[00:01] igc: Hey. [00:01] k [00:01] hi fullermd [00:01] igc: Do I remember right that sometime back in the Dark Ages of N months ago you had some stuff toward recursive upgrade? [00:01] igc, which would be similar to the main web site [00:03] fullermd: I do [00:03] fullermd: there was a branch somewhere ... [00:03] but it missed the 2.0 cut and ... [00:04] fell down my list of priorities [00:05] fullermd: https://code.edge.launchpad.net/~bzr/bzr/smooth-upgrades === awilkins_ is now known as awilkins [00:13] Was it particularly incomplete, or just didn't get wrung out and landed? [00:16] fullermd: the patch was almost ready for re-review IIRC [00:17] fullermd: I don't think I addressed all of poolie's internal design issues [00:17] fullermd: external design was largely ok I think [00:18] Cool. [00:18] The hope for that soon enough is one of the things holding me back from feeling able to declare "OK, time, we're moving $STUFF to 2a". [00:20] So, if you come up with a "nearly done stuff we should finish" list while sprinting, keep it in mind :) [00:28] (I mean, I know how bored you all are, with so little to do...) [00:41] Odd_Bloke: Could you comment on https://bugs.edge.launchpad.net/bzr/+bug/183559 - "Fix Committed" where? [00:41] Launchpad bug 183559 in bzr "bzr switch should have a -r option" [Undecided,Fix committed] [00:42] maxb: is it in the "related branch"? [00:43] Hmm.... could be.... though that doesn't correspond to my understanding of "Fix Committed" [00:43] maxb: http://bazaar-vcs.org/BugGuidelines [00:44] Oh, stale link. [00:44] maxb: http://doc.bazaar-vcs.org/bzr.dev/developers/bug-handling.html [00:44] maxb: specifically, http://doc.bazaar-vcs.org/bzr.dev/developers/bug-handling.html#bug-status [00:44] ah. different to how ubuntu use the status. confusing :-/ [01:47] poolie1: hi, do you have the password for the udd list? (I think I'm an admin too?) [01:47] poolie1: if so, could you send me it, I'll add it in to listadmin and so help with the moderation [01:51] * igc lunch === timchen119 is now known as nasloc__ [03:23] Hi [03:23] I'm trying to merge from trunk to my branch and after a log list of changes and conflicts I get a 'bzr: ERROR : [Error 5] Access is denied", even though I have admin rights.. [03:24] any ideas why? [04:24] GungaDin: on windows? is it possible you have one of the files open in an editor or something? [04:28] yup, on windows [04:28] hmmm... maybe [04:28] yeah [04:29] good point [05:59] spiv, igc, can you tell me again the topics you wanted to add for next week? [05:59] i don't see them on the wiki [06:10] poolie: the two I have are the 2.1/Ubuntu topic (which is already there), and maybe also talking about what the bzr team needs to do with the code import system [06:10] k thanks [06:10] i'd like you to present a bit of a summary about what's already in the UDD specs and already done [06:10] for those who may not have read it too much [06:11] Ok, that makes sense. [06:11] and also to try to think of an interesting lightning talk [06:11] doesn't have to be particularly relevant, in fact refreshingly irrelevant ++ [06:11] There's a lot of spec to wade through, and the current status of a lot of it is pretty hard to figure out. [06:12] Hmm, irrelevant lighting talk... [07:38] hi all [07:38] vila: good evening [07:39] hehe, last tiem you do that one for a couple of days :) [07:39] :) [08:37] * igc1 dinner [08:58] Hello everybody. I have a question. When I'm in "bzr qlog", clicked the "Diff" button, and ticked "Complete", [08:59] is there a quick way to go from diff hunk to diff hunk? My case is that the file is huge, with some changes here and there, so I hit PgDown a lot; but I still need to see the complete file, to get better context... ? [08:59] guilhembi: there is no way at the moment [08:59] luks: ok, thanks. [09:07] guilhembi: Could you just use a really big context? === GaryvdM_ is now known as GaryvdM [10:42] Hi GaryvdM [10:43] Hi bialix [10:43] I want to share with you quote from my today's mail [10:43] ok [10:44] one mane wrote: "I've chose DVCS from Mercurial and Bazaar, and QBzr was the reason I've selected Bazaar" [10:45] it's a rough translation from russian [10:45] GaryvdM: ^ [10:45] Woot! [10:45] Indeed! [10:45] :-) [10:56] bialix, GaryvdM: nice :) [10:57] hi spiv, thanks [10:58] spiv: quick question [10:58] hi, i have a question about translation of qbzr [10:58] what is needed to add traffic reports for ftp transport? [10:59] stefano-k: ask [11:00] spiv: what is needed to add traffic reports for ftp transport? [11:01] bialix: hmm [11:01] bialix: add calls to self._report_activity everywhere the code reads or writes data I guess [11:01] ok, I understand [11:02] I've started to actively using ftp and want to improve it a bit [11:07] hullo [11:07] is there any way to push parts of a private repo into a public one? [11:08] bialix: Is that just for qbzr, or also in bzr cli? [11:12] GaryvdM: um? [11:12] bialix: traffic reports for ftp transport [11:12] ElMonkey: using fast-import-filter but it will create second branch completelly unrelated to original private one [11:13] GaryvdM: CLI [11:14] bialix, so changes from the original private branch couldnt be pushed after the creation of the new repo? [11:15] yep [11:15] only by diff + patch [11:15] or you can drop the history but preserve file-ids [11:16] in the latter case you can cherrypick changes from original repo [11:16] but you won't share history [11:16] hmm, tricky [11:17] the sitation is this: i have a private repo with all my projects, and i'd like to open source one of them [11:17] the thing is, that there's also some shared files between projects [11:17] ElMonkey: Are the different projects in different branches? [11:17] GaryvdM, no, its one repo [11:18] if it was different branches, i guess i wouldn't have this problem [11:18] ElMonkey A branch does not = repo [11:18] ElMonkey: you can have more than one branch in a shared repo [11:20] well, its just some folders in a repo [11:20] ElMonkey: in a branch? [11:20] ElMonkey: To check, go into the folder and type bzr info [11:21] ElMonkey: show us output of `bzr info` [11:21] Standalone tree (format: 1.9-rich-root) [11:23] it's a branch [11:23] by repo in bzr usually understand shared-repo [11:24] ElMonkey: I'd say it's a bad idea keep all your projects in the one branch [11:24] ElMonkey: maybe bzr-split does that - I'm not sure though. [11:24] bzr-split does not filter out history [11:25] well, whats the alternative? the reason for it being so as the projects share quite a few source files [11:25] well, as things grow [11:25] well, well, well [11:25] let me guess: you need nested trees, right? [11:26] i dont know :) [11:26] well, if there is only few common source files then you can add them in every branch with the same file-ids and cherrypick changes [11:26] i've been trying to figure out the proper way of doing this last year, but i didnt get anywhere [11:27] or you can extract common files in the separate branch and then merge them back to every project and therefore share them [11:28] ElMonkey: what is your background before you switch to bzr? svn& cvs& [11:28] ElMonkey: what is your background before you switch to bzr? svn? cvs? [11:29] cvs, then svn, in the meantime a bit of clearcase agony :) [11:29] ElMonkey: unlike cvs bzr can't share separate files, only group of files called trees [11:30] unlike svn bzr still has no support for equivalent of svn:externals [11:30] well, i havent really run into this problem before with the other RCSs [11:30] the latter called nested trees in bzr [11:30] never had the need to publish just parts of something before [11:30] ElMonkey: The idea with nested trees, is that you can have a branch for proj A and a branch for proj B, and a branch for shared, at it all works nicely. Nested trees is not implemented yet, but smc-proj alternative. Note: this is something you needed in the beginning. To get there, you can use bialix's suggestion: fast-import-filter [11:31] scmproj [11:33] bah, i should just write a shell script that pushes source tarballs onto the server occasionally :) [11:35] thanks for the tips, though [11:35] i'll have to do a bit more research to decide what to do === GaryvdM_ is now known as GaryvdM [12:45] moin === mrevell is now known as mrevell-lunch === verterok|lunch is now known as verterok === mrevell-lunch is now known as mrevell === weigon__ is now known as weigon [14:18] lifeless: ping? got a wave account by any chance [14:19] i'd like you to take a look at the revised distributed testing sketchup at https://wave.google.com/wave/#restored:wave:googlewave.com!w%252BezH3Z8aIB [14:54] hmm, I don't think it's possible to link waves like that [15:06] Tak: works fine [15:07] Tak: wfm [15:07] ah - the ua check breaks it [15:07] bialix: ??? [15:07] Tak: what? [15:08] wfm = works for me [15:08] ah [15:47] bialix: You have wave? [15:48] bialix: If possible, please invite me :-) [15:52] GaryvdM, to what address do you want the invitation? [15:52] garyvdm@gmail.com [15:53] Hi beuno [15:53] :-) [15:53] GaryvdM, hi [15:53] and sent [15:53] it takes a few days sometimes to materialize [15:54] but it's in the tubes === EdwinGrubbs2 is now known as EdwinGrubbs [15:55] Thanks beuno [15:56] GaryvdM, np. Are you going to be at UDS this time? [15:57] beuno: No :-( [15:57] beuno: Are you going to the bzr sprint? [15:58] GaryvdM, no, I'm off tomorrow to a different sprint [15:58] I see. [15:59] (lazr-js, Canonical's javascript library) [16:01] GaryvdM: just yesterday I've got invitation [16:01] GaryvdM: I'm still don't understand how to work with this thing [16:02] when I figure out how can I made invitation I'll do [16:02] bialix: beuno has sent me an invitation. [16:02] ah, ok, thanks beuno [16:02] bialix: I think you have to wait a while before you can send an invite. [16:02] you need to wait several days [16:03] before you got invitation [16:03] I waited ~ week [16:03] ok [16:03] it works very slow for me as it full of ajax [16:04] approx as slow as google reader [16:04] maybe it's firefox [16:04] * bialix still not sure what is wave for === deryck is now known as deryck[lunch] === deryck[lunch] is now known as deryck [17:03] bialix: for bug 418471 - Do you think we should have a "?", or a "x" for the icon? [17:04] Launchpad bug 418471 in qbzr "treeview: will be nice to show some custom icon for deleted entries" [Wishlist,Confirmed] https://launchpad.net/bugs/418471 [17:10] Or maybe trashbin [17:15] Going to go with trashbin... === EdwinGrubbs is now known as Edwin-lunch [17:31] GaryvdM: I've thought about sheet of papaer and red X over it [17:31] and similar thing for folders [17:31] GaryvdM: sorry, I'm tired today; trashbin is ok [17:32] * bialix gonna home [17:41] hello. I'm trying to commit changes to a branch that I initially checked out using http apparently. I've imported my ssh key onto this system and logged in to launchpad with bzr launchpad-login. but it's still trying to use http. How can I get it to change to ssh without editing anything in .bzr/ ? [17:45] ronny_: I do [17:45] robert.collins [17:52] ahh bzr bind is what I needed [18:10] help! how to setup a simple bzr repository for development, we dont need editors [18:14] elroboto: bzr init - But I recomend that you do some reading first: http://doc.bazaar-vcs.org/latest/en/mini-tutorial/ [18:18] thanks bro [18:34] bzr: ERROR: Revision {[('term/i386/pc/at__keyboard.c', 'svn-v3-single1-dHJ1bmsvZ3J1YjI.:d0de0278-0dc1-4c01-8a07-af38b3205e46:trunk%2Fgrub2:2613')]} not present in "KnitVersionedFiles(_KnitGraphIndex(CombinedGraphIndex()), )". [18:34] anyone can give me directions on how to debug this? [18:36] note that there's no at__keyboard.c file in our reposiroty, it's at_keyboard.c (one underscore) [18:36] nyu, have you run "bzr check" and "bzr reconcile"? [18:38] check raises an internal error: [18:38] bzr: ERROR: bzrlib.errors.SHA1KnitCorrupt: Knit corrupt: sha-1 of reconstructed text does not match expected sha-1. key ('svn-v4:d0de0278-0dc1-4c01-8a07-af38b3205e46:trunk/grub2:2681',) expected sha 81411fbdd0c3cdb2041adeb7761a4a16d28f8fe8 actual sha d698afceb95a6f9fe71e7462fddd4843d0d3ecb8 [18:38] what is this "svn-v*" information? [18:39] I didn't expect anything from svn needed to be preserved [18:39] seems bzr-svn was used [18:39] yes [18:39] it was [18:42] beuno: is it feasible to attempt recovery of this repository, or is the only debugging option trying to figure out when and how it reached this state? [18:42] nyu, I don't know much around this area of bzr [18:43] abentley or jam may be able to help [18:43] beuno: do you think this problem is linked to the fact bzr-svn was used? [18:44] nyu, it could of been a faulty version of bzr-svn [18:44] jelmer, around? [18:44] but I'm stabbing in the dark here [18:45] beuno: that's what we thought initially, as we were using bzr-svn 0.4.10. but then I rebuilt the repository from scratch using 1.0.0 [18:45] and hit the same problem [18:45] unfortunately it doesn't happen inmediately, only after a few days of using it [18:46] nyu, what version of bzr? [18:46] beuno: 2.0.1 [18:46] with --format=1.14-rich-root shared root [18:47] nyu, I'd try the mailing list [18:49] ok === Edwin-lunch is now known as EdwinGrubbs === BasicPRO is now known as BasicOSX [20:08] hi [20:15] Hi moldy [20:16] GaryvdM: did you ping me? [20:16] lifeless: no not recently. [20:17] kk [20:17] kk === Leonidas1 is now known as Leonidas [20:29] . === BasicPRO is now known as BasicOSX [21:31] help!! ive initilized a direcotry, and merged with the main branch, but i cant see the changes on the main branch? [21:32] im using the windows Gui [21:35] yohoo [22:07] HELP!!! [22:07] how to create a simple branch without [22:08] just simple push and merges [22:08] no editions [22:11] how to make a bzr tree standalone in linux? [22:17] you make a branch with 'bzr branch' [22:17] ohhhhhhhhh [22:18] bob2: i mean a standalone branch, i dont want to control [22:19] not sure what you mean [22:19] I don't understand the question [22:19] i just want to share , some code between webdevs [22:19] if you're trying to create a new empty branch to start a new project in that's just "bzr init" [22:20] if you have a branch and want others to have access to it that usually involves "bzr push"-ing it it to a location they have access to [22:20] but the manual covers all that [22:21] you may have to explain what you're trying to do in more detail [22:21] this is the scenario [22:21] (possibly including what you've already tried and how it didn't work) [22:21] 1. we need a central repository [22:21] 2. there is 5 developers [22:21] they just want to share code, without control [22:21] I don't know what "without control" means [22:22] for example, when we try that with Bzr windows GUI [22:22] mzz: Dogs and cats living together, mass hysteria. [22:22] it works fine [22:23] but when we try to store it on linux, we do "bzr init" , the changes are not reflected on the peers [22:23] "bzr init" is not normally a thing that shows up elsewhere. [22:23] A. commits and pushes [22:23] B. pulls [22:23] elroboto: one person does 'bzr init', once [22:23] it is possible you're using a checkout on windows and a standalone branch elsewhere, but it is very hard to tell (even more so because I've never used that windows gui) [22:23] C. Merges [22:24] elroboto: then everyone else pushes and pulls from that [22:24] one time somebody told me a command to create a "share" bzr repository [22:24] so far I do not exactly understand what this thing is that works on windows and not elsewhere [22:25] but I suspect you're clicking a button in the gui and issuing a command on linux, and those two don't do the same thing [22:25] in which case you'll have to explain what button you're clicking [22:26] yeah [22:26] (if I understand correctly so far your question is "I'm doing something in the windows gui and can't figure out how to do it on the linux commandline", but you haven't explained *what* you're doing in the windows gui in a way I can understand) [22:26] they dont do the same [22:26] exactly! [22:26] im doing in windows BZR init [22:26] and in linux im typing bzr init [22:27] "bzr init" is not something that shows up remotely (unless you're doing it in a network share or something) [22:27] brb 2 hours [22:27] so unless the windows gui is nuts I don't understand the question [22:28] Nor is it something that generally makes sense to do twice on one project. I think he's confused about what he's trying to do... [22:30] well, not having used the windows gui, my best guess is his "init" button isn't actually equivalent to "bzr init", it's a more general "get me a branch" button that can also do "bzr branch" and/or "bzr checkout" [22:31] either that or his current branch is actually on a network share mapped by all those 5 developers, so he's only had to "bzr init" that one branch and hasn't had to branch it or check it out at all yet. === tchan1 is now known as tchan [22:32] but that's just guesswork, so lemme try to switch topics: [22:33] I'm using bzr-builddeb and now have a bzr diff with "=== renamed directory 'oursqlx' => 'oursqlx'" in it. Does that make any sense? Is this some quirk of bzr-builddeb or can it have something to do with non-richroot vs 2a format conversions? [22:35] Not offhand. You're not running on some sort of case-insensible filesystem or something? [22:35] ext4, so no [22:36] jelmer: can you look at https://answers.edge.launchpad.net/launchpad/+question/88128 ? [22:37] the rest of the diff looks sane to me, including bzr noticing the bogus permission changes that happened. But that dir doesn't have either a rename or a permissions change I can see. [22:37] Hello, all. May a repo internal structure (like indexes) get corrupted because of bzr-svn? Is there a way to regenerate all indices? [22:37] and even if it *had* a rename bzr-builddeb shouldn't have been able to see it, because it's importing a tarball. [22:38] phcoder: I'm no expert, but did you run "bzr check" yet? [22:39] phcoder: iirc bzr-svn also has an sqlite-based cache somewhere. Removing that cache may be reasonable, but don't take my word for that. [22:40] it's in ~/.bazaar/svn-cache [22:40] yeah, sometimes it helps. but not this time :-( [22:41] I'm bzr-checking my local repo [22:42] nyu: have you tried bzr reconcile on our repo? [22:44] phcoder: nope. the description sounded uhm.. dangerous [22:45] you can try it on a copy [22:45] you're right ;-) [22:45] * nyu tries [22:46] 87 inconsistent parents [22:46] I have three machines, one is running bzr 2.0, one being the central server running 1.7 and one running 1.3. how can i push from the 2.0 machine to the server so that the 1.3 machine still can play with the rest? downgrading to 0.92 doesn't seem to work (can't downgrade to non richt-tree)? [22:46] is it a problem? [22:47] bzr: ERROR: An inconsistent delta was supplied involving '', '723@d0de0278-0dc1-4c01-8a07-af38b3205e46:trunk%2Fgrub2:' [22:47] reason: Parent not in inventory. [22:47] Standart: use --format when creating a shared-repo in the server [22:47] (I think) [22:48] Standart: --rich-root-pack should be ok [22:48] do i have to throw away my rich-tree branch? [22:49] no [22:49] rich-root-pack is readable by 1.0 and compatible with 2a [22:50] ok [22:50] nyu: can you send me a tar.lzma of our damaged repo, perhaps my bzr is newer and can reconcile it [22:50] its just a lot slower [22:50] the 1.3 server says bzr: ERROR: Unknown branch format: 'Bazaar Branch Format 7 (needs bzr 1.6) [22:50] which format ist the best to be used with 2.0, 1.17 and 1.3? [22:51] phcoder: the second one? [22:52] I have both [22:52] Standart: rich-root-pack [22:52] nyu: yes [22:52] lifeless: i thought there are different versions of that [22:52] ok [23:02] with --create-prefix i can push to a location that I don't have shell access to. is there a way to choose a format? [23:08] not on push [23:08] you can do this: [23:08] bzr init --format temp [23:08] cd temp [23:08] bzr push --create-prefix [23:08] cd .. [23:08] rm -rf temp [23:08] bzr push --overwrite [23:09] I see. so it's always the format of the one I'm pushing from [23:11] Standart: converting data is slow, we try to avoid it. [23:11] Standart: if you can, upgrade your server to 2 - much much much faster [23:12] i can't. don't even have shell access [23:12] Standart: that doesn't mean that you can't arrange for it to happen ;) [23:12] amd i need to checkout with bzr 1.3. can't help it [23:13] the existing repos are 0.92 but i cannot downgrade my new one, even if i wanted to, if i understand this correctly [23:14] looks like 1.14-rich-root can be used by all my versions. [23:14] 2a can not be converted to 0.92 (unless you create new history - essentially a full export-import, which won't be able to be merged with) [23:15] Standart: 1.14-rich-root can't be used by < 1.14 [23:15] Standart: rich-root-pack is the format you need [23:18] lifeless: ok, seems that i converted my local repos to 1.14-rich-root and pushed it to a server that took it as rich-root-pack [23:23] is there a format and push compatibility matrix somewhere? === BasicPRO is now known as BasicOSX [23:29] blehhhh, I just wish the world would magically upgrade to 2a already [23:30] * mzz redoes some stuff so he can actually push it to launchpad without "different serializers" failure, hopefully [23:36] Agreed, but I'd do many other things with magic first. Like, world hunger, or 1 Gbps fiber for everyone. [23:37] Increase Dollhouse's ratings.... [23:37] heh [23:37] Also, I'd like a marshmallow-pooping unicorn. [23:38] ok, poor wording there. Still, annoying. [23:38] I dunno how you can be so cruel as to want world hunger, though. [23:38] IP over wormhole, to decrease my RTT to Launchpad. :D [23:39] fullermd: Whoops. s/wo/ solve wo/ [23:39] I keep forgetting to check for non-richroot formats before I init repos or branches [23:43] lifeless: thanks. everything rich-root-pack...