/srv/irclogs.ubuntu.com/2011/05/04/#ubuntu-motu.txt

=== medberry is now known as med_out
evaluateHello.07:04
evaluateThere is a fix for a pretty serious bug for a package in natty. I currently can't upload it to Debian because of the perl transition. Can someone update it in Ubuntu directly or should I best mail the maintainer of the package in Ubuntu?07:05
micahgevaluate: which package?07:06
evaluatemicahg, http://packages.ubuntu.com/natty/clipit07:07
evaluateThe bug which I'm talking about is bug #70231607:07
ubottuLaunchpad bug 702316 in clipit (Ubuntu) "Generic Libindicate fallback support breaks applications (such as clipit) on non-Unity WMs/DEs" [Undecided,Fix committed] https://launchpad.net/bugs/70231607:07
evaluateAnd it is fixed in the 1.3.13 release, which is available on http://sourceforge.net/projects/gtkclipit/files/Version%201/07:08
micahgevaluate: best to file a bug, prepare debdiffs for oneiric and natty-proposed and subscribe ubuntu-sponsors07:09
evaluateI wasn't talking about oneiric, I rather thought about a SRU for natty...07:10
micahgevaluate: SRUs have to go to dev release first :)07:12
evaluateoic07:13
micahgevaluate: it would be the same debdiff, just a different version/suite07:13
evaluateIs this the procedure you're reffering to: https://wiki.ubuntu.com/PackagingGuide/Recipes/Debdiff ?07:13
micahgevaluate: yeah07:13
RhondaCan we get the next release done sooner? I fear I might mistype the codename far too regularly for it to be useful.07:13
micahgevaluate: version is ubuntu2 for oneiric and ubuntu1.1 for natty-proposed07:13
micahgRhonda: 5.5 months isn't soon enough?07:14
evaluateok, I'll try to go through that...07:14
Rhondamicahg: I fear I will have to type it too often in that time. :)07:14
* micahg guesses debian is easier since it's almost always unstable or experimental07:15
Rhondaalmost always fits pretty well, though I am one of these rare persons who seem to have uploaded to every single other possibility out there, too ;)07:21
RhondaPreparing security updates yourself to reduce the workload of the security team seems to entail that somehow.07:22
nigelbRhonda: heh07:23
Rhondanigelb: Hey, I like to give everything a try on my own, to form my own opinion instead of being guided by prejudices handed by others.07:24
nigelbRhonda: No no, i was smiling about the bit were you wanted the release early ;)07:24
RhondaThe one time I tried emacs it though ended in a reboot of the server I was sitting infront of.07:24
nigelbvim ftw07:25
RhondaI wondered how to exit it, and a "helpful" collegue said "I think it's control, meta …  and me thinks delete"07:25
RhondaAh, that one. :)07:25
nigelbheh07:25
evaluateWhat should I use in the changelog, natty or natty-proposed (or oneiric)?07:28
nigelbnatty-proposed I guess07:30
nigelbbut shouldn'd you fix it first in oneiric?07:30
micahgevaluate: natty-proposed for natty, oneiric for oneiric07:30
evaluateSo do I need to create two debdiffs?07:31
micahgevaluate: yep07:31
micahgevaluate: same debdiff, just change the version/changelog07:31
evaluateok, got it.07:31
evaluateok, could someone please have a look at the debdiff and tell me if it's fine before I submit it: http://pastebin.ubuntu.com/603123/ ?07:36
Bachstelzeshould be ubuntu1.2 I think07:40
micahgevaluate: version should be ubuntu1.1, for oneiric ubuntu207:41
Bachstelzealso you should fill out the DEP3 info in the patch07:41
evaluateBachstelze, which information is needed exactly?07:44
Bachstelzeevaluate: lines 39-45 of your paste07:44
micahgactually, he already filled it out in 31-3207:45
evaluatehttp://pastebin.ubuntu.com/603125/07:48
evaluateBug, Bug-Debian and Reviewed-By don't apply.07:48
evaluateNot sure why it says 'debian-changes-1.3.12-1ubuntu2' on the last line though. Is that normal?07:48
Bachstelze[08:48] < evaluate> Bug, Bug-Debian and Reviewed-By don't apply. <= delete them, then07:51
evaluateYou mean the whole part between lines 36-47?07:51
Bachstelzethe lines that don't apply07:52
Bachstelzethough since Origin is upstream, you should probably have a Bug line07:52
evaluateWhat do you mean with Bug line please?07:52
evaluateThe bug only applies to Ubuntu, since it's indicator related...07:53
Bachstelzethe Bug line should give the URL of the bug in the upstream bug tracker07:53
Bachstelzethen why does Origin say upstream?07:53
evaluateIsn't Origin reffering to the source of the patch?07:54
Bachstelzeyes07:54
evaluateWell, the bug was fixed upstream, so I guess that should be fine then...?07:55
Bachstelzethen you shoukd out an URL to the bug in the upstream bug tracker in Bug07:55
Bachstelzeand normally the URL to the upstream VCS commit in Origin07:56
Bachstelzeit's all there really: http://dep.debian.net/deps/dep3/07:56
Bachstelzeput*07:59
evaluateWell, there's no bug in the upstream bug tracker about this. It's just in launchpad....08:00
RhondaIs an int signed or unsigned by default?08:01
nigelbI think signed08:03
Bachstelzeyes08:05
evaluateBah, this is too complicated. I'll just request a sync once it gets into Debian.08:06
RhondaThat's the spirit! :)08:13
micahgevaluate: to get into natty, you still need a debdiff/test case08:42
evaluateWell, I don't. I just wanted to help fix that bug, if you're making such a simple patch so complicated, then you can fix it.08:44
micahgevaluate: Debian doesn't have requirements about DEP-3 headers on patches?08:46
evaluatemicahg, it does, but the way to place them there is more sane. If I run debuild -S they are generated automatically and I don't have any influence about what gets put there apparently.08:47
=== yofel_ is now known as yofel
micahgevaluate: you should be able to use DEP-3 headers the same way in Ubuntu as Debian, it just requires a Bug-Ubuntu instead of a Bug-Debian, they're auto-generated, but that's meant as a guide, you can remove the auto-generated ones if you don't need them08:49
evaluateAlso, since the bug was fixed upstream, I don't see the logic behind also creating a patch for Ubuntu. Why not just import the package from upstream?08:50
micahgevaluate: we can't take a new upstream version in a stable release without an SRU exception08:50
micahgevaluate: would be fine for oneiric though08:50
evaluateAnd wouldn't a complete freeze of X justify a SRU exception?08:51
micahgevaluate: not if it can be fixed with a simple patch08:51
micahgevaluate: I think the only thing missing was a link to the "upstream" commit, that would go on the Origin line following upstream,08:55
RhondaAnd actually, the Description that debuild -S produces isn't really sane.09:01
=== Quintasan_ is now known as Quintasan
RhondaThe whole created thing screams "I'm a template - please adjust me!" in most of its part.09:02
evaluatemicahg, would this be acceptable: http://pastebin.ubuntu.com/603147/ ?09:07
micahgevaluate: that looks good to me09:08
evaluateDo I just attach it to the bug report or should I also ping a list or something?09:09
micahgevaluate: attach to the bug, subscribe ubuntu-sponsors09:09
evaluateAlso, about oneiric, do I need to do it for oneiric too? Wouldn't it be wiser to just sync the patch from Debian once it's there?09:10
micahgevaluate: you can attach a second one to the bug for oneiric, just changing the version to 1.3.12-1ubuntu2 and the suite to oneiric09:10
evaluateSince oneiric isn't limited to any SRU exceptions or anything...09:10
micahgevaluate: yeah, you can sync from Debian if you want to wait for that, but it needs to go in oneiric before ubuntu SRU will accept to natty09:10
evaluateohh well09:10
micahgevaluate: you could prepare a new version upload for oneiric if you like as well, but that's up to you09:11
micahgevaluate: seems like the least amount of work would be to get this patch uploaded to oneiric and then sync from Debian once it's ready there09:12
evaluatemicahg, ok.09:13
evaluateGah. "You cannot make any changes at this time. Launchpad is undergoing maintenance and is running in read-only mode."09:14
micahgevaluate: oops, yeah, should be over by 9:30 UTC09:16
micahgevaluate: thanks for your work on this :)09:18
evaluatemicahg, sure. It's my package after all... :-)09:19
micahgevaluate: you should find that after the first few bumps, it should be clear sailing to get stuff sponsored, if you have any ideas for making the process easier/better, please let us know09:22
evaluatemicahg, I don't see anything wrong with the process of having something sponsored, but rather with the idea behind preffering a ubuntu patch rather than a new upstream version in a SRU...09:25
micahgevaluate: this might help with that: https://wiki.ubuntu.com/StableReleaseUpdates#Why09:32
evaluatemicahg, https://bugs.launchpad.net/ubuntu/+source/clipit/+bug/702316 -- hope it's ok now09:43
ubottuUbuntu bug 702316 in clipit (Ubuntu) "Generic Libindicate fallback support breaks applications (such as clipit) on non-Unity WMs/DEs" [Undecided,Fix committed]09:44
=== hannesw_ is now known as hannesw
ScottKmicahg: DEP-3 is not required in Debian.12:27
ScottKAnd I think this is a good example of why.12:27
RhondaActually that patch is only a good example that debuild -S creates only a template and requires a fair amount of tweaking afterwards. :)12:45
c2taruncan anyone please tell me how can I edit this page and write comments? https://merges.ubuntu.com/universe.html12:46
geserc2tarun: there is a hidden input field in the comment column12:54
c2tarungeser: in natty there was a linker problem due to binutils-gold linker. is this also problem in oneiric?12:56
c2tarunhere is a debian/changelog file http://paste.kde.org/50683/ in line 14 its adding a patch, but when I checked into the file the patch is all ready applied. How is this possible?13:10
geserc2tarun: yes, oneiric uses the same linker defaults as early natty13:11
c2tarungeser: can you please look at the problem I just posted?13:12
geserc2tarun: see line 10; the package now uses v3 (quilt) where package are automatically applied after unpacking13:12
c2tarungeser: I also thought so, but when I checked by quilt top I got no patches applied.13:13
geserhmm13:13
c2tarungeser: I got this error while applying the patch Patch 01_local_hiscores_file_check.patch can be reverse-applied13:14
c2tarungeser: I grabbed the package by grab-merge.sh script13:16
c2tarungrab-merge.sh bastet13:16
c2tarungeser: I am getting similar error with the other patch mentioned in line 1713:18
c2tarungeser: ping13:19
c2tarungeser seems to be busy, can anyone please help me?13:22
Laneyyou don't need to ping specific people for questions like that13:23
Laneythat message usually means that the patch has been applied upstream or in debian, and you can remove it13:23
Laney(or you are attempting to apply it twice)13:23
geserhmm, does MoM the right thing for v3 packages?13:25
c2tarunLaney: in changelog its mentioned about adding the patch, how can that be applied upstream?13:25
geserLaney: see the changelog entry c2tarun linked; that patch got added in the current Debian revision13:26
Laneysure, I was giving a generic answer13:27
c2tarunI just pulled-debian-source bastet and patch is not applied there.13:30
c2tarungrab-merge.sh can only apply the ubuntu1.patch and no internal patches13:31
* c2tarun confused, what went wrong and where?13:31
c2tarunLaney: are you suggesting that I should drop the patch?13:31
geserc2tarun: do you have a .pc/ directory in the package directory?13:32
c2tarungeser: yes but its empty.13:33
c2tarungeser: nope its not13:33
c2tarungeser: it contains three files .  ..  .quilt_patches  .quilt_series  .version13:33
geserperhaps it would be best if you take the (unmodified) Debian package and merge manually13:34
Laneydoes it need a merge?13:35
Laneychangelog suggests not13:35
c2tarunLaney: I too think so, let me check it13:35
Laneythis http://launchpadlibrarian.net/65435629/bastet_0.41-6_0.41-6ubuntu1.diff.gz was the previous diff13:37
Laneyand the bug logs and debian changelog suggest that that is now upstream13:37
Laneyso it looks like a sync to me13:37
c2tarunLaney: sorry I got disconnected, did you replied to my query?13:49
Laneyc2tarun: didn't see it13:49
c2tarunLaney: how can I request for sync? I tried with reqeuestsync script but dont know why its not working. Is there any particular format for sync bug for LP?13:53
Laneyc2tarun: did you try with requestsync --lp13:57
Laney?13:57
Laneyotherwise, https://wiki.ubuntu.com/SyncRequestProcess13:58
c2tarunLaney: I posted a sync bug 777083  please check and tell me if more details are required?13:58
ubottuLaunchpad bug 777083 in bastet (Ubuntu) "Please Sync package bastet-0.43 from debian unstable" [Undecided,New] https://launchpad.net/bugs/77708313:58
Laneyusually the debian changelog is included13:58
Laneyand it'd be nice if you said that the only fixes were build fixes which are now incorporated upstream13:58
c2tarunLaney: ok, I'll do that, and sorry I requested a sync for the first time so its little bit less in details13:59
Laneyno problem at all13:59
c2tarunLaney: complete debian changelog or latest entry?14:00
Laneythe new ones14:01
c2tarunLaney: please take a look now.14:01
Laneyc2tarun: looks good, please subscribe ubuntu-sponsors and that's it14:06
=== med_out is now known as medberry
=== aalex is now known as aalex-home
debfxhow can I remove a merge request from the sponsoring overview?15:07
debfx(the equivalent to setting a bug status to invalid)15:07
Laneyset the status of the merge request15:09
debfxto work in progress?15:09
=== Tonio__ is now known as Tonio_
Laneydebfx: sounds good15:34
Laneybarry: regarding your winpdb upgrade request, have you seen that debian has 1.4.8 now?15:46
Laneyit has in fact been autosynced15:47
c2tarunthere is a package named dares, there was a patch to fix binutils-gold error. Debian developer just included this patch in its newer version. No other changes and patch is also not applied in source code, just included as a patch. Is there any need of sync in this case?17:03
Laneyc2tarun: might as well while you are looking at it, then it'll be autosynced in future17:53
c2tarunok, this question may not be relevant but there was a guy with IRC nick persia, I haven't seen him from very long time on any channel. why so?17:54
c2tarunI asked because he was very with everything we do on this channel17:55
c2tarunI asked because he was very good with everything we do on this channel17:56
nigelbc2tarun: he's in japan, he's having internet trouble for a while.17:56
c2tarunohh...17:56
c2tarunI have a question about grab-merge.sh script. This script downloads the source code from debian and then applies the ubuntu.patch file to it? is there any way to make it stop applying that patch?19:34
micahgc2tarun: just unpack the debian source separate if you want to see it, it'll be in a different dir19:41
c2tarunmicahg: got it. in previous ubuntu version, a patch was introduced to fix ftbfs due to binutils-gold, that patch is not included in new debian release. This should be a merge. What should I write in changelog, patch imported from previous ubuntu version?19:42
micahgc2tarun: list the patch in remaining changes19:43
yofeldoesn't debian use the gold linker too?19:44
c2tarunyofel: dont think so. :/ because new version is giving same error due to gold linker :(19:44
yofel:/19:44
micahgthe patch should be forwarded if they're affected, but that doesn't mean they'll include it19:44
micahgyofel: not yet by default19:45
yofelah ok19:45
c2tarunone more problem, whenever i try to use the grab-merge.sh script, none of the patches in debian/patches folder apply. On executing quilt top i see no patches applied, but on pushing the patches none of them apply :( Even the patches introduced in the same version also didn't apply. why so?19:49
yofelis the series file intact? I just merged kaffeine, and there the patch got applied without quilt and the series file got renamed to series.UBUNTU19:51
c2tarunyofel: on series file isn't intact I have to modify it to include all the patches. can you please try this on your system. grab-merge.sh dma19:52
c2tarunyofel: then go inside the folder and check whether any patch is applied or not? then modify the series file in proper format and then try to apply the patches.19:52
yofelyeah, it seems the ubuntu patch is applied now and the series file has conflicts19:55
yofeleveryone: what's the right way to fix that?19:55
c2tarunyofel: conflict is because ubuntu patch is not included and few new patches are introduced. We have to include ubuntu patch as well all the debian patches19:56
yofelthat is true, what I'm not sure is how to handle that fact that 38_fix_ftbfs_binutils-gold.patch is already applied to the source19:57
yofelunless all patches were automatically applied..19:57
c2tarunyofel: well there is no information in changelog that 38_fix_ftbfs_binutils-gold.patch is applied.19:58
yofelnono, I meant that it was applied by MOM19:58
c2tarunyofel: and try to pop the patches, none of them are applied :/ or something weird is happening which I have never seen before.20:01
c2tarunhmmm.... thats new to me. I didn't know that MOM also applies the patch.20:01
yofelnah, I'm wondering if MOM just doesn't consider that dpkg-source now automatically applies patches with 3.0(quilt)20:01
c2tarunyofel: what should I do here? get the debian source, copy the debian folder from dma-0.0.2010.06.17-10ubuntu1 and try to build it on oneiric?20:02
yofelno idea, I'm pretty new to merging myself and was just asking myself the same..20:02
dupondjethere is a magic tool for this20:03
yofelin this case you could probably just revert that one patch in the Makefile since it's just a one line patch and then rename the patch file and update series20:03
dupondjegrab-merge package20:03
dupondje:)20:03
dupondjedoes alot by itself20:04
yofeldupondje: yeah, and that gets us a source with applied quilt patches and quilt says no patch applied20:04
dupondjeeuh, but thats because no patches are loaded in quilt ?20:04
c2taruncan anyone please help us here?20:04
yofeldupondje: yes, but it seems that the patches were applied by MOM before merging20:05
dupondjewhat package are we talking about ,20:05
yofelor MOM just doesn't check if dpkg-source auto-applies the patches20:05
c2tarundupondje: what does that mean that no patches are loaded in quilt?20:05
yofeldma20:05
yofelit's the 38_fix_ftbfs_binutils-gold.patch from ubuntu20:06
yofelthe series file has conflicts, but the patch is already applied20:06
yofeland that's not because it's in debian or upstream20:07
dupondjegrab-merge dma20:09
dupondjefix conflict in debian/patches/series20:10
dupondjeand that does it no ?20:10
dupondje:)20:10
yofeldupondje: yes, but why is the patch (that needs to be kept) already applied?20:10
yofelthat's not how quilt works...20:10
micahgyofel: source format 3 and grab-merge will not work well together20:11
yofelthat's what I feared...20:11
c2tarunmicahg: that means we have to merge it manually.20:12
c2tarunI tried and encountered this error http://paste.kde.org/51187/20:13
c2taruncan anyone help me with this error ^^20:16
geserc2tarun: which package is this?20:16
yofelstill dma20:17
c2tarungeser: dma20:17
dupondjethat patch is applied20:18
dupondjebecause its fixed in the package itself20:18
dupondjesee ftp://ftp.debian.org/debian/pool/main/d/dma/dma_0.0.2010.06.17.orig.tar.gz20:18
c2tarundupondje: I extracted fresh tarball and then applied all the patches and then on buliding I got this error.20:19
yofela) it's not applied, b) all patches were applied, it's just that MOM can't handle source format 3.0(quilt)20:19
c2tarunyofel: was that ^^ for me?20:20
yofelno, re dupondje20:20
yofelbut I got the same error if it helps (no, not really...)20:20
c2tarunyofel: you got some different error?20:20
yofelno, same20:21
yofelcrypto.c:92:7: error: assignment discards 'const' qualifier from pointer target type [-Werror]20:21
yofelnot sure if debian builds with -Werror20:21
micahgyofel: they don't20:21
yofelgood, then we have the problem at least20:22
geserc2tarun: change crypto.c:87 to "const SSL_METHOD *meth = NULL;" (add the const to the beginning)20:22
c2tarungeser: should I create a patch for this?20:24
geseryes, as the package is in v3 (quilt) format20:25
c2tarungeser: it worked :) thanks20:27
gesernp20:27
dupondjebtw: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=62205220:36
ubottuDebian bug 622052 in src:dma "dma: FTBFS: crypto.c:92:7: error: assignment discards qualifiers from pointer target type" [Serious,Open]20:36
c2tarundupondje: ohh.. I should close this bug in my changelog . Thanks :)20:38
c2tarunno I cant20:38
c2tarunits a debian bug I guess.20:38
dupondjeye :)20:38
dupondjebut you could reply to it with the fix for example :)20:38
dupondjethe old ftbfs (38_fix_ftbfs_binutils-gold.patch) got fixed this way in debian20:38
geserc2tarun: I sometimes add "Closes: #xxx" to my changelog entries, even if it does nothing in Ubuntu but it good for a reference20:39
geserit helps checking later if the Ubuntu delta needs merging or can be dropped (because the bug got fixed in Debian too)20:40
c2tarunok then, I'll add "Closes: #debian-bug" to my changelog entry, should I send it to debian as well?20:40
geserthe patch yes20:41
dupondjeadd the diff to the bug yep20:41
dupondjealways usefull for debian :)20:41
dupondjegeser: you have upload rights in universe?20:45
geseryes, like any other MOTU too20:46
=== cyphermox_ is now known as cyphermox
dupondjewant to check https://bugs.launchpad.net/ubuntu/+source/toonloop/+bug/775588 ?20:46
ubottuUbuntu bug 775588 in toonloop (Ubuntu) "Merge toonloop 2.0.6-1 (universe) from Debian unstable (main)" [Wishlist,New]20:46
micahgdupondje: you can ask jdstrand, he's piloting in #ubuntu-devel20:48
dupondjealso for universe packages ? :)20:49
geserfor everything in the sponsoring queue20:49
dupondjeoh ok20:49
Laneybigon: I feel like looking at merging epiphany 3, mind if I take it over from you, or are you planning on doing it yourself?21:36

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