/srv/irclogs.ubuntu.com/2010/10/15/#ubuntu-motu.txt

TheMusoNatty is open.00:15
micahg\o/00:16
=== persia changed the topic of #ubuntu-motu to: Maverick is released. Lots of SRUs to be done. | Natty is open | Want to get involved with the MOTU? https://wiki.ubuntu.com/MOTU/Contributing | Sponsor queue: http://is.gd/2y76G | http://qa.ubuntuwire.com/ftbfs | http://people.canonical.com/~ubuntu-archive/NBS | http://qa.ubuntuwire.com/bugs/rcbugs/ | Congratulations to new MOTU: debfx.
highvoltageTheMuso: whohoo!01:08
coolbhavihello...isnt PPA for natty activated yet?04:21
persiaQuite possibly not, but ask #launchpad: they host PPAs (and we don't even all use them).04:23
persiaYou can set up a pbuilder or sbuild today though.04:23
coolbhavisure thanks persia04:23
nigelbhyperair: HAPPY BIRTHDAY!05:06
hyperairthanks =)05:06
hyperairhow did you know?05:06
nigelbThe Force :p05:06
nigelbFacebook, you added me remember? ;)05:07
hyperairahahahahaha05:07
hyperairright05:07
hyperairnow i remember05:07
* hyperair got flooded with facebook mails this morning05:07
* nigelb adds to the flood05:08
* hyperair gets flattened by facebook mails05:09
nigelbhahaha05:11
micahglucidfox: ping06:52
=== ivoks-afk is now known as ivoks
lucidfoxmicahg, pong07:26
micahglucidfox: hi, did you want to merge chmsee?  I noticed you were the last uploader07:26
lucidfoxEh... to be honest, I'm done with it. If you want to upload, feel free to proceed07:32
micahglucidfox: k, thanks07:33
persiaAnd this is why we don't have maintainers in Ubuntu : makes it easier to just fix stuff with certain confidence that someone else on the team will pick it up later if you're done with it.07:33
micahgpersia: we're still supposed to ask, right?07:34
persiaDepends.07:34
bilalakhtarmicahg: I have uploaded chmsee!07:34
bilalakhtarmicahg: Please don't!07:34
persiaIf you've a burning need to do something, because it's blocking your work, same rules apply for merges and non-merge uploads.07:34
micahgbilalakhtar: ?07:34
bilalakhtarmicahg: no, wait, I haven't uploaded it yet07:35
bilalakhtarmicahg: Its lying in my directory07:35
persiaOnce we hit DIF, anything not done is essentially up for grabs.07:35
micahgbilalakhtar: ah, you grabbed it...that's one of the few things I can upload at the moment :(07:35
persiaWe tend to follow TIL rules between archive-open and DIF just because most folk who are TIL are willing to do the merges.07:35
bilalakhtarmicahg: Do you want to do it? I have done the merge, but will do a test-build, run and upload07:36
micahgbilalakhtar: nah, go ahead07:36
persiabilalakhtar, As much as I harp on Ubuntu-does-not-have-maintainers-!!!! It's considered polite to ask the last uploader about it between archive-open and DIF: lots of times people have other out-of-archive stuff they are doing, and quick merges interfere with their work.07:36
persiaOnce we reach DIF, all bets are off.07:36
bilalakhtarohk07:37
* micahg would also think MOTU should focus on unseeded unless it's a pet package07:37
persiaIndeed, but the LP stuff to block MOTU uploading to packagesets isn't done yet, sadly.07:38
micahgpersia: I don't think they should be blocked necessarily, just focus on the unseeded07:39
persiaAnd the UI for MOTU to *know* what isn't in a packageset is still fuzzy.07:39
micahgyeah, that's the bigger issue :)(07:39
persiamicahg, Intent is to block it: packages that have interested folks aren't MOTU responsibility.  MOTU who happen to be interested should join the relevant devteam.07:39
micahgpersia: ah, ok, so me going for MOTU will then be in addition to mozilla stuff :)07:40
persiaIdeally, MOTU grants no upload rights, but that takes a lot more active developers :)07:40
persiamicahg, Right.  MOTU is an additional grant, and it's supposed to be about dedication to archive quality.07:40
persiaAnd for work concentrated in areas that don't have any group caring for them.07:41
bilalakhtarI am interested in desktop packages, some of which are in universe. However, before touching a desktop package, I ask in #ubuntu-desktop07:41
persiaWhen you say "desktop package", how are you defining that?07:43
micahgis the current preference to file merge bugs or not to file merge bugs?07:45
persiamicahg, I'm still a fan of filing merge bugs, just as a means to declare who is merging it.07:46
Rhondamicahg: It's to check wether the difference still makes sense and rather to help Debian release, otherwise natty won't get any new packages.07:46
persiaThat said, LP bugs are now so cluttered, I don't think most folk even look at bugs for a package when merging (sadly), making this less useful.07:47
persia(plus what Rhonda said)07:47
micahgRhonda: how do merge bugs help to see if the diff makes sense or help Debian?07:48
persiamicahg, The point being that it's probably better to focus on squeeze release than worry about merges.07:49
micahgpersia: ah, right, ok07:49
persiaThe chances that any merge done now will not have to be redone when squeeze comes out and everyone uploads is tiny.07:49
persiaSo the faster we get squeeze released, the sooner it makes sense to do any merges.07:49
Rhondamicahg: The other way round. If the diff doesn't make sense there is no need for a merge bug, rather sync request. :)07:49
micahghmm, makes sense, but I'm not sure how much I can do to help squeeze release, alas, I'll give it another whirl07:49
persiaA better focus is probably UEHS stuff, for folks who don't want to help Debian.07:50
micahgUEHS?07:50
micahgpersia: should I unmark the merges I marked for myself or just claim them and do them when the new version come out?07:50
persiamicahg, Your choice :)07:51
persia!uehs07:51
persia($#"W~)("#07:51
persiahttp://qa.ubuntuwire.com/uehs/07:51
RhondaI'd like to throw in this URL which gives a list of bugs that are in need of getting addressed: http://udd.debian.org/bugs.cgi?release=squeeze&patch=ign&pending=ign&security=ign&notmain=ign&notsqueeze=ign&merged=ign&done=ign&fnewerval=7&rc=107:51
=== ivoks is now known as ivoks-afk
persiaubottu, uehs is <reply> the Ubuntu External Health System is hosted at http://qa.ubuntuwire.com/uehs and lists many upstream projects with which Ubuntu is not currently in sync.07:52
micahgpersia: I didn't know about that one, thanks07:52
micahgpersia: I want to help squeeze release, I guess I need to dig deeper in the RC list to see if something is fixed upstream or something I can do07:53
dholbachgood morning! :)07:57
ajmitchhi07:58
dholbachhey ajmitch07:58
ajmitchyay for friday evening07:58
* ajmitch can sleep now07:58
nigelbdang, timezones are unfair07:59
nigelbanother 10 hours at least before I can sleep07:59
ajmitchwell it's only 8pm here, but I just walked home after a long day :)07:59
nigelbaha07:59
* micahg is going to sleep now, timezeones are interesting08:01
persianigelb, So, um, consider what time ajmitch gets up.08:03
nigelbpersia: Considering what time I got up today, makes not much of difference :/08:03
* micahg got up 19 hrs ago, definitely time to go to sleep08:04
* persia is amused when 7:30 of timezone difference makes little effective difference in diurnal cycles08:04
* ajmitch isn't really going to sleep yet :)08:05
bilalakhtarpersia: What's the time at your place?08:11
persiabilalakhtar, UTC+9 today08:11
bilalakhtaroh08:12
=== ara__ is now known as ara
ari-tczewwhen autosync is starting?10:22
persiaA few minutes ago.10:23
ari-tczewcan I see which packages are autosyncing right now? is it in  Latest uploads  ?10:33
persiahttps://launchpad.net/ubuntu/natty/+queue?queue_state=3 lets you see what just got processed.10:33
ari-tczewthanks persia10:34
ari-tczewpersia: hmm, syncs are not coming :/10:55
persiaSome will, but not so many, because of the squeeze freeze.10:56
ari-tczewpersia: I see that 1700+ packages should be synced.10:57
ari-tczew(autosynced)10:57
persiaAre you working on something that is blocked on one of them/10:58
persias:/:?:10:58
ari-tczewpersia: I would wait to moment, when all packages are synced. then I can test build packages on fresh build-depends11:10
persiaMost build-depends ought be done: that's the point of the toolchain freeze.11:11
ari-tczewaha11:11
persiaIt is believed that anything not already uploaded isn't supposed to affect how stuff builds significantl.11:12
persiaThis may not be a correct assumption, but it's safe for you to also assume it.11:12
ari-tczewpersia: ok, now packages are going to be uploaded by autosync.11:27
persiaYep.11:28
ari-tczewbuilders will be clogged11:28
persiaYep11:29
gusnanpdebuilding my GTK program I get the following: http://paste.ubuntu.com/513757/ during build - does someone know anything about this? It doesn't help to depend on libgdk-pixbuf2.0-dev oddly enough...11:54
gesergusnan: try adding -I/usr/include/gdk-pixbuf-2.0 to the CFLAGS13:02
gusnanah, that seems to work... Thanks geser!13:10
jariqWhen i run "dh_make -s" it generates rules file with just three lines (#!/usr/bin/make -f \ %: \ dh $@).. I am not sure if this can be used for my applicaiton :) How can i get the "old style behaviour" with normal rules file ?13:39
tumbleweedjariq: there's an easy way to find out if the minimal dh rules will work :)13:41
tumbleweedjariq: also (#ubuntu-packaging)13:41
jariqtumbleweed: minimal dh rules does not work for me.. how can I generate longer version ?13:41
tumbleweedjariq: don't know of an easier way than writing them out by hand. dh_make doesn't ship them any more.13:43
tumbleweedjariq: why don't they work?13:43
jariqtumbleweed: i am installing into non standard location13:45
jariqtumbleweed: your answer "dh_make doesn't ship them any more" is enough for me.. thank you very much13:46
tumbleweedjariq: you know about dh overrides, right?13:46
tumbleweedso much for that...13:46
ari-tczewhuh, builder is blocked till 2 days, at the moment14:14
persiaheh, lots of packages, and powerpc is unfortunately not very fast right now.14:19
persiaBe nice to put new powerpc hardware in the DC, but that kinda requires someone to make some and sell it at reasonable prices.14:19
AlanBellhow do I find out who should review this? https://code.launchpad.net/~alanbell/dasher/bugfix-lp-579181/+merge/3752814:24
AlanBellso I reported a bug, took a branch, made the fix, done a merge request . . . now what? How do I make it happen?14:24
vishAlanBell¦ set the sponsor team as the reviewer14:26
vishdasher is in Ubuntu right?14:26
* vish checks..14:26
AlanBellyes, it is in ubuntu14:27
AlanBellok, done that I think14:27
AlanBell!info dasher14:28
ubottudasher (source: dasher): A graphical predictive text input system. In component universe, is optional. Version 4.11-1 (maverick), package size 364 kB, installed size 984 kB14:28
vishAlanBell¦ hmm, wait, i think you might have fixed the wrong branch..  thats a vcs import branch14:28
AlanBellerr ok, not sure what that means14:28
vishAlanBell¦ https://code.launchpad.net/~vcs-imports/dasher/trunk , thats automatically imported14:29
vishAlanBell¦ if you want to fix only in Ubuntu, grab the Ubuntu branch14:29
AlanBellfound that from here https://launchpad.net/dasher14:30
* persia notes that bzr diff may produce a patch suitable for application into the Ubuntu branch14:30
AlanBellsorry, I am going round in circles here, where is the ubuntu branch?14:31
persiaUsually at lp:ubuntu/${package}14:32
vishAlanBell¦ https://code.launchpad.net/ubuntu/+source/dasher14:32
* persia thinks, but doesn't use bzr for packaging, so may be mistaken14:32
vishAlanBell¦ so its for lp:ubuntu/dasher , that the sponsors can upload14:32
vishalso, i think you'd need to file a bug too..14:33
persiaNo, it shouldn't need a bug.14:33
AlanBellhmm, ok, most of the links on that page take me back to the upstream dasher14:33
AlanBellthere is a bug14:33
persiaA merge proposal should show up in the sponsoring report.14:33
persiaIf there's a bug, just link the branch to the bug and subscribe ubuntu-sponsors.14:34
AlanBellbug 57918114:34
ubottuLaunchpad bug 579181 in dasher (Ubuntu) "Newline (¶) only selectable after punctuation or space" [Medium,Confirmed] https://launchpad.net/bugs/57918114:34
persiaYou'll still need to base the branch on the Ubuntu tree, but that'S separate.14:34
vishpersia¦ i meant a bug for mentioning what we are changing and for the changelog..14:34
persiavish, No point.  Best to use the bug that is the problem.14:35
persiaAdding new bugs just for sponsoring is a waste of bug IDs.14:35
vishnah, i dint mean new bugs :)14:35
persiaAnd we only have a limited number of those.14:35
persiaOK. :)14:35
AlanBellsubscribe sponsors to the bug, or the branch?14:36
vishAlanBell¦ branch14:37
vishAlanBell¦ if you have made the changes for lp:ubuntu/dasher , they can merge/upload14:37
vish lp:dasher they cant..14:37
AlanBellhmm, odd14:38
vishthink of lp:dasher as the upstream branch, its just a mirror now of the svn from gnome14:38
vishsponsors can upload upload to Ubuntu branches14:39
vishupload only* to14:39
persiaWell, sponsors can upload whatever they want to Ubuntu, whether it has any relation to an Ubuntu branch or not.14:40
persiaBut sponsors can only process merge proposals against Ubuntu branches.14:40
persiaand upstream patches tend to get sent upstream rather than sponsored.14:40
AlanBellok, so I need to bzr branch lp:ubuntu/dasher, redo my changes, commit, push and do a new merge request?14:40
vish yup14:41
vishAlanBell¦ would sending the patch upstream not be helpful?14:41
AlanBellwhat have I just done then?14:41
vishAlanBell¦ the branch will fix the bug in Ubuntu alone, [if sponsored]14:42
AlanBellisn't the issue that I *did* send it upstream?14:42
vishthe upstream code you pulled is actually mirrored from  http://svn.gnome.org/svn/dasher/trunk/14:43
persiaAlanBell, You based the patch on upstream code (good), but didn't send it to actual upstream (maybe needed, maybe not).  You're now working to reapplyy the patch (bzr diff is good to extract it) to the Ubuntu branch with the intent to request someone to upload it to Ubuntu.14:44
vishso i'm *guessing* they file their bugs in gnome bugzilla, not sure though14:44
AlanBellok, so where does that pull from?14:44
AlanBellhttp://svn.gnome.org/ says GNOME has changed to using Git for version control. Current GNOME sources can be found on git.gnome.org. All content on this site is obsolete14:45
vishfrom " https://code.launchpad.net/~vcs-imports/dasher/trunk   "This branch is an import of the Subversion  branch from http://svn.gnome.org/svn/dasher/trunk. "14:46
vishlast sync was 30mins ago14:47
vishimport rather14:47
AlanBellindeed, so why is it pulling from a site that claims to be obsolete14:47
vishright,  and the Ubuntu package is also a sync from debian.. 0.o14:48
vishauto-sync14:48
AlanBellhttp://www.inference.phy.cam.ac.uk/dasher/Develop.html14:52
AlanBellso looks like the real upstream is indeed on gnome14:52
AlanBellbut should be git clone git://git.gnome.org/dasher14:53
vishif the svn is obsolete then https://code.launchpad.net/~vcs-imports/dasher/trunk might be a waste ?  why does that keep trying there?  maybe we should close/stop that?14:53
vishpersia¦ ^14:53
AlanBellthey might be syncing git to svn and launchpad is pulling from that14:53
vishheh! musical chairs! ;p14:54
AlanBellbut debian will get releases from gnome, ubuntu gets packages from debian, I can't see where that lp:dasher is actually used14:54
AlanBellright now it is just causing multi-cycle delays in getting fixes done14:55
persiavish, better to just ask in #launchpad to get it updated14:56
vishyea..14:57
vishAlanBell¦ phew! » https://bugzilla.gnome.org/browse.cgi?product=dasher15:01
vishthe bugs seem to go there, and sending the patch there would probably nice as well, but seeing that the package has not been updated since 04/2009  not sure where the log is, debian or bugzilla15:02
persialog?15:03
persialag?15:03
persiaquite possibly related to the squeeze freeze.15:03
visherr, yeah, lag15:04
AlanBellok, so I need to register and file a bug in gnome or something?15:06
ScottKIf it's upstream Gnome, it won't be the squeeze freeze that's the cause of delay I wouldn't think.15:09
vishAlanBell¦ yup, file a bug in gnome and send the patch there,  also trying to find out why the package was not updated for a while might be the real clue ;), the latest version /might/ have the problem fixed..15:09
AlanBelldoubt it, given what the issue is15:12
AlanBellwhen using IRC to control dasher it is very hard to type /window 3<return>15:12
vishAlanBell¦ there is dasher 4_11 in git15:13
AlanBellbecause digit followed directly by return isn't in the training data15:13
vishhttp://git.gnome.org/browse/dasher/tag/?id=DASHER_4_11_015:13
vishhrm! but seems nothing new in it, just a welcome message :D15:15
AlanBellhttp://git.gnome.org/browse/dasher/log/ wow, lots of stuff happening15:16
vishnope, someone is working on it15:16
vishheh , yeah, was just looking at the same link15:17
AlanBellok, so the current ubuntu situation is a bit sucky15:18
AlanBellbugs get reported, they are not being passed upstream, it was *hard* to discover where the upstream actually was15:18
bilalakhtarA package is facing the DSO link failure problem and I am trying to fix it. The package uses cmake. Any idea how do I get it to use the -lX11 flag when it calls gcc ?15:19
AlanBelland I am fairly sure when I fixed that bug in the first place I used ground control to fix the bug number and it downloaded the vcs branch for me, totally inappropriately15:20
bilalakhtargot it15:20
vishAlanBell¦ yeah, just noticed that none of those bugs got triaged15:22
vishAlanBell¦ adopt upstream? < enter jcastro »15:22
jcastroI can help!15:23
jcastrogive me a few minutes to finish this call15:23
vishj-castro¦  https://bugs.launchpad.net/ubuntu/+source/dasher seems to be fairly important for accessibility, but has got no attention15:24
bilalakhtar| ?15:25
bilalakhtarWhy the pipe character, vish?15:25
vishbilalakhtar¦ just so that people ask about it! :)15:26
AlanBellvish: ¦ is not | or : but somewhere inbetween15:26
vishAlanBell¦ yup, i came across it while we were looking for the Rs sign ;)15:27
AlanBellah, you mean ₹15:27
bilalakhtarvish: Ubuntu supports the rupee symbol15:28
vishAlanBell¦ yup, the other day, so you are to be blamed! ;p15:28
AlanBell#blamealan15:28
=== ivoks-afk is now known as ivoks
jcastrovish: the bug tracker seems set correctly in launchpad to me?15:37
AlanBellhttps://bugs.launchpad.net/dasher says that bugs are tracked in gnome, and also ubuntu15:40
AlanBellso do the dasher packages come from debian, or from gnome?15:42
jcastrothe code comes from gnome, the packages from debian most likely15:43
AlanBellso where do I file a bug, and where do I fix it, and how should I have found this out for myself?15:46
tumbleweedAlanBell: you file a bug in Ubuntu if it's something we want to track in ubuntu (i.e. important or visible). You fix it in gnome. If we can't wait for the fix from gnome, we fix it with a patch in Ubuntu (or better, Debian) but still pass the patch to gnome15:48
AlanBellright, OK, so I did file a bug in Ubuntu, and then got in a mess trying to fix it in ubuntu against the pseudo upstream branch at lp:dasher15:49
tumbleweedAlanBell: https://wiki.ubuntu.com/Bugs/HowToFix covers some of this15:50
AlanBellnow I have the git tree from Gnome and I am fixing it there15:50
vishjcastro¦ yea, but AlanBell seems fairly interested in the package, so i thought you could sweet talk him into adoption it ;)15:50
vishadopting*15:50
jcastroindeed15:50
jcastroAlanBell: I would fix it in gnome, and then ping the DD and a MOTU about carrying it back down15:55
AlanBellyay, I have a git patch16:00
persiayay!16:00
jcastroAlanBell: please make a mental note of how brutal it will be getting the fix back down, I am always interested in that kind of data16:04
AlanBellgnome bug 63222616:06
ubottuGnome bug 632226 in core "Dasher is hard to use to operate IRC clients such as irssi" [Normal,Unconfirmed] http://bugzilla.gnome.org/show_bug.cgi?id=63222616:06
jcastro\o/16:07
AlanBellbug filed in gnome, ubuntu bug attached to gnome bug, patch attached to gnome bug16:08
PendulumAlanBell: I think I owe you a drink next time I see you :)16:16
vishPendulum¦ hey! AlanBell owes me that drink! i literally handheld him through it! ;p16:27
Pendulumvish: or I owe you both drinks :P16:36
vish\o/16:40
=== ivoks is now known as ivoks-afk
=== ivoks-afk is now known as ivoks
micahgSpamapS: you might want to wait for me to merge 1.6.3 before you request the backport18:46
micahgSpamapS: actually, since you TIL, were you planning on merging mongodb?18:47
SpamapSmicahg: oo that would be sweet actually. :)18:59
micahgSpamapS: which one?18:59
SpamapSmicahg: 1.6 in backports18:59
micahgSpamapS: k, do you want me to merge it, or do you want to?19:00
SpamapSmicahg: I don't mind merging it at all. I'm a little bit perplexed at what they're planning for 1.7 though.. mongodb is planning to embed pcre and boost in their distribution.19:00
SpamapSmicahg: and they're going to be statically linking by default.19:00
SpamapS"rapidly changing xulrunner and boost versions cause too many problems for our users"19:01
SpamapSThey really don't understand that they are cutting off their nose to spite their face.19:01
micahgSpamapS: k, well, they can switch to webkit :-/19:01
ScottKBoost and stable ABI are oxymorons.19:02
SpamapSmicahg: does webkit have a more stable js library?19:06
micahgSpamapS: er, I don't know for sure, but I think they do19:07
micahgchrisccoulson: can you speak to the stability of the webkit JS API?19:10
chrisccoulsonmicahg - i'm not sure about that19:19
SpamapSmicahg: Oh, and if I wasn't clear.. yes I will do the mongodb merge19:22
micahgSpamapS: feel free to ping me when you're ready for sponsorship19:23
SpamapSmicahg: will do, thanks. :)19:23
=== ivoks is now known as ivoks-afk
=== yofel_ is now known as yofel
ari-tczewbdrung: hey, which command can I use for fakesync?23:13
SpamapSmicahg: should I bother to subscribe ubuntu-sporsor for the mongodb merge, or do you just want to take it? bug #66151323:29
ubottuLaunchpad bug 661513 in mongodb (Ubuntu) "Please merge mongodb 1.6.3-1 (universe) from Debian unstable (main)" [Undecided,In progress] https://launchpad.net/bugs/66151323:29
micahgSpamapS: nah, that's ok, just subscribe me and I'll upload Sat night23:30
SpamapSseems like the 'subscribe someone else' link on launchpad has gotten slower lately :(23:31
SpamapSmicahg: thanks!23:31
micahgnm, I'm already subscribed :)23:31
micahgoh, well, all good23:32
micahgSpamapS: np23:32
micahgSpamapS: only comment on your debdiff is you fix up the dep3 patch headers for the ubuntu_hacks workaround23:44
* micahg will be back in 25 hours23:45
SpamapSmicahg: oh, those are the result of the 3.0 source format.. I always forget to fix those. ;)23:46
ScottKSpamapS: Here's a hint - Read your debdiff before you ask for sponsorship.  Then this kind of stuff leaps out at you.23:47
SpamapSI did read it. I don't ever notice those, because they're just boilerplate stuff. The 3.0 format does make it kind of weird. The Debian maintainer didn't remove their boilerplate stuff either. ;)23:53
SpamapSin fact.. I'm not really even sure how to edit the generated patches23:54

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