/srv/irclogs.ubuntu.com/2009/04/28/#bzr.txt

=== dereine is now known as dereine[OFF]
spivHmm, entries in NEWS aren't being alphabetically sorted lately.00:15
igcmorning01:02
garyvdmigc: evening ;-)01:02
lifelessmwhudson: so01:09
lifelessin idnars branch01:09
lifelesscheck last-revision01:09
lifelessand then iter_revision_history01:10
lifelessloggerhead is thinking that they are equivalent01:10
mwhudsonlifeless: are they not?01:15
lifelesstry it on the branch in question01:16
lifelessspiv: https://code.edge.launchpad.net/~txawsteam/txaws/trunk for your ec2 usage01:18
mwhudson>>> b.repository.get_revision(b.revision_history()[0]).parent_ids01:18
mwhudson['mithrandi@mithrandi.net-20090320033718-jxiqtdpm4qru31x2']01:18
mwhudsonthat's a big strange, isn't it?01:18
lifelessspiv: bin/aws-status specifically01:18
lifelessmwhudson: unusual, but we may see it with ghosts in arch conversions01:19
lifelessmwhudson: so not _wrong_. But unexpected with a native originated branch yes01:19
mwhudsonlifeless: right, but i don't think that's what's happened here :)01:19
lifelessmwhudson: I've explained what happened already :)01:19
mwhudsonoh right01:20
mwhudsonwell, i still think it's _probably_ a bzr bug01:20
lifelessour data structures permit it; we don't try to create it and its likely a bug that its being permitted01:21
mwhudsonlifeless: bzr log --short http://bazaar.launchpad.net/~txawsteam/txaws/trunk is quite funny01:21
lifelessregardless, lp and loggerhead should agree about their links. and right now loggerhead *doesn't* agree with bzr about the revno.01:21
mwhudsonlifeless: *bzr* doesn't agree about the revnos01:22
mwhudsonlifeless: if you log --include-merges you get different revnos than if you log --short01:23
mwhudson(and this more or less explains why the scanner and loggerhead disagree, in fact)01:23
lifelessmwhudson: ugh01:23
lifelessmwhudson: we need a bug about this01:23
mwhudsonlifeless: not going to disagree about that01:24
cody-somervilleWhat happens when you update a branch when pushing another branch that is using the former branch to stack off of?01:30
lifelesscody-somerville: you update the branch01:32
cody-somervilleokay :)01:32
* igc offline for a few hours01:47
TheColonialHi guys. I've just installed bzr on my Vista32 box, and can't seem to find the location of bazaar.conf. it's not where it says in the docs.02:39
lifelessit doesn't exist by default02:40
lifelessbzr --version will list the path that bzr will be looking for it in02:40
TheColoniallifeless: i see. so where does it store my user info when i use "bar whoami ..." ?02:40
TheColoniallifeless: thanks that's helpful!02:40
lifelessspiv: alive?02:43
spivlifeless: yep02:57
spivlifeless: just got a couple of vaccinations, so alive but with two sore arms :)02:58
lifelessouch03:00
lifelessflu?03:00
TheColonialhow do i get the difftools plugin to point at my external diff tool?03:00
lifelessTheColonial: I'm not sure sorry; is there a README in it? or perhaps bzr help difftools?03:00
TheColonialbzr help difftools was what i was looking for. thanks again. sorry ffor the noob questions03:01
spivlifeless: yeah, and tetanus/whooping cough/diptheria.03:02
lifelessthe flu one had my arm sore for about 3 days, couldn't type one of them03:06
spivInteresting.  So far the tetanus arm is hurting more, but early days...03:07
wgrantTetanus always hurts most, IME.03:08
spivYeah, that was my memory from 10 years ago.03:10
lifelessis tetanus a live virus, or deactivated?03:13
lifelessanyone know if PEP 383 considers normalisation issues?03:15
mwhudsontetanus isn't a virus at all, is it?03:21
lifelessGram-positive, obligate anaerobic bacterium Clostridium tetani03:22
lifeless^ apparently03:22
spivThis vaccine contains "tetanus toxoid".03:25
=== abentley1 is now known as abentley
johnjosephbachir'ello03:40
johnjosephbachirif i have A is a branch of B is a branch of C, and I want to make A a branch of C, how do I do that?03:42
SamByou could do pull --remember C, I guess03:43
mwhudsonjohnjosephbachir: i'm not really sure what that means03:45
mwhudsonjohnjosephbachir: A is already a branch of C, in most reasonable definitions of the word03:46
lifelessjohnjosephbachir: if you want a direct relationship, as SamB says, pull --remember C  (or merge --remember C)04:09
johnjosephbachirSamB, mwhudson, lifeless -- appologies, got pulled away for something -- will be back later, thanks for your answers04:09
lifelessspiv: ping04:45
TheColoniallifeless: I see you're an Aussie. Used to use internode myself, pretty nice ISP. Where you based?04:59
lifelessI live in Sydney04:59
TheColonialDid university there05:00
=== jfroy is now known as jfroy|slave
TheColonialso what makes you a fan of bzr as compared to other dvcs?05:00
=== jfroy|slave is now known as jfroy|worker
spivlifeless: pong05:02
lifelessTheColonial: well, I'm one of the core developers/architects05:04
TheColoniallifeless: didn't know that :) sorry.05:04
lifelessTheColonial: nothing to be sorry about05:04
lifelessbut it means my answers are not typical :P05:04
TheColonial i noticed ;)05:04
TheColoniallifeless: I've only played with it a little, and doday is the first time i've had a go. it certainly seems nice. output is certainly a lot easier to digest than in the likes of git and hg.05:05
johnjosephbachirlifeless, SamB: after pull --remember, after the next time i push, will the parent branch for the remotely hosted branch also be changed?05:05
lifelessjohnjosephbachir: the parent branch won't, but it won't have a reference anyway05:05
lifelessTheColonial: glad you're liking it05:06
lifelessspiv: if you look at my most recent patch, I need to get the token for an already locked repo; I couldn't see an obvious api for that05:06
TheColoniallifeless: me too :) So is bzr something you work on in your own time? of so,what do you do full time?05:06
lifelessbzr is what I'm paid to work on at the moment05:07
spivlifeless: I think token = r.lock_write(); r.unlock() is all we have05:07
lifelessspiv: thats ~= what I did ;). Anyhow - 9 round trips05:07
spiv:)05:07
lifelessspiv: also did you see the ec2 status gui I wrote?05:08
TheColoniallifeless: excellent stuff. well i wish you good luck. thanks for the help. i shall continue to play with it.05:09
lifelessTheColonial: thanks :). Just ask if you have any questions05:10
spivlifeless: no, sounds shiny though!05:11
TheColoniallifeless: will do :)05:15
lifelessspiv: we probably need to examine some of the other ACF reports05:16
lifelessspiv: see if there are other causes05:16
spivlifeless: I've been looking at them as they come in, so far there's no solid evidence of other causes that I can see.05:19
spivBut it's hard to rule out.05:19
spivThese bugs where cause and symptom happen far apart are PITA to diagnose.05:20
BasicOSXbzr tags works for anyone in r4307?05:21
=== bob2 is now known as Guest5952
SamBcrash early and often, I say!05:22
spivBasicOSX: seems to for me05:22
BasicOSXI get nothing :-(05:23
spivWell, that's expected if you're running it on the bzr.dev branch (unfortunately).05:23
BasicOSXjust tried it on fresh check out of 1.14 too05:23
spivYes, the PQM-managed branches don't have tags.05:24
spivI'm not sure why; it's a known issue that hasn't been tracked down yet.05:24
BasicOSXoh!05:24
BasicOSXI thought I really borked my local stuff :)05:25
BasicOSXthanks for the quick answers05:25
fullermdI blame Celine Dion, myself.05:25
SamBlol05:25
SamBthat's a pretty strange person to blame05:25
fullermdThat's exactly why she thought she could get away with it!05:26
BasicOSXspiv:  how do I check out 1.13.1 if there are no tags in PQM managed branches :-)05:26
spivIf you can't blame a former winner of the Eurovision Song Contest who can you blame?05:26
BasicOSXWell, if you are American, there is always the Republican Party05:27
fullermdYeah, but then the other guys feel all left out.  It's so much work being sure to balance the blame evenly all the time.05:28
SamBI don't think the Republicans are smart enough to cause such a problem!05:28
SamBdemocrats neither!05:28
spivBasicOSX: by reading the bzr log of the 1.13 branch I guess.05:29
BasicOSXok, that is what I was doing, thanks.05:30
* spiv -> lunch05:31
=== Guest5952 is now known as bob2
johnjosephbachirlifeless: when i do bzr info <remote url>, it reports a parent branch. that parent branch is going to disappear from existence... shouldn't i change what it points to?05:51
fullermdjohnjosephbachir: Depends on why you want to change it.05:52
fullermdThe parent branch is mostly cosmetic.  It's used as the default for a few commands like 'pull'.05:52
johnjosephbachirfullermd: well for one thing, it will be a bit meaningless/useless after that parent branch disappears05:52
johnjosephbachirfullermd: well exactly, i want to change the default pull location : )05:53
fullermdBut it doesn't imply any deeper data-model-ish link between the branches.05:53
fullermdWell, you just use --remember next time you pull there.05:53
johnjosephbachirand for the rest of the life of the project? for years and years the default pull location won't exist?05:53
johnjosephbachirseems like there should be a solution for this...05:53
fullermdNo, when you use --remember it'll be changed.05:53
johnjosephbachirremotely?05:54
* fullermd suspects we're talking past each other.05:54
johnjosephbachiri mean, for years and years, contributors for the project will have to change the default pull location...05:54
johnjosephbachirhaha05:54
johnjosephbachirperha05:54
johnjosephbachirss05:54
fullermdHuh?  No.05:54
fullermdPull locations are ENTIRELY local.  The pull location on $REMOTE_BRANCH is of meaning only to $REMOTE_BRANCH.  It doesn't mean anything to any other branches you make based on it.05:54
johnjosephbachirfullermd: look at the output of: bzr info bzr+ssh://bazaar.launchpad.net/%7Ejohnjosephbachir/lyceum/trunk05:55
fullermdSo if you "cd $REMOTE_BRANCH ; bzr pull", the parent loc matters.  If you "bzr branch $REMOTE_BRANCH local ; cd local && bzr pull", the $R_B's "parent" doesn't mean anything.05:55
fullermdRight, but you never ssh to launchpad, cd into that branch, and run 'bzr pull'.05:55
johnjosephbachirahh, okay05:55
johnjosephbachirgotcha05:55
spivThe parent location does not propagate.  e.g. When contributors branch from your branch (say lp:~johnjosephbachir/lyceum/trunk), then their branch will get lp:~johnjosephbachir/lyceum/trunk as their parent location.05:56
fullermdHow did it get a parent loc in the first place?05:56
fullermdOh, I see.  That's LP internal bits.05:56
johnjosephbachirspiv: that's what i thought, but was also wondering what the meaning of parent branch was05:57
fullermdI guess you COULD go in with sftp and edit the branch config file manually if you really wanted to.  But it wouldn't affect anything to you from the outside.05:57
fullermdAnd LP may assign internal meaning to it, in which case it would probably be Bad(tm).05:57
mwhudsonpull -d lp:whatver whatever-else --remember would also change it, i guess05:57
mwhudsonfullermd: LP doesn't, for sure05:57
* johnjosephbachir nods05:58
fullermdWell, I see a URL that starts "lp-hosted:///", I run the other direction from touching it  :)05:58
spivfullermd: it's actually just an accident of how LP publishes hosted branches, I think.05:58
fullermdjohnjosephbachir: Anyway.  In a technical sense, the parent branch only matters when you actually cd into that branch and do various ops.  You won't in this case, so there's no technical reason to change it.05:59
spivfullermd: to the extent that the owner of the branch can't edit the relevant branch.conf...05:59
johnjosephbachirk05:59
fullermdjohnjosephbachir: You may WANT to change it for cosmetic reasons, but there's no in-bzr reason to.05:59
spivmwhudson: LP probably shouldn't be cluttering the branch info with that though.05:59
johnjosephbachirfullermd: yeah, to avoid user confusion06:00
mwhudsonspiv: hey, launchpad is just calling .clone!06:00
johnjosephbachirmwhudson: that command-- is that done on the server, or do you mean, i can do that remotely?06:00
mwhudsonjohnjosephbachir: i think you can do it remotely06:00
spivmwhudson: sure.  The end result is still a bug though :P06:01
mwhudsonspiv: i guess06:01
mwhudsonspiv: want to file it?06:01
spivSure.06:01
* fullermd stabbies all the progress bar fragments left on his terminal from that 'info'...06:01
* johnjosephbachir reads about the -d flag on pull06:01
johnjosephbachirOIC06:01
johnjosephbachirspiv, mwhudson: a bzr bug, or a launchpad bug?06:02
mwhudsonlaunchpad-bazaar i guess06:03
spivFiled: https://bugs.launchpad.net/launchpad-bazaar/+bug/36838306:06
ubottuUbuntu bug 368383 in launchpad-bazaar "Public copies of hosted branches have nonsense parent location" [Undecided,New]06:06
lifelessmmmm, not sure I agree06:11
lifelessstill, worth noting it caused confusion06:11
johnjosephbachirwell, i know my newbie question was a good once, since it caused several veterans to embark on intense debate and file a bug report :-D06:22
johnjosephbachirs/once/one06:24
lifelessback soon, popping up to chemist06:51
vilahi all06:59
lifelessspiv: I'm done for the day; 3 patches up08:00
spivlifeless: :)08:04
BasicPROPQM 1.13.2, success, attempting to re-pull bzr.1.13 and I get this: http://pastebin.com/m61bd628c08:04
spivlifeless: I guess I should do more reviewing then :)08:04
lifelessspiv: 9 rtt :)08:04
lifelessspiv: and 15 for stacked08:04
spivBasicPRO: Hmm.  That command worked for me.08:08
BasicPRO bzr branch http://code.launchpad.net/~bzr/bzr/bzr.1.13  bzr-1.13.2 ?08:08
spivRight.08:08
BasicPROwhich version of bzr? :-)08:08
BasicPRO$ bzr --version08:09
BasicPROBazaar (bzr) 1.15dev08:09
BasicPRO    revision: 430708:09
spivBasicPRO: same08:09
BasicPROhmm, /usr/bin/python 2.5.2 ?08:10
lifelessshared repo?08:10
spivAlthough I do notice that that branch has the AbsentContentFactory bug!08:10
BasicPROlifeless:  http://code.launchpad.net/~bzr/bzr/bzr.1.1308:11
BasicPROtrying it on pristine hardy+updates08:13
spivBasicPRO: Python 2.6.2 (jaunty)08:13
spivBasicPRO: are you branching into a shared repo or standalone?08:14
BasicPROpython revision should not have anything to do with it? standalone08:14
BasicPROissuing  bzr branch http://code.launchpad.net/~bzr/bzr/bzr.1.13   bzr-1.13.2 and bzr-.13.2 does not exist on local filesystem08:15
spivYeah, python version should be irrelevant.08:16
lifelesswe need to get pqm running 1.14 final08:16
spivBasicPRO: is there a local shared repo?08:16
spivlifeless: yes please!08:16
lifelessbut thats not the issue08:17
lifelessthe issue is probably that lp's mirror-branch-pull makes bad branches08:17
lifelessmwhudson: ^08:17
spivI think LP is planning to upgrade to 1.14 more-or-less as soon as its out.08:17
=== BasicPRO is now known as BasicOSX
lifelessfor now foo.py on the 1.13 branch is probably a good idea08:18
davidstraussIs there a release date scheduled for 1.14 yet?08:19
BasicOSX"soon" :-)08:19
davidstraussBasicOSX: Is there a listing of current blockers?08:19
BasicOSXjust 1, my time08:19
davidstraussBasicOSX: So there aren't any critical bugs left?08:21
BasicOSXnone that I know of, but that normally does not get revealed until I put the last call for merge for release to the mailing list08:22
davidstraussBasicOSX: lol08:22
lifelessBasicOSX: Just Do It08:25
BasicOSXlifeless:  ok, I'd like to get 1.13.2 out before switching gears08:25
davidstraussBasicOSX: I'm asking because I'm eager to use the new join support for vendor branching.08:25
spivBasicOSX: can you pastebin the traceback from your .bzr.log for that error?  (Or use -Derror)08:31
spivBasicOSX: it probably won't tell us much, but just in case...08:31
spivBasicOSX: (but don't let that stop you from doing the release!)08:31
BasicOSXI opened bug report, it's all in there Bug 36841808:34
ubottuLaunchpad bug 368418 in bzr "ERROR: Revision X not present" [Undecided,New] https://launchpad.net/bugs/36841808:34
mtaylorGAAAAAAAAAAAAAAAHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH08:36
mtaylorDAMMMMMMMMMMMIIIIIIIIIIIIIIIIIIIIIIIIITTTTTTTTTTTTTTTTTTTTTTTTT08:36
mtaylorlifeless: why does revert with no arguments not present a warning?08:37
lifelessmtaylor: because its safe?08:37
* wgrant points out that 'bzr revert' backs things up first.08:37
mtaylorlifeless: um, no08:37
lifelessmtaylor: it backs everything up08:37
mtaylorit does?08:37
mtayloroh thank god08:37
wgrant*.~N~08:37
lifelessmtaylor: except for things that are unaltered08:37
wgrantIt's not like Subversion.08:37
mtaylorhow do I un-revert08:37
lifelessmtaylor: thats a little harder, [its not automated]08:38
mtaylorlifeless:  that's fine ... I just reverted about 3 hours of hacking08:38
lifelessmtaylor: also, if you want it to be able to be configured to warn, like rm -rf can be, that would be fine08:38
mtaylorin the wrong window08:38
lifelessbut it shouldn't be the default, its an unbreakme option :)08:38
mtaylorlifeless: that would be nice... this is not the first time I've reverted08:38
mtayloragree08:38
lifelessbecause, if it did ask, you'd hit Y by muscle memory and have the problem anyway08:38
mtaylorso... are there docs somewhere as to how to un revert?08:38
lifelesswell you have the list of changes in the output from revert08:39
lifelessjust look for those backup files and put em back08:39
mtaylorlifeless: gotcha. rock!08:40
mtaylorlifeless, wgrant: you have saved my life (as usual)08:47
lifelessanytime (that I'm awake :P)08:50
davidstraussreverts backups are so annoying08:55
davidstraussrevert's*08:55
lifelessyou can configure it I think08:55
davidstrausslifeless: I guess I could alias to no-backup08:56
davidstrausslifeless: but i much prefer the shelve all workflow08:56
davidstrausslifeless: and then clearing the shelf08:56
bob2wtf09:05
bob2svn obliterates uncommited working tree changes on up09:05
BasicOSXspiv:  1.14rc2 branches just fine, I don't feel comfortable releasing 1.13.2 (2nd regression) unless I can get some dev feed back that the issue is local to just me09:08
spivBasicOSX: I'm about to update the bug09:09
spivBasicOSX: I can reproduce too when I branch to a standalone branch.09:09
BasicOSXok, well, 3am, real job in 5h, so bed time for me, will work on it again tonight09:10
spivBasicOSX: ok, sleep well09:10
BasicOSXyes, dreams of swine flu :-(09:10
spivBasicOSX: *oinksneeze*09:13
spivBasicOSX: ok, I've narrowed it down some, it's to do with that change I landed in bzr.dev today.09:22
lllamaMorning all. Anyone using bzr with trac? I'm having a weird problem where trac only shows me the files from my last checkin regardless of what branch I'm browsing.09:48
=== thekorn_ is now known as thekorn
yogsotothHi all, I'm trying to push on a mounted webdavfs, unfortunately, I believe somethings goes wrong with the latency...10:34
yogsotothKnowing rsync works correctly but not cp -R10:35
yogsotothis there someone who managed to work with webdav?10:35
bob2there's a webdav plugin for having bzr push via wedav10:36
yogsotothI know, but there is a bug with iDisk10:37
yogsotothAnd I can't use it10:37
yogsotothThis is why I tried to mount directly the webdav disk10:38
yogsotothBut with no more success :-(10:38
lifelessBasicOSX: its not a problem with the tip, its the other ongoing issue we're debugging10:44
lifelessBasicOSX: I suggest not blocking on it10:44
lifelessnight all13:49
cdecarlohey, I want to use trac with bzr, but I don't want to install the .deb b/c the package manager want to install alot of other packages, any idea how I can just install trac-bzr14:07
LeoNerdAre they actual requirements, or just recommends?14:10
LeoNerdIIRC apt now installs recommends by default.. you can ignore them14:10
mrbrocolidoes bzr support something like svn:external?15:10
Peng_mrbrocoli: Not yet.15:11
Peng_It's been sort of partially experimental since...1995?15:11
mrbrocolihehe15:11
beunomrbrocoli, abentley is working on finishing the implementation15:12
beunoI hear that it will be widely available in the next release or two15:13
beunohi Peng_   :)15:13
Peng_Hello. :)15:13
mrbrocolinice, thanks for the info!15:13
Peng_Wait, "nosmart+lp" works? I totally didn't know that!15:47
=== jelmer is now known as Guest30662
=== jelmer is now known as Guest10759
=== Guest10759 is now known as jelmer
=== ja1 is now known as jam
=== thekorn_ is now known as thekorn
=== jelmer is now known as Guest94894
=== dereine[OFF] is now known as dereine
dereinewhats the best way to work on 2 seperate features  at the same time16:54
Peng_Two separate branches?16:55
dereineah never thought of this16:55
dereinethx!16:55
Ngsay I had a branch for project/ and a separate branch for project/docs, with no common ancestor, is there any way to consume docs into the project branch without losing all the history?17:33
Peng_Ng: "bzr join". You might have to upgrade to an experimental disk format, or you might not. I don't remember.17:34
Nginteresting17:34
Nghow experimental? will any puppies be endangered? ;)17:35
Ng(and could the resulting branch merge back into a non-experimental one?)17:35
james_wNg: it would need to be rich-root version of a format, of which there are vegetarian varieties. However, they are a trap-door format (with good reason), so they couldn't be merged back17:38
james_wthere is a merge-into plugin that John wrote to do this17:38
james_wit doesn't require a format bump, but works best at one-off conversions17:38
Ngjames_w: thanks :)17:40
Peng_rich-root formats are not experimental. Everyone'll need to upgrade eventually anyway.17:43
=== beuno_ is now known as beuno
=== dereine is now known as dereine[OFF]
=== dereine[OFF] is now known as dereine
Peng_mwhudson: BTW, Loggerhead memory usage is staying far lower than it used to; like 60 MB. However, it still creeps up slightly over time. That may be okay or it may not; I dunno.18:26
=== dereine is now known as dereine[OFF]
Peng_Running "bzr check" after upgrading bzrtools to rich-roots, there were 23 inconsistent parents there weren't before.19:59
=== dereine[OFF] is now known as dereine
=== dereine is now known as dereine[OFF]
=== dereine[OFF] is now known as dereine
=== dereine is now known as dereine[OFF]
=== dereine[OFF] is now known as dereine
=== dereine is now known as dereine[OFF]
=== dereine[OFF] is now known as dereine
mwhudsonPeng_: good to hear, thanks21:16
Peng_:)21:18
=== beuno_ is now known as beuno
mwhudsonBasicOSX: is 1.14 out yet? :)23:22
=== dereine is now known as dereine[OFF]
BasicOSXmwhudson:  ran into bug with bzr.dev and ghosts which slowed me down, but lifeless  said don't let it block me, so working on it right now23:54
BasicOSXbummer make dist-check fails, investigating23:55

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