=== doko__ is now known as doko [02:03] Is there a reason why bzr for os x doesn't give any feedback during a branch operation? [02:25] PyCon registration is open: http://us.pycon.org/2008/registration/ [02:25] awmcclain: no good reason, it's a bug [02:26] awmcclain: (a bug that happens on all platforms) [02:26] spiv: When I branched on linux, i got a status bar. At least when branching from svn... [02:27] awmcclain: hmm, I don't think there should be any difference between mac and linux for status bars. [02:27] spiv: I'll look into it. [02:43] hello kiko === Verterok-laptop is now known as Verterok [02:52] hey poolie [02:52] hi, are you home now? [04:16] trying to install with macports, but installing macports itself doesn't seem to work. (port: command not found) [04:16] any advice? [04:19] guillaumebokiau, i think you need to install the macports dmg [04:19] which i did :| [04:19] guillaumebokiau, http://www.macports.org/install.php -- then you'll have a port command [04:19] hm [04:19] is it on your $PATH? [04:19] you might need to start a new terminal window? [04:20] i even rebooted [04:20] what is $path ? [04:20] sorry, not very good at those things :/ [04:21] please type 'echo $PATH' [04:21] into your terminal [04:22] also, 'echo $0' [04:22] (and press enter) [04:22] \Library\Frameworks\Python.framework\Versions\Current\bin:\bin:\sbin:\usr\bin:\usr\sbin [04:22] see, http://guide.macports.org/#installing.shell [04:23] guillaumebokiau, ok, that's probably your problem [04:23] try the instructions there [04:23] ok, thx [04:30] guillaumebokiau, did that work? [04:30] trying to figure this out :) [04:31] guillaumebokiau, just for curiousity, why are you using macports rather than a dmg? [04:31] did we not have one that would suit? [04:31] isn't the dmg for ppc ? [04:31] i'm on intel [04:31] 10,4 [04:32] ok [04:32] yes, that's it [04:32] the dmg should be universal I think [04:32] i mean, i wondered if that was the problem [04:32] we need someone with a intel 10.4 mac to rebuild it [04:33] i tried with the ppc dmg anyway, but it didn't work [04:36] guillaumebokiau: yes, checking again, the 10.4 installer is only ppc only right now sorry [04:37] there's no other way to install it? [04:37] than with macports? [04:37] guillaumebokiau: running straight from source is really easy actually [04:38] simply untar and put the directory on your path [04:38] you'll need to install Python 2.4 or 2.5 separately [04:38] installed that [04:38] in not already installed [04:38] s/in/if/ [04:38] er, what's my 'path' ? [04:39] (i'm really new to all this) [04:39] it's where your terminal looks for commands you type [04:39] that : \Library\Frameworks\Python.framework\Versions\Current\bin:\bin:\sbin:\usr\bin:\usr\sbin ? [04:39] you can create a file in your home directory to set it each time you start a terminal session [04:40] guillaumebokiau: when you type "bzr", your shell looks in each of the directorys in the PATH variable (:-seperated) for that command [04:40] without setting a path just yet, let's test it works [04:41] unpack the tar.gz [04:41] change to the created directory and .. [04:41] type "./bzr version" [04:42] have a hard time following. where do I unpack? [04:42] anywhere you like ... [04:42] If you double-click ... [04:42] ok, it's on my desktop :) [04:43] cool [04:43] in a terminal, change into that directory [04:43] try "./bzr version" [04:45] guillaumebokiau: did that work? [04:46] hang on :) [04:46] the directory to change into is something like "Desktop/bzr-1.1" [04:47] ok, i'm there [04:47] ~/Desktop/bzr-1.1 [04:48] great [04:48] does "./bzr" work? [04:48] yes! [04:48] sweet [04:48] cool [04:48] now let's put it on your path [04:48] type this: [04:48] yeah [04:49] PATH=~/Desktop/bzr-1.1:$PATH [04:49] then change to another directory and try "bzr" [04:49] it"s on the path when i echo it [04:50] working [04:50] great [04:50] ok, let's make it permanent [04:50] type "cd" to return to your home directory [04:50] done [04:50] should i copy the folder there? [04:51] with your text editor, create a file called ".prfile" if it doesn't already exist [04:51] oops ... [04:51] make that .profile [04:51] inside it put this text ... [04:52] PATH=~/Desktop/bzr-1.1:$PATH [04:52] export PATH [04:52] (that's it) [04:52] to test ... [04:52] start another Terminal, in a tab or new Window [04:52] type "bzr" [04:52] sound ok? [04:53] bzr command not found :( [04:54] ok ... [04:54] type "echo $PATH" [04:54] it's not in it [04:54] maybe ~/.bashrc would be better, .profile is only sourced on "login" [04:54] good point [04:55] depending on what os x considers login to be [04:55] i hav a .bash_profile file [04:56] guillaumebokiau: try putting those 2 lines in there then [04:56] k [04:57] ok, i got it working [04:57] pfew [04:57] good [04:57] i think the correct line was PATH="~/.bzr-1.1:${PATH}" [04:58] anyway, i thanks a lot [04:58] that ok [04:58] either syntax should work [04:58] you saved me :) [04:58] let us know if you need any more help [04:58] no problem [05:15] New bug: #184733 in bzr "Internal error running bzr selftest for bzr 1.1 on MacOS X 10.5.1" [Undecided,New] https://launchpad.net/bugs/184733 [05:24] where does bazaar save local copies ? [05:24] into .bzr in the top of the branch [05:25] not sure you understood the question :) if I run bzr branch http://bazaar.launchpad.net/~manfre/xpattern/xpattern-1.0 where does it land ? [05:26] ah, found [05:26] ok, sorry, i'm officially stupid [05:27] right, into your working directory [05:27] no problem, questions are welcome === n2diy_ is now known as n2diy [05:52] how come it says Unable to load plugin 'bzr_svn_0_4_6' from '/home/ian/.bazaar/plugins' ? [05:52] when trying to do a bzr svn clone [05:52] I do have the plugin there [05:53] eean: look in ~/.bzr.log [05:54] eean: it should have a more detailed error [05:54] ImportError: The Subversion plugin must be installed as bzrlib.plugins.svn not bzrlib.plugins.bzr_svn_0_4_6 [05:54] so I'll rename the directory I guess [05:54] Right. [05:55] there it goes [05:55] I hear it doesn't have a memory leak anymore [05:55] appears so, so far [05:56] hmm [05:56] it is increasing in memory steadily :/ [05:57] eean: it depends on how new your subversion bindings for python are [05:57] aaaah [05:58] spiv: do you happen to know how new they have to be? [05:58] it is not using crazy amounts of memory like it used to, but its certainly still leaking. [05:59] eean: SVN version of the as of about or week or so ago, IIRC [05:59] that would make sense that the subversion bindings would be to blame, I didn't understand how a python script could leak. so its probably something in the c bindings [05:59] haha crap [05:59] eean: I think the bzr-svn README links to the relevant bug [05:59] riddell blogged before then that it had been fixed in bzr-svn [05:59] eean: I think newer bzr-svn tries to workaround the worst of it, but obviously it can only do so much... [06:00] I see [06:00] eean: yeah, see the first part of the FAQ in the bzr-svn source :) [06:00] "bzr-svn can use a lot of memory cloning big branches when older [06:00] versions of the python-subversion bindings are used. This memory leak [06:00] has been fixed in the trunk of Subversion (r28544) and has been [06:00] proposed for inclusion in Subversion 1.4.7. [06:00] " [06:01] oh wow, its subversion itself to blame? [06:01] eean: it's possible this fix has already been included hardy, I haven't checked... [06:01] well I use gutsy [06:01] eean: right, the python bindings are made by the subversion guys. [06:01] ah ok [06:02] I can probably make due if I quit and redo every 100000 commits [06:02] Yeah :) [06:03] before it would eat all my memory at about 1000 [06:03] git svn is giving me shit, so this is a good time to switch ;) [06:04] :) [06:05] I guess this bug has probably made bzr svn clone pretty robust at restarting [06:05] git svn couldn't deal with the svn server timing out, after about 30 minutes of cloning :/ [06:07] pfft, or not. "bzr: ERROR: sqlite3.OperationalError: SQL logic error or missing database" [06:08] eean: ooh, I haven't seen that one before. [06:08] http://pastie.caboo.se/141399 [06:08] on resuming [06:08] eean: could you file a bug report? I expect jelmer would like to see that. [06:08] /dev/sda8 44G 42G 1.7M 100% /home [06:08] might explain it :) [06:08] eean: ah :) [06:24] will I be able to share this checkout with other devs? [06:24] I'm kind of worried that it appears to be downloading into ~/.bazaar [06:25] eean: yes [06:25] eean: the stuff in .bazaar is just a cache of the mapping between data the SVN repo and the data converted to bzr [06:26] eean: if the cache is missing or incomplete, bzr-svn will just regenerate it as needed. [06:30] bah my laptop power cord got loose [06:30] spiv: did you say anything after 'yes'? [06:30] eeanm: I did, I'll paste: [06:30] 17:25 < spiv> eean: the stuff in .bazaar is just a cache of the mapping between data the SVN repo and the data converted to bzr [06:30] 17:26 < spiv> eean: if the cache is missing or incomplete, bzr-svn will just regenerate it as needed. [06:31] um that mapping the hard part [06:31] um *isn't that mapping the hard part === jamesh_ is now known as jamesh [06:35] at least when you're cloning a 2.5k LOC, 200 commit module from a SVN with >700000 commits [06:35] yes, but people branching from your bzr repository won't need it, all they need is in the bzr repository [06:35] the mapping is a big chunk :| [06:36] eeanm: people only need the mapping if they want to use the SVN repo [06:36] well yea [06:37] eeanm: I can e.g. take your bzr branch, and make a branch off that, make some commits, and send those back to you, all without touching the upstream SVN repo. [06:37] suppose I could do it that way, but it'd be nice if folks could push to svn themselves [06:37] They'll need the mapping then. [06:38] won't be such a big deal once this memory leak fix is availabe [06:38] You can probably share the cache from your .bazaar directory. [06:39] sounds like bzr has their act more together then git, we try publishing a git svn repo, someone commited something that they didn't end up pushing to SVN, it totally screwed it up [06:39] would be nice to have feature branching in bzr that could be easily synced with svn [06:41] (that's what we were trying to do with git) [06:42] eeanm, well, that should work through bzr-svn [06:57] dang the cache is up to 771 megabytes [06:58] how big is the svn repo? [06:59] >700000 commits [06:59] 764k [06:59] I'm up to 421k [07:07] sqlite> select count( * ) from changed_path; [07:07] 3578513 [07:08] probably has something to do with it :) [07:25] 529215/764151 woot [07:54] * igc dinner [08:33] the svn clone is soo near completion, and now its just hanging apparently :) [08:33] erm [08:33] :( [08:38] eean: Is the Subversion repository in question public? If it is, I'm sure people will be willing to attempt to replicate the problem. [08:39] yea, its KDE [08:39] ah ok cool [08:39] nevermind [08:39] now its "analyzing layout" [08:40] about 4 minutes there when it didn't respond, I was worried :) [08:40] heh analyzing layout is going to take an hour or so it looks like. at least it doesn't leak memory. [08:41] ah no, that's not true, more like 15 minutes I bet, its already 1/7 done [08:49] hi [08:55] jelmer: sorry, missed that in backlog, only saw your ^^, what was it ? [08:56] eean: You might want to reply to poolie's RFC about finer grained progress reporting on the mailing list today and note your experience of long waits without feedback [09:30] hi all [09:32] hi [10:43] abentley: hi [10:43] abentley: how is sydney :) [10:43] Quite nice. [10:43] Weather's definitely superior to Toronto's. [10:44] Although pretty similar to Toronto in summertime. [10:44] lifeless: it is hot and muggy [10:44] too hot and too muggy [10:44] but that's from my Dunedin point of view [10:44] This is the first sprint I've ever done without you, though, which is a bit wierd. [10:45] thumper: Toronto in the summertime can be quite muggy, too. We're on the edge of Lake Ontario, one of the Great Lakes. [10:45] thumper: unedin is easily as humid; its the heat that does it [10:49] Greetings chaps. [10:51] Are the London sprint dates set in stone? Just it clashes with SXSWi in part. [10:51] Though in any case I would try a pop along on the Monday if I can [10:52] muffinresearch: what is SXSWi? [10:52] south by south west [10:52] http://sxsw.com [10:56] Is coming to the sprint, a good idea for people who want to start getting more involved? [10:58] http://ifaedi.insa-lyon.fr/~asabil/bzr-viz-tags.png [11:00] Or is it more of a closed event for current core developers? === cprov-afk is now known as cprov === mrevell is now known as mrevell-lunch === mrevell-lunch is now known as mrevell [13:44] does bzr svn not follow svn moves? [13:44] * eean looks at the website again [13:45] svn does not have move [13:46] um, yes it does [13:46] it has copy, iirc. [13:46] and move [13:46] but anyways [13:46] its possible either way [13:46] it has sime kind of add with history? [13:46] does bzr not follow them? [13:46] Akheron: yes [13:47] eean: svn only has copy+delete [13:47] that sounds awfully like moving to me [13:47] and there is a svn move command [13:47] eean: in order to be able to convert a copy+delete to a rename you have to prove there is only one copy of a file and that the original was removed [13:47] I don't care if its seen as a rename or not === brilliantnu1 is now known as brilliantnut [13:48] "svn move" is an alias for "svn copy"+"svn delete" [13:48] but it thinks this 8 month old project is only a day old [13:48] since it was copied+deleted yesterday [13:48] it should follow copies of branches [13:49] (if you copy /trunk to /tags/2.0 it should recognize that) [13:49] ah ok, well it didn't. I'll double check the svn log [13:50] is this a public repository? [13:50] yes [13:50] svn://anonsvn.kde.org/home/kde/trunk/KDE/kdemultimedia/dragonplayer [13:51] well svn log tracks its history through all the reneames [13:51] ah, the KDE repository [13:52] this project is entirely post-CVS if that's what you're thinking :) [13:52] I heard that conversion messed some things up [13:52] At the moment the copy tracking only works for the current "branching scheme" [13:52] ah [13:52] by default that's the standard /trunk ; branches/* ; tags/* structure [13:53] jelmer: is that what you wanted : http://ifaedi.insa-lyon.fr/~asabil/bzr-viz-tags.png ? [13:53] jelmer: can you add this as a known problem? [13:53] eean: We're trying to make that deal with other situations as well, but that won't happen until 0.5.0 [13:53] cause I just wasted 3 hours last night, restarting bzr svn constantly :| [13:53] eean: There's already a bug for it [13:54] (iirc I had a discussion yesterday with you and Schierbeck) [13:54] I mean, on the website [13:54] asabil: Ah, that looks quite nice [13:54] eean: will do, thanks [13:55] jelmer: want a branch to give it a try? [13:55] cool thank you [13:55] asabil: any chance you can send an email to the bzr-gtk list? [13:55] jelmer: I am not subscribed :/ [13:55] asabil: Should be easy to do :-) Otherwise I can manually approve your email [13:56] jelmer: I was subscribed a while ago [13:56] but the list is very high traffic for me [13:56] asabil: it's a lot quieter these days :-/ [13:56] jelmer: do you plan on optimizing how you use the sqlite database? I suppose personally I don't optimize until after I'm finished, so I can't blame you for not. :) [13:57] cause for KDE its 1.9gigs [13:57] eean: the focus so far has mostly been on correctness [13:57] eean: hopefully we can cache less in the near future [13:58] corrections is paramount: performance is a bonus. Especially in this sort of product. [13:58] right [13:58] s/corrections/correctness/ [14:04] jelmer: https://code.launchpad.net/~asabil/bzr-gtk/viz-tags-fancy [14:04] jelmer: please feel free to mail the list about it, if you want [14:04] I am busy with other things right now sorry [14:05] I edited the branch summary to point to the screenshot [14:32] asabil: hello! [14:33] hey schierbeck [14:33] schierbeck: https://code.launchpad.net/~asabil/bzr-gtk/viz-tags-fancy [14:33] i've just seen it, great job! [14:33] are you on the mailing list? [14:35] nop schierbeck [14:35] please feel free to mail the list about it if you want [14:35] and thanks :) [14:36] i'm currently looking into some small changes, like decreasing the font size, having the tags stack horizontally so they're all visible, and using a single color for all tags [14:37] decreasing the font size is as simple as [14:38] tag_layout.set_markup("" + tag + "") [14:38] instead of tag_layout.set_text(...) [14:38] (but am not sure if you want to use tags) [14:39] making the tags stack horizontally is a matter of changing x0 and leaving y0 fixed [15:05] asabil: ok, i've made a few changes, they should be at http://bazaar.launchpad.net/~bzr-gtk/bzr-gtk/viz-tags [15:10] schierbeck: did you try the code with many tags ? [15:10] there is too much space between the tags I think [15:10] asabil: only with two -- i'll check it out now [15:11] but it looks nicer indeed [15:11] there is just one issue [15:11] I am not sure about using markups [15:11] yeah, it screws up [15:11] as it might blow up if ^evil^ developers decide to use pango markup in their tags :D [15:12] asabil: we'll just escape the tag text [15:12] we do that with the summary already [15:12] if it is just about making the text smaller [15:12] no, it's also about tooltips and such [15:12] you can set the font directly [15:12] oh ok [15:13] how can I apply a bzr diff between releases 239 and 240 for example? [15:13] "apply"? [15:14] bzr revert -r239..240? [15:14] cr3: bzr merge -r 239..240 [15:14] or -r 240..239, depending on what you want [15:15] dato: cheers, exactly what I was looking for [15:16] New bug: #183821 in bzr-svn "bzr branch fails with AssertionError in parse_revid_property" [Undecided,New] https://launchpad.net/bugs/183821 [15:17] asabil: try pulling again [15:18] it seems to be working now [15:18] imho it looks better [15:19] :) [15:19] asabil: would you be able to make the label have a single, common color? [15:19] like a yellow, post-it-ish one? [15:19] I am not sure if it is good idea [15:20] i think it would be more aesthetically pleasing [15:20] so we need to decide on a color then [15:20] yup [15:20] a hardcode color :p [15:20] :D [15:21] You could make it a red/blue offset, and sell 30-cent cardboard/gel glasses, so people can see them in 3-D! [15:21] :x [15:22] there may lie a few usability issues right there... :) [15:22] although it would be kick-ass! [15:23] I can just see the reviews, though... "I liked bzr overall, but it felt like the tags were attacking me." [15:23] fuck, i always seem to get excited about these projects when i'm studying... my exam is on wednesday! [15:24] Heck, that's _days_ away... plenty of time ;> [15:24] fullermd: we'd still be less scary than git... [15:24] Yeah. git would be 3-d, but would sneak up from behind you. [15:26] yup, and it would hold you down until you iterated all available git commands [15:26] schierbeck: That sounds familiar... [15:27] "Say 'git-fmt-merge-msg'! *SLAP* SAY 'git-fmt-merge-msg'!!!" [15:27] for some reason the time I'm most productive is the days before I have a deadline or an exam or something [15:27] jelmer: me too, it's just that i'm not studying [15:27] everything else seems so damn interesting in comparison! [15:28] Well, I've got this project you could do some work on... would cure that little problem RIGHT up :p [15:28] zzzzzZZZzzz [15:28] huh.. what? [15:29] :) [15:30] :-) [15:30] jelmer: have you checked out my additions to asabil's whoop-cake awesome work? [15:30] http://bazaar.launchpad.net/~bzr-gtk/bzr-gtk/viz-tags [15:30] not yet [15:31] it's on the team repo, so you guys can write to it [15:31] thanks, will have a look [15:31] :) [15:34] schierbeck: you can pull from my branch [15:34] the colour is yellow now [15:36] cool! [15:38] asabil: that looks great! i've merged it into the ~bzr-gtk branch [15:38] you should be able to pull [15:38] schierbeck: ? Your branch you mean? [15:39] http://bazaar.launchpad.net/~bzr-gtk/bzr-gtk/viz-tags [15:48] Hi all. I have a question regarding 'commit --local' in 0.90.0. When I use it, I keep getting conflicts, and now I am wondering if I am using it in the correct way. [15:48] Currently I do: [15:48] bzr co sftp://[snip] [15:48] bzr ci --local [15:48] bzr ci [15:48] and this is where I get conflicts [15:48] schierbeck: can you pull again please ? [15:49] asabil: sure [15:49] not very sure if it looks better [15:49] fell free to test it [15:49] also maybe you want to delete the tag that I added and forgot to remove : [15:49] :D [15:49] the Woot tag [15:50] cool! [15:50] the holes could be a bit bigger, though [15:50] jelmer: new patch sent to the ml, it's pretty trivial [15:50] schierbeck: they don't look nice when bigger [15:51] maybe you can try [15:51] ok [15:57] asabil: okay, i've made them a bit larger [15:57] pull them in [15:57] yep that's better :D [15:58] do you think we can adjust the position of the hole? [15:58] it's just a tad too high [15:58] I think it looks perfect to me [15:58] it is right in the middle [15:58] yeah, could be i'm wrong -- such things depend on the screen and whatnot [15:58] oO how can it be high ? [15:59] asabil: many such things are psychological -- sometimes you need to bend things to make them look even [15:59] :/ [15:59] :) [16:00] oh and btw, I am not sure you needed a bzr merge [16:00] you could have used bzr pull [16:00] nah, you forgot to pull my merger [16:00] i tried [16:00] because there was no divergence [16:01] look at the graph -- i had already merged with your branch once, hence there was a revision in my branch that wasn't in yours [16:02] oh ok sorry [16:02] didn't notice [16:27] np :) [16:46] New bug: #178108 in bzr-svn "Implement SvnRepository.find_branches()" [Wishlist,Fix committed] https://launchpad.net/bugs/178108 [16:59] asabil: do you think the branch is ready to be merged into trunk? [16:59] schierbeck: yep, I don't see any big issue [16:59] then i'll make a merge request [17:00] cool :) [17:00] next time I might make the tags corner rounded :D [17:00] * asabil is a bling addict [17:04] asabil: i thought about it, too -- it would be cool [17:04] btw, sign up for the mailing list [17:05] And it should be a cube! That you can spin! [17:05] schierbeck: the ML is too high traffic for me [17:05] schierbeck: I was subscribed [17:06] fullermd: lol [17:06] it's not that high traffic atm... [17:06] ok ok I will subscribe again [17:06] it's about 40 messages per month these days [17:06] jelmer: hi, sorry I haven't gotten you on pqm yet [17:07] 'morning lifeless [17:07] lifeless: no hurry; would be nice to be able to use pqm at some point though [17:07] jelmer: I'm in London this week; so we can chat in the same timezone (FSVO same) [17:08] ah, ok [17:09] schierbeck: I am subscribed [17:09] to bazaar@lists..... [17:10] nah, it's bzr-gtk@lists.canonical.com [17:12] ah oki [17:14] done [17:28] is there a command to view the gpg-signature of a commit? [17:30] welterde: not yet afaik [17:32] jelmer: hmm... how strong is the authentication of the commits? eg. do they depend on a hash of the commit-data, on the previous commits? [17:32] (like git for example) [17:34] welterde: afaik both [17:35] hmm.. so manipulating a older revision would be detected? [17:35] yes [17:35] at the moment, there isn't any functionality to check signatures though [17:37] but gpg-signatures are checked afair, or not? [17:37] I don't think they are either [17:38] ah.. is an extra plugin [17:50] welterde: commit signatures do not transitively verify older history [17:50] welterde: this is deliberate [17:51] welterde: as it allows partial history environments to be verified [17:53] Just wondering, is it possible to store the whoami value on a per project basis? [17:56] sure; you can probably set it in branch.conf; or url based in locations.conf [17:58] lifeless: so no protection against a manipulated bzr branch? [17:58] welterde: uh, sure there is protection; just sign the commits for the history as well [17:59] hmm... yeah... but besides that there is none? [18:00] I'm not sure what you're getting at [18:01] if someones stops by and manipulates the unsigned history, that wouldnt be detected, right? [18:02] (manipulates in a way the checkout changes) [18:14] does bzr leak ip-addresses on any operation? [18:19] welterde: 'leak'? [18:19] lifeless: like, being noted somewhere [18:19] inside of commits or something like that [18:20] it may record them in your !/.bzr.log; and the commiter name is derived from your username and hostname by default [18:20] which can be changed with bzr whoami [18:22] right [18:23] lifeless: but signing one commit doesnt automaticly "sign" all of the commits before :/ [18:23] right [18:23] you can bzr sign-my-commits though if you haven't been signing them [18:24] and when the person, of whom the commits are, is not reachable? [18:24] thus cant sign his commits? [18:27] lifeless: and can i sign commits which just have a name? [18:27] eg. from a cvs-import [18:28] sure; I'm off to dinner now, have fun [18:31] welterde: You might also want to change your bazaar.conf/locations.conf/branch.conf to require signatures. See the docs. [18:32] Peng: yeah... know that [18:52] is there a way to get some stats out of bazaar... Like how many lines of code has been changed the last day or period? [18:54] Enquest: apart from 'bzr info' there is nothing that I know of. [18:54] maybe that would be a cool thing to get into it. [18:55] However the example you mention could be approximated by piping something like 'bzr diff -rdate:yesterday' to diffstat [18:56] I agree it would be interesting. It would work very nicely as a plugin I expect. [18:56] Enquest, and there is a stats plugins which shows mostly per-user stuff: http://bzr.arbash-meinel.com/plugins/stats [19:31] Copying repository content as tarball... [19:31] bzr: ERROR: Tags not supported by BzrBranch5('file:///home/james/scratch/trunk/'); you may be able to use bzr upgrade --dirstate-tags. [19:31] why does bzr hate me? [19:32] It's your shoes. [19:42] wtf [19:42] if I rsync the entire dir, and bzr branch locally it works [19:42] bzr makes me CRY [19:43] elmo: Bzr version? Smart server? Shared repo? [19:43] Smart server, I suspect. [19:44] bzr 1.0, bzr+ssh (so I guess smart?), and yes, shared repo I think [19:44] An incarnation of bug 173002. [19:44] Launchpad bug 173002 in bzr "Branching from hpss doesn't preserve non-repository formats" [High,New] https://launchpad.net/bugs/173002 [19:44] Ah. [19:44] Does that affect bzr+http too? [19:45] I presume it affects bzr+anything. [19:45] shall I followup to the bug? [19:45] I dunno why your bzr 1.0 would try to make a branch5 branch, though. [19:46] It should go with the default, which is branch6. [19:46] Please. [19:48] You may have been the person who said "no" last time I asked, but is there an easy way to "upgrade" from branc6 to branch5? [19:48] branch6* [19:48] I didn't create the repo on the other end [19:48] You can pull. It may even not bother giving you an error if there are no tags. [19:51] Heh! morning. [19:52] Anyone in here on OS X Leopard? (I presume most of you are on Ubuntu/Debian.) [19:53] Jelmer I updated the wiki page several times last week. I never did get it 100% right. Hopefully I can keep bugging people on Leopard to try it out.... [19:53] (it being bzr-svn) [19:53] bah. sorry, I somehow ended up on gutsy bzr instead of 1.0 [19:53] 1.0 works [19:53] Well, 1.0 works by coincidence. [19:54] It happens to use branch6 in its default format, so you don't hit the same problem. [19:54] God, Canonical owns our souls. Bazaar, Ubuntu, Launchpad... [19:54] It's till a bug that it doesn't preserve the remote format. [19:54] fullermd: ah, ok - so still follow up? [19:54] Peng: and that is ontopic how? [19:54] dato: It's not. :) [19:54] He's fishing for bribes to keep quiet about it. [19:55] Maybe the common criticism is wrong; bzr needs MORE format changes, to keep those sort of bugs in the forefront and high on the priority stack ;) === kiko is now known as kiko-afk [20:15] jelmer: ping [20:25] New bug: #184898 in bzr "AttributeError: 'NoneType' object has no attribute 'abort'" [Undecided,New] https://launchpad.net/bugs/184898 === jw2328_ is now known as james_w [21:19] are gpg-signatures pushed yet? [21:21] *over existing non-signed commits [22:24] Is there a way to test bzr's email plugin? I think my upstream smtp server is rejecting relaying, but I want to confirm. Some sort of debug output from the plugin? [22:30] BasicOSX: you could look in ~/.bzr.log [22:50] morning [23:00] hello everybody. [23:00] is there a way (simpler) to change the commit messages? (4 or 5 revisions) [23:01] I asked yesterday about the location of the latest debs. Set my repositories to PPA. The problem is there does not seem to be a *.deb for Dapper. [23:03] dysinger: thanks for the wiki updates [23:03] jelmer np - I just wish it worked perfectly - [23:04] dysinger: what's not working? [23:04] for some reason ssl svn access through bzr dosen't work [23:04] neon installs, svn 1.5 installs, svn 1.5 + ssl works [23:04] but not bzr-svn ssl [23:04] even using the svn+https:// prefix? [23:04] n [23:05] I am just hoping for more people trying it soon. [23:05] It dumps out of python. [23:05] oh, what error? [23:06] some stack trace - I didn't study it closely. I can go back to it soon and give specifics - it's nothing that was real clear to me what the problem was. [23:07] You can see from the wiki, I am using bzr-svn "stable" repo - is this ok? [23:07] dysinger: yeah, "stable" is just a symlink to the 0.4 branch [23:07] dysinger: seeing the backtrace may help debug this problem === thumper_laptop is now known as thumper [23:14] ok hold on 2 minutes jelmer, i'll get it. [23:21] ok [23:25] I am installing again fresh [23:25] I spent time reading through the svn/neon INSTALL docs [23:25] so I think I have that part right [23:26] I think this is some sort of problem with Apple's python or something. [23:26] if it' [23:26] s giving you a python backtrace, chances are it may be a bzr-svn bug [23:26] ah ok [23:38] OK I have svn 1.5 + neon installed with openssl support - all check-swig-py tests pass. svn --version reveals 1.5 with https support. I can checkout https://svn.collab.net/repos/svn/trunk with svn 1.5 no problem. No for the stack dump on bzr-svn [23:46] So just like it says in the wiki bzr checkout --lightweight http://svn.collab.net/repos/svn/trunk test works [23:46] jelmer notice no ssl [23:46] one sec and running ssl [23:50] jelmer I get [23:50] bzr checkout --lightweight \ [23:50] > https://svn.collab.net/repos/svn/trunk test [23:50] Assertion failed: (g->gc.gc_refs != _PyGC_REFS_UNTRACKED), function instancemethod_dealloc, file Objects/classobject.c, line 2285. [23:50] Abort trap [23:52] The apple core dump pop up gives me this [23:52] http://pastie.caboo.se/141720 [23:57] that's all - I don't know what to do know. [23:57] now [23:57] s/know/now/g [23:57] lolz