/srv/irclogs.ubuntu.com/2008/07/19/#bzr.txt

daugustine_laphello, i want to set up a bazaar server with public checking out but then authenticated committing... is this possible? and if so, how do you set up the users, can they be the users already on the os?00:29
Pengdaugustine_lap: Usually you would use http for anonymous checkouts, then bzr+ssh for. committing.00:31
daugustine_lapalright... and that uses the current system users cause it goes through ssh?00:32
PengYep.00:32
daugustine_lapalright, thanks00:32
alex-weejis there something like giggle/gitk for bazaar?00:41
datoalex-weej: `bzr viz` from bzr-gtk00:47
alex-weejoo00:48
* alex-weej checks it out00:48
awilkinsRight, I'm off on my holidays, back in 2 weeks01:03
=== bigdo3 is now known as bigdo2
=== cprov is now known as cprov-zZz
ozzloyi want to push to a windows 2k3 server.  what's the minimum stuff i need to install to do that?01:54
ozzloyon the win 2k3 server that is01:54
ozzloyit doesn't even have ftp installed right now01:54
ozzloyand if i can, i'd like to avoid.  is there a way to use bazaar itself to listen for connections?01:55
bob2you can use ssh/sftp/webdav(experimental), too01:57
bob2not sure if bzr:// itself has any auth yet01:59
ozzloybob2: what if i'm ok that it doesn't have auth?01:59
ozzloyhow do i make it so i can go bzr push bzr://192.168.1.10002:00
bob2then you can just run 'bzr serve' to allow read-only access02:00
ozzloyoh, so bzr:// without auth won't write02:00
ozzloyok then.  i'll see aboutsome other method02:01
bob2no, bzr help serve02:01
bob2I just don't really want to suggest something so insecure :)02:01
ozzloyit's insecure, yes.  but i'm only doing it because my boss doesn't understand that every copy of the repo is completely functional02:02
ozzloyand so having it on 5 or so computers isn't enough backup02:02
ozzloyit has to be on the server that gets "backed up"02:02
ozzloyaha, bzr serve will probably do02:03
ozzloythanks02:03
daugustine_laphow do you get svn+ssh to work..., if anyone could help me out or point me to the right place, i'd appreciate it bunches, i'm having the damnedest time02:04
antoranzHi, Guys!02:09
antoranzHave you heard that bzr has problems running on 64-bit boxes?02:09
bob2no02:09
bob2are you having a problem?02:10
antoranzToday I saw the same kind of problem on two different boxes... one running SLES10 (unpathced, I think) and the other an OpenSuSE 10.2 (pathed... I think)02:10
antoranzpatched, i mean02:10
antoranzI'll get the logs on monday and submit a bug then02:10
antoranzThere were no problems besides those two on the two boxes.... they were both Turion Boxes.02:11
bob2I've not had any issues in 10 months running it on amd64, and I'm pretty sure the whole test suite has to pass on amd64 for every commit to trunk02:12
antoranzcould it be a distribution problem then?02:13
antoranzwhat ditro do you use?02:13
bob2debian and ubuntu02:13
antoranzme two.. and I haven't seen the kind of problem I saw today02:14
antoranz(ubuntu, mostly, by the way)02:14
antoranzanyway.. I'll get the logs on monday and tell you how it was02:14
antoranz(post a bug report, anyway)02:15
cody-somervillebzr check shows:      1 inconsistent parents02:30
cody-somervilleshould I do a bzr reconcile?02:30
Peng_cody-somerville: That will fix it, but it's not like it's a big deal.02:31
spivcody-somerville: yeah.  Although it's not a very serious problem.02:32
Peng_If you've gone to the effort to run check, you might as well reconcile.02:32
cody-somervillePeng, well, I've noticed a performance loss lately and my branch is 46mb while if I export it, the actual stuff only takes up 1.4mb02:34
cody-somervilleWell, 2.4mb to be correct.02:35
antoranzbut export will only export a revision02:37
antoranzthat's not the repository per se02:37
Peng_cody-somerville: One inconsistent parent is not going to give you bad performance.02:38
Peng_cody-somerville: You're using packs, right?02:38
cody-somervilleI just upgraded tonight from knit02:39
cody-somervillemy branch only has 100 some commits but it has quite a bit of branch merging02:39
spivAfter an upgrade you might have a bit of junk in .bzr/repository/obsolete_packs.  You can save some space by emptying (not deleting!) that directory.  That won't make any difference to performance though.02:41
Peng_cody-somerville: You might as well run reconcile, then. I don't think it will take long.02:42
cody-somervillespiv, that'll improve initial branch times, no?02:43
pickscrapeWeird. My bash prompts is suddenly slower and shows bzr branch details if I'm in a branch directory.02:43
cody-somervillepickscrape, mine too :(02:43
pickscrapeSomething new in 1.6?02:44
cody-somervillewow!02:45
cody-somervilleI02:45
cody-somervilleAfter doing the reconcile, bzr branch flew! :)02:45
Peng_pickscrape: Yeah. It was an accident that it got installed.02:45
Peng_pickscrape: Delete /etc/bash_completion.d/bzrbashprompt.sh to get rid of it if you want to.02:46
pickscrapeIt's would actually be a pretty cool feature if it didn't cause such a drag on general bash performance02:46
Peng_Agreed.02:47
pickscrapeAnyone know if 1.5 for hardy will make it back into the PPA before Monday? We're about to migrate to bzr so that's a complication I wasn't planning on having to deal with :)02:48
pickscrapeI suppose they could just try specifying feisty in sources.list02:49
Peng_pickscrape: Well, using 1.6b3 shouldn't cause any problems (except with plugin compatibility.)02:49
pickscrapeAnd constantly complaining about getting the server upgraded :)02:50
Peng_Heh, that too.02:50
spivcody-somerville: no, it won't affect initial branch times, the obsolete_packs are basically ignored by everything.02:52
cody-somervillePeng_, spiv: Well, regardless, now that I've done the reconcile the branch took 43 seconds (and that included me entering my password) in comparison to 941.563 seconds02:54
Peng_What took 941 seconds?02:55
cody-somervillePeng_, initial branch02:57
cody-somervillefrom launchpad02:57
Peng_Ooh.02:58
docgnomeI'm have a bzr repo on my desktop that I keep all my emacs settings in. mostly I'm using it like a glorified rsync. I want to whack that machine though. If I unbind one of my checkouts on another box, rebuild my desktop machine and do a checkout on it, can I tell my other checkouts to rebind to the new place?02:59
bob2as long as you remember to commit and update03:02
docgnomeon the checkout?03:05
bob2commit on the desktop03:06
bob2and udpate on the other machine03:06
docgnomeah yeah.03:07
cody-somervilleHow do I revision control bzr hooks if they exist under .bzr?03:16
Odd_Blokecody-somerville: You want the hooks to be part of the branch's history?03:30
cody-somervilleOdd_Bloke, well, I want the hooks to be downloaded with the branch so that they're always run03:30
Odd_BlokeWell, off the top of my head, I'm not sure of a good way to do that.03:31
Odd_BlokeI'd suggest posting something to the ML or asking a question on LP.03:31
pickscrapeI think there are security implication with that03:32
Odd_BlokeAs it's 3:30am here, I'm going to bed. :)03:32
Peng_You can set up something so "make hooks" or whatever installs them.03:33
Peng_And tell people to run it.03:33
Odd_BlokeThat seems reasonable.03:33
Odd_Bloke*GONE*03:33
Peng_Odd_Bloke: Good night.03:34
docgnome/part/part03:39
docgnome...03:39
docgnomefail03:39
Peng_I love when people fail to exit. It gives you a chance to get a last word in.03:45
alfonsodgHello05:53
alfonsodgi have a problem....05:54
alfonsodgwhen i try to push a content in a new branch with: bzr push bzr+ssh://alfonsodg-pe@bazaar.launchpad.net/~alfonsodg-pe/edukt/main05:55
alfonsodgi receive a message: bzr: ERROR: Transport operation not possible: readonly transport05:55
alfonsodgany ideas?05:55
YurimHi, I just shot myself in the foot with bzr. Is there anybode awake willing to help me reanimate my branch?  After some hours of work, i moved files from a subdirectory into the parent directory, removed the subdir and got now "ERROR: exceptions.AssertionError: Could not find target parent in wt: w3mimg/fb"05:59
Yurimnow i can't branch from this dead branch and "bzr st" raises an exception. how do i get the last revision?06:01
* Yurim sighs.06:06
Peng_Yurim: This isn't great, but if you rename .bzr/checkout/, bzr should forget that you have a working tree.06:13
Peng_Yurim: (Don't delete it though.)06:13
Peng_Yurim: Are you using bzr 1.5 or 1.6b3?06:13
=== sdboyer__ is now known as sdboyer
Yurim@peng: ah, somebody listening. bazaar 1.3.1 from ubuntu06:14
Yurimi can reproduce the bug06:14
Yurimmkdir -p test_repo/subdir1; touch test_repo/subdir1/file1; cd test_repo; bzr init .; bzr add .; bzr commit -m "initial commit. 1 subdir, 1 file"; rm subdir1/file1; bzr rm subdir1; bzr st06:14
Peng_Yurim: I can confirm that with bzr.dev. You should file a bug, if there isn't one already.06:16
Yurimok, i'll do that immediately06:16
Peng_I think I remember hearing about something like that a few months ago.06:18
Yurim@Peng_ Thank you very much. Removing .bzr/checkout was the thing i needed.06:19
alfonsodgsorry .... any ideas about my problem?06:20
Peng_"bzr branch" stopped working too? I wonder why it cares about the working tree?06:20
Peng_alfonsodg: Err.06:20
Peng_alfonsodg: Sorry, I hadn't noticed your question. If you're sure you're using bzr+ssh, I dunno.06:22
Peng_alfonsodg: Are you using a modern version of bzr?06:22
alfonsodgbzr 1.3.106:22
alfonsodgPeng_: included with ubuntu hardy06:23
alfonsodgPeng_: i must to upgrade, right?06:23
Yurimright, "bzr branch" stopped working too. that was the first thing i tried and after i saw the exception i began to panik06:23
Peng_alfonsodg: It would be a good idea to upgrade, but I don't know if it will help here.06:24
bob2have you tarred up the whole thing now so fddling won't break it further?06:24
Peng_alfonsodg: (You can use the apt repo -- https://launchpad.net/~bzr/+archive )06:25
Peng_Does the PPA seriously not have a version of bzr for Hardy right now? Oops.06:26
alfonsodgPeng_: updating.....06:29
alfonsodgPeng_: sure... i will install from sources.....06:30
Yurim@bob2 sorry, i didn't see your question. no i haven't but after copying the whole directory and removing .bzr/checkout i could branch from this new directory and copy my changes from the old one. everything's back to normal again06:46
alfonsodgPeng_: nothing... still the problem.......06:50
Peng_alfonsodg: Can you push anywhere else? Are you sure you aren't accidentally using http? What's the exact command you're running?06:51
alfonsodgbzr push bzr+ssh://alfonsodg@bazaar.launchpad.net/~alfonsodg/edukt/main/06:52
alfonsodgbzr: ERROR: Transport operation not possible: readonly transport06:52
alfonsodgPeng_: with this works: bzr push sftp://alfonsodg@localhost/~/example_dok06:53
alfonsodg bzr push sftp://alfonsodg@bazaar.launchpad.net/~alfonsodg/edukt/main/06:54
alfonsodgbzr: ERROR: Permission denied: "/~alfonsodg/edukt/main": [Errno 13] mkdir failed06:54
Peng_alfonsodg: Can you push to any other branches on Launchpad?06:54
Peng_Ah-ha!06:54
Peng_alfonsodg: That's a mirrored branch. Push to your server, not Launchpad.06:55
Peng_alfonsodg: https://code.launchpad.net/~alfonsodg/edukt/main says it's a mirror of http://cosperu.com/EduKT , which doesn't exist. If you didn't intend to make it mirrored, it's easy to change.06:55
alfonsodgPeng_: thanks.... in fact, the problem was in the launchpad side.....07:02
lifelessmoin07:39
Peng_Good morning.07:39
Jc2kmoin07:41
* Peng_ wonders why a pull of bzr.dev has been running for a couple minutes.07:46
Peng_Still going . . .07:56
Peng_If I'm DoSing bazaar-vcs.org, well, it's your fault. :P07:56
PengOh, good, it exited with "No revisions to pull" after 1030.762 seconds.08:25
PengMan, Mercurial's really got you guys beat. :P08:25
RAOFThat seems to be unusually poor performance for a noop.08:26
PengYes, yes it does.08:26
RAOFOh, I'm pulling from launchpad, which may make a difference, but mine is 17s08:27
PengI know it's weird. It's usually only a few seconds.08:27
bob2is your local branch the same format as /bzr/bzr.dev?08:27
lifelessPeng: wtf08:27
lifelessPeng: can you grab your ~/.bzr.log for the session and file a bug?08:28
fullermdPshaw.  hg isn't THAT much faster.  Why, it took me almost 10 seconds to pull 3.5 months worth of mutt changes.08:28
Peng_Haha, "wtf" is a good reaction.08:28
Peng_I just tried again. It took 3.081 seconds.08:29
Peng_lifeless: I was kinda pulling 4 things at once, so my .bzr.log is messy.08:29
fullermdI did have a bzr.dev pull go off into the weeds once a month or two back.  It spent several minutes using no CPU and doing no network transfer.  Killed it, restarted, and it ran normally.08:30
fullermdI marked it up to "bizarre network crap".08:30
Peng_Heck, I can't even tell which is which. .bzr.log should include the CWD.08:30
Peng_fullermd: Yeah. It was probably something like a connection died and it took forever to time out.08:31
bob2could pull trigger a local repack even if nothing is pulled?08:53
fullermdI don't think so...08:55
Peng_I doubt it.08:55
Peng_If it would, it would've repacked last time it added revisions.08:57
Peng_Also, it didn't trigger one.08:57
lifelessuws: rebase bug fixed in a branch for you09:13
lifelessjelmer: bug is fixed09:32
Peng_Ooh, what bug?09:48
jelmerlifeless: which bug ? :-)10:04
lifelessjelmer: rebase slow10:17
jelmerEMISSINGBUNDLE10:18
Peng_Nice.10:18
jelmer:-P10:18
lifelessjelmer: EATTACHEDTOBUG10:18
lifeless:P10:18
jelmerat least lp didn't mail me about anything being attached to that bug...10:19
lifelesshttps://bugs.edge.launchpad.net/bzr-rebase/+bug/24982310:19
ubottuLaunchpad bug 249823 in bzr-rebase "speed ftw" [High,In progress]10:19
jelmerah, it's a branch10:20
jelmermalone doesn't appear to email if a  branch is attached10:20
* jelmer files bug10:20
lifeless-> conf10:28
lifelessjelmer: where is commit-notify?!11:30
LarstiQlifeless: bzr-gtk?12:52
kiorkyjelmer: any chance, you around?13:08
ChristopheTHi.  One can use "bzr ci --fixes x:id".  And then what?  It is visible with "bzr cat-revision" but can it be searched upon (to answer the question "in which revision was bug x fixed?")?14:29
bob2ChristopheT: not sure if it is exposed by the bzr ui, but launchpad can use it (and maybe trac etc)15:04
Peng_Launchpad uses it. Trac certainly *could*, but who knows if it does.15:22
kiorkyjelmer: it seems the progress function is bugged somehow15:28
kiorkyjelmer: i putted a "return;" at first statement in ra.c:py_progress_function as the first statement and then i was able to check out something15:29
uwslifeless: rock on15:31
alfonsodgPeng_: ping16:55
Peng_alfonsodg: Pong.16:59
Peng_peng ping pong17:01
alfonsodgPeng_: in reference to https://bugs.launchpad.net/bzr/+bug/250022.... what's the procedure for fix a bug?... i found the problem but is my first bug with bzr /launchpad17:01
ubottuLaunchpad bug 250022 in bzr "bzr crashes after directory removal" [Undecided,Confirmed]17:01
Peng_alfonsodg: Branch bzr.dev, start hacking?17:03
Peng_But that might be a big one for someone new...17:03
alfonsodgPeng_: nope...... the code is really easy.... great work... by the way... joining to bzr.dev17:04
Peng_:)17:05
Peng_You shouldn't direct that at me; I'm hardly a developer.17:05
Peng_I fix the occasional typo in the documentation but that's it.17:05
alfonsodgPeng_: ok.... but at least...... need some guides...17:06
Peng_Well, there's http://bazaar-vcs.org/Documentation and the ./doc directory in the source tree.17:07
alfonsodgPeng_: looking.... right... thanks17:07
alfonsodgPeng_: i think that the bug was for bzr-gtk which doesn't have enough information.....17:08
alfonsodgPeng_: my mistake17:08
colbracIn bzrlib: I have a workingtree object and want to merge from a bundle. For a branch you use wt.merge_from_branch(branchobject). Where is the merge_from_bundle? :)19:23
beunoVerterok, mornin'.   If you don't have anything fun to do today, you may want to try upgrading knits -> packs on xmloutput: https://code.launchpad.net/~guillo.gonzo/bzr-xmloutput/trunk19:38
Verterokbeuno: mornin'19:39
Verterokbeuno: ups, I already upgraded it, but only xmlrpc branch :P19:39
* Verterok upgrading trunk19:40
beunoVerterok, :)19:40
* beuno is on a crusade to upgrade knits -> packs everywhere19:42
Verterokbeuno: bzr: ERROR: The branch format Bazaar-NG meta directory, format 1 is already at the most recent format.19:44
beunoVerterok, are you upgrading using sftp?19:59
beuno(you can only upgrade through sftp)19:59
Verterokbeuno: bzr upgrade --pack-0.92 lp:bzr-xmloutput/0.419:59
beunoVerterok, bzr upgrade sftp://guillo.gonzo@bazaar.launchpad.net/~guillo.gonzo/bzr-xmloutput/trunk20:00
Verterokok, thanks20:00
beunowe should really warn people when trying to upgrade via bzr+ssh...20:01
Verterokindeed :)20:01
Verterokbeuno: done, the upgrade just finished. thanks for the tip.20:13
beunoVerterok, cool, one down, 12 to go20:13
Verterok:D20:13
Verterokgood luck20:14
bialixluks: ping20:40
pygiI see you're pushing KDE to bzr :)21:22
lifelesshola22:55
Jc2klo23:05
Mecha25Anybody here have time to help with debugging a BZR upload?  my client says it's uploaded properly, and launchpad says it has the structure, but that the directories are empty23:17
Mecha25I tried reuploading, but it said all files were up to date, even though they're not there23:17
Mecha25 anybody out there?23:18
Mecha25115 people in the room and everybody's afk23:20
Mecha25jeez23:20
cody-somervilleI23:20
cody-somervilleI'm not afk.23:20
Mecha25do you have time to help me debug an upload?  I've spent 2 weeks working on a plug-in for Gnome-do, and it'll be for nothing if I can't upload it23:21
lifelessMecha25: sure23:22
lifelessMecha25: run 'bzr info URL -v" where URL is the place you uploaded to23:23
Mecha25URL including the LP: at the front?23:24
lifelessyes23:24
Mecha25standalone branch, format unnamed23:25
Mecha25and gives me a location23:25
Mecha25I uploaded the code before, it says it worked, but launchpad lists the branch as empty23:25
lifelessMecha25: whats the branch ?23:25
Mecha25there's 2, I'll give you the simple one, I'll do the complicated one later23:26
Mecha25lp:~mechaphoenix25/do-plugins/Grep-Plugin23:26
lifelessMecha25: have you done 'bzr commit' ?23:28
Mecha25no23:28
Mecha25bzr commit lp:~mechaphoenix25/do-plugins/Grep-Plugin ?23:29
lifelessMecha25: if you haven't committed there is nothing to upload :)23:29
lifelessMecha25: three key commands:23:29
lifelessMecha25: bzr st23:29
lifelessbzr diff23:29
lifelessbzr commit23:29
Mecha25ok23:29
lifelessst shows you what will be added/deleted/renamed/altered etc23:29
lifelessdiff shows you the content changes of each file23:29
Mecha25I've never used a version control system before, I know how to use diff though.  Thanks, it's been really confusing so far23:30
Mecha25so I need to   bzr commit lp:~mechaphoenix25/do-plugins/Grep-Plugin ?23:30
Mecha25from the directory containing the code?23:30
lifelessjust 'bzr commit'23:31
lifelessafter you use 'bzr st' and 'bzr diff' to be sure its going to commit what you want committed23:31
lifelessafter teh commit, you can do 'bzr push'23:31
Mecha25ok... did the commit, and it threw me into vim23:31
lifelessright23:31
Mecha25exit it?23:31
lifelesswrite a message describing your commit23:32
Mecha25alright, then bzr st23:32
Mecha25and bzr push lp:... etc23:32
Mecha25?23:32
lifelessyes23:32
Mecha25alright, one sec23:33
Mecha25it says it pushed the revision23:34
Mecha25ok, now what does a commit actually mean?23:34
lifelessa commit records a directory structure and the content fo the files within it23:34
lifelessa bit like a tar file23:34
Mecha25ok, then what does bzr add do?23:34
Mecha25I had to do that first I think23:34
lifelessbzr add selects files to have them be recorded23:35
lifelessfiles that have not been added are 'unknown' to thbzr23:35
lifeless*bzr823:35
Mecha25ok23:35
Mecha25so I create the branch, add the files, commit the tree, then push it?23:35
lifelessyes23:35
lifelessand then you can make further changes23:35
Mecha25um... I did the push, it's still showing up as empty23:35
lifelesscommit the tree again and push again23:35
lifelessMecha25: run bzr log URL23:36
lifelessMecha25: where URL is the lp url23:36
lifelessand look http://bazaar.launchpad.net/~mechaphoenix25/do-plugins/Grep-Plugin/23:36
Mecha25says it uploaded, has my message23:36
lifelessits all operating as normal23:36
Mecha25um.... it still says empty on my end23:37
lifelessMecha25: what do you mean23:37
lifelessuws: please let me know if the performance issue in rebase is better23:37
Mecha25refreshing the page says it's still empty, but the link you sent me worked23:37
Mecha25weird23:38
Mecha25your link is what I'm viewing, and it says it's empty when I view it normally, and working when I view it from your link23:38
Mecha25odd23:38
lifelesswhat do yo umean 'view it normal'23:38
Mecha25no, wait, you're on bazaar.launchpad, I'm on code.launchpad23:39
Mecha25I mean go in from my launchpad profile23:39
lifelesscode.launchpad.net takes a few minutes to update its history listing normally23:39
Mecha25d'oh23:39
Mecha25alright then, I can wait23:39
lifelessthere is a issue right now I think, something backlogged23:39
lifelessbut it has worked23:39
Mecha25wahoo!  I'll upload the other one, and chuck it on the mailing list23:40
Mecha25thanks a ton, this open source development thing is finally paying off23:40
Mecha25peace23:40

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