[00:01] * Hobbsee blecks at that 'diff' containing the entire source code and debian/ of the project. [00:03] I've been reading https://wiki.ubuntu.com/PackagingGuide/Recipes/PackageUpdate to learn how to update a package. After I update the package do I make a bug on launchpad? The howto doesn't say. Should I subsribe the universe sponsors? [00:03] hello, I am currently trying to get some software I wrote into ubuntu. Is it important to use make for compiling? Currently it compiles using a script which runs a few g++ commands (its a fairly simple compile), but if I want it packaged for ubuntu should I switch to make? [00:04] I was directed to ask that question here by someone in #ubuntu, btw. [00:05] shenron: It's not important to use make at all; all that's required is that it builds from source, and this build can be automated. [00:05] oh ok, so since there is a script that builds it that should be fine? [00:05] jsmidt: Yes, you make a bug on launchpad, attaching the new diff.gz [00:05] shenron: Right. [00:06] w00t, awesome. should I include an installer which will put something in /usr/bin as well? [00:06] raof, thanks. [00:06] raof, do I need to kink to the new orig tarball? [00:07] link [00:07] shenron: That's generally nice for other distros, but is not absolutely necessary. [00:07] jsmidt: No; have a watch file that works ;) [00:07] alright, thanks raof [00:23] maxb: thx. shot myself in the foot with that one hehe [00:23] * porthose goes and fixes === _neversfelde is now known as neversfelde [00:34] porthose: see comments on REVU, sorry, I've given you a lot to think about :-) [00:43] I have a debian rules file with a slice command. What is that? Pbuilder is complaining slice: Command not found. [00:45] jsmidt: Try 'dpkg -p slice' [00:45] Though clearly the package is wrong if it uses a command and doesn't Build-Depend on it [00:48] maxb, it does build depend on slice. Yet, right off the bat I get this error [00:50] hrm, peculiar [00:50] that doesn't make any sense :-/ [00:50] ya [00:50] do you want to pastebin a buildlog? [00:52] maxb, http://pastebin.com/d3b659090 [00:53] Ohhhh [00:53] See line 14 [00:53] ya [00:53] What warning is that? [00:54] unmet build-deps [00:54] I thought pbuilder took care of adding build-depends [00:55] pbuilder has two modes, in the mode you're running it in, it builds a source package *outside* the chroot, copies(?) the source into the chroot environment and builds it [00:56] How do I run the other mode? I typed "pdebuild " which has seemed to work with everything else. [00:57] --use-pdebuild-internal. The man page warns "This option will not protect the working directory and its parent directories from the build scripts." I get why it doesn't protect the working dir, since it's being bindmounted into the chroot. I don't understand how the parent dirs would be visible, though [00:58] unless you can actuall ../../ up out of a bindmount!? [00:58] weird [01:00] jsmidt: The problem is that slice is required for the clean target; IE: you need slice to build the _source_ package. [01:00] okay [01:24] does anyone here have experience making a program link with xulrunner? [01:27] hyperair: asac does. [01:31] asac: ping [01:31] raof: thanks [02:53] Is there any documentation for interacting with Lanuchpad over the command line? Open/closing bugs, adding attachments, etc... Do I have to use the web interface? [02:53] You can use gpg-signed mail. [02:54] Okay, is there documantaion for how to open a bug using mail for example? [02:56] Yes. There's a page on launchpad, or you can email help@bugs.launchpad.net, I think, to get an automated thingy. [02:58] raof, thanks, I think I have found it. [03:06] Do I need to write patch description anywhere? Changelog? [03:10] You do need to write a patch description _somewhere_. If you're using dpatch, I'd stick it in the dpatch header. [03:10] You also need to describe what the patch _does_ in changelog. [03:10] raof: simple-patch-sys? [03:10] raof: patch comments on top + changelog? [03:11] That should be OK, yeah. [03:11] raof: is it required for initial releases? [03:12] Is what required for initial releases? [03:12] entries in changelog regarding patches [03:14] hyperair: Yes. All patches should be documented. [03:14] ScottK: even if there's a header? [03:14] hyperair: Yes. [03:15] i see [03:15] on a side note, if i've got a package which requires *-gtkmozembed, which package should i b-d on? [03:16] i tried using libxul-dev, but it seems that it's an older version of xulrunner [03:16] xulrunner-1.9-dev refuses to be compiled without -R in LDFLAGS [03:20] xulrunner-1.9-dev is, I believe, the right header. [03:20] s/header/package. [03:31] OT: Are there any doxygen-literate people here? I'm faced with a problem and I don't know where to ask it. [03:32] I've been asked to fix up some PHP files (ugh) and I'm doxygenizing them now. For *one* of the files, a completely unremarkable file, it is refusing to document the functions (but not the @file). [03:34] (The line endings are the same, BTW.) [03:39] raof: but the libraries are all in /usr/lib/xulrunner-1.9.0.5/, and ld can't find them to link without -R. if i chrpath -d the executable after compiling, ld can't find the library to link with [03:39] hyperair: That's right. You either need a wrapper that sets LD_LIBRARY_PATH, or you need a RPATH. [03:40] raof: i'd leave the RPATH intact, but it seems that debian policy doesn't allow it [03:40] If anyone feels like taking a look, the file is here: http://gunkies.org/stuff/problematic_php.txt [03:40] hyperair: There's a _reason_ the world + their dog is moving to embedding webkit rather than gecko. Gecko is a pain in the arse. [03:40] It sees the function, but not my comments. [03:42] raof: you don't really have a choice if the package only has support for embedding gecko [03:42] well there's the choice of not embedding it at all [03:42] i mean not packaging it at all [03:43] Right. [03:44] * hyperair grumbles [03:44] i thought patching the thing to use gtksourceview2 instead of 1 was the end of my problems [03:44] at least it compiles with libxul-dev [03:44] and runs. just that you get the old chiselblock interface that reminds you of firefox 2 [04:33] Hi. When fixing a FTBFS, should I also fix warnings in lintian? Or I should just patch the FTBFS? It's for openmovieeditor in jaunty [04:35] (I'm getting warning/errors on watch file, out-of-date standard 3.7.2 and build-depends-on-obsolete-package [04:35] ) [04:36] fabrice_sp: Standards version you should not change at all. [04:36] There is Ubuntu Policy on that. [04:36] Fixing the watch file and build-dep-on-obsolete package (make sure it works with the new one) are both good. [04:36] Those are also changes that should go back to Debian. [04:37] ok (I didn't remember is Policy apply to development version). As soon as I've got a working package, I'll report the FTBFS to debian [04:37] thanks [05:20] Hello, I would like a MOTU to perform a merge for Jaunty [05:23] How should I format a copyright file if none of the headers have copyright attributions? [05:24] stochastic: If its obvious that they're all under the same license as found in COPYING or whatever, then just as if they all had the same copyright header. [05:24] stochastic: If it's not obvious what license they're under, then you get to play "poke upstream until Ubuntu can redistribute". [05:25] raof, it's clear it's GPL, but no years are mentioned [05:25] diwic1, is there a bug in launchpad? [05:26] fabrice_sp: #317254 and duplicates [05:26] bug #317254 [05:26] Launchpad bug 317254 in audacity "Please sync audacity 1.3.6-2 (universe) from Debian unstable (main)." [Wishlist,Confirmed] https://launchpad.net/bugs/317254 [05:26] stochastic: Then, I guess, you don't have copyright years in debian/copyright. But I'm not entirely sure. [05:27] <_16aR_> Hello, any admin of revu there ? [05:28] diwic1: it's enqueued for archive admin processing; please be patient. [05:31] raof, I'm updating libphat0 (and it's dev counterpart) to the upstream version, and neither have anything in the copyright field of the debian/copyright file but lintian gives me warnings about this, should I just leave it without (it came from upstream that way) [05:31] dtchen: Okay, thanks. Is there a way I can find status or know more about this "archive admin processing"? [05:32] stochastic: If you don't have date information, I don't see what else you can do; you might want to get other opinions, though. Also, libraries are a bit more complex; have you read the library packaging guide and done all the necessary ABI checks etc? [05:32] diwic1: one of the archive admins will look at the bug in due time. i'm not an archive admin, so i don't have an ETA. [05:33] raof, I'll take a look, I'm new to packaging, so I'm learning as I go... === hyperair1 is now known as hyperair [05:35] stochastic: You'll want to read at least the library packaging guide http://www.netfort.gr.jp/~dancer/column/libpkg-guide/libpkg-guide.html [05:36] raof, thanks === hyperair1 is now known as hyperair [05:41] Any REVU people about? I'm troubleshooting the upload of hexdiff for _16aR_ (lp:dolanor). The GPG key in use changed recently, and the package is repeatedly rejected, although the signature matches the key in LP. [05:41] Anyone know how to force key resync? [05:41] wait 24 hours? [05:41] wait for a REVU admin to come along? [05:43] coppro, I am a REVU admin. I just haven't tried a resync since the LP integration. Do you know that 24 hours fixes it? I thought that might have changed with the LP integration. [05:43] oh. [05:43] in that case, no clue, since I do not have resync access. But IIRC, REVU resyncs automatically every 24 hours [05:44] It certainly used to be that way, then it was stopped, and we did it manually roughly once a day. I don't know the current state. [05:44] nhandler, ? [05:45] jpds: ^ [05:51] speaking of revus could someone review my package http://revu.ubuntuwire.com/details.py?package=sigx? [05:53] no, MOTUs don't review your packages; they just let them simmer and hope they get better :P [05:54] T_T [05:57] well i can't think of anything else i need to do, so i'd either like guidance or an advocate === hyperair1 is now known as hyperair === hggdh is now known as hggdh|away [06:12] Is there a website that keeps track of packages in debian and not Ubuntu? [06:13] And outdated packages in Ubuntu versus upstream? [06:13] good morning [06:14] jsmidt: We've got an Ubuntu external health thingy. [06:15] raof, found it, thanks === hyperair1 is now known as hyperair [06:19] was there anybody in the Italian "Getting Started" session yesterday and could post the log at https://wiki.ubuntu.com/MeetingLogs/devweek0901/GettingStartedIT ? [06:25] jsmidt, You've found UEHS, or mdt, or both? [06:27] I found http://qa.ubuntuwire.com/ which seems to have everything [06:27] hey warp10 [06:27] warp10: do you have the log of the Italian Getting Started session? :) [06:28] morning dholbach! [06:28] warp10: could you add it to https://wiki.ubuntu.com/MeetingLogs/devweek0901/GettingStartedIT if you have it? :) [06:28] dholbach: I have about one hour and half of the whole log, my network had troubles in the remaining half hour [06:28] warp10: no worries, I can ask quadrispro or gaspa once they turn up :) [06:29] warp10: how did the session go on the Italian end? [06:29] dholbach: anyway, one of the attendant logged the whole sessione, he should be online soon [06:29] warp10: super, thanks! :) [06:30] dholbach: very well, IMO. We had 4 or 5 people attending the session, and we covered the whole program: LP, GPG, setting up the tools, some basic information and a lot of questions! :) [06:30] so we have a bunch of new Italian MOTUs soon? ;-) [06:30] dholbach: I really hope so! :D [06:31] woohoo :) [06:31] * dholbach hugs warp10 [06:31] * dholbach is really happy with how UDW is kicking off this time [06:31] * warp10 hugs back dholbach [06:31] OT: Are there any doxygen-literate people here? I'm faced with a problem that looks a lot like a bug and I don't know where to ask it. [06:32] dholbach: what about the other LoCo sessions? Lot of people attending everywhere? [06:32] Well everyone, thanks for the help. I learned a lot today. [06:32] I was able to work on 10 bugs. I've never been able to do that many. [06:32] warp10: I think we had like 25 people in the German channel - what I liked very much was that they were all asking lots of questions in the sessions afterwards, so I'd say they found their way in (during the getting started session) quite easily [06:33] But I've tried a lot of new stuff so sorry in advance. :) [06:33] jsmidt: Keep up the good work! :-) [06:33] Try not to send to many angry emails. [06:33] Again, thanks for the help. [06:33] toresbe: I used doxygen many years ago - I'm not sure I can help, but can take a look [06:33] dholbach: thanks. [06:34] I've been asked to fix up some PHP files (ugh) and I'm doxygenizing them now. For *one* of the files, a completely unremarkable file, it is refusing to document the functions (but not the @file). [06:35] dholbach: oh, a *lot* of people! well, great! I hope we will keep on with these MOTU-Loco activities. The italian development team will be happy to partecipate [06:35] toresbe: do you have that file, your Doxyfile and the log somewhere? [06:35] warp10: that's exactly what I wanted to hear! :) [06:35] dholbach: The log is uneventful; it parses the file no problem. [06:35] :D [06:35] http://gunkies.org/stuff/problematic_php.txt [06:35] that's the PHP file. [06:35] warp10: let's try to plan a developer-loco-something day in a few weeks [06:37] dholbach: http://gunkies.org/stuff/problematic_php_doxy.txt [06:38] dholbach: Oops, ignore the colourful comment at the top of the code, please. :) [06:38] dholbach: cool! Feel free to ping me if you need help on the italian side :) [06:39] warp10: super! :) [06:40] <_ruben> ok .. this is weird .. im trying to backport bind 9.5 to hardy .. the ./configure command fails when run by pbuilder, but the same ./configure command runs just fine when run in the shell that i got via a pbuilder hook .. guess next step would be to see how debian/rules would go about things [06:40] toresbe: does it document other files in the directory? does it stop somewhere? is there NO output at all about that file? [06:42] dholbach: It documents other files no problem. It runs through the entire thing. The file is documented, but the *functions* are not. [06:42] dholbach: subsequent @warnings in the same file *are* included; so it doesn't stop parsing. It just doesn't see @fns for *that* file. [06:45] toresbe: very weird... I'm sorry I'm not of much help - it's been ages since I used it. [06:45] toresbe: no idea [06:46] dholbach: It's driving me crazy, documenting this code is something that I was supposed to be done with yesterday. [06:47] toresbe: do the doxygen folks have an IRC channel? [06:47] dholbach: Nope. :( === RAOF is now known as raof [06:47] That's why I'm asking in random channels wherein I expect to find clever people :) [06:47] toresbe: I'd try there mailing list then [06:48] dholbach: Sigh, I haven't got many good experiences with those. [06:48] <_ruben> hmm .. even debian/rules build seems to work just fine .. must be some odd $ENV thingie messing stuff around .. oh well .. lets wait for the results first [07:19] toresbe, Check your autotools configuration. [07:39] persia: How could that change the behavior of an already-compiled program? [07:40] toresbe, It oughtn't: it's just one of the places I tend to look when ./configure doesn't do what I'm expecting. [07:41] Since ./configure is often run at package-time or build-time, I'm probably confused. [08:18] Hi everyone [08:19] I try to use start-stop-daemon in an init script, with -m option (to create the pidfile), but it does not write the right pid. I think the process is forking as soon as it starts (cf http://paste.ubuntu.com/107252/) [08:19] consequently, my init script fails :/ [08:24] where can I find documentation about .update-notifier files, the ones that go in /var/lib/update-notifier/user.d/ ? [08:24] I can pgrep to erase the wrong pid, but well... I would prefer another solution [08:24] didrocks: does the program not create its own PID? [08:25] stefanlsd: unfortunately not. That's why I use the -m option [08:25] didrocks: is this about an init script/daemon? [08:25] didrocks: yeah, it prob forks after starting. have u tried the -b option [08:26] savvas: yes [08:26] stefanlsd: not sure, let me check the manpage [08:26] stefanlsd: yes, I used the long form. And same pid retrieved [08:28] didrocks: mm. yeah. -m -b should hopefully do it. i can see its not thou. -v help u anything [08:30] stefanlsd: unfortunatly yeah, it seems that there is no solution with start-stop-daemon option. That's why I supposed to use pgrep to create it in the init script. Does this make sense? [08:31] didrocks: yeah. not sure why start-stop-daemon isnt doing it properly, maybe we could investigate that more, but pgrepping and writing the PID and telling start-stop-daemon the PID file should be ok [08:33] stefanlsd: ok, I will try to add more options (and to be in contact with upstream to see if he can implement it). If not possible, I will pgrep though :/ [08:33] thanks :) [08:33] didrocks: suuure. good luck :) [08:33] didrocks: you're trying to "daemonize" a bash script, right? [08:34] savvas: not a bash script, it's a C program [08:34] dholbach: I found the problem, FWIW. [08:34] didrocks: I have something here I use for a bash, maybe it'll work for you: http://bazaar.launchpad.net/~timekpr-maintainers/timekpr/trunk/annotate/head%3A/debian/timekpr.init [08:35] savvas: thanks. I will have a look at it :) [08:36] I use -m -b and --startas (no idea why, but it made it work) :P [08:36] savvas: why --test and then, execute it. Is it mandatory? [08:38] savvas: ok, I see it with the man page [08:39] thanks, will give it a try [08:39] didrocks: --test tests if the file is present [08:39] there was an example init script somewhere in /etc/ I think, but I can't find it :\ [08:39] savvas: no pb, I will work from your example and see if I can use it [08:40] ok great :) [08:41] toresbe: great - what was it? [08:44] dholbach: I had functions with the same name in previously-interpreted files. All info was added to those files, and not to the file the function actually was in. [08:44] ahhhhhh [08:44] toresbe: that was hard to spot [08:44] maybe file a bug report at doxygen? would be nice to have a warning for cases like that :) [08:45] I'll do so [08:46] super [09:09] gaspa: do you still have the log of the Italian Getting Started session? if you do, could you please post it here: https://wiki.ubuntu.com/MeetingLogs/devweek0901/GettingStartedIT ? [09:10] dholbach: i was just waiting for them, in order to put on the page. [09:11] gaspa: great, thanks :) [09:17] morning ! [09:31] dholbach: done. should I remove all time references, login/logout notifications and so on? [09:31] gaspa: no, it's fine as it is [09:31] thanks a bunch [09:31] ;) [09:32] it was amazing [09:32] it was [09:32] great work everybody :) [09:32] and now I need an oculist, after an hour of writing without breaks... :P [10:02] oin #ubuntu-at [10:13] Koon: I am trying to package the ActiveMQ message broker which is Java software .. what is a good existing Java software package to look at? [10:15] al-maisan: I might be biased :) [10:15] no problem :) [10:16] al-maisan: I like to think tomcat6 was packaged correctly, but others might think otherwise. [10:16] Thanks! [10:16] al-maisan: ActiveMQ is a library rather than a server, right ? [10:16] I am a beginner and need an example to look at. [10:16] ActiveMQ is a server actually [10:17] It is a message broker [10:17] a JMS implementation [10:17] al-maisan: it runs standalone ? It should be started at the end of the package installation ? [10:17] Koon: yes, ideally. [10:18] al-maisan: then I'd recommend looking at tomcat6, yes. Libraries are easier, and the commons-* are good examples for that [10:18] Koon: good point, very much appreciated :) [10:18] al-maisan: also you might find #ubuntu-java helpful. [10:18] ah, OK. [10:18] pfft, java [10:18] ;) [10:18] didn't even know such a channel existed :) [10:18] directhex: everyone has a cross to carry :) [10:19] 'one cross each' [10:19] * al-maisan is not a big Java fan but that specific piece of software was written in Java [10:19] carry cross once, get nailed anywhere? [10:19] :P [10:26] Koon: Welcome back. :-) [10:27] slytherin: heh thx === fabo_ is now known as fabo [10:42] Could someone review my package: http://revu.ubuntuwire.com/details.py?package=webstrict, I've added a debian/watch file as required [10:50] <_ruben> hm .. when building a package using pbuilder, how does one ensure the resulting .deb is signed with my key ? [10:56] _ruben: You mean .changes, right? .debs are never signed directly [10:56] <_ruben> maxb: then what does apt-get check when downloading the debs from a mirror? [10:57] Checksums for the .debs are listed in the Packages file. [10:57] Checksums for the Packages file are listed in the Release file [10:57] The Release file is signed [10:58] <_ruben> ah .. that should be enough to get me going for a bit .. thanks [10:58] _ruben: the .changes file (the upload control file) contains checksums for the .debs, and the .changes file is signed by the uploader to authenticate themselves to the archive they are uploading to [10:59] ScottK: you were right about the sponsoring entries in the HoF (uploads) - thanks again [11:00] morning dholbach [11:00] (or should I say evening?) ;) [11:00] hiya pochu [11:00] dholbach: I uploaded the logs right after you left [11:00] no :) [11:00] pochu: thanks a lot [11:00] yw [11:00] <_ruben> maxb: ah .. so the .changes files would be signed by uploader, and the Release file by the repository ('s maintainer's special key) [11:01] In practice, usually the repository's automatic signing key, rather than any person-associated key, but yes [11:01] <_ruben> maxb: that's what i realised during typing, hence the odd sentence ;) [11:02] <_ruben> thanks .. time to re-do some stuff properly now :) [11:03] Incidentally, you don't have to sign .changes and source packages at time of build. You can just run "debsign" on them later (e.g. after you've tested them - or, even, on a different host to where you build them, if your key can't/shouldn't be on the build host) === azeem_ is now known as azeem [11:18] I'm looking at Twisted right now; there's been a new upstream release that's apparently not packaged yet in the archives. I'm very new to this, but given the state of the debian release/etc, would it be helpful to try and update the packaging in Ubuntu? In what form/where would I send the work when it's done, given that (old version of) the package is currently coming unmodified from Debian? [11:18] WalterMundt: doko_ maintains it, you should talk to him [11:19] okay, good plan; I don't want to duplicate effort [11:20] (more generally, I wonder how much interest the various Debian maintainers have in running updates through Ubuntu in the interim while the Debian release process does its thing.) [11:20] Debian has experimental available for use [11:21] hmm, true [11:38] how can I get at the config.log file from a pbuilder environment? [11:38] stochastic: you probably need to suspend it and chroot into it or something [11:39] I don't think you can get to it after the build [11:39] There's a hook you can set up to remain in the environment if a build fails [11:39] I don't have it, though [11:39] maybe CDBS should cat config.log if configure fails === hyperair1 is now known as hyperair [11:45] Does anyone know where can I find documentation about .update-notifier files, the ones that go in /var/lib/update-notifier/user.d/ ? [11:45] There is some doc in the update-notifier source package, at least. [11:54] stochastic: https://wiki.ubuntu.com/PbuilderHowto mentions the "keeping shell in pbuilder after fail" hook thing btw [11:54] Laney: e-mailed doko_; may be back later depending on response [11:54] excellent, good luck [11:56] hm... [11:56] I found some in kubuntu wiki https://wiki.kubuntu.org/InteractiveUpgradeHooks === hyperair1 is now known as hyperair [12:06] persia: Did you need somethign? [12:06] s/somethign/something/ [12:26] nhandler, _16ar_ changed the GPG key in LP recently, and it differs from the one in the LP database. I don't know the command to run on spooky to fix this. Do you? [12:28] Err, ...differs from the one in the REVU database ... [12:33] persia: Sorry, I don't. I don't even have access to spooky yet [12:34] https://launchpad.net/ubuntu/jaunty/+source/kde4bindings/4:4.1.96-0ubuntu1/+files/kde4bindings_4.1.96-0ubuntu1.dsc [12:34] bah [12:34] nhandler, You've sent a request to UWSA? [12:35] nhandler, Also, become a REVU Admin or REVU Hacker at your option: you don't really need to be either to be REVU Coordinator, as long as you get those of us who do have those roles (and Reviewer and Contributor) doing stuff. [13:21] http://primates.ximian.com/~miguel/pictures/moon-obama.png [13:21] bah [13:21] shameless plug: http://ubuntuforums.org/showthread.php?p=6584115 [13:25] How do I disable the executable flag in a image which executable in upstream? [13:28] MaduserTr, do you mean that the .jpg (or whatever format it is) ends up with execute permissions in the .deb? normally debhelper normalizes permissions, but if not, you can add a command to debian/rules to change the permissions yourself [13:31] liw, Yes this I mean. I use cdbs on a python program (wine-doors). Can I hook in the build to do so? [13:32] persia: Have you relogged into REVU with the new key on LP? That's how keys get synced in. [13:32] MaduserTr: install/binary-package-name:: [13:32] MaduserTr: chmod 0644 debian//path/to/jpg [13:32] jpds, the user did relogin, but it didn't resync. I had them log out and log in again, and forced the rejected .changes back into /srv/uploads [13:33] The rejected .changes was signed by the key shown as the right key on LP. [13:33] persia: Oh, did it work after that? [13:33] No. [13:33] hyperair, thanks I will try to get it [13:33] MaduserTr: if you've got cdbs installed on your comp, read the cdbs doc here: file:///usr/share/doc/cdbs/cdbs-doc.html [13:33] That's why I'm asking for guidance: the last docs I have on being a REVU admin are out of date, and so I don't know how to fix this. [13:35] Hmm, no idea either. [13:35] persia, you have docs?? [13:35] persia: I'd like to see those, even though they are out-of-date [13:35] mok0, Yep. sistpoty sent some out to ubuntu-motu about a year ago. [13:36] persia: I'll search the archive then [13:37] mok0, March 2008 [13:37] persia: thx [13:38] found it [13:39] shoudn't we have that on the wiki? [13:39] Well, no. We should probably have a section on the wiki that's accurate. [13:39] At least I'd find it helpful, but I think we need help from REVU Hackers to do it. [13:39] persia: err yes, that's what I mean :-) [13:40] In that case, Yes :) [13:41] * hyperair wonders if anybody is willing to revu some packages [13:42] sorry hyperair, trying to get some work done... [13:42] okay [13:42] no prob [13:42] i'll wait another day =p [13:43] I am going early to watch the inauguration of Obama :-) [13:44] =O [13:44] in 75 minutes [13:49] =O === LucidFox_ is now known as LucidFox [13:50] well before you go, do you know what's required to make a package that links fine with libxul-dev to link fine with xulrunner-1.9-dev? it uses gtkmozembed [13:50] -R works, but using an rpath is forbidden it seems [13:51] and using the XPCOM glue causes ld to throw up a hissy fit about undefined references [13:51] hyperair: I guess you will need to patch makefile. [13:52] slytherin: i've done lots. patch Makefile.am, gecko.m4, among other things [13:52] hyperair, Have you asked the Mozilla team? [13:52] slytherin: and configure.ac [13:52] eh no? [13:52] where can i contact them? [13:52] hyperair, liw, thanks got it [13:53] MaduserTr, you're welcome [13:53] MaduserTr: you're welcome === mib_65qsfnbw is now known as shankhs [13:53] MaduserTr, we thank you for choosing #ubuntu-motu for your development questions and hope that you will choose us again :) [13:53] (do I fly too much?) [13:54] hyperair: https://edge.launchpad.net/~mozillateam [13:54] :) [13:54] hyperair: #ubuntu-mozillateam [13:54] liw: maybe :P [13:55] otherwise we do/get the same good service as we should :) [13:55] james_w, slytherin: thanks [13:56] How can I download the source code of geany http://bazaar.launchpad.net/~vcs-imports/geany/trunk/changes contains a lot of files which gonna take a lot of time to download,moreover I dont know which files to download which not.I am a newbie in development. [13:56] Please help [13:56] shankhs: bzr branch lp:geany [13:57] Laney: thats not working giving some erroe like:bzr: ERROR: socket.error: (111, 'Connection refused') and asking to file a bug report [13:57] maybe something's wrong with LP [13:57] You can get it from geany's site too [13:57] shankhs: are you behind some firewall of some sort? [13:57] I am behind a proxy which needs authetincation [13:58] aha [13:58] what should I do? [13:58] Does it not need to be lp:~vcs-imports/geany/trunk ? [13:58] maxb: I copied it from LP [13:58] Laney: "bzr branch lp:geany" ... that's the coolest reply I've seen in a while :-) [13:58] \o/ [13:59] maxb: ya its that /geany/trunk [13:59] then how to make "bzr branch lp:geany work? [14:00] look into how to use bzr behind a proxy [14:00] python is broken behind proxies [14:00] bzr branch lp:geany/trunk [14:00] use the full url instead [14:01] "bzr branch bzr+ssh://youruser@bazaar.launchpad.net/~vcs-imports/geany/trunk" [14:02] proxies... an invention of nazi sysadms [14:02] mok0: I disagree [14:03] btw is there any security concerns with bazaar? [14:03] like what? [14:04] I guess proxies are needed to protect machines running Microsoft Windows... [14:04] shankhs, not in particular: every program that accesses data from the network is always a security concern, if you're paranoid enough, but bzr doesn't do anything particularly worrying [14:04] I am repeatedly getting bzr: ERROR: Connection closed: please check connectivity and permissions (and try -Dhpss if further diagnosis is required) [14:05] liw: I got it [14:05] shankhs: try using tsocks [14:05] hyperair: and now what is that tsocks??? [14:05] shankhs: can you "ssh youruser@bazaar.launchpad.net"? [14:06] james_w: no one can [14:06] shankhs: it's a wrapper, using LD_PRELOAD or something to wrap all outbound connections through a SOCKS proxy [14:06] mok0: sure you can [14:07] james_w: no its saying: Permission denied (publickey). [14:07] james_w: well I don't get a connection [14:07] it doesn't give you a shell, but it's a connectivity/key etc. test [14:07] shankhs: ah, have you registered an ssh key with launchpad? [14:07] :( no [14:08]