[00:11] anybody have any suggestions for good "stastics tools" for bzr branches? [00:14] latexer, there is a stats plugin written by jam, but I'm not sure what kind of statistics you're looking for [00:15] latexer, bzr branch lp:bzr-stats ~/.bazaar/plugins/stats [00:15] that should get it installed :) [00:15] yeah, i checked that out, it's pretty slim on stats though. [00:15] i'm thinking something web based, show commit timelines, LoC added/removed maybe, etc. [00:16] latexer, well, not currently, no [00:16] I don't see it very far away, or too complicated even [00:16] but it hasn't been done [00:16] fair enough. === mw is now known as mw|out === samurai is now known as samiam [04:00] Hi [04:00] I have been playing with the gtk front end of bzr [04:01] and I could not get the nautilus integration working [04:03] does someone know how to get the plugin to work? [04:12] ischnura__, what version of bzr-gtk? [04:13] it was enabled in 0.93 [04:13] er, disabled [04:13] so you should try 0.94 === rockstar` is now known as rockstar [04:22] i find bzr a little strange at times :/ rename a file and it lists it as 'removed' name it back and re add it and its both removed and added ;| [04:23] I am back sorry for the delay [04:24] I am using Hardy with the version from the repositories [04:24] 0.93 [04:26] beuno, Is there some configuration that I need to do in nautilus? [04:27] Kamping_Kaiser: that's what bzr mv is for [04:27] bob2, yeah. i'm still getting used to having an rcs (and as such not using coreutils) [04:28] (not using bzr mv does fracture the files history by storing a delete and an add) [04:28] drat. [04:28] ischnura__, well, I'd use 0.94, where it's enabled by default, and many performance issues fixed [04:29] beuno, thanks [04:29] I will get the latest version then [04:29] 'bzr mv --after oldname newname' will fix it if you haven't commited [04:32] i had commited, but i've reverted that change. bzr seems happy with it now for some reason. [04:33] for reference, if you haven't commited anything else, 'bzr uncommit' might be simpler [04:33] neat [04:39] Kamping_Kaiser: See also 'bzr mv --after'. [04:40] thanks. [04:40] another idea might be to alias commit to 'commit --strict' [04:40] looks like bzr has everything covered, if only i stopped to think and look in doco *heh* [04:41] every vcs except for rcs and cvs have mv commands, fwiw [04:43] And most other VCSs don't track moves at all. [04:49] beuno, I have read the README file in the 0.94 version [04:50] they talk about installing a the python-nautilus package [04:51] and to paste the installation files in .bazaar/plugins/gtk and the nautilus python script in ./nautilus/python-extensions ... [04:52] the problem is that it is still not working did you do any other extra configuration? [04:52] beuno, did you do any extra configuration? [04:54] ischnura__, try restarting nautilus [04:54] sudo killall nautilus [04:54] but still not work... [04:55] I will try to restart the system... [04:55] beuno, I will let you know if it works [05:39] beuno, no luck [05:39] I still can not get the nautilus integration to work... [05:40] beuno, did you have to do any other configuration to get it to work? [07:20] morning [07:20] what do you think about 1st item in http://bramcohen.livejournal.com/52148.html ? [07:24] gour: well, having two sets of half-done work is generally worse than having one finished thing [07:25] So from that perspective, it's good to minimise the number of things you are doing at once. [07:26] But I think he's too extreme. Sometimes it's unavoidable. Sometimes you do have a long-lived experiment, or need to collaborate with someone in a different timezone (e.g. "does this change fix the bug for you?"). [07:27] In fact, one of the great things about one-branch-per-feature is that if work on one feature isn't mature yet, that doesn't interfere with landing a different feature. [07:28] He says in the comments "I just plan out what I'm going to do and then start typing, with very little experimentation and patch management." [07:28] right. branches are also not so expensive when we have shared repos [07:29] and can be easily discarded when feature is merged back [07:29] That's fine if you're the only person working on the code, and you don't have process where all code is reviewed by someone else before it lands. [07:29] * gour is accustomed to darcs where branching is quite normal operation [07:30] yes, that's what i was doing with darcs, not pushing stuff 'out' which is not stable === ndim_ is now known as ndim === phanatic_ is now known as phanatic === weigon__ is now known as weigon [09:59] abentley: Is the BundleBuggy "This change has been merged." message a new feature? I like it. :) [10:06] will new bzr-gtk still have dependency on seahorse? [10:39] if it does, i hope seahorse learns to pop up in front of the window it's being used in, instead of hiding away [10:42] I think that's focus stealing prevention. In my experience it happens under compiz but not metacity. [10:45] why seahorse and some extra gnome-stuff? gpa is not enough? [15:48] What's up, guys! [15:48] anybody awake? [15:49] I wanted to ask how to "efficiently" handle "new lines" between diferent architectures? [15:49] hello [15:49] ideally, use an editor that will not mess them up [15:50] but is there a way to make bazaar "figure out" how to handle the noe lines by itself? [15:51] it handles them fine [15:51] perhaps with an option when run? [15:51] Yesterday I made a brnach of my project on a M$ mox [15:51] if you're asking for it to translate between platform native line endings, that's under development [15:52] and when a line was edited (on wordpad) when you sae the diff of the file, it was replaced from top to bottom (because of the new lines) [15:52] right, notepad messed it up by not using the existing line ending format [15:52] afaik auto translation is under development but close [15:53] notepad is not even an editor. :-) [15:53] well, wordpad [15:53] and wordpad is close to noet being one either. :-D [15:53] so it's under development. [15:54] Is there a timeframe to have it on the stable or trial versions? [15:54] now... a question not directly involved with bazaar but close..... [15:55] let's say I will stick with the other guy working on windows. [15:55] and have his file replaced everytime. [15:55] i could "merge" his changes (which will make bazaar think the file has been rewritten from top to bottom) [15:56] it's reallyt not a platform issue per se, it's an editor one [15:57] (yes... that's hat I told the guy.... that he needed to set up his editor to use Unix new lines.... and you know what he replyed? Hold on to your chair: He was very happy using windows and won't install GNU/Linux... HUH?!?) [15:57] (I don't know when it will be released, but there's been increased chatter about it on the list) [15:58] so... as I was saying.... I get his files replacing mine completely.... is there a command (or series of commands) that I could run on GNU/Linux to replace all the M$ new lines for unix ones? [15:59] Just curious [15:59] sure, dos2unix [15:59] or even better: series of command on windows he can run so he doesn't mess up the files? say just before committing [15:59] antoranz: is your friend using notepad for editing? [16:00] on the flip side, most unix editors are competent enough to just not damage files with windows line endings [16:00] so if you conceded defeat and use them, you should be ok [16:00] I just don't know what he will be using... it was just a "trial baloon" [16:00] if so, I strongly recommend something like notepad 2.0 which I believe can work with unix line endings [16:01] ok... you make it an editor problem and that's it. I don't know how I will be coping with that... I guess I can work dos2unix to "correct" HIS problem on my side. [16:01] no, you misunderstand [16:02] your editor will probably be happy with the line endings [16:02] Oh.. I know... kate has no problem with that at all. [16:03] but when he save (not saves... save) a file, if he's not careful, he will mess up bazaar and it won't be able to see line changes but file replacement instead [16:03] how so? [16:04] say... I make a file... 100 lines long and put it in my repository [16:04] unless you convert back on your end, it'll be fine [16:04] he "merges" it into his repo [16:04] opens it.... say.. in wordpad [16:04] he adds two lines [16:04] saves it [16:05] if he does a diff, bazaar won't see the two-line change [16:05] ah, yes, that will suck, if you want this scheme to work, you'd need to add files with windows line endings, too [16:05] he will think the file as it was was completely removed and added 102 lines of code to replace it [16:07] I know.... that's exactly what we're taling about. The problem is that he won't install GNU/Linux so he can have the lines the way I want (man... I'm still wondering at his lack of understanding of the problem :-S) [16:07] :-) [16:07] hehe [16:07] Take off and nuke the site from orbit. It's the only way to be sure. [16:08] well, at the moment you can either: use windows line endings yourself, or insist he uses a competent editor or use dos2unix a lot [16:08] ok... bottomline: bazaar agnostic new lines recognition is on the way [16:08] at some point :) [16:09] I can use dos2unix to "correct other people's carelessness" [16:10] is there a tool that does the opposite for the windows side? [16:10] well, if you run it after every merge from him and before he sees your tree, yeah [16:10] unix2dos [16:11] for windows? OK..... that's will do it... at the worst, I could add a batch file to the project that he could run before he commits. [16:11] man.... do I hate people at Microsoft [16:11] Could be worse; you could have Mac newlines mixed in too :p [16:12] yeah! ;.) [16:14] I have never programmed for the CMD. [16:14] how can I build a while read line; do ; ; ; ; done < filelist.txt [16:15] if anybody knows, I mean [16:16] antoranz: if he's got python i could help you :D [16:19] well.... If he didin't have it, bazaar installed it for him, cause bazaar is working [16:20] and that's obvious..... use python to do the job [16:21] as a matter of fact, I've been trying to get my head in python for a while... perhaps it could be a nice exercise to begin with [16:21] :-) [16:21] antoranz: should i write a file ending replacer for multiple files for you? [16:22] nope... don't worry... I'll do it myself. ;-) [16:22] antoranz: okay.. [16:22] that's anyway, name [16:22] it's not really hard anyway :) [16:22] look at os.listdir [16:23] and just replace ;) "\r\n" with "\n" or counterwise [16:23] if you got any questions, just ask me, query if you want [16:23] ok. :-) [16:23] let me give it a try. ;-) [16:23] ofcourse ;) [16:24] os.walk would be simpler [16:25] bob2: ah right, forgot that it could be multiple dirs :) [16:27] I won't start that high... I'll setup a file where the files that have to be checked will be. [16:27] I will think of simpler ways later [16:27] (once I get a grip on python) [16:27] http://diveintopython.org is good [16:29] (if you have some programming knowledge) [16:29] if tou can't get this person to understand the issue, you might be better off just doing the work on your side [16:30] bob2: Well... after his response that "he won't install GNU/Linux", I'm starting to wonder if I'm better of by myself. :-D [16:30] lol [16:31] but well.... I'll just create this script just in case I'm stuck with him. :-) [16:31] and learning python is always good ;) [16:32] I know.. I know... I was just waiting for one excuse to come by... and it's here... so.... let's not waste anymore time, right? ;-) [16:32] let's see how I go,. [16:32] antoranz: nopaste it so i can check how you did it :) [16:33] ok... I'll call dos2unix (to begin with). [16:33] (i mean... for my python script) [16:33] I'll do the whole job from the script later [16:33] lol [16:34] don't laugh! :-) [16:34] I mean.. I'll create a script that will call dos2unix [16:34] a python script.... I'll change it later to do the converion by itself. [16:34] what's the point of the script them o.O [16:35] give me a break, guys! :-D [16:35] that I can't make a while read line < file_with_filenames on cmd... can I? [16:35] :-P [16:35] antoranz: think the python way, query? :D [16:36] what? [16:36] :-) [16:37] I'm sorry: say what? [16:37] antoranz: if i can query you for that is offtopic here [16:37] if you can query me for what? :-S Man... I'm dizzy already [16:38] antoranz: about your script [16:38] yea, my script [16:38] wan't to see how it's going? [16:38] I've made huge leaps forward [16:39] http://www.pastebin.ca/1027994 [16:40] antoranz: WOW! [16:40] Told you! :-D [16:40] i can't speak spanish tho [16:41] well.... the first line is not important... but the rest of the script speaks for itself.... doesn't it? [16:42] do you recomend to use tabs or spaces for indentation? [16:42] antoranz: that file contains the files to be converted, eh? [16:43] yes... (to begin with) [16:43] i think i just wrote your script ;) [16:43] I already know that python uses indentation to figure out "blocks" [16:43] so which do you recommend to use? [16:44] i use spaces, but with the tab key [16:44] my ide does it well :) [16:44] ok [16:44] pep8 (the python style guidelines) say 4 spaces [16:44] but like name says, it is your ide's problem to sort that out when you hit tab [16:45] WingIDE for the rule! [16:46] what do I have to import to use fopen? [16:46] nothing [16:46] use open [16:46] or file, it's the same [16:48] it worked with open [16:48] antoranz: loops are not the same as in C, for is not the same as C for... [16:49] give me a minute,... I think I almost got V1.0 [16:50] lol i sent u a paste o.O [16:50] what do you "pass" dos2unix? the name of the file and it will reconvert it? [16:50] no new file is created? [16:50] u don't need dos2unix, just use replace of python [16:51] come on, guys... that's V2.0 [16:52] it's easier to do anyway ^^ [16:52] messing with subprocesses is not easy for windows [16:52] it should be pretty easy with the subprocess module [16:54] V1.0: http://www.pastebin.ca/1028003 [16:54] let me commit it in the repository ;-) [16:55] why in gods name so complicated o.O [16:55] man... I'm a rookie! Give me a break [16:56] in a few minutes i'll have it working _your_ way [16:56] using external tools is just as hard or easy as using python internals [16:56] i tell you i've had loads of problems with subprocesses, namely wesnoth.exe for windows [16:57] let it go, man.... com eon... breath... count to ten. :-) [16:57] i just warn you. [16:57] I'll do the change in a few minutes... hold on [16:59] i had hundres of running processes with windows and loops and subprocess, tho it worked fine with linux [17:01] *hundreds [17:02] basically what u want is open(file, "w").write(open(file).read().replace("\r\n", "\n")) [17:03] no guarantee for that tho :) [17:03] as I'm reading the filename from another file, I have a \n at the end of the filename [17:03] how can I remove it? [17:03] antoranz: using readlines [17:04] file.readlines() [17:07] same thing: No such file or directory: 'interfaz_standalone.php\n' [17:07] then use .strip on the string, tho it should not happen [17:07] I hate python already. :-D [17:08] antoranz: it just hates you ;) [17:08] show me your code [17:08] and i'm sure it'll be PEBKAC [17:08] ;-) [17:09] archivo = archivo.replace("\n", "") [17:09] I'm not stupid, man... I know how to program, man... I just don't know my "resources" nor sytax with python... take it easy [17:10] yes, that's what i mean [17:10] but that replace will make it hard to tell between the lines [17:12] no no.. that's just for the name of the file I have to process [17:12] see ho it's going [17:12] sure but aren't there multiple files in that file? [17:12] yes, there are [17:12] see how it's going [17:12] let me put it in the pastebin [17:14] http://www.pastebin.ca/1028011 [17:15] gour: the gtk dependency will be removed [17:15] gour: it should really be 'optional usage' but its buggy right now === andrea-bs_ is now known as andrea-bs [17:17] guys.... I have to go for a while... I'll come back later to continue working on it [17:17] thanks for your kind help [17:17] name: thank you, man [17:17] np, look at the code i sent you in query [17:18] I saw it... but I have to go to by some groceries [17:18] okay hf [17:18] I'll come back to work on it later... I promise [17:18] lol [17:18] take care. [17:27] lifeless: you mean seahorse dependency? what is bzr-gtk without gtk? [17:29] gtk doesn't depend on seahorse [17:29] Pieter: bzr-gtk was crashing without seahorse [17:30] my question is seahorse is really required for bzr-gtk 'cause it brings some gnome-stuff deps, instead of e.g. gpa [17:33] s/is seahorse/is if seahorse/ [17:38] gour: The explicit dependency on seahorse is a bug [17:38] it's going to be fixed in the next release [17:40] jelmer: thanks. that's i wanted to hear ;) [17:41] Odd_Bloke: Indeed. Just implemented yesterday. [17:44] gour: I thought that was what I said ? :P [17:47] lifeless: "the gtk dependency will be removed" is not quite the same [17:47] lifeless: how can bzr-gtk be without gtk? [17:48] meh :) [17:48] I mean the erroneous dependency *of* bzr-gtk [17:49] np ;) [19:04] OK... I'm back! Hell... I'm very sleepy [19:04] |-) [19:25] OK... got it... perhaps it's a little rough... but it does the trick (just tried it) [19:27] http://www.pastebin.ca/1028099 [19:30] thanks for your help, guys! [19:31] I'll be coming once in a while to see what I can help with. [19:31] C ya! [20:24] If I've just pushed to a launchpad and my friend hasn't changed anything since he last pushed, should he do bzr merge or bzr pull? [20:24] or are they both equivalent in this case? [20:26] Pilky: he should just pull [20:26] ok, cool [20:31] hmm [20:31] my repo seems horked [20:31] I think I unknowingly shut down during a very long operation [20:32] any recommendations? [20:38] bug #234647 [20:38] Launchpad bug 234647 in bzr "bzr dies with parse error on corrupted history" [Undecided,New] https://launchpad.net/bugs/234647 [20:38] ubottu: thanks [20:38] You're welcome! But keep in mind I'm just a bot ;-) [20:40] ubottu: it's nice that you help people here ;) [20:40] gour: Error: I am only a bot, please don't think I'm intelligent :) [20:41] ahh, i expected some more intelligence [22:07] how can I pull all branches at once? [22:07] Pieter: if you have bzrtools installed, bzr multi-pull [22:08] right, thanks === mario_ is now known as pygi [22:57] bzr won't add my directory! I do bzr add dirname and nothing happens [22:57] it's listed as unknown in bzr status