/srv/irclogs.ubuntu.com/2010/08/20/#bzr.txt

=== beuno_ is now known as beuno
pooliehi all00:57
spivGood morning.01:11
pooliehi there spiv01:57
pooliespiv does https://bugs.edge.launchpad.net/bzr/+bug/620684 suggest anything obvious to you?02:47
ubot5Launchpad bug 620684 in Bazaar "ERROR file id "None" is not present in the tree when checking the root entry (affected: 1, heat: 6)" [Undecided,New]02:47
spivpoolie: huh, no.  Strange upgrade, maybe?02:50
* poolie is writing up a mail of annoyances in ppa updates03:41
mwhudsondid i just dream that there was a bzr init --standalone option?04:05
fullermdmwhudson: There's a *branch* --standalone.  Maybe that's what you're dreaming of?04:28
mwhudsonfullermd: ah, probably04:28
fullermdMe, I'll stick with dreaming of Cindy Crawford, but...04:28
poolieGaryvdM: hi, can you tell me which branches you used for your bzr-gtk packages07:19
poolienm07:24
vilahi all !08:04
fullermdWassat?  Morning again already??08:05
vilafullermd: yeah, and end of vacations too, wake up NOW :)08:05
fullermdVacations?  Heck, I don't even REMEMBER what those are...08:06
vilafullermd: You really should, there are nice places around the world you really should visit while you still can ;)08:06
fullermdThat would require leaving home.  To say nothing of putting on pants.08:06
fullermdWell.  I guess the REALLY nice places don't need the pants...08:06
vilafullermd: yup, including the under water world...08:07
pooliehi vila08:08
vilapoolie: heya !08:08
pooliei'm going to change tack on the ppa and try to get everything just at least working on maverick08:09
vilapoolie: what isn't working so far ?08:10
pooliea bunch of litle annoyances08:10
poolieone being that in the debian bzr packaging, they've cut out python-configobj08:10
pooliebecause of a sensible policy that packages should not ship their own copies of libraries08:10
pooliebut that's not packaged on hardy, and it's a little hard to make it work there08:11
poolieit could all be fixed but it just seems like a time sink08:11
vilawell, not shipping copies of libraries shouldn't apply if the said libraries aren't available methinks...08:12
vilahm, restart required bbib08:12
fullermdYeah, if we shipped testtools, I could run selftest  ;>08:12
poolieit's completely fixable it's just one thing after another08:14
poolieseparated by long pauses for soyuz to decide08:14
* fullermd is unpleasantly familiar with such tasks...08:14
poolieanyhow istm the main thing is to get it all working on lucid08:15
poolieyou might think it would make sense to start with the oldest one and work forward08:15
pooliebut i now think this is probably wrong08:15
vilapoolie: waitaminute, I thought we modified configobj lightly but enough to not be able to use the stock version no ?08:25
poolieoh, really?08:25
pooliethat's such a bad idea08:25
pooliei hadn't heard of that08:26
vilaIMBW or the changes may just be cosmetic...08:26
GaryvdMHi poolie08:30
GaryvdMpoolie: lp:~bzr/bzr-gtk/packaging-$DISTRO08:30
GaryvdMI may not have push up all my changes yet, as I've not finished the hardy build. Let me push up the rest.08:31
GaryvdMpoolie: I've also merged them with the debian unstable branch :-)08:32
GaryvdMThat's why hardy is failing...08:32
poolieGaryvdM: got it now08:32
vilapoolie: hmm, scratch that, if we used a modified version and debian deleted it, we would have heard about bugs08:34
poolievila, that's what i thought too08:34
pooliei don't feel like going to look for trouble there; i have enough already08:34
vilapoolie: ok, but may be we should file a bug about using our own copy only if none is available08:35
spivThere is at least one bzr-local change to ours, to improve import speed :/08:36
spiv(It imports the 'compiler' module for a feature we never use)08:37
pooliehm, that's just the kind of thing we wouldn't get bugs about too08:37
pooliehello spiv, btw, you've been a quiet mouse :)08:37
vilahi spiv, I was just seeing your comment in configobj.py :D08:38
pooliespiv, i we should file a bug against bzr, debian, and ubunut08:38
poolieand maybe try to send the patch upstream to configobj08:38
spivAFAIK we'd run ok with stock configobj, I think most of the other changes that bzr log reports are cosmetic.08:39
spivpoolie: I've been chipping away at old todo items before they go entirely stale... I'm currently looking at a draft of a LEP about logging (and contemplating how e.g. 'bzr serve' could change to do it better).08:43
vilapoolie, spiv: even comparing with actual configobj-4.7 (we have a copy of 4.6) I see nothing worth worrying. As spiv said, there is the compiler loading and then 4.7 differs slightly for some errors handling08:45
spivvila: that's a relief :)08:46
* bialix waves at GaryvdM09:21
GaryvdMHi bialix.09:23
pooliehi bialix09:33
poolieppas should be all ok once the packager finishes updating09:33
pooliewell, by that i mean the lucid proposed ppa09:33
bialixhi poolie09:33
bialixdoes anybody know which version of bzr-explorer is packaged in ppa and lucid?09:35
pooliehttps://edge.launchpad.net/~bzr/+archive/proposed?field.series_filter=lucid has 1.1.0~beta109:35
poolieand lucid itself has09:35
poolie1.0.1-0ubuntu109:36
bialixwhy not 1.0.2...?09:36
poolieand maverick also has 1.1.0~beta109:36
pooliei don't know09:37
bialixrhetorical question09:37
pooliei would guess that nobody updated it09:37
poolieif it's only bugfixes, it still could go in09:37
bialixsorry, I've meant maverick09:37
bialix1.0.2 has released after lucid09:37
pooliemaverick of course is the one that's coming out in october09:37
bialixyep09:38
GaryvdMbialix: https://wiki.ubuntu.com/StableReleaseUpdates09:38
bialixthat's fine then09:38
* bialix notes09:38
GaryvdMbialix: It's quite a process to get it updated in an existing release.09:38
pooliewe should do something about the sru policy09:38
bialixyes, I understand09:38
pooliespiv it would be nice if you could do that next week09:39
bialixI'm not ready to force SRU right now09:39
pooliewe'll try to keep the ppa up to date09:39
GaryvdMbialix: The ppa will have 1.1.0~beta soon.09:40
bialixGaryvdM: thank you!09:40
GaryvdMpoolie: the latest bzr-explorer, and qbzr are compatible with bzr 2.1.0, so I may as well copy them to the main ppa now.09:41
poolieok09:41
poolieso i think we're now all done09:41
GaryvdMpoolie: what about git, svn loggerhead?09:42
GaryvdMgit has new version for lucid, but not other distros.09:43
GaryvdMSame for dulwich09:43
GaryvdMSame for svn09:43
GaryvdMpoolie: I think we still have some way to go.09:44
poolieGaryvdM: i was wondering if we should have some enlightened lazyness about builds on old distros09:44
GaryvdM:-)09:45
pooliei found i was bogging down a lot09:45
poolietrying to get them all through, and once i focussed on just lucid it was much easier09:45
GaryvdMI know the feeling.09:45
poolieit ought to be mechanical but there are proportionally more snags across the old ones09:45
pooliei'm going to try to get some data on how much people actually use them, or want upgrades there09:45
poolieboth from the download logs, and perhaps we'll see if anyone objects on the list09:46
GaryvdMpoolie: Lets copy for lucid now, and not copy for the others...09:46
pooliemy sense is that people mostly do not stick with old ubuntu releases for a long time (except for lts)09:46
poolieunless they specifically want the machine not to move very much09:46
pooliebut this is ... wishful thinking :)09:46
GaryvdMpoolie: How come your uploads show as no signer?09:50
vilapoolie: the reasoning sounds fine, if I want a newer bzr I'm likely already tracking a newer Ubuntu. But even if I track an older Ubuntu *and* I've subscribed to the bzr ppa, I still prefer stability over new features so I may accept fewer updates on the bzr ppa. I.e. we can start proposing an up-to-date ppa for lucid and maverick and update hardy/jaunty/whatever when time permits (including when our process become more automatic)09:54
bialixI can't remember the name of the plugin to run test suite on each commit, aka poor-man pqm. anyone?09:57
bialixtestrunner09:58
poolieGaryvdM: i copied the binaries from the main archive; i guess that's why09:59
GaryvdMpoolie: Ah - ok10:00
GaryvdMpoolie: I just read your mail.10:00
poolievila, are you talking about maybe having a 2.1, 2.2,  etc ppa?10:00
GaryvdMpoolie: This was the frustration I was trying to describe at uds - You described it much better than I could...10:01
vilahmm, no I was thinking about just the stable and proposed ones10:01
pooliei'm just glad it's done; the combination of lags plus tiny failures is so annoying y10:02
poolieyou can't concentrate and just get it done but you can't concentrate on anything else10:02
poolieor at any rate i can't10:02
poolievila: oh you're saying people on hardy may be happy to just stay on 2.0 or whatever?10:02
pooliei agree10:02
pooliei agree they may10:02
GaryvdMI upload, play a game of solitaire, and then go and check if it succeeded. :-010:03
poolieexactly10:03
poolienice for 20m, not a good way to spend two days :-{10:03
poolieor more :{10:03
vilapoolie: yes, they may accept to upgrade to 2.2 later10:04
vilaerr, they may accept that the upgrade is delayed10:04
GaryvdMpoolie: I'm going keep at it though. Next up - for me: bzr-builddeb10:04
poolie?10:04
pooliei think they're all done10:04
poolieassuming the publisher runns...10:05
lifelessawesome10:05
GaryvdMpoolie: For all distros10:05
poolieoh, ok10:05
poolieif you really want to10:05
pooliewow, launchpadlib pulls in a bunch of stuff10:06
pooliei think it's used by builddeb10:06
poolieincluding a mail server10:06
pooliestrace10:06
GaryvdMpoolie: The other thing is that merging the debian unstable packaging branches as caused more complication, but I think it will make it easier in the long term. Packages for which I did this when I did 2.1 were easier this time round.10:07
bialixpoolie: q about pad.lv: why you've used .lv domain for it?10:10
poolieit's cheap, and it sounds like "pad love"10:16
pooliei'm not latvian10:16
pooliei think spiv is ½ latvian though, or lithuanian10:16
bialixfunny10:19
pooliegood night and good luck!10:20
GaryvdMGood night10:20
bialix:-)10:20
GaryvdMi386 ppa build queue > 4hrs :-(13:50
jelmervila: Thanks for the review, very much appreciated!13:50
vilajelmer: as is your work in this area, you're definitely on something here and I was just wondering if the factory idea shouldn't be pushed even more to clearly define how one create repo/branch/wt for a given format13:52
vilajelmer: they are all created from the format or its associated dir so far and without looking into the details, I wonder if we have inherited that from the early days of bzr...13:53
vilajelmer: I was quite surprised that you were able to produce such a clean patch there, I would have swear that things were too mixed-up for that... So, well done ! ;)13:54
jelmervila: Thanks :-)14:37
jelmervila: History shows there indeed, although I think it's a slightly disconnected problem.14:37
jelmerI'll followup to the branch tonight and see if I can get some of the plugins working with this new API then :-)14:38
vilagreat !14:38
GaryvdMjelmer: My I ask you some debian packaging questions?14:46
jelmerGaryvdM, yeah, sure14:46
GaryvdMjelmer: I would like to submit a patch to the packaging config for qbzr. Should I submit a branch with just the change, or should I also merge the upstream, and update the changelog?14:48
GaryvdMMaking the change, and updating the changelog with out merging upstream feels odd.14:49
jelmerGaryvdM: You should definitely update the changelog to mention what you've fixed14:49
jelmermerging upstream isn't necessary, but you might as well while you're there14:50
GaryvdMAlso one of the changes I want to make is specific to the upstream version (minimum bzr version)14:50
GaryvdMjelmer: And then push to lp, and send mail to pkg-bazaar-maint@lists.alioth.debian.org ?14:51
jelmerGaryvdM: yep14:53
GaryvdMOk :-)14:53
starenka__hi, sorry for noob question but is there a way how to show all modified files since rev x?15:19
vilastarenka__: bzr status -r x15:20
starenka__neat15:20
starenka__thanks15:20
mgzgaryvdm: re your pretty html table of package versions on the mailing list15:51
GaryvdMYhea - sorry about the html.15:51
mgzI think testtools really wants to be at either 0.9.2 or 0.9.515:51
mgzwe probably can't do much about debian unstable, but who would need bugging about maverick?15:52
mgz^needing to do a table is a good reason for html email, the mess people make trying to do them is plain text is worse15:53
GaryvdMmgz: Why not do any thing about debian unstable. We have 2 dd's amongst us!15:54
mgzwell, they're deeper in freeze than ubuntu is I believe15:54
mgzbut sure, jelmer/lifeless, if you can find a way to get the release managers to take the next testtools version, that'd be great15:55
GaryvdMmgz: I don't think that would be required for sid, only for squeeze.15:56
mgzyup.15:57
GaryvdMmgz: Ok - so we should try get sid/mavrick updated to 0.9.5 before updating the others in the ppa.15:58
GaryvdMThats ok, because I want to get the more user facing packages updated first.15:59
mgzgreat, thanks.15:59
=== mvo is now known as mvo|undercover
mgzmy busy week is over now so I should be around if you need help with anything15:59
jelmerhi Gary, Martin16:04
jelmermgz, GaryvdM: Rob is the person to talk to, he's the only maintainer of testtools in Debian.16:04
mgzthanks jelmer, I'll bug him later.16:05
=== lamont` is now known as lamont
=== Ursinha is now known as Ursinha-lunch
zygahello16:40
zygais 2.2 officially released ?16:40
jelmeryeah16:41
zygajelmer, then why does the website state that 2.1 is stable release?16:42
jelmerzyga: That's a good point..16:43
jelmerjam, vila: ^16:43
=== zyga is now known as zyga-afk
* zyga-afk needs to go away for a few hours :/16:44
vilajelmer: I think we wait for the installers before releasing officially16:44
vilajelmer: and also an updated ppa but this should be good now thanks to poolie and garyvdm16:45
=== ddaa1 is now known as ddaa
vilazyga-afk: so 2.2 is really *close* to be officially released16:45
=== deryck is now known as deryck[lunch]
MTecknologyI tried doing this --> bzr pull lp:pressflow --remember  and got this --> bzr: ERROR: Cannot lock LockDir(lp-70598736:///~pressflow/pressflow/6/.bzr/branchlock): Transport operation not possible: readonly transport17:45
MTecknologywhy?17:45
vilaMTecknology: bzr info ?17:47
vilaMTecknology: may be your branch is bound to lp:pressflow ?17:48
MTecknologyvila: that'd explain it :D17:50
MTecknologythanks17:50
rockyjelmer, what's the latest version of bzr-rebase/rewrite i should use with bzr 2.2.x ?18:25
=== Ursinha-lunch is now known as Ursinha
jelmerrocky, the latest version f bzr-rewrite (not sure off the top of my head what that is)18:26
rockywell baseically i need rebase that works with bzr-svn and bzr 2.2.018:26
rockyso what do i get? :)18:26
rockyjelmer, ^^18:28
jelmerrocky: as I said, the latest version of bzr-rewrite18:29
rockyjelmer, oh sorry i misunderstood... the reason i'm asking is because the latest download on http://wiki.bazaar.canonical.com/Rewrite?action=show&redirect=Rebase says it's for bzr 1.13 and higher... hwich makes it seem old18:29
MTecknologyIs there any easy way to take one branch abd run bzr pull lp:newbranch --remember --ignore-conflicts ?19:09
MTecknologyThe only way I can think of is to move the directory, then grab a fresh copy - but not ideal19:10
jelmerrocky: the APIs it uses haven't changed, so there's no reason to change the minimum bzr version19:35
rockyk19:35
MarcWeberI dowloaded two blender branches. How can I visualize when one was forked off and how many commits are different?20:20
dashMarcWeber: 'bzr missing' will tell you the differences between the branches.20:20
dash'bzr vis' or 'bzr qlog' can show you a pretty view of the history, if you have the gtk or qt extensions installed20:21
MarcWeberdash: I prever pretty views :) Are those extensions included in the main source ?20:22
dashthey're plugins.20:22
MarcWeberDoes bazaar store multiple branches in one directory as git does?20:30
MarcWeberI have a blender branch on lp I'd like to compare with current unstable.20:31
dashMarcWeber: no, it doesn't20:31
dashyou can use a shared repo so that branches don't have duplicate copies of the revs they have in common20:32
MarcWeberSo I checkout twice and then use diff ../other-dir or such ? or diff lp-url ?20:32
dashMarcWeber: sure20:33
MarcWeberor missing ../other-dir etc?20:33
dashyep20:33
dashif you start with 'bzr init-repo .' then it'll create a repo in the current directory20:33
dashand so checking out a second branch in that dir will only fetch the extra revs20:33
MarcWeberbranch 'lp:~diresu/blender/blender-command-port-002'20:34
MarcWeberand branch lp:blender20:34
MarcWeberCan I make a shared repo out of a checked out one?20:41
dashMarcWeber: to do that you do init-repo as above, then clone your existing branch into it20:42
dashso 'bzr init-repo .; bzr branch mybranch mybranch_new' will do it.20:42
MarcWeber( --trees) ?20:43
dashif you like :)20:44
eydaimonif I do a bzr up -r 123 and then a bzr revno, it shows the head revno, not the one the repo is at. How can I get the revision the repo is at?21:32
=== Ursinha is now known as Ursinha-bbl
maxbeydaimon: You don't mean repo, you mean tree. Use bzr revno --tree23:18

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