[00:31] anyone running lucid alpha yet? [00:45] hi folks does anyone want to revu my package rssedit? I need some advocates please [01:36] Ok. I want to use monodevelop-vala, so it's time to check out those merges :) === dous_ is now known as dous [03:32] Hi! How do I find the package maintainer of a community managed project to ask him to provide an update as the default install segfaults on karmic. Project is renpy and bug is here: https://bugs.launchpad.net/renpy/+bug/467730 [03:32] Ubuntu bug 467730 in renpy "Segfault when launching" [Medium,In progress] [03:36] BLUG_Fred: I'll mark it triaged as the upstream maintainer said that it's fixed in a new upstream version [03:36] micahg: thanks a lot === MTecknology is now known as MTeck-ricer === dous_ is now known as dous [05:20] slangasek, ping, when you have time please have a look at bug #486157 thx :) [05:20] Launchpad bug 486157 in xtide "Sync xtide 2.10-2 (universe) from Debian testing (main)" [Unknown,Fix released] https://launchpad.net/bugs/486157 [05:23] * porthose hates it when life gets in the way of his ubuntu development fun ;) === cemc1 is now known as cemc [07:53] good morning [07:53] good morning and you are an hour late [07:54] I know, I'm sorry [08:00] are non-free packages autosynced to multiverse? [08:01] I'm interested in the cuneiform ocr package in debian non-free, but it's not yet in ubuntu [08:13] pwnguin: I don't think they are auto synced. You will have to request and justify the sync. === noodles775_ is now known as noodles775 [11:36] I'm looking for a bit of advice from a Java packaging wizard. I'm going to be doing a course which requires me to use javanns (http://www.ra.cs.uni-tuebingen.de/downloads/JavaNNS/) and sb-mase (http://sourceforge.net/projects/sbmase/). Both of these supply the original Java source code and not just the compiled bytecode. I guess I'm looking for a good example package that I can download and use as an example. I'm familiar [11:38] Also... does anyone know of a link on how to get your PPA packages built for multiple series? I tried googling this. Is it just a case of "dch -i" and setting the series to have you want and put a changelog entry of * Backport in before uploading again? [11:48] I've just taken a look at azureus source package and think I'll use that... [11:51] azureus is a BIG example [11:51] i'd poke ttx if i were you [11:51] which may mean waiting for timezone alignment [11:53] directhex: thanks. Strangely... seems like the smaller packages are put together more hap-hazardly with no build systems and stuff all over the place [11:56] you play the hand you're dealt [11:56] and java upstreams can be haphazard themselves [12:03] directhex: Yeh. I need to look at setting up a build system for it. Looks like ANT might be quite easy to do this from the azureus example... but I've poked ttx and will wait for some expert guidance on it. [12:05] pmcenery, typically in my experience, if upstream provides NO build system, then don't go overboard in plugging the gap. a manual "compile.sh" which just runs javac won't offend anyone if upstream didn't give you better [12:06] Do you know offhand of a package which uses such a script? [12:11] not for java, no [12:14] directhex: thanks. I'll have a play and see if I can build it first. Looks like I just need to get it to build a single jar and put that in /usr/share/, and do all the desktop links etc. These are fairly small apps [12:15] /usr/share//name.jar rather [12:22] pmcenery: you should have a look at https://wiki.ubuntu.com/JavaTeam/LibraryPackaging [12:23] pmcenery: if the project doesn't provide any kind of build system, I'd write a debian/build.xml and use ant CDBS [12:24] See the geronimo-*-spec packages for a minimal example of that (for example geronimo-interceptor-3.0-spec) [12:25] slytherin: we should probably update the examples to use the mh_ magic now... [12:26] that would remve the dh_link part [12:26] ttx: thanks. I'll check that link. The one I'm looking at the moment is sb-mase, which doesnt appear to have "built" to a .jar. Its just got a classes folder and a start.bat in it I've executed the commands manually and it seems to actually run. Should I be trying to make this all into a .jar in /usr/share/pkgname/, or should I be putting all the classes there? [12:27] either a jar in /usr/share/java, or classes directory in /usr/share/pkgname. [12:27] I'd recommend the former if that jar is intended to be used by something else [12:29] ttx: I think these classes are only used by the app itself. One of the "classes" is also just the icons and stuff used by the application. I'm guessing I should opt for just the classes in /usr/share/pkgname/ and patch the source if necessary to find them. [12:30] pmcenery: then yes, eithe ship a classes directory or a jar in /usr/share/pkgname [12:31] ttx: thanks for the advice. I'll let you kwow when I got something remotely build-able === pace_t_zulu_ is now known as pace_t_zulu [13:59] \sh: ola? [14:07] Hi, this chat room wouldn't happen to be about the Mark of The Unicorn or MOTU company of audio interfaces would it? [14:08] ttx: ever played with jetty configuration? need some quick help with that. [14:08] pepsifx357: nope [14:08] pepsifx357: no, sorry. [14:08] I'm afraid not. :) It's about "The Masters of the Universe" that maintain the universe package repository of Ubuntu [14:08] ok thanks [14:08] !motu | pepsifx357 [14:08] pepsifx357: motu is short for Masters of the Universe. The brave souls who maintain the packages in the Universe section of Ubuntu. See http://wiki.ubuntu.com/MOTU [14:08] Oh ok, thanks [14:08] slytherin: not really. I plan to, in late January. [14:09] ok [14:13] * slytherin feels it is time we got rid of compiz from default install, it's just eye candy [14:55] Heya gang [14:56] huhu bddebian :) [14:56] Heya sebner [14:56] Hi bddebian and sebner [14:57] Hi geser [15:00] dholbach: hi! would you mind adding ubuntu-security-sponsors to http://people.ubuntu.com/~dholbach/sponsoring/ ? [15:01] jdstrand: not at all [15:01] ahoi geser :) [15:01] * sebner waves at jdstrand dholbach [15:01] hi sebner :) === alkisg1 is now known as alkisg [15:03] dholbach: thanks! [15:14] I have a bad feeling about the exFAT thing. [15:17] indeed [15:24] jdstrand: pushed, should be part of the list after the next cron run [15:24] jdstrand: just seems empty right now [15:24] * jdstrand hugs dholbach [15:25] * dholbach hugs jdstrand back :) [15:25] dholbach: yes it is, I only just created the team a few minutes ago :) [15:25] dholbach: going through https://blueprints.launchpad.net/ubuntu/+spec/security-lucid-sponsorship-review right now [15:25] nice :) [15:25] 5 [15:25] o/ [15:26] could someone update MoM please? it seems to be frozen for over a week [15:41] Maybe Adri2000 needs to put a copy of it up somewhere .... === vorian is now known as Obama === Obama is now known as LordObama [16:12] randomaction: what I've heard MoM doesn't like the new format 3.0 packages, don't know if it's fixed or not [16:38] against what package should bugs be filed about the way the alternate cd has been built? [16:39] try debian-cd [16:40] greetings all! I am trying to help the Ubuntu Studio developers build packages and find myself stymied again and again my build errors [16:40] that I cannot solve due to my ignorance, is there a place to solicit for a mentor to help me learn more about building packages? [16:41] geser: thanks for your help before with my lucid pbuilder environment, works like a charm now :) [16:41] s/my/by [16:42] why not ask your questions here and get input from the whole MOTU community? [16:43] thanks ogra [16:46] LucidFox, about fuse-zip, please bump Standards up to 3.8.3 [16:47] geser: I can do that :) but I'll have to wait until I'm in front of my home computer [16:47] quadrispro: If this is for Ubuntu, unless the package isn't in Debian, please don't. [16:48] It should be 3.8.3... [16:48] ok, but debhelper build-dep should be at least 7.0.50 (due to the use of override_*) [16:51] Well, I mean... it *is* 3.8.3 there. [16:51] this may sound cliche or dippy but I have learned gads of information already and I feel very fortunate to have these resources available from which to learn [16:51] ScottK, why shouldn't non-Debian Ubuntu packages have 3.8.3? [16:52] BTW, the package looks good to me, LucidFox did you get in touch with debian ITP bug's owner? [16:52] LucidFox: I meant the other way around. If we are merging from Debian, don't bump the standards version from what's in Debian. It's pointless and ubuntu-policy says not to. [16:52] Sorry if I misunderstood the situation. [16:53] I don't see any ITP in Debian, only an RFP. [16:53] ScottK> It's a new package on REVU, lintian complains that the version is 3.8.3 rather than 3.8.0. [16:53] Due to an old lintian on REVU [16:55] LucidFox, ok, if you are looking for a co-maintainer to maintain it even in Debian, feel free to ping me :) [16:55] * LucidFox nods [16:55] So you're going to advocate and upload it? [16:57] yep, I can change those little details and upload it [16:58] Little details? [16:58] Standards, dh build-dep and maintainer field (should be set to Ubuntu Developers ) [17:01] oh, standards is right and REVU is wrong :) [17:05] LucidFox, looks really good to me, http://home.alessiotreglia.com/lucid/pool/fuse-zip_0.2.8-0ubuntu1/ [17:05] advocating and uploading [17:05] Danke. [17:14] Bitte [17:34] Hi, I'm using the resolvconf package, which currently completely breaks DNS due to bug #448095. Since I have to fix the package anyway, (to get a working DNS again, ) I wondered if I should, (instead of a local quick&dirty hack,) try to really fix it and try to get it into the official updates. Is this a good idea? What is the correct procedure? (Just attaching a debdiff to the bug?) [17:34] Launchpad bug 448095 in resolvconf "resolvconf starts after ifupdown, does not pick the dns-nameserver and dns-search lines up from /etc/network/interfaces" [Undecided,Confirmed] https://launchpad.net/bugs/448095 [17:35] michas: https://wiki.ubuntu.com/StableReleaseUpdates [17:36] * syn-ack trips micahg for the heck of it [17:36] I'd like to say good morning to you today, sir is all. :) [17:36] hi syn-ack [17:48] micahg, I would call it a "Bug which represent severe regressions from the previous release of Ubuntu", hence I guess it would justify an SRU. But independently of the SRU, the next step should be a debdiff to fix it in lucid, right? [17:49] michas: correct [17:50] michas: it's already been marked regression-release, so there will be an update at some point [17:50] michas: but, if you can make a debdiff for lucid, that would speed things along [17:55] micahg, alright, I'll do so. thanks for your help. === yofel_ is now known as yofel === yofel_ is now known as yofel [19:39] i have been trying to build zynjacku with pbuilder and I get an error "pythondir is NOT in your Python's sys.path" so I log into my pbuilder environment and see [19:39] that I do have the directory "/usr/local/lib/python2.6/site-packages" but when I check "python -c "import sys; print sys.path"" it is not in my python path [19:40] how can I add this into my path within my pbuilder environment so that I can then build zynjacku in launchpad for inclusion for Ubuntu Studio? [19:49] ScottL_: You shouldn't be using /usr/local in a package for the archives. [19:57] ScottK: so perhaps I can set the pythondir with DEB_BUILD_OPTIONS in the rules files? [19:57] That sounds likely [19:59] ScottK: if so then can you suggest a more proper place? /usr/lib perhaps? [20:00] ScottK: sorry for my vast inexperience, but I'm learning rapidly :) [20:01] ScottL_: Sounds like you need to include --install-layout=deb on the setup.py command line [20:01] ScottL_: The Debian Python policy (which we use) is in the midst of a major update. Have a look at http://lists.debian.org/debian-python/2009/12/msg00009.html and that should have most of the python specific answers you are looking for. === Pici` is now known as Pici [20:02] I don't think DEB_BUILD_OPTIONS is meant for that... [20:02] ScottL_: yes [20:05] maxb and ScottK: thank you for the answers, now I'll have to research them [20:06] pochu: I'm a little confused by your repsonses ? [20:07] I don't think setting pythondir in DEB_BUILD_OPTIONS will have any effect [20:07] but what maxb said :) [20:08] pochu: ah, thank you [21:03] jdstrand: hola [21:03] o/ [21:03] 14:56 < jdstrand> kees, mdeslaur: ok, I updated SecurityTeam/UpdatePreparation#Submission to point to SponsorshipProcess. it, among other things, points to SecurityTeam/SponsorsQueue [21:04] kees, mdeslaur: so, as per our UDS discussion and this blueprint: https://blueprints.launchpad.net/ubuntu/+spec/security-lucid-sponsorship-review [21:04] kees, mdeslaur: we are all tied into the normal sponsorship process [21:05] kees, mdeslaur: it would be great if at some point you guys could read SecurityTeam/SponsorsQueue [21:05] jdstrand: nice! looks really good. [21:05] kees, mdeslaur: it basically follows the MOTU/Sponsorship/SponsorsQueue process, and differs mainly in that ubuntu-security-sponsors remain subscribed to the bug in most situations. [21:05] reading SecurityTeam/SponsorsQueue now [21:05] s/and/but/ [21:06] kees: oh, I was just going to say 'no rush' :) [21:06] I'm at a nice pause while making lunch for my mom-in-law. :) [21:08] jdstrand: 'The Status should be "New" for sync requests' is that accurate for -security? [21:09] jdstrand: as in, we have "sync" requests? we have the debian fake-syncs, is that what's meant there? [21:09] kees: it is meant to refer to fake syncs [21:09] ok [21:09] (I forgot to add a link to that) [21:10] ah, I see the "Syncs" section now. heh [21:13] jdstrand: what do you think of adding a "7." to the "Notes for Contributors" that asks the contributor to comment on the runtime testing that has been done? [21:14] also, do you think linking to or including the "Issues that warrant a security update" from the https://wiki.ubuntu.com/SponsorshipProcess to the "Requesting Sponsorship" section of https://wiki.ubuntu.com/SponsorshipProcess ? [21:14] (so that someone can attempt to self-determine which pocket to target?) [21:14] kees: sure. we can link back to the new, softer and gentler SecurityTeam/UpdatePreparation#Testing [21:15] kees: that sounds good too [21:15] hehe [21:15] * jdstrand goes to do it [21:15] this is very cool, nice. [21:18] kees: '7' is done and I cleared up the fake sync dealie [21:18] * jdstrand goes to SponsorshipProcess [21:22] kees: done === elky is now known as Guest85008 [22:04] Hi there [22:06] I have a question regarding getting a package into universe so hope this is the place [22:08] What are the process when I have a Launchpad project with sourcecode and i whant to build the package from there, and not just build a source package on my laptop and send it to revu. [22:12] Our process is throught REVU. [22:13] The fact that upstream may be hosted on Launchpad doesn't affect that. [22:26] I have attached a debdiff with the fix for a bug, I should susbcribe ubuntu-universe-sponsors for the sponsorship process right ? [22:27] thx ScottK [22:29] joaopinto: yes [22:44] /win/win c [23:30] earlier I asked for help about a python sys.path issue, but now that I read the warning again I was mistaken about the location: [23:30] this is where the scripts are located: Python script dir : /usr/lib/python2.6/site-packages not/usr/local so that is good [23:30] but I find the other warning: pythondir is NOT in your Python's sys.path which means you should add it to PYTHONPATH at runtime, or add a .pth file to one of your Python's existing sys.path directorie [23:30] and : Alternatively, re-run ./configure with PYLIBDIR set to override the default pythondir, or adjust prefix to match the prefix used to build your Python. [23:31] is there something I can do in the rules file dh_make created to solve this? [23:32] ScottL_: Debian/Ubuntu packages are expected to install into dist-packages not site-packages from Python 2.6 onwards [23:32] maxb, is this something I should talk with nedko (upstream dev) about? [23:32] More conventional distutils-based builds will do this properly when passed --install-layout=deb. If this thing has a nonstandard buildsystem, it'll require patching [23:34] I searched through various config files and make files but couldn't find anything that called for setup.py which you suggested I pass the "--install-layout=deb" to :( [23:34] Can you point me to the source of this thing in something web browsable? [23:35] uno momento, por favor [23:36] https://launchpad.net/~slavender/+archive/lucid this is my build in my ppa for it....you can see the buildlog for the warning [23:37] https://launchpad.net/~slavender/+archive/lucid/+build/1383726 perphas this is a quicker, more direct link [23:41] ScottL_: Ah, ok. So it's a traditional autoconf build, rather than a Python distutils one. [23:41] The problem is that its paths are wrong. Where they say site-packages, they need to say dist-packages. [23:41] Find a command-line parameter or a way of patching the package to make that the case [23:47] maxb, thank you, I will dig into it for a while but I will probably end back up here tomorrow :) [23:49] I am imagining that I can simply edit his configure files to replace 'site-packages' with 'dist-packages', no? [23:49] Hard to say without having looked at where it gets those paths from [23:49] then find the Ubuntu wiki/help patch page to help me create a patch and send it upstream? [23:50] The patch will not be suitable for upstream. The dist-packages thing is unique to Debian/Ubuntu [23:52] hmmm, learned something else [23:52] well, later tonight I'll try to replace 'site' with 'dist' and see what happens then, thank you immensely for your help