[00:01]  * awilkins cackles evilly because he has just written some thoroughly evil yet powerful Java generics code
[00:04]  * jdong thinks awilkins could've sliced out "yet powerful" and "generics"
[00:23]  * awilkins doesn't care, it nearly worked first time around :-)
[00:23]  * awilkins is going to bed
[00:48] <lamby> jelmer: dbus-fu now works? :)
[00:50] <jelmer> lamby: yep, just packaged it
[00:51] <jelmer> with a bit of luck I can get lifeless to upload it tomorrow
[00:52] <lamby> jelmer: I must admit, that was one of the main things that lured me into packaging bzr-gtk.
[00:53] <lamby> (Maybe bzr-gtk could be uploaded too, ooi?)
[00:54] <jelmer> we're also working on a new bzr-gtk
[00:55] <lamby> Oh neat. What's the motivation behind that?
[00:56] <beuno> lamby, among other things, fixing nautilus integration  (jelmer is discussing as he falls asleep)
[00:57] <Odd_Bloke> So I was anticipating finding an AfC in my room...
[00:57] <beuno> Odd_Bloke, and you found what?
[00:57] <Odd_Bloke> No AfC.
[00:57] <beuno> is that a metaphore?
[00:57] <Odd_Bloke> No, AfC is Andrew Cowie. :)
[00:58] <beuno> right, was just trying to block the continued gpg discussion out of my head
[00:58] <beuno> but that joke got cut off too
[00:58] <beuno> must be time for bed
[00:58]  * beuno is off
[01:02] <lamby> 'sup Odd_Bloke
[01:02] <lamby> jelmer: nn ;)
[01:03] <jelmer> lamby: g'night
[01:05] <Odd_Bloke> lamby: o/
[05:30] <indraveni> hi all
[05:30] <indraveni> i want to use trac as front end for my bzr site
[05:30] <indraveni> I need some help on this
[05:30] <indraveni> i installed trac and trac-bzr
[05:30] <indraveni> and when I am accessing the trac, browse code, page
[05:30] <indraveni> the following is the error i am seeing
[05:31] <indraveni> http://pastebin.ca/928643
[05:32] <indraveni> could some one please help me in resolving this issue
[05:34] <bob2> are you using the recommended trac, trac-bzr and bzr versions?
[05:36] <ubotu> New bug: #193253 in launchpad-bazaar "sockets being leaked in branch puller tests" [High,Confirmed] https://launchpad.net/bugs/193253
[06:07] <indraveni> bob2, recommended ? I am using debian os. thus done installation through aptitude
[06:36] <indraveni> I now downloaded trac-bzr frm launchpad from the traunk
[06:36] <indraveni> but I am not knowing how to install it or use it
[06:36] <indraveni> could some one here , help me out
[06:46] <ubotu> New bug: #198645 in bzr-dbus ""bzr commit" does not trigger DBus signal" [Undecided,New] https://launchpad.net/bugs/198645
[06:48] <tbnorth> hi all - how can I change the path to the parent branch?  It should be using /home/user/... but instead it's using /mnt/250GB/home/... which is only valid on one machine.
[06:50] <fullermd> Does it matter?  It wouldn't have any meaning except on one machine anyway.
[06:52] <bob2> ('bzr pull --remember /home/...' or edit .bzr/branch/branch.conf)
[06:52] <james_w> indraveni: there is a a README in the source that should help.
[06:52] <indraveni> its only saying about after installation
[06:53] <indraveni> james_w, nothing about how  to install or ue
[06:53] <indraveni> use
[06:54] <tbnorth> thanks bob2
[06:55] <tbnorth> fullermd: I clone my whole file system between machines, but sometimes I forget to check if I'm on the common symbolic path of the machine specific mount path
[06:56] <huslu> someone here knowledgeable about debian's bzr package?
[07:01] <ubotu> New bug: #198646 in bzr "Invalid http response ... Expected a boundary" [Undecided,New] https://launchpad.net/bugs/198646
[07:10] <james_w> huslu: what about it?
[07:11] <huslu> james_w: in lenny i have to install a lot of extra packages to be able to use recent bzr, (x11-common, etc)
[07:12] <james_w> huslu: that sounds a little odd. I don't know the reason though.
[07:12] <james_w> I don't have time to investigate now, sorry.
[07:12] <james_w> There will be other knowledgeable people on in the next couple of hours I think.
[07:12] <huslu> ok, np. etch-backports has it ok, just lenny.
[07:13] <james_w> Has it started recommending bzr-gtk?
[07:13] <huslu> 'suggesting', yes
[07:14] <huslu> but that's not forced to install with it
[07:16] <bob2> are you using aptitude?  there's a recommends chain bzr -> bzrtools -> graphviz -> libxt6 -> x11-common.
[07:17] <huslu> no, just most up to date apt
[07:17] <huslu> apt 0.7.11
[07:18] <huslu> buy yes, with apt-get it forces to install graphviz, libxt6, and so on.
[07:19] <bob2> maybe APT::Install-Recommends is on by default these days
[07:21] <huslu> that may be very likely
[07:22] <huslu> i have to see where's the settings for Install-Recommends
[08:11] <corevette> my bzr says: Unable to obtain lock file:///home/corevette/www/.bzr/repository/lock held by corevette@gmail.com on host corevette-desktop [process #12320]
[08:12] <Peng> The repo is locked.
[08:13] <Peng> Do you have another bzr process running (presumably with the PID 12320)?
[08:14] <corevette> my PID's only go up to around 8000 peng
[08:16] <Peng> Wait, are you working on that machine or pushing to it from another?
[08:17] <Peng> Anyway, if you're 100% sure nothing else is going on (do a ps on both machines), use "bzr break-lock $location".
[08:19] <corevette> peng, nevermind, i just pulled in a different location...and if it fixed it
[09:18] <mrevell> phanatic: Could we talk about Olive today?
[09:22] <jml> Odd_Bloke: did you take notes on the Admin breakout?
[09:23] <Odd_Bloke> jml: http://bazaar-vcs.org/SprintLondonMarch08/Brainstorms
[09:24] <phanatic> mrevell: sure, it'd be great
[09:25] <mrevell> phanatic: Lovely. I'm busy from around 11.00 until 13.00.
[09:52] <mrevell> phanatic: Sorry dude.
[09:53] <phanatic> mrevell: may i join you? :)
[09:53] <mrevell> phanatic: sure, please :)
[10:08] <Lllama> Morning all. I've just done a merge and got some conflicts. I'm editing the BASE, THIS and OTHER files and can fix the changes, but I'm not sure what I need to be doing to resolve the conflict. I.e. which file will be the master version and whether I need to delete the others. A bit of a newbie question but any pointers will be gratefully received.
[10:13] <beuno> Lllama, just rename your file once you've solved them
[10:14] <beuno> to it's original name
[10:14] <beuno> (I suppose that should be in the docs...)
[10:15] <spiv> mwhudson_: well done on the release
[10:16] <Lllama> beuno: Ah, so I take the THIS, BASE and OTHER, create a resolved version and then save that as the original file. (the original file is still in the directory, so I'll be overwriting it.)
[10:18] <beuno> Lllama, right
[10:18] <beuno> you just edit the original one to what you want
[10:18] <beuno> or overwrite it
[10:21] <Lllama> beuno: Excellent. Thank you.
[10:23] <beuno> Lllama, welcome
[10:37] <Zindar_> Lllama: or you can just ignore THIS, BASE, OTHER... edit the original.. it contains markers on conflicts, solve them...  then run bzr resolve on the file...
[10:37] <Zindar_> I never open THIS, BASE, OTHER... I do use them for auto resolve tools thoguh, like bzr extmerge
[10:39] <Lllama> Zindar: cool.
[10:42] <bob2> where can I set bzr_Remote_path?  the [DEFAULT] block in ~/.bazaar/locations.conf seems to be the wrong one.
[11:06] <snod> ös
[11:36] <lifeless> bob2: not sure you can in a config yet; seems to be an open bug
[12:56] <LarstiQ> Zindar: are you still in London today?
[12:56] <Zindar> yes.. but leaving at 1600
[12:56] <Zindar> so.. just in a few hours
[12:58] <LarstiQ> ah.
[13:01] <Zindar_> crap network
[13:16] <soren> If someone e-mails me a patch and I'd like to import that into bzr somehow (e.g. using the text of his e-mail as the body, and his sender address as the committer info), how to do?
[13:17] <soren> It's fine if I need to copy/paste the body of his e-mail into the commit message, but isn't there a command line option to use another committer identity or something?
[13:17] <fullermd> You probably want --author
[13:18] <fullermd> The best solution would be for the submitter to use bzr and send you a bundle instead of course   :)
[13:18] <soren> Sure.
[13:19] <soren> Er.. Yeah, --author is totally what I want.
[13:19] <soren> How could I have missed that? *shrug*
[13:19] <fullermd> It's relatively new I think; 2 or 3 versions, maybe.
[13:25] <soren> fullermd: Yeah, well, I looked at "bzr help commit" just before I asked my question :)
[13:28] <fullermd> Oh.  Well, in that case, it's REALLY new, and was pushed to all installations via a secret backdoor bzr installs right after you asked.
[13:29] <soren> I suspected that :)
[13:34] <muszek> hi
[13:37] <asabil> hmm, seems like launchpad is getting crazy again :/
[13:37] <asabil> locks are being held
[13:38] <muszek> I have a problem... suddenly two files "stopped being added" (are not versioned anymore)... I try to add them again - I don't get any message and these files still show up as "unknown" when I do bzr status.  In the past I modified them many times and had no problems.
[13:38] <muszek> using 1.1.0.candidate.1 on Debian Sarge
[13:39] <asabil> anyone having this problem :
[13:39] <asabil> Unable to obtain lock lp--1217851700:///lock
[13:39] <asabil> held by asabil@bazaar.launchpad.net on host vostok [process #17382]
[13:39] <asabil> locked 4 seconds ago
[13:39] <asabil> using bzr break-lock will just create another lock
[13:39] <beuno> asabil, do it twice
[13:40] <beuno> or maybe 3 times, LP sometimes hast multiple process waiting to lock it
[13:41] <asabil> beuno: thx, fixed my problem
[13:42] <muszek> someone please help: http://pastebin.us/?show=d82d276b
[13:45] <muszek> I've figured out the problem...
[13:46] <beuno> asabil, :D
[13:46] <fullermd> Different files than you thought were unknown?
[13:48] <muszek> fullermd: I ran a db dump script from grandparent dir... and it placed those sql files in that grandparent dir (I didn't predict that)... and I kept trying to add those already added (grandparent) files and bzr status kept telling me, that grandparent files are unknown
[13:49]  * fullermd nods.
[14:34] <asabil> hmmm
[14:34] <asabil> can anyone take a look at this :
[14:34] <asabil> https://code.launchpad.net/~easy-radio/easy-radio/trunk
[14:35] <asabil> seems like a bzr/launchpad issue
[15:15] <mathrick> hiya
[15:16] <mathrick> what is the easiest way to convert a CVS repo to bzr, without having a local copy of CVSROOT?
[15:18] <Zindar> I don't know.. but I created a very simple thing using cvsps once...
[15:18] <fullermd> AFAIK cvsps-import and tailor can both use cvs remote.  It'd be a buttload slower than local, though.  And probably run up the same or more network traffic/server load...
[15:20] <mathrick> fullermd: cvsps yells at me that :pserver: is not supported and I need a local copy
[15:20] <fullermd> Are you using --use-cvs?
[15:22] <mathrick> no
[15:23] <mathrick> I tried that, same result
[15:35] <salgado> hey guys, what's the new record command (from looms) for?
[15:37] <salgado> people have been merging from a loomified branch of mine and they were able to see the threads even though I never ran 'bzr record'
[15:58] <CardinalFang> Hi all.  Should we expect a new version of Bazaar very soon?  I don't want to trouble my sysadmin to update twice.
[16:04] <beuno> CardinalFang, a few weeks time
[16:04] <beuno> there is one every month
[16:04] <CardinalFang> Gracia.
[16:04] <beuno> :D
[16:16] <Odd_Bloke> beuno: There's something of a gap this time, because of the sprint.
[16:19] <beuno> Odd_Bloke, a gap where?
[16:19] <awilkins> Whe, just got bzr approved as part of an internal project.
[16:20] <LeoNerd> Is there a way to use bzr against CVS repos, like there is bzr-svn for SVN ones?
[16:20] <awilkins> Well "not objected to", which is nice
[16:20] <LeoNerd> I'd love to be able to do things like 'bzr shelve' in checkouts
[16:20] <awilkins> LeoNerd: AFAIK there is no round-trip support for CVS
[16:20] <LeoNerd> :/ Boo
[16:20] <Odd_Bloke> beuno: There'll be an extra week or so before the release.
[16:20] <Odd_Bloke> awilkins: \o/
[16:21] <awilkins> LeoNerd: You can get a bzr repo to follow a CVS repo (I think) but pushes to "home" would have to be patches
[16:21] <awilkins> Is there PQM for CVS?
[16:21] <LeoNerd> Hmm? Not sure I get what you mean there
[16:21] <beuno> Odd_Bloke, right, hence, pushing it a few weeks. It was suppose to be out the first days of march IIRC
[16:22] <awilkins> LeoNerd: I mean, is there an automatic thing that can accept patches which you've generated from your bzr tree and commit them to a CVS repo for you.
[16:22] <beuno> actually
[16:22] <beuno> I'm wrong
[16:22] <beuno>  1.3 final  21 March
[16:23] <awilkins> Verterok: I may be devoting more effort to bzr-eclipse simply because I'll be wanting to use bits of it.
[16:23] <LeoNerd> awilkins: Well, I wouldn't mind if it had to be a bound tree.. that every commit operation pushed it to the CVS repo immediately
[16:26] <awilkins> LeoNerd: http://bazaar-vcs.org/pushcvs?highlight=%28cvs%29  ??
[16:27] <LeoNerd> Mmmm
[16:29] <awilkins> LeoNerd: Is this a CVS repo you have influence over?
[16:29] <awilkins> LeoNerd: Could you get them to migrate it to.... well, SVN at least ...
[16:29] <LeoNerd> Heh.. It's been a point of contention for a while now :P
[16:29]  * awilkins only uses CVS when he is forced to
[16:29] <LeoNerd> Personally I'd vote to move the whole lot to bzr.. But.. naturally, some corporate inertia here ;)
[16:30] <awilkins> My VCS experience started with Visual SourceSafe and moved onto Subversion
[16:30] <Verterok> awilkins: nice :)
[16:30] <awilkins> I dallied a little with monotone, git and hg, but I seem to be gravitating to bzr
[16:31] <awilkins> Verterok: I have this horrible project to work on which needs revision control ; trunk-> branch merges are part of the deal, as are renames.
[16:32] <awilkins> If I use use SVN I'll have to automate ALL of the revision control because I need merge tracking
[16:33] <Verterok> awilkins: merge is in the next-features-to-land
[16:33] <awilkins> I've been checking out the Roadmap :-)
[16:34] <Verterok> awilkins: the conflicts resolution integration might take a bit more of time/effort
[16:34] <awilkins> Verterok: Meh, even worse for me, I'm dealing with models not text files.
[16:35] <awilkins> Verterok: Fancy shchmancy not-editing-the-xml-by-hand features like that are firly in "Stage 2"
[16:36] <ubotu> New bug: #198793 in bzr "Problem accessing branches via bzr+https" [Undecided,New] https://launchpad.net/bugs/198793
[16:36] <ubotu> New bug: #198798 in bzr "doc.bazaar-vcs.org needs a search form" [Wishlist,Triaged] https://launchpad.net/bugs/198798
[16:36] <Verterok> awilkins: it' good to have this present during the design, bzr-eclipse could provides a extension point for the conflict resolution thingy ;)
[16:37] <awilkins> Verterok: I'd be unsurprised to discover that there was already such a thing in there
[16:37] <awilkins> Verterok: Eclipse has extension points for the kitchen sink
[16:38] <awilkins> Question : Is it acceptible to modify content in a pre-commit hook?
[16:38] <Verterok> awilkins: sure, but bzr-eclipse should provides the hooks for the bzr side
[16:40] <fullermd> I don't think it's _possible_ to modify content pre-commit currently.
[16:45] <awilkins> fullermd: The hook document says you mustn't modify delta_tree, but does that mean "you mustn't touch delta_tree" or "You mustn't do anything that would changes delta_tree" or even "you mustn't modify ANYTHING even if it wouldn't delta_tree because you'd just be tidying one of the files it describes as having changed"
[16:46] <igc> abentley: here's the email - https://lists.ubuntu.com/archives/bazaar/2007q4/035340.html
[16:46] <awilkins> So e.g. you couldn't read delta_tree and run a tidy on files it mentions
[16:52] <awilkins> The way I read it, the only way for a pre-commit hook to abort a commit is to throw an exception, that right?
[16:54] <Prodoc> good afternoon
[16:54] <Prodoc> is it possible to change a commit message afterwards?
[16:56] <beuno> Prodoc, no. You'd have to bzr uncommit
[16:56] <beuno> and commit again
[16:56] <Prodoc> darn
[16:57] <Prodoc> I was hoping to prevent that
[16:57] <awilkins> The commit log is probably hashed into the revision ID like everything else....
[16:58]  * awilkins possibly displays lamentable ignorance of internal working here, presumes it's rather like monotone in some respects
[16:59] <abentley> igc: tx
[17:15] <LarstiQ> jam: 17:46:12 < igc> abentley: here's the email - https://lists.ubuntu.com/archives/bazaar/2007q4/035340.html
[17:33] <ubotu> New bug: #198821 in bzr "switch of lw checkout shouldn't require force when branch moved" [Medium,Confirmed] https://launchpad.net/bugs/198821
[17:38] <spiv> jelmer: http://rafb.net/p/nNwbWU65.html
[17:40] <henke> How can I downgrade a branch to an older format?
[17:41] <spiv> henke: use "bzr upgrade --foo", where "foo" is the format you want to downgrade to.
[17:42] <TFKyle> spiv: how much does that help performance? I wouldn't expect refering to a variable there would be that slow but you never know
[17:42] <henke> spiv, I tried that, but it tells me that it is already in the latest format
[17:42] <spiv> henke: what formats do you want to downgrade from/to
[17:42] <spiv> TFKyle: about 6x
[17:42] <TFKyle> wow
[17:42] <henke> spiv, well, it might not be necessary to downgrade now. Found some backport of a newer bzr for a friend
[17:43] <spiv> TFKyle: i.e. "bzr svn-import" on Twisted's SVN takes ~20 sec per branch, rather that about 2 minutes.
[17:43] <spiv> henke: ah right, for compatibility with old clients, I see.
[17:54] <Odd_Bloke> abentley: The code is at https://code.launchpad.net/~daniel-thewatkins/+junk/bzr-mirror if you get time to glance over it.
[17:56] <beuno> +junk sounds like it's promising
[17:57] <Odd_Bloke> beuno: It's not a bzr branch, and I can't be bothered to set up a new project yet. :p
[17:57] <Stavros> hello
[17:57] <Stavros> say i checked out rev -2, made some changes and want to check in again
[17:57] <Stavros> how would i do that?
[17:58] <beuno> Stavros, commit?
[17:58] <beuno> if it's a checkout, it will end the changes automatically
[17:59] <Stavros> will it let me commit something from two revs back?
[17:59] <Stavros> because i hate subversion so ****ing much right now
[18:00] <fullermd> You can't commit onto a non-head rev...  you can only make a new branch from that and merge it.  A checkout might do that behind your back, not sure.
[18:01] <Stavros> well, what i want to do is check out something two revs back, make some changes and recommit, overwriting those two revs
[18:02] <fullermd> Well, you can dump revs via uncommit or pull or some such.
[18:02] <luks> why not just uncommit them and commit the new change?
[18:02] <fullermd> (assuming they're not out in the wild where other people are basing work off 'em, of course)
[18:03] <mathrick> it seems to me that you could use bzr-rebase for that
[18:03] <beuno> Stavros, or, you can branch, so bzr revert -r REVID
[18:03] <luks> mathrick: no
[18:03] <beuno> and you would have your commit history
[18:03] <beuno> mathrick, no
[18:04] <Stavros> bzr-rebase sounds like what i need
[18:05] <luks> no, it doesn't
[18:05] <beuno> Stavros, you won;t be able to merge back
[18:05] <Stavros> will it make it look like i'm in HEAD without changing files?
[18:05] <Stavros> beuno: i don't want to merge anything
[18:05] <Stavros> i just want to commit my files
[18:05] <beuno> Stavros, never again?
[18:05] <beuno> bzr rebase will distroy your history
[18:05] <Stavros> oh
[18:05] <beuno> (or destroy)
[18:05] <Stavros> that's not it then :P
[18:05] <beuno> I'd say either uncommit
[18:06] <beuno> or revert and commit
[18:06] <beuno> both will do it for ya'
[18:06] <Stavros> won't revert change my files?
[18:06] <beuno> one will keep history that you reverted the change
[18:06] <beuno> Stavros, revert will leave the files at the exact stage they where in htat commit
[18:06] <beuno> so you don't need to checkout a few revisions back
[18:06] <Stavros> beuno: destroying my changes, then
[18:06] <beuno> Stavros, branch somewhere else
[18:07] <beuno> do bzr revert
[18:07] <Stavros> yeah, i was looking for something automatic
[18:07] <beuno> copy over whatever you want
[18:07] <beuno> and commit
[18:07] <Stavros> ideally, i could just make bzr think i'm in head
[18:07] <luks> Stavros: what should be the final result of this?
[18:07] <luks> two branches with different heads?
[18:07] <Stavros> luks: no, revisions -1 and -2 completely gone
[18:07] <mathrick> humm, I have problems with upgrading a branch in dirstate format, upgrading to pack-0.92 works just fine, but when I try rich-root-pack, it fails with a missing revision error
[18:07] <luks> or just one branch, with 2 revisions removed and one new
[18:07] <Stavros> luks: the latter, yes
[18:07] <luks> Stavros: then uncommit is what you want
[18:08] <luks> uncommit twice
[18:08] <mathrick> it's a CVS repo imported by tailor, btw
[18:08] <luks> and then commit
[18:08] <Stavros> luks: ah, but i have already done the changes, will that matter?
[18:08] <fullermd> Uncommit probably isn't what you want, if you want to wipe out the changes in those revs.
[18:08] <luks> Stavros: make a new branch and move them aside (merge --uncommitted)
[18:08] <Stavros> aha, thanks
[18:08] <fullermd> That'll just lose the commit granularity, not the changes.
[18:08] <Stavros> ah, good
[18:08] <Stavros> thanks a lot for your help
[18:08] <fullermd> I'd try pull, myself.  That merges WT changes, though I've never tried it backward.
[18:09] <Stavros> fullermd: oh
[18:09] <Stavros> that might actually work
[18:09] <fullermd> (merging changes backward, that is; I've used it to reset head often)
[18:09] <Stavros> fullermd: well, when you pull from other revisions and change something, how do you recommit?
[18:09] <Stavros> assuming that's what "reset head" means
[18:10] <fullermd> Well, what pull does is essentially reset the head of the branch.
[18:10] <fullermd> In normal usage, it's to a descendent of the previous head, but it can do it to an ancestor too, or an unrelated rev.
[18:10] <mathrick> http://pastebin.com/m2e5c84e1 <-- any idea what might cause that, and how I can fix it?
[18:10] <Stavros> oh
[18:10] <Stavros> i didn't know you could pull to a release and ignore subsequent ones
[18:10] <fullermd> At least when going forward, it will try to merge WT changes across (like a 'cvs up' with local changes)
[18:11] <Stavros> ah, i see
[18:11] <fullermd> I presume it will at least try to do so when it's moving "backward" too; I don't know how successful it would be in any given case, but it can't hurt to tar up a backup and try.
[18:17]  * mathrick attempts to trick someone into looking at his error
[18:23] <bobbo> does anyone know how i could fix http://pastebin.ubuntu.com/5315/plain/ ?
[18:25] <beuno> bobbo, bzr break-lock
[18:25] <beuno> bzr break-lock bzr+ssh://bobbo@bazaar.launchpad.net/~bobbo/+junk/bzr-grab
[18:25] <beuno> (you might need to do it twice)
[18:26] <LarstiQ> beuno: why is that?
[18:26] <LarstiQ> beuno: repo, branch, tree?
[18:26] <LarstiQ> but not tree because remote?
[18:26] <beuno> LarstiQ, LP spawns multiple smart server threads for some reason and keeps getting locks
[18:26] <beuno> not sure *why*
[18:26] <beuno> but it's very common
[18:27] <beuno> so you have to kick all of em in the nuts, or be fast enough before it locks it again
[18:27] <bobbo> beuno; thanks that fixed it
[18:27] <Odd_Bloke> I've found that breaking the locks via sftp rather than bzr+ssh tends to reduce the chances of screwage-up.
[18:58] <phanatic> sprinters: anybody up for a nearby pizza hut maybe?
[18:58] <jelmer> phanatic: anything that's food & relatively quick sounds good to me
[18:59] <phanatic> jelmer: there's a restaurant near the tower, and near the hotel as well (ca. 15 mins by foot)
[19:07] <james_w> Hi all. Is anyone still at the office?
[19:07] <Odd_Bloke> james_w: Yeah, there are people around.
[19:07] <james_w> Hi Odd_Bloke. Is cprov there?
[19:07] <Odd_Bloke> Though we're heading off fairly imminently (I think).
[19:08] <Odd_Bloke> james_w: I have no idea, I'm afraid.
[19:09] <james_w> Odd_Bloke: no problem. If you see him, can you tell him he needs to ask for a new key at reception, as they had to invalidate his to give me one.
[19:09] <james_w> Odd_Bloke: it's not that important though.
[19:09] <james_w> See you all tomorrow, got to dash.
[20:52]  * awilkins suspects that everyone is eating a large restaurant dinner
[20:52]  * awilkins had to mak edo with leftover pizza
[20:53]  * fullermd isn't...
[20:55] <Odd_Bloke> awilkins: I ate room service, the other guys went for pizza.
[20:55] <Odd_Bloke> And there are some guys in a pub somewhere.
[20:56] <awilkins> Meh, I'm at home listening to NIN and drinking Vokdy-tonic
[20:56] <awilkins> I contemplated coming, but it was a bit short notice, my python is not strong, and I'm too busy at work.
[21:01] <grantgm> is there a good way to go about tracking the splitting of files?
[21:07] <awilkins> grantgm: Use git :-)
[21:08] <awilkins> Copy the file and delete the bits you don't need?
[21:16] <grantgm> awilkins: alright...that's what I thought the answer might be.
[21:16] <grantgm> does git have support for that?
[21:16] <grantgm> it would be nice if I could 'bzr cp', but I'm sure that would cause all sorts of confusing-ness for merging later on
[21:17] <awilkins> Git (AFAIK) tracks _content_ rather than _files_, so it naturally tracks file splitting and joining.
[21:17] <awilkins> It snapshots whole trees as revisions
[21:18] <awilkins> Although you might ge ta more coherent response in a git channel, where the people are either not in a huge pizza orgy in London, or drunk in their office.
[21:19] <grantgm> mmm...pizza orgy...
[21:19] <ferringb> disturbing imagery
[21:19]  * awilkins would settle for a pizza quickie right now.
[21:19] <fullermd> Sounds like a sausage-fest.
[21:19] <abentley> The pizza was totally into it, so I don't know what your problem is;.
[21:20] <grantgm> oh noes...tomato sauce EVERYWHERE!
[21:20] <awilkins> At least you don't have to deal with that anchovy smell
[21:23]  * awilkins stalled the conversation againb
[21:23] <abentley> awilkins: After a while, you don't even smell the anchovies.
[21:24] <awilkins> Does bzr have a (per file) merge hook / configurable file merge support ?
[21:25] <abentley> No, but tomorrow I'll be helping Odd_Bloke get started on implementing that.
[21:26] <awilkins> Can you have per-branch hooks (ooh, becoming petilent here)
[21:26] <awilkins> *pestilent
[21:26] <abentley> awilkins: Sure.
[21:27] <awilkins> abentley: Where would you put a per-branch hook?
[21:27] <fullermd> You can put 'em in branch.conf, and probably in locations.conf too.
[21:27] <abentley> In ~/.bazaar/plugins
[21:27] <awilkins> How about somewhere in the branch so they propagate?
[21:27] <fullermd> That I more doubt.
[21:28] <abentley> awilkins: There are a few security issues with that.
[21:28] <awilkins> (with a config option to switch it on so you canb't hack people so easily with it)
[21:28] <awilkins> Hah, was just getting there
[21:29] <awilkins> ... maybe you could just permit signed revisions of hook that have been user-approved to run
[21:29] <awilkins> ("revision" and "signed" being redundant of course)
[21:30] <awilkins> Meh, you could even have branch plugins I suppose
[21:30]  * awilkins is getting power-mad
[21:30]  * fullermd unplugs awilkins.
[21:30] <awilkins> Hooks are effectively special plugins anyway
[21:31] <radix> I would love to have something that said "this branch has a branch hook, as shown. Would you like to enable it?" upon branch
[21:31] <abentley> awilkins: I don't think remotely-configured hooks are on anyone's roadmap.  Sane, secure patches accepted.
[21:32] <awilkins> Remote config was always a feature people kept asking for in SVN
[21:32] <awilkins> All the hooks are already in the repo with that of course
[21:33] <awilkins> Things like automatic mime-type attributes were a local config option which meant that redistributing the was a PITA when you had a lot of users
[21:34] <awilkins> (esp when your users have a habit of surfing your repo directly to check content, and they hit a page with no mim-type in Firefox, which respects them and treats it as text/plain)
[21:34] <awilkins> (IE was fine because it's a presumptive pile of crap)
[21:58] <beuno> vila, http://permalink.gmane.org/gmane.emacs.devel/91263
[21:59] <beuno> vila,
[22:20] <Odd_Bloke> beuno: That whole thread makes for interesting reading.
[22:21] <Odd_Bloke> Apparently one of the things git has going for it is a large and active developer base...
[22:21] <Odd_Bloke> Anyhow, bed time.
[22:22] <jelmer> Odd_Bloke! \o/
[22:22] <jelmer> Odd_Bloke: You're missing all of the interesting discussions
[22:53] <mwhudson> oh god, of course loggerhead 1.2 had to have a catastrophic embarrassing bug
[22:56] <Peng> What is it?
[22:56] <awilkins> What, it outs people at random?
[22:57] <mwhudson> the changelog view is randomly ordered
[22:57] <awilkins> Unstable sort algorithm?
[22:57] <mwhudson> well, just no sort algorithm
[22:58] <awilkins> Ah. hashtable?
[22:58] <mwhudson> i changed the way that filtering ghosts was done, and a side effect was that the order of the results of a particular method ceased to be related to the order of the input
[23:07] <awilkins> Anyone else get a copy of "Ghosts" by Nine Inch Nails?
[23:10] <paty__l> i'm thinking about participating at gsoc doing bazaar integration for eclipse. with whom i can talk about it?
[23:11] <awilkins> Verterok
[23:11] <awilkins> aka Guillermo Gonzalez
[23:11] <paty__l> thanks =)
[23:12] <Verterok> hi paty__l :)
[23:12] <paty__l> hi =) does this integration with eclipse has already been done or it is still open?
[23:13] <Verterok> I'm working on this, but all contributors are more than wellcome
[23:14] <Verterok> paty__l: take a look at: http://bazaar-vcs.org/BzrEclipse
[23:15] <paty__l> great.. i'll have a look at this. is bazaar going to participate at gsoc this year?
[23:16] <Verterok> paty__l: yes, are you only interested in Eclipse integration or in any other IDE integration, like Netbeans or IDEA?
[23:17] <Verterok> awilkins: thanks for the hilight ;)
[23:19] <paty__l> i have some experience with building eclipse plugins and i dont have this experience with Netbeans or IDEA. but if it is possible to learn it untill the program begins, i'm interested
[23:21] <awilkins> Has Jelmer Veernooij been at Sprint?
[23:22] <Verterok> paty__l: oh, great. If you want to work in the eclipse integration there is plenty of work to be do
[23:23]  * awilkins cheers
[23:23] <Verterok> paty__l: also, if you are interested in learning other IDE integration, you are wellcome too :)
[23:25] <Verterok> awilkins: yes, he's leaving tomorrow
[23:25] <awilkins> I have a win32 test log from bzr-svn for him
[23:28]  * awilkins reads PEP-8 and immediately gets annoyed that it recommends spaces over tabs for indents
[23:28] <paty__l> great :) i'm interested in working in the eclipse integration. where can i learn more about what work still need to be done in this integration?
[23:28] <awilkins> Try the bzr-eclipse roadmap... http://bazaar-vcs.org/BzrEclipse/Roadmap
[23:29]  * Peng gapes at awilkins. Tabs?!
[23:29]  * awilkins prepares for TAB JIHAD
[23:29] <paty__l> sure.. i'll have a look at that
[23:31] <Verterok> paty__l: if you have any questions, don't hesitate in contact me you can find my email, Jabber, etc at launchpad.net/~guillo.gonzo
[23:34] <paty__l> oh, great.. thanks :) i'll have a look and if have any questions, i'll contact you
[23:35] <Verterok> great :)
[23:43] <mlh_> http://lists.gnu.org/archive/html/emacs-devel/2008-03/msg00234.html  - awesome
[23:45] <awilkins> Heh, someone posted the Stallman quote from that thread earlier
[23:46] <awilkins> "We should all snuggle up to Bzr because it want's to be a GNU too" (or words to that effect)
[23:47] <jdong> mlh_: that's awesome
[23:48] <jdong> I might... start using emacs.... because of this :)
[23:49]  * awilkins would rather not have to pay a surgeon to unknot his fingers
[23:50] <awilkins> Get Bram Moolenaar to change vim to bzr.