[00:17] <Jerky_san> is there a way for when you do a push to set what permissions you want the files to have?
[01:09] <markh> if you have a bundlebuggy account, does it offer a simple way to find/filter all of your own requests?
[01:13] <lifeless> yes
[01:13] <lifeless> mine
[01:13] <lifeless> and my pending
[01:14] <markh> how do I get an account then? :)
[01:14] <markh> mail aaron?
[01:17]  * markh tries to get a few off his 'resubmit' queue...
[02:41] <Peng_> markh: You rock for working on the "up -r" thing. :)
[02:41] <Peng_> (That was you, right?)
[02:42] <fullermd> Mmph.  Does anybody else see 'missing' over bzr+ssh being a lot slower with 1.6 than 1.5?
[02:52] <fullermd> Heck, missing took longer than pull.
[02:55] <jam> fullermd: "time bzr1.6 missing" 0.663s, "time bzr1.5 missing" 2.766s
[02:55] <jam> Seems faster to me
[02:55] <jam> it does use a different algorithm, so there are cases where it is slower
[02:55] <fullermd> Is that against a 1.6 server?
[02:55] <jam> fullermd: this is localhost stuff
[02:56] <jam> and 3 extra revisions locally
[02:56] <fullermd> Well, I'm talking about with matching servers.  1.6 client was REAL slow against a 1.5 server, since it stopped using some verb or other.
[02:56] <jam> fullermd: for "bzr missing bzr+ssh://" I get 6.8s bzr1.6,  9.9 for bzr 1.5
[02:57] <fullermd> This is across 100ms.  1.5 on both ends used to be <10s, for much larger than the 3 revs I just got in 20s with 1.6.
[02:57] <jam> no verbs changed for missing, that I know of
[02:57] <fullermd> Oh, there was something.  1.6 would churn for 30 seconds or so before it started showing anything, then list the revs at the same speed.  1.5 went right into the listing.
[02:58] <jam> fullermd: are you sure about that, 1.5 spends a long time grabbing the full ancestries before it continues
[02:58] <fullermd> Against a 1.6 server?
[02:58] <jam> fullermd: against all
[02:58] <jam> there is no smart missing functionality
[02:58] <jam> it loads the whole graph
[02:58] <jam> and then does set ancestry diffs
[02:58] <jam> in 1.5
[02:59] <jam> the only thing I could possibly see
[02:59] <jam> is if loading the whole graph was fast
[02:59] <jam> but that is not usually the case
[02:59] <fullermd> Well, the other week, the server was running 1.5.  Missing ~20 revs, 1.5 listed them out in 15 or 20 seconds.  1.6 churned for upwards of 30 seconds before it started listing them in (in 15 or 20)
[02:59] <fullermd> That was repeatable over multiple runs, back and forth.
[02:59] <jam> well 1.6 *searches* the graph, 1.5 just loads it and does set diff
[03:00] <jam> it is possible for the search to take longer
[03:00] <jam> generally depends on the amount of extra ancestry you have
[03:00] <jam> stuff "around" the search locations
[03:01] <jam> So... in all the common cases I've tried, 1.6 is faster, but maybe every 50th try or so 1.5 is faster
[03:01] <jam> same thing for "bzr status" showing pending merges
[03:05] <fullermd> % time \bzr missing
[03:05] <fullermd> 2.508u 0.760s 0:28.12 11.5%     212+3647k 0+0io 0pf+0w
[03:05] <fullermd> (1.6 both ends)
[03:05] <fullermd> % time env BZR_REMOTE_PATH=~/src/bzr-1.5/bzr ~/src/bzr-1.5/bzr missing
[03:05] <fullermd> 1.799u 0.599s 0:11.23 21.1%     215+3123k 0+0io 0pf+0w
[03:05] <fullermd> (1.5 both ends)
[03:06] <fullermd> There are 2 missing revs.
[03:06] <fullermd> In both casts, it takes about 4 seconds to display them after the "You are missing 2 revision(s):" line.
[03:06] <fullermd> But with 1.6, it churns for ~23 seconds before showing that line, and 1.5 goes for ~7.
[03:06] <fullermd> (multiple runs of both; everything on both sides will be warm cache)
[03:08] <fullermd> ~100ms round trip.  Not bandwidth contrained; it peaks up to a top of ~50k/s, while I can pull over 300 between these systems.
[03:22] <ElianaTamerin> first time using a RVS, I've created my BZR repo and project under my windows platform, do I need my project files in that dir prior to the bzr add?
[03:25] <ElianaTamerin> nevermind, figured it out ;)
[03:25] <markh> jam: I've got a bzr-setup-1.6.1rc1.exe - after I upload it I will followup your announcement - anything else you'd like to me to?
[03:27] <jam> markh: that sounds good, thanks for putting it together on a weekend
[03:27] <markh> np
[03:33] <ElianaTamerin> i'm trying to setup an ftp push system, I've added my files and done a bzr add, what do I need to do before pushing to FTP
[03:41] <ElianaTamerin> hello?
[03:49] <Peng_> ElianaTamerin: "bzr commit"
[03:49] <Peng_> ElianaTamerin: Have you read the tutorial?
[03:49] <ElianaTamerin> commit stalls for me
[03:50] <Peng_> ElianaTamerin: Um, are you adding tons of large files or something?
[03:50] <ElianaTamerin> Peng_: largest file is 61kb
[03:51] <Peng_> ElianaTamerin: What version of bzr?
[03:51] <ElianaTamerin> 1.6
[03:51] <ElianaTamerin> stable
[03:52] <Peng_> ElianaTamerin: I don't know much about debugging bzr. Maybe you should pastebin exactly what commands you're running.
[03:53] <ElianaTamerin> I did a cd to the directory I wanted, made a bzr init and created the repo, created the project and then dumped in my files. Then did a bzr add and tried a bzr commit, which froze the console
[03:54] <ElianaTamerin> it did say it committed the files, and popped up the log
[03:54] <ElianaTamerin> here's the log: http://devira.pastebin.com/d1acb1da8
[03:55] <ElianaTamerin> and no prompt returns, the console merely sits after it's added the last file with a blank line afterwards
[03:56] <Peng_> ElianaTamerin: You need to write a commit message, save and exit the text editor.
[03:57] <ElianaTamerin> oh, lol!
[04:00] <ElianaTamerin> I'm too slow tonight
[05:08] <Peng_> I wonder how Googlebot found its way to "/loggerhead/a/branch/revision/unknown".
[05:09] <Peng_> Hmm, it seems it's been trying that URL for a while now.
[06:27] <cammoblammo> I'm trying to install bzr-git. As far as I can tell all I need to do is ./setup.py build then ./setup.py install as root. Before I risk hosing my system, is this the correct procedure?
[08:47] <asabil> cammoblammo: isn't checking it out enough ?
[08:49] <asabil> cammoblammo: cd ~/.bazaar/plugins && bzr branch lp:bzr-git git
[08:49] <cammoblammo> Doesn't seem to be. I ran the setup script and it copied a few bits and pieces into /usr/lib somewhere. Now I'm trying to figure out exactly how to use it.
[08:49] <asabil> that should be it
[08:49] <cammoblammo> Ah, I'll look at that.
[08:54] <lifeless> cammoblammo: bzr branch local-git-repo new-dir
[08:56] <cammoblammo> lifeless: Cool! That was my next question...!
[08:56] <asabil> bzr vis doesn't seem to work with git repos
[08:56] <asabil> :/
[08:57] <cammoblammo> So I git to clone a repo on my box then use bzr to operate on it?
[08:57] <AfC> asabil: (that seems worthy of a bug report)
[08:57] <lifeless> cammoblammo: its early days for bzr-git
[09:00] <cammoblammo> lifeless: Yeah, I get that. I'm only looking to use it for local stuff for now. The bzr UI seems a little easier to use.
[09:01] <AfC> Just a little.
[09:01] <lifeless> :P
[09:02] <lifeless> ended up chatting to a git user last night; one that uses the index
[09:02] <lifeless> walked him through why its totally unnecessary etc - hes going to give bzr a go :P
[09:02] <AfC> lifeless: (speaking of last night, it was delightful catching up with you for dinner. You'll be pleased to know that the Shiraz from "out back" didn't give me a hangover (or a heart attack, for that matter)).
[09:03] <lifeless> excellent
[09:03] <lifeless> I enjoyed the catch up as well
[09:03] <lifeless> the wine was scary  :P
[09:03] <AfC> Scott and I went to a right proper wine bar after and more than made up for it :)
[09:04] <AfC> lifeless: incidentally, I really like your idea about one list with multiple topics. If you can possibly figure out the mailman foo for that it would be superb.
[09:05] <lifeless> will try
[13:54] <gustavonarea> Hi. By mistake I've added some files to the branch, and now I have bzr to ignore them *without* removing them. What should I do? Thanks in advance
[13:59] <gustavonarea> OK, it's with "bzr revert [myfile]"
[17:23] <nekohayo> I am a bit scared by the concept of merge conflicts
[17:23] <nekohayo> what does manually resolving a conflict achieve?
[17:23] <nekohayo> does that mean "I did magic. Bazaar, please ignore whatever happened here and only care about diffs in the next commits/merges" ?
[17:40] <nekohayo> oh wow, I just discovered "bzr gconflicts"
[17:41] <nekohayo> it starts a 3-way comparison in meld with .BASE on the left, .THIS in the middle and .OTHER on the right
[17:41] <nekohayo> I am assuming that I must "fix" the one in the middle and save?
[19:48] <sohail> hi, I just moved a repository on my server to a public/ subdirectory, how do I tell my clone locally to use the new one?
[19:50] <luks> use for what?
[19:50] <sohail> luks, for pull and push
[19:51] <asabil> sohail: bzr pull -- remember newlocation
[19:51] <luks> bzr pull <URL> --remeber
[19:51] <luks> and the same for push
[19:51] <sohail> ah, k
[19:51] <sohail> thanks
[19:52] <sohail> How to get rid of this warning: This transport does not update the working tree of: <url>
[19:52] <sohail> the help doesn't tell me
[19:52] <luks> I don't think you can't
[19:53] <luks> *can
[19:53] <sohail> oh that's annoying
[19:54] <jelmer> there's a plugin that can take care of running update on the server side
[19:54] <sohail> but I don't want any updates server side, do I?
[19:55] <asabil> well, sohail, do you actually need a working tree on the server side ?
[19:55] <sohail> I'm just using the server as a central point for the master repo
[19:56] <asabil> sohail: if you don't need a tree on the server, you can do a bzr remove-tree URL
[19:56] <asabil> that will make the warning disappear
[19:57] <sohail> great
[19:57] <sohail> ah crap
[19:57] <sohail> I just removed my local tree
[19:57] <asabil> :)
[19:57] <luks> bzr checkout .
[19:57] <sohail> I had mods that I had forgotten to commit
[19:57]  * sohail cries
[19:58] <sohail> oh theyre still there
[19:58] <sohail> yay
[19:58] <luks> hm, remove-tree doesn't check for a clean tree
[19:58] <luks> ?
[19:59] <asabil> it does iirc
[19:59] <luks> ah, it just keeps the modified files there
[19:59] <luks> maybe an error would be better
[20:00] <sohail> bzr: ERROR: You cannot remove the working tree of a remote path
[20:00] <luks> because if you do "checkout ." to such a branch, you get conflicts
[20:00] <sohail> somehow I have bzr version 1.3.1
[20:00] <luks> can you do it from the server?
[20:00] <sohail> ya
[20:01] <sohail> ok warning gone
[20:01] <sohail> thanks
[21:15] <hsn_> how to remove parent branch from related branches, i have there for some reason bogus branch
[21:31] <uws> hsn_: use 'bzr pull --remember' with the right url
[21:32] <hsn_> isnt pull branch different from parent branch?
[21:32] <uws> hsn_: (or edit .bzr/branch/branch.conf)
[21:32] <uws> hsn_: "pull" uses the parent location. "push" uses the "push" location.
[21:32] <uws> update/bind/unbind/checkout has "bound" location
[21:32] <uws> and "send" has a "submit" location
[21:33] <uws> hsn_: see .bzr/branch/branch.conf  (usually you don't edit this since it's inside .bzr/)
[21:36] <hsn_> is there simple way to look into bzr protocol? i.e. commands sent over socket?
[21:36] <uws> hsn_: wireshark/tcpdump
[21:37] <hsn_> i upgraded protocol to bzr+ssh and SVN is still much faster
[21:37] <uws> that's my experience as well for some reason :(
[21:37] <hsn_> well, but wireshark will not help much since it is encrypted
[21:38] <uws> hsn_: you can use bzr:// with a branch published using "bzr serve" on localhost to test
[21:39] <hsn_> local bzr speed is fine, but network speed is still too slow. I never used git over network so i cant compare
[22:23] <funkyHat> I read an article (on a planet I think) about using bzr for web development, and a command that could push just a single revision to a server via ssh... but I've lost it now
[22:24] <funkyHat> I've been poking around trying to figure out how to do it myself but not got anywhere, anyone know which command I should be using? Perhaps it's a plugin I need to add or something?
[22:33] <jelmer> funkyHat, bzr upload
[22:35] <funkyHat> thanks jelmer that's it :)
[22:43] <funkyHat> :( the plugin seems to be broken though
[22:46] <jelmer> funkyHat, works fine here
[22:46] <jelmer> funkyHat, what's not working?
[23:14] <funkyHat> jelmer: the problem was I was using bazaar from hardy, I've upgraded to the bazaar ppa version now and it's all good :)
[23:27] <sohail> funkyHat, how do you do that?
[23:31] <funkyHat> sohail: https://launchpad.net/~bzr/+archive
[23:33] <sohail> funkyHat, thanks
[23:33] <funkyHat> :)
[23:47] <Odd_Bloke> luks: remove-tree not warning when you have uncommitted changes is bug 74101.
[23:50]  * ToyKeeper wonders when bzr 1.6 will land in debian/testing, or even unstable
[23:55] <Peng_> Hun. I get a KeyError when pulling from http://people.samba.org/bzr/jelmer/bzr-svn/0.4/
[23:55] <jelmer> Peng_, CAn you pastebin it?
[23:57] <Peng_> Err, hold on
[23:59] <Peng_> jelmer: From .bzr.log, http://paste.pocoo.org/show/VgqQj9CdOoFkDfZkvswz/
[23:59] <Peng_> (I'm running latest bzr.dev as of 20 minutes ago.)