HOME, i get this: http://pastebin.com/m77f36d0212:48
hyperairdirecthex: oh you can do that?12:48
slytherinA|i: You should file a bug. It is behaving like for installing openjdk it needs openjdk installed.12:48
directhexhyperair, sure12:48
hyperairdirecthex: i've never really tried it before; i looked at the docs and decided against even touching it12:49
* ScottK thought you uploaded an rpm spec file and among other thing got .debs back.12:49
ScottKOne can easily imagine how well that works.12:49
hyperairalien?12:49
directhexScottK, nay12:50
ScottKOK.12:50
savvasdholbach: here? can you revert gnote back to 0.4.0 in karmic? there's a data corruption bug I missed: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=53496912:50
ubottuDebian bug 534969 in gnote "gnote 0.5 damages notes" [Important,Open]12:50
hyperairis that the one where gnote modifies the noets so they'er incompatible with tomboy so that you can't go back?12:51
directhexbut there IS some magic for re-using patches and orig from an rpm build12:51
directhexsee http://en.opensuse.org/Build_Service/Deb_builds12:51
directhexhyperair, no, tomboy was patched to support the gnote note format12:52
hyperairdirecthex: ah right.12:52
directhexhyperair, this one looks a lot like a problem i saw in 0.5.0 with a fresh install (no ~/.tomboy or ~/.gnote)12:52
dholbachsavvas: no, sorry12:52
savvashyperair: ah no, they made an sync plugin for that - this one is much worse unfortunately :\12:52
hyperairhmm12:52
dholbachsavvas: and I didn't update - it was done already12:53
dholbachI just closed the bug12:53
savvasdholbach: ah, ok thanks :) I'll just wait for the update12:53
=== AnAnt_ is now known as AnAnt
stefanlsdRainCT: i've fixed that dep5 problem up. anything else you've noticed before i push back to revu?12:57
AnAnthello, has anyone seen the sl-modem issue that I sent by email ? any ideas ?12:58
RainCTstefanlsd: don't make lines longer than 80 characters in README.Debian13:05
RainCTstefanlsd: (personally I'd remove the "designed to be used o both Google and non-google" sentence and in the second paragraph list them all together; I'd also change the first word from "Gears" to "Google Gears".. but well that's just me, feel free to ignore)13:06
RainCTI'm test-building it now13:07
A|islytherin, is it possible toinstall openjdk without ca-certificates-java? Debian Lenny does it13:07
directhexGoogle(tm)(r)(c) Gears(c)(r)(tm)13:07
RainCTdirecthex: ehm.. are you okay? :P13:09
directhexRainCT, delerious due to dieting, if that counts13:10
jonnymindhello all.13:11
slytherinA|i: that is because ca-certificates-java is recommends and not depends for openjdk in lenny.13:11
stefanlsdRainCT: i have a long line in README.debian of 92 chars, that doesnt make sense to wrap (its a string they need to remove) - would that could issues or should i wrap it anyways?13:12
RainCTstefanlsd: no, you can leave those. technically it doesn't make a difference at all, it's jsut that 80 char lines are easier to read from the terminal13:13
stefanlsdRainCT: thanks. changes done as suggested.13:14
PecisDarbshi people, is there list of packages which are planned to package for next Ubuntu versions? I would like to find out if anyone plans to package fprintd13:24
stefanlsdPecisDarbs: typically search for a bug in launchpad and debian for fprintd13:29
PecisDarbsahh, right13:30
RainCTPecisDarbs: For Debian, see http://www.debian.org/devel/wnpp/being_packaged and http://www.debian.org/devel/wnpp/requested13:31
stefanlsdPecisDarbs: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=50213813:31
ubottuDebian bug 502138 in wnpp "ITP: fprintd -- Fingerprint scanning over D-Bus" [Wishlist,Open]13:31
slytherinthat bug is pretty old.13:33
stefanlsdPecisDarbs: yeah. you maybe wanna contact the reporter and ask if he still intends to package it, if he's done anything that you can use and you can do it together, or if you can take over doing it13:36
PecisDarbsI see13:36
PecisDarbswell, I will try to do that after exams13:36
_rubenbah .. cant wait for dhcp4 to hit the repos .. other dhcpv6 implementation just pretty much suck imo :p13:52
joaopintovbox 3.0 is out today, hope it can be ready for karmic14:06
joaopintois there an SRU pending for pidgin, regarding the yahoo problem ?14:12
RainCTstefanlsd: I'm uploading :)14:19
stefanlsdRainCT: i've got some changes...14:20
stefanlsdi havent pushed yet :)14:20
RainCTokay14:20
=== dholbach_ is now known as dholbach
RainCTah right, my suggestions aren't up yet14:21
RainCTstefanlsd: poke me once you've uploaded14:21
stefanlsdRainCT: awesome, pushing to revu now :)14:22
iulianjoaopinto: Hmm, I think I saw a discussion regarding that issue.  If I remember correctly, pidgin is still under testing.14:29
joaopintoI saw someone metion it on the morning it would go into backports14:30
joaopintobut wasn't clear for me if that means there will be no SRU for the yahoo protocol problem14:30
=== jonnymind_ is now known as jonnymind
Laneyjoaopinto: it went to jaunty-proposed today14:30
iulianLaney: Do you have a bug#?14:31
Laneyiulian: not to hand, check LP14:31
Laneyor the changelog for the upload14:31
iulianLaney: Ah, right.  I've just checked the changelog but I've somehow missed your upload :-)14:34
Laneyit might not be published yet14:34
iulianIt is.14:34
iulianPublished in jaunty-proposed 4 hours ago14:34
RainCTstefanlsd: still uploading?14:35
stefanlsdRainCT: yeah. sorry. its running now. I had a bzr dev come to my office, so i was bugging him with bzr questions!  :)14:39
stefanlsdRainCT: Its up14:48
=== evanrmurphy_ is now known as evanrmurphy
RainCTargh not again that bug, the gears upload to Ubuntu is hanging at the last KB :/15:15
stefanlsdgaryvdm: https://wiki.ubuntu.com/PackagingGuide/Recipes/PackageUpdate15:19
stefanlsdRainCT: still stuck?15:27
persiaporthose, Congratulations!15:27
RainCTstefanlsd: yep15:27
porthosepersia: ty15:27
RainCTUploading gears_0.5.21.0~svn3334+dfsg.orig.tar.gz: 13811k/13812k15:27
stefanlsdRainCT: heh. do we do anything, or just wait, or poke someone?15:30
RainCTstefanlsd: just wait15:32
Railstefanlsd: congrats!15:36
stefanlsdRail: heys! congrats to you to! thanks for all your help!15:37
persiaIt uploaded?  Cool!15:37
Railstefanlsd: FF 3.5 is awaiting ;)15:37
stefanlsdpersia: RainCT busy uploading it... :)   (its kinda sticking, but yeah. got someone upstream to fix that license issue!)15:38
persiaIt's *huge*, but nice work with the licensing.15:39
stefanlsdpersia: thanks! (and thanks for all your help also!) i cant believe how long its taken.15:42
persiaWell, it's a huge package.  Few people would have dared to tackle it.15:43
stefanlsdyeah, hindsight is 20/20  :)15:44
stefanlsdgaryvdm: http://www.debian.org/doc/debian-policy/ch-relationships.html15:45
c_kornwhy does the package text say that it is scilab-5.1-0ubuntu2 but it is scilab-5.1.1-6 (see the files at the right)? http://packages.ubuntu.com/karmic/scilab15:49
garyvdmstefanlsd:https://bugs.launchpad.net/ubuntu/+source/qbzr/+bug/39389015:54
ubottuUbuntu bug 393890 in qbzr "Update qbzr to 0.11.0" [Undecided,New]15:54
RainCTstefanlsd: I'm away for some hours, will leave dput running16:03
stefanlsdgaryvdm: http://wiki.debian.org/DebianPython/NewPolicy16:04
stefanlsdRainCT: ok. thanks! :)16:04
=== rgreening_ is now known as rgreening
dholbachCan I interest somebody in giving a Packaging Training session at 02nd July, 06:00 UTC? https://wiki.ubuntu.com/Packaging/Training has lots of ideas for sessions16:26
Laney# Dealing with hideously packaged packages hahaha16:32
LaneySerious tact needed there16:32
dholbachLaney: interested in something on the list?16:34
Laneydholbach: not this weekend I'm afraid, maybe one after16:34
Laneywait, it's even earlier than the weekend16:35
Laneyeven worse!16:35
* Laney has to run now anyway, bye16:35
dholbachLaney: thanks for considering it :)16:35
dholbachtake care16:35
DktrKranzdholbach: I could sign up for a Python packaging one, I need to match some times, though.17:00
dholbachDktrKranz: match some times?17:01
DktrKranzdholbach: 18 UTC should be fine if I manage to get home in time, other times are no-go for me17:02
dholbachDktrKranz: ah ok - that sounds like August then17:04
dholbach:-)17:04
DktrKranzyeah17:04
DktrKranzand probably have some holidays, then :P17:04
dholbach## * 16th August, 18:00 UTC, '''Python Packaging''', `DktrKranz`17:05
dholbachthanks muchly Luca!17:05
DktrKranzdholbach: I could eventually split apps from modules or extensions, since they should be handled separately17:06
DktrKranzand double my sessions17:06
dholbachDktrKranz: sounds fantastic17:06
dholbachDktrKranz: do you just want to update Packaging/Training with that info?17:07
DktrKranzI'll do17:07
* dholbach hugs DktrKranz!17:07
* DktrKranz hugs dholbach back17:07
* DktrKranz wanted to do a session on his birthday, but missed by four days :)17:07
dholbachah damn17:08
dholbachwhich day is it?17:08
DktrKranzaug 12, but I'll probably be on holiday with limited internet access, 16th is fine17:08
dholbachcool17:11
dholbach:-)17:11
RoAkSoAxIs there a way to know which values should go under Depends when packaging a pure python app?17:13
DktrKranzRoAkSoAx: do you mean how to discover which modules it imports/requires?17:15
RoAkSoAxDktrKranz, yes17:16
keesmok0: hi! sup?17:26
DktrKranzRoAkSoAx: you can use http://svn.debian.org/viewsvn/python-modules/tools/, if pacakge is not complex, I use grep :)17:35
RoAkSoAxawesome thanks DktrKranz17:37
DktrKranzRoAkSoAx: if you have some Python related questions, ping ;)17:38
=== itnet7_ is now known as itnet7
=== dpm is now known as dpm-afk
AmpelbeinDktrKranz: hi. may I ask your opinion on bug #377871 ? Should we go with the debian way and don't specify the interpreter directly?18:10
ubottuLaunchpad bug 377871 in keysafe "Please merge keysafe 0.4.1 (universe) from debian unstable (gnome)" [Undecided,Confirmed] https://launchpad.net/bugs/37787118:10
* POX marks 16th August in the calendar ("show up on #ubuntu-classroom and make DktrKranz sweat" ;)18:29
pochuDktrKranz: are you giving a session?18:37
pochuoh18:37
* pochu just read backlog :)18:37
=== asac____ is now known as asac
CarlFK1how do I use "cvs diff' to make a .dpatch ?18:45
CarlFK1it doesn't seem to like:  diff -u -r1.70 dvgrab.cc18:46
CarlFK1--- dvgrab.cc15 Jan 2009 08:12:37 -00001.7018:46
CarlFK1missing header for unified diff at line 11 of patch; can't find file to patch at input line 1118:46
DktrKranzpochu: yep, do you want to co-host it?18:53
CarlFK1apparently it doesn't like the datetime format18:54
DktrKranzPOX: erm... I cancelled it... go for a walk instead :D18:54
pochuDktrKranz: I wouldn't mind, but I don't know if I'll be available that day19:03
pochuDktrKranz: remind me about it a few days before it and I'll tell you ;)19:03
DktrKranzpochu: so we can be two to swarm at POX's expert eyes :)19:03
DktrKranz*sweat19:04
cabreyis firefox-3.5 being packaged and is the firefox meta package going to point to the new FF?19:10
CarlFK1cabrey: firefox-3.5 is in the repo - duh no the 2nd 119:11
CarlFK12nd Question19:11
cabreythe firefox-3.5 in the repo right now is beta419:11
CarlFK1oh so it is.  missed that19:12
CarlFK1never mind me :)19:12
cabreyi don't understand why it is so bad to upgrade to FF3.519:12
cabreywill provide a better experience to users19:13
directhexcabrey, define "stable"19:13
cabreysoftware that is declared stable19:14
Pici!ff35 | cabrey19:16
ubottucabrey: Firefox 3.5 Final is currently being built and tested for Jaunty.  If you install the firefox-3.5 *beta* package from the repos, it will be updated when testing is complete. See also http://is.gd/1jkNY19:16
PiciDoesnt answer #2, but clarifies #119:16
cabrey:)19:16
=== cabrey_ is now known as cabrey
=== cabrey_ is now known as cabrey
POXDktrKranz: if you want me to proof read your speech or something, just ping me19:30
RoAkSoAxDktrKranz, what do you grep when searching python modules for Depends ?19:33
quadrisprohi devfil19:33
devfilquadrispro, o/19:33
DktrKranzRoAkSoAx, I do a grep * -lroie "import", this is good for small packages, or you'll be flooded19:34
RoAkSoAxDktrKranz, and what about for Build-Depends ?? Should I just try to install and add there the packages need while installing?19:35
jpdsDktrKranz: Why not uniq it?19:37
DktrKranzRoAkSoAx, do as for normal packages, add them if you need them19:38
POXRainCT: if upstream is sane, you can search for setup_requires in setup.py (for build dependencies) and install_requires (or package-egg/requires.txt) for runtime dependencies19:39
DktrKranzjpds, it works, but you still have to remove spaces19:39
POXRainCT: sorry, wrong nick19:40
POXRoAkSoAx: ^^19:40
RoAkSoAxDktrKranz, ok thanks :)19:40
RoAkSoAxPOX, cool thanks :)19:40
quadrisprojdstrand: are you around? :)19:40
cpscottiHey... I'm running out of coffee here.. could someone please check my package (I thing I already cleaned it for all "machine detectable" errors).. Its a python GUI for computer vision code generation. here is it: http://revu.ubuntuwire.com/p/harpia20:17
cpscottimok0 warp10: since you are from the science team, you could help even more!20:20
DktrKranzcpscotti, tag it as python, I usually process them20:36
cpscottion motu?20:40
cpscottiDktrKranz: how can I tag the package at revu?20:41
DktrKranzcpscotti, Tags: [edit], at the bottom of the page20:58
cpscottiDktrKranz: I see "Tags: python" but no "edit" link. Have you added the python tag? Does everybody have permissions to do that?21:11
gaspageser: are you expert of launchpadlib?21:32
RoAkSoAxDktrKranz, when packaging a python app that for example installs things in /etc/httpd/conf.d instead of /etc/apache2/conf.d, can I modify the setup.py to fix this things?21:42
cpscottiRoAkSoAx: are those "things" data files or py programs?21:48
RoAkSoAxcpscotti, config files that allow the app to be accessed via apache21:49
cpscottiRoAkSoAx: i think you can relate them on the data_files list of setup.py21:50
cpscottisomething like: data_files=[ ('/etc/httpd/conf.d', 'path2Theconfig')),...]21:51
RoAkSoAxcpscotti, for example, I have this: (wwwconf,  ['config/cobbler.conf']), where wwwconf is /etc/httpd/conf.d21:54
cpscottiRoAkSoAx: yeah.. that will do it21:58
RainCTstefanlsd: omg it's still hanging on the last KB :/22:52
stefanlsdRainCT: heh. maybe u should just abort it and try again?22:53
Laneydput?22:53
RainCTLaney: yep22:54
Laneyknown bug, check lp :(22:54
droolpalI am trying to test the new Phatch version, but I can not upgrade because of unmet deps: phatch:23:02
droolpal Depends: libtiff-tools  but it is not installable23:02
droolpal Depends: libjpeg-progs  but it is not installable23:02
droolpalhow do I find what is messing up these dependencies23:02
droolpal?23:02
cjwatsondroolpal: 'sudo apt-get install phatch libtiff-tools libjpeg-progs'. Repeat down the chain until you get a useful message.23:21
=== jono_ is now known as jono
RainCTstefanlsd: same again :/23:50
* ajmitch mutters about buggy insecure packages23:52

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