[00:01] bjp_: those are daemons [00:09] so will configuration file support be dropped eventually? it seems like it has a lot more options than just serve-branches [00:10] hmpf. okay, I now see how to fix it, but still not sure *why* r4781 broke my code [00:15] ah, now I do. before stopTest was being run before the final finally, which replaces the TestCase __dict__ and now it's being run afterwards [00:15] not the easiest diff in the world to read. [00:27] what is causing this? : Unsupported configuration: PasteDeploy not available, but loggerhead appears to be behind a proxy. [00:38] bjp_: you haven't installed python-paste-deploy [00:40] rar: sorry :) [00:40] what was syour code doin that this has broken [00:44] replacing a test with another one, but keeping the original test around for later [00:44] If a svn repository has switched from one layout to another during its lifetime, is it likely to completely confound even bzr-svn's sophisticated mappings? [00:44] rar: ?! [00:44] you also fixed the never-not-applicable bug though, so I could remove my workaround for that [00:44] maxb: I don't know - give it a go [00:44] Well, it's dumping crash reports at me :-) [00:45] maxb: file bugs ;) [00:45] against the mindset of people who rearrange their svn repositories? :-) [00:45] yes:) [00:46] basically it's a thing for raising a test result without having to run the test, as some of them blow up in violent, suite-destroying ways [00:47] rar: I am totally confused; is there code I can see? [00:47] I don't understand how a test can destroy a suite [00:47] well, test_breakin_harder likes hanging the process [00:48] it should be a lot more robust nowadays, I hammered on it recentl. [00:48] and there are some others that blow the stack on ironpython, and I've yet to track down what exactly is making jython die [00:49] ic [00:49] at any rate, it's a hack, but a useful one for the moment [00:49] rar: oh, you're working on other vms - cool [00:51] maxb, it shouldn't cause a crash at least, at most it should cause your history to be thinner than you would expect [00:53] anyway, the rev is fine, it just meant I needed to discover attrs_to_keep (understanding all of bzrlib.tests is... not a one sitting thing) [00:55] rar: I still don't get why you need to poke at that level === iBasic____ is now known as iBasic [00:55] well, another for instance, currently if any of the test modules has an import that fails, the whole test run stops [00:56] some of the test modules import bz2 or signal or something, that other imps don't have [00:56] file a bu on that please [00:56] (the test suite not running) [00:57] I'm not really sure it's a bug, and the best fix I have has issues [00:57] I'd still like a bug :) [00:57] taggest selftest [00:57] the bits of this that aren't nasty workarounds I will try and get in [00:59] anyway, monkeypatched in a bit so it registers as a kind of failure instead, eg: [00:59] http://float.endofinternet.org/temp/bazaar_j_noncpython_bt.test_#bt.test_smart [00:59] (and the wide red bars just under that) [02:37] is there a way to add a new file with the history of an existing one? [02:37] for example having one file that you are breaking into two? [02:37] or is that not possible? [02:40] thumper: You mean something like "bzr cp old new", right? I'm fairly sure that's not (yet) supported, but is something that might be worked on at some point. [02:41] RAOF: thanks [04:07] jml: the room is open, btw [06:43] Hello. [06:43] Is there any plan to translate help strings? (ex: bzr help, bzr help commit, etc...) [06:44] does the bug #375013 fixed? [06:44] Launchpad bug 375013 in rosetta "Cannot commit directly to a stacked branch" [High,Triaged] https://launchpad.net/bugs/375013 [07:05] SuperMMX: no [07:18] mwhudson: thanks, just asking [07:18] SuperMMX: np, i understand it's trickier than you might thing [07:18] *think === loxs_wrk is now known as loxs [10:16] Hi GaryvdM [10:17] GaryvdM: I won't mind if you rename trunk2a [10:18] bialix: Ok - I'll do that now. [10:18] wait a sec [10:20] GaryvdM: I've just replied to your mail [10:20] I have concern re stacked btanches. [10:20] * GaryvdM goes to read it [10:20] GaryvdM: does they will re-stack automatically? [10:20] should we upgrade lp:qbzr/0.14 as well? [10:22] poor roots - lol [10:22] bialix: I think that stacked branches on launchpad will automaticly point to the renamed branch. [10:22] But I'll try double check [10:23] poor roots (c) fullermd [10:23] GaryvdM: ok, perhaps we just need to try and if things going wrong we need to rename them back [10:23] yes [10:24] bialix: But I'll try find out anyway before I make the change [10:27] ok [10:34] GaryvdM: I've sent mail to luks and ask about admin rights for you [10:34] Thanks [10:36] let's increase bus factor ;-) [10:40] bialix: renaming trunk did break the stacked branches, so I reverted it. [10:40] * GaryvdM => lunch [10:41] GaryvdM: can you write it in mail please, so our beloved core devs will be aware of this problem? [10:41] bon appetit [10:41] yes - I will. [12:33] verterok: hi, we had a big blackout here, sorry [12:33] verterok: I worked in the test failures, so I improved the code: http://bazaar.launchpad.net/~renatosilva/bzr-java-lib/encoding-fixes/revision/206 [12:34] verterok: tests are all ok here, both in mvn and eclipse: http://img195.imageshack.us/img195/2136/testsok.png [12:35] verterok: would be nice if you merge rev206 in your branch to test in your environment [12:36] RenatoSilva: I was reading the irc log, and I saw that you said it you be nice if qlog has a search function. [12:36] GaryvdM: yes [12:36] GaryvdM: it has, but not for what I want [12:36] GaryvdM: it does not search in the actual changes [12:37] RenatoSilva: Ah - It does integrate with bzr-search. Install that, index your branch, and it will also search in the file content. :-) [12:38] GaryvdM: file content or file changes? [12:39] Um - not sure. Same as what bzr-search does. I think content. [12:40] verterok: also, your hudson server successfully built the bzr-java-lib-xp (encoding fixes) :) [12:41] GaryvdM: what I want is for example bzr search "print 'hi'", then get all the --revisions-- which have added or removed such a line [12:42] RenatoSilva: I see. - Please will you log a bug in bzr-search for that. [12:42] GaryvdM: you are a bzr-search developer? [12:43] No - but I'm getting in to it. [12:43] I've submitted my first merge proposal. [12:44] RenatoSilva: lifeless is the author. If you read the comments in the code, you will see that he was thinking about that idea. [12:57] GaryvdM: have you looked at https://code.launchpad.net/~craighewetson/qbzr/tag_from_qlog/+merge/13096 ? [13:02] GaryvdM: bug 480684 [13:02] Launchpad bug 480684 in bzr-search "Return matching revisions" [Undecided,New] https://launchpad.net/bugs/480684 [13:11] hi - is it possible to add something to .bzrignore to ignore symlinks [13:11] thrope: nope [13:12] ok thanks [13:13] presumably a precommit hook could filter them out though [13:13] yep, but you need to explicitly ignore them then [13:17] Is there some way to subscribe to early notifications when a new version have been uploaded to launchpad? [13:19] new version of what? [13:19] bzr [13:21] there is bazaar-announce ML, not sure it will serve you well in the case of "early" notifications [13:22] No, that's late.. want to get notified when the release is available for download so I can push it to the Fedora updates-testing repository, ready to be pushed out as an update when the update is formally announced. [13:24] I'm monitoring main bzr ML for the messages like this: bzr 2.0.x gone gold === mrevell is now known as mrevell-lunch [13:24] you need to ask for core devs maybe they can suggets something or improve their process [13:25] for example 2.0.2 was uploaded to launchpad a week ago, but not yet announced. [13:26] Guess I need to do the same.. === mrevell-lunch is now known as mrevell [14:00] hi all [14:01] I want to swipe my whole office to use bzr [14:02] should I setup a bzr server on my machine to poeple could branch my code and push code to it ? [14:02] I don't want a central repository [14:03] I see on the docs there's no special need for a server, but everybody here is used to svn [14:03] so.. [14:06] sivang, Why not peruse http://doc.bazaar-vcs.org/migration/en/foreign/bzr-on-svn-projects.html and see about persuading them that way? [14:07] awilkins: I'd rather have bzr working for me, show them how easy it is and then be accepted without too much persuation === herb__ is now known as herb [14:08] sivang, I'm using bzr on a few projects that still use SVN servers ; you can demonstrate all the advantages without divorcing yourself from the SVN server [14:08] awilkins: right, I will read now the documentation about this [14:20] awilkins: where the general guidlines and starting documention for bzr ? is James Black still working on docs ? [14:23] sivang: try http://bazaar-vcs.org and then use the Documentation link at the top [14:24] sivang: no, James Black no more participate in Bazaar project [14:24] now Ian Clatworthy is our champion [15:13] is there a way to shelve all changes? instead of going through one by one? [15:14] jsmith23: bzr shelve --all [15:14] bzr shelve --all ? (bzr help shelve) [15:15] gioele: smartgpx: sure enough, first line of the help [15:15] thank you [15:36] does anybody use Eclipse to develop bzr? === rockstar` is now known as rockstar [15:48] how come the launchpad plugin now uses launchpad_username instead of [Launchpad] inside authentication.conf? [15:50] gioele: I think it uses both [15:53] bialix: it just complained that the login name was not set (it has been in auth.conf since ages). Using "bzr launchpad-login gioele" created a brand new entry in bazaar.conf (my ~/.bazaar/ is bzr-managed ;)) [15:54] without looking into lp code I'd say entry in bazaar.conf used as flag [15:54] or maybe for lp: access? === khmarbaise_ is now known as khmarbaise [16:09] I found some off-by-one whitespace problems in builtins.py, should I correct them and send a patch? [16:27] is there a way in bzrlib to send a message to the user? I'd like to ask her "Are you sure to continue? [Y/n]" [16:29] gioele: see ui package [16:29] there is ask_boolean === bialix_ is now known as bialix [16:34] bialix: thank you! (get_boolean, btw) [16:35] my memory as sieve, silly me! [18:02] Is there any programmatic way to query the format of a *remote* branch? [18:03] maxb, via http you can get it, so I expect you need to look at a specific file [18:03] maxb: `bzr info remote` or use BzrDir [18:04] LarstiQ: bzr info lp:bzr, for example, just says format: unnamed [18:04] maxb, because it goes via bzr+ssh [18:04] if you use http:.. [18:04] it will tell you the truth [18:04] it;s something about the smart server [18:04] hm [18:05] maxb: how about -v? [18:06] LarstiQ: Just says the format is a remote one [18:06] Is there any way to override ~/.bazaar/bazaar.conf settings on the command line? [18:06] you need to fallback to vfs to read it [18:06] maybe you can specify that in the API call [18:09] Hmm. HOME=/dev/null bzr info lp:bzr [18:09] hacky, but works [18:09] Is is a shame there isn't a lphttp: [18:11] You can use nosmart+bzr+ssh (and I think nosmart+lp too), which falls back to VFS methods. [18:12] Yeah, nosmart+lp expands through to nosmart+bzr+ssh === lamont is now known as lamont` === lamont` is now known as lamont [18:35] fullermd: yay, that's exactly what I need, thanks :-) [19:45] Hello all. I have a question: I have a situation where I just finished merging in some changes from what we will call our "trunk" into a development branch. I went to push it to that branch and got an error. After investigation, I determined someone already tried to make the same change to the branch. So I pulled and got diverging branches as expected, but what I want to do is [19:45] drop everything that is not mine, rather than merge [19:45] Is there a way I can do this? [19:46] I know that's the correct solution, I'm just not sure how exactly to go about doing it [19:46] Basically I have diverging branches and I just want to kill one of the branches [19:50] So currently I have a revision showing up as "missing" and I just want to kill that revision so it never gets merged in, I guess that's what I want to do [19:53] Shirik: push --overwrite [19:53] or uncommit [19:54] that worked, thanks [19:59] if 2.1.0b2 is gold, why isn't it in the bzr-beta-ppa? [20:01] also the bzr nightlies don't seem to be building either === jam3 is now known as jam [22:02] I'm writing a test for a little addition I made to cmd_commit. I expect bzr to ask me a question via get_boolean. How can I check that? === AfC1 is now known as AfC [22:35] gioele: there is a testing UI factory [22:35] lifeless: hi [22:35] gioele: where you can supply the inputs it gets [22:35] lifeless: who is responsible for the PPAs? [22:35] spyuz [22:35] spyuz [22:35] bah [22:35] soyuz [22:36] lifeless: who uploads the bzr stuff to the ppa? [22:36] as the daily and beta are out of date [22:36] and I've been bitten by a bug that jam has fixed [22:36] bzr keeps crashing on me [22:36] is very frustrating [22:36] jamesw does the daily, johnf is working on fixing the beta, currently sorting out a bunch of packaging differences to make it more scriptable [22:37] jamesw is on leave [22:37] I may just grab bzr.dev and use that [22:37] for now [22:37] :( [22:37] not done that yet on this computer [22:37] lifeless: what dependancies to I need to build bzr.dev fully? [22:38] apt-get build-dep bzr [22:38] + python-pyrex [22:39] bbiab [22:44] lifeless: I had a look around that but I could not find method of TestCommit that I could use to assert "now this question is asked and a Y/n prompt is presented to the user" [23:02] lifeless: poos [23:02] lifeless: my crash is still in bzr.dev [23:02] thumper: whiss [23:11] lifeless: I'm getting something like bug 461643 [23:11] Launchpad bug 461643 in bzr "Can't commit, StaticTuple is not a tuple" [Undecided,Invalid] https://launchpad.net/bugs/461643 [23:12] lifeless: but this is with a clean bzr.dev and I ran make [23:12] * thumper files a bug anyway [23:14] WTF??? [23:15] lifeless: I thought I was running my new bzr.dev, but it seems it isn't [23:16] g'ah [23:16] ok, is fixed in dev [23:18] * thumper needs new deb badly [23:33] gioele: I will try to look it up for you, at a meeting at the moment [23:34] lifeless: thank you [23:35] lifeless: anyway, I'm leaving but I sent a RFC on a tiny patch on the mailing list; I asked the same question in that mail === lifeless changed the topic of #bzr to: Bazaar version control system | 2.0.2 and 2.1.0b2 have gone gold! time to build those installers| try https://answers.launchpad.net/bzr for more help | http://bazaar-vcs.org | http://irclogs.ubuntu.com/ | Current pilot: -coming soon- [23:39] morning all [23:39] I have a plea for help =] [23:40] Long story short, a local branch got accidently bound to a remote branch, and then updated; all the changes in the local branch were lost [23:40] the local branch was in a shared repository, so hypothetically the data should still be in there somewhere [23:40] does anyone know how I could find/get to it? [23:41] They're there whether it's a shared repo or not. [23:42] You can probably use the heads plugin to take a guess at the revid. [23:42] even though the branch itself is now bound to a different branch? and updated? [23:42] *looking for heads plugin* [23:42] And then use pull to set that as the head (after unbinding of course) [23:43] If you bound and update'd, but then didn't revert or anything, it's actually sitting right there as a merge rev, but there's no way to sweet-talk the UI into telling you that revid. [23:43] the plugin is called "heads"? === mm-mysql is now known as mark|dinner [23:43] It's included in bzrtools these days. [23:44] ahh; I see it. running it now. [23:44] thank you [23:48] I think I found the revision I need; how do I get to it now? using bzr pull? [23:52] actually, no, it isn't listed there [23:52] none of the heads I need are listed :-( [23:52] rather none of the listed heads are the one I need [23:55] Are you running head --dead-only?