[02:18] jelmer, good thanks. break from bzr/coding over xmas? ;-) [02:18] hi wgz [11:54] hey [11:55] Similarly to what I can do with git's "checkout", can I change the branch I'm currently working on with Bazaar, so when I "bzr pull", it will pull from the branch I'm working on, and I don't have to indicate it? Thanks. [12:27] Yes and no. Some terminological confusion there. [12:28] git checkout is about changing what branch the working tree is 'hooked up' to. bzr has a 'switch' command that does that. [12:28] But bzr pull is about updating one _branch_ to match another. That it also updated the WT is sorta a side effect. [12:29] If you want to change [semi-]permanently which branch your local branch is tracking, pull has a --remember arg that will change the remembered location; that may be what you want. [12:57] Hm fullermd [12:57] thanks fullermd [12:57] bzr pull lp:my-branch --remember [14:16] Hi. I'm having some trouble understanding the bzr API. I'm trying to check whether a file is changed, so I'm doing tree.get_file_sha1(id) == tree.basis_tree().get_file_sha1(id) but they always differ [14:16] I thought basis_tree was supposed to be the tree of the latest revision, whereas tree is the current working tree [14:16] (I got the 'tree' object via WorkingTree.open_containing(dir)) [14:57] Ah, it looks like this works when I flush the changes I made to the file === yofel_ is now known as yofel [18:45] * jelmer waves [18:46] hi jelmer [19:16] What's the recommended way to publish a bzr plugin? [19:17] I made a very small and simple plugin to help out with a very specific task. Should this have its own launchpad page? [21:13] hi folks [21:13] hi Noldorin_ [21:13] hi [21:15] I have a bazillion unrelated checkouts of various projects in ~/src/ [21:15] I'd like to see if any of them have modified and uncommitted files [21:16] with svn checkouts I can do svnst * 2>/dev/null|grep -v ^? [21:16] (svnst is an alias for svn st) [21:16] bazaar doesn't like 'bzr st ~/src/*' [21:17] and for i in ~/src/*; do bzr st $i; done doesn't show me the particular $i [21:17] also it recognizes svn checkouts and starts the slow conversion-into-cache process, any way to disable that? [21:17] do echo $i; bzr st $i [21:17] i use 'mr' for such jobs though, so i don't know what The Bzr Way is [21:18] doesn't mr require you to have a config file explicitly listing all your checkouts? [21:18] yes [21:18] is there a bzr --disable-plugins or something? [21:19] Doesn't stat give useful exit codes depending on changes vs. not? I'd go with that over output... [21:19] it doesn't seem to [21:20] ah, found 'bzr help global-options': there's a --no-plugins [21:20] You could use -V and pipe that to wc -l and see if it's bigger than zero, or something [21:21] * mgedmin ends up with for i in *; do test -d $i/.bzr && { echo + $i; bzr --no-plugins st $i; }; done [21:22] dang [21:22] hm, is there a way to get loggerhead tarball downloads to extract to only 1 directory named after the project? And can you use tags instead of revision numbers? [21:22] with subversion I can filter out unknown files with |grep -v ^? [21:22] how do I do that with 'bzr st'? [21:22] mgedmin: -V does that [21:22] ah! [21:22] so _that's_ what "only show versioned files" means! [21:22] heh [21:23] read it, didn't understand it [21:34] Does anyone know how I can see raw files without knowing their internal ID? https://bugs.launchpad.net/loggerhead/+bug/605775 says it should be possible [21:34] Launchpad bug 605775 in loggerhead "Loggerhead doesn't support linking to the raw content" [Medium,Fix committed] [21:35] (or should I ask in #launchpad?) [21:37] * sjamaan tries there [21:45] * mwhudson rediscovers loggerhead's context_url "feature" [22:05] jelmer: around-p? [22:06] behold my greatest creation: ~/bin/whatsmodified -- http://pastie.org/3150920 [22:40] duh, how did I miss bzr st -S? [22:43] I doSn't knoSw. MaySbe Syou haSSve trSouble SsSeeing capSital sS'Ss... [22:43] :) [22:43] And he kept trying: bzr st - [22:44] IT JUST WON'T WORK! [23:09] soo [23:10] bzrlib.delta.report_changes will tell you that the executable bit changed [23:10] but not in which direction [23:15] I think I've grumbled about that at the UI level in the past. [23:25] yeah, i presume that's why status is as it is [23:30] It annoyed me in log. It's _really_ hard to understand what happened to +/-x over history. [23:31] diff displays it nicely