[00:05] <lifeless> I think --xml is in a plugin
[00:05] <lifeless> written for IDE integtation
[00:22] <mnn> hi everyone... has been there any progress on this?
[00:22] <mnn> https://blueprints.launchpad.net/bzr/+spec/tls-and-ssl-support
[00:24] <Noldorin> mnn: yeah, sounds highly advisable. the python 2.7 argument doesn't quite work for me...
[00:24] <Noldorin> it can always be conditional functionality
[00:24] <Noldorin> on the build or version of python installed
[00:25] <mnn> and what about gnutls along with python-gnutls?
[00:25] <mnn> that way python 2.7 won't be required
[00:25] <Noldorin> py2.7 seems like the more elegant way
[00:25] <Noldorin> but yeah, that's feasilbe
[00:25] <Noldorin> especially as a plugin
[00:26] <Noldorin> mnn: there way forward with this, as with most experimental/new technology in bzr, is a demonstration plugin
[00:26] <Noldorin> as proof-of-concept at the very least
[01:54] <mnn> Noldorin: I have little knowledge about ftp/ssl but I could try it... anyway I have more pressing concern:
[01:54] <mnn> https://bugs.launchpad.net/bzr/+bug/1027529
[01:54] <mnn> seems like old issues with SFTP transport are crawling back:
[01:54] <mnn> https://bugs.launchpad.net/bzr/+bug/681047
[01:55] <mgrandi> i haven't had any problems with sftp on windows or mac...
[01:56] <mnn> however, I believe the sftp server I'm using is not hosted on windows
[01:57] <mnn> most likely linux
[03:07] <seany> is it possible to `bzr log` all revisions that aren't part of the branch anymore?
[03:07] <seany> i.e. do not have revno anymore
[03:38] <lifeless> seany: bzr heads
[04:40] <ls3> hello. Is it safe to remove some older /tmp/bzr-index-* files ?  (2 wks+)
[04:40] <ls3> I'm trying to read about what they are but all I've gained is they are used for caching
[04:41] <ls3> i just have 440 3M files clogging up /tmp ..
[05:20] <spiv> (in case ls3 comes back): anything bzr leaves in /tmp is safe to remove, and probably also a bug (in the bzr-index plugin maybe?)
[07:45] <seany> lifeless: thanks!
[08:03] <jelmer> moin
[08:24] <mgz> morning
[13:18] <bzrnewb> hi guys
[13:19] <bzrnewb> Anyone around?
[13:21] <jelmer> hi bzrnewb
[13:21] <bzrnewb> hey jelmer
[13:22] <bzrnewb> are you well acquainted with bzr hooks by any chance?
[13:22] <bzrnewb> i require very specific assistance and google seems to be failing me today
[13:23] <jelmer> bzrnewb: a bit
[13:23] <jelmer> bzrnewb: please just ask your question - if there's somebody around who can help they'll hopefully speak up
[13:23] <bzrnewb> sweet thanks
[13:23] <bzrnewb> well
[13:25] <bzrnewb> is it possible when pushing to run a post_push hook only if there are changes in a specified directory
[13:25] <bzrnewb> ?
[13:25] <jelmer> bzrnewb: inside of your hook you could check if there were any changes?
[13:27] <bzrnewb> ah that's not a bad idea.
[13:27] <jelmer> bzrnewb: the hook params should contain the old and the new revid
[13:27] <jelmer> so it should be a matter of just checking that those are different
[13:28] <bzrnewb> Would you do it file by file?
[13:28] <jelmer> bzrnewb: what do you mean exactly by "changes in a specified directory" ?
[13:28] <bzrnewb> excuse my ignorance here, I've only just started setting up a continuous integration environment at my new office
[13:28] <jelmer> bzrnewb: committed changes that were pushed that weren't present remotely before the push?
[13:28] <bzrnewb> yes
[13:29] <bzrnewb> but only for one of the directories in the branch
[13:30] <jelmer> bzrnewb: you could probably call Repository.revision_tree() for old-revid and new-revid, and then use new_tree.changes_from(old_tree) to get the list of files that have somehow changed
[13:30] <bzrnewb> great stuff
[13:30] <bzrnewb> I'll go investigate
[13:30] <bzrnewb> Thanks jelmer