| poolie | igc, hi, i'm setting up a new Analytics account for the bzr team | 00:15 |
|---|---|---|
| poolie | in which we should have more configuration control | 00:16 |
| GungaDin | I have a pending merge (lots of conflicts) that I still haven't committed... how do I get rid of it? | 00:29 |
| spiv | "bzr revert" | 00:30 |
| spiv | Or "bzr revert --forget-merges" if for some reason you just want to revert the pending merge and not the conflicts. | 00:31 |
| poolie | hello spiv | 00:37 |
| === spm_ is now known as spm | ||
| maxb | Are there any special subject-line conventions for discussing subprojects on the bazaar mailing list? (i.e. bzr-fastimport, for example) | 01:57 |
| lifeless | nope | 01:57 |
| poolie | igc i'm a bit disturbed by the number of "if wt.supports_content_filtering" special cases getting scattered through the code | 01:59 |
| poolie | it really smells like something not sufficiently well separated | 02:00 |
| === Peng_ is now known as Peng__ | ||
| === Peng is now known as Peng_ | ||
| === Peng__ is now known as Peng | ||
| igc | poolie: IIRC, some of them are there just to ensure "zero" overhead on early formats. I'll certainly think about a bit more | 02:52 |
| ricardokirkner | hi there. I am having issues pushing a branch to launchpad. I keep getting 'Permission denied (publickey).' error messages | 03:07 |
| ricardokirkner | any ideas what might be going on? (I added my ssh key to my profile) | 03:09 |
| === timchen1` is now known as nasloc__ | ||
| mwhudson | ricardokirkner: perhaps you need to tell bzr your launchpad user name? | 03:10 |
| ricardokirkner | nay, I did that too | 03:10 |
| SamB_XP | you added the right key to the profile ? | 03:11 |
| SamB_XP | and entered the *right* username ? | 03:11 |
| ricardokirkner | I actually deleted any registered keys, created a rsa key, and added it | 03:11 |
| ricardokirkner | what's the right username? I entered *my* username | 03:11 |
| SamB_XP | ricardokirkner: well, I mean, it would be bad if it was misspelled ;-) | 03:12 |
| lifeless | spiv: that log bug - backport to 2.0? | 04:40 |
| spiv | lifeless: the merge proposal was for 2.0 :P | 04:42 |
| AfC | Enjoyed reading jam's email about a possible "improved_chk_index" | 04:43 |
| lifeless | cool | 04:43 |
| lifeless | EODing. | 05:09 |
| lifeless | ring me if needed :) | 05:09 |
| spiv | Oh, gar, I remembered to run the blackbox tests to see what other bugs my test fix uncovered, then forgot to actually look at that terminal to see if anything failed. | 05:12 |
| fullermd | Well, every bit of advice out there just tells you to run the tests, so surely you did the important bit :p | 05:16 |
| AfC | test driven development. Doesn't that mean that the tests fix things for you? | 05:25 |
| fullermd | Depends on whether your bugs are stronger. It's like an e-Thunderdome. | 05:25 |
| igc | back | 05:27 |
| spiv | jam: still around? want to do a quickie review of some near-trivial fixes needed for my bzr log fix? http://bazaar.launchpad.net/~spiv/bzr/log-objectnotlocked-bug-445171/revision/4697 | 05:53 |
| fullermd | spiv: Something in the discussion about that reminded me of a bug I filed a while ago... | 05:56 |
| * fullermd digs. | 05:56 | |
| fullermd | Hm, looks like a cluster of semi-related bugs... | 05:58 |
| fullermd | bug 149270, bug 144421 (closed?), bug 144300 | 05:59 |
| ubottu | Launchpad bug 149270 in bzr "revisionspec in_history calls fetch, which requires the branch to be writable" [Medium,Confirmed] https://launchpad.net/bugs/149270 | 05:59 |
| ubottu | Launchpad bug 144421 in bzr "Using branch: revspec in stat blows up" [Undecided,Fix released] https://launchpad.net/bugs/144421 | 05:59 |
| ubottu | Launchpad bug 144300 in bzr "bzr log -r branch|ancestor attempt to fetch data in a read only transaction" [High,Triaged] https://launchpad.net/bugs/144300 | 05:59 |
| fullermd | It sounded from the discussion like those may be fixed in that branch? | 05:59 |
| spiv | No, in fact it exacerbates that slightly :( | 06:01 |
| spiv | But they are all related, yeah. | 06:01 |
| fullermd | Well, you can pretend they're fixed; just run the tests and don't look at the results 8-} | 06:02 |
| spiv | Heh. | 06:02 |
| igc | bbiab | 06:06 |
| vila | hi all | 07:10 |
| Peng | Good morning. :) | 07:12 |
| * fullermd waves at vila. | 07:13 | |
| fullermd | vila: BTW, big thanks for your help keeping the DWIM stuff moving! | 07:18 |
| * Peng goes to sleep. | 07:19 | |
| vila | fullermd: always happy to help (TM) | 07:36 |
| vila | fullermd: I had a question about it though, I suppose you used the dwim revspec for quite some time now, what is your feedback as a *user* (trying to forget your also are the implementor) ? | 07:37 |
| vila | james_w: ping, ready for you whenever you are | 07:38 |
| fullermd | Oh, I haven't used it much beyond testing. | 07:39 |
| fullermd | It wasn't so much a "gaah, I need this" as a "wtf, why don't we have this yet, how hard can it be?" | 07:39 |
| vila | hmm, I see | 07:40 |
| fullermd | I mean, we had discussions about "Yeah, -r should be able to just take a revid and DTRT with it", like, 3.5 years ago. | 07:40 |
| fullermd | I figured it would be, like, a half hour change. Obviously, my estimation skills are subpar, but... | 07:41 |
| vila | yeah, half our changes... I knew a guy who never made estimations *below* one man month :) | 07:42 |
| fullermd | Well, it's probably only actually an hour or so for somebody comptent :p | 07:42 |
| fullermd | Full day or so for me... | 07:43 |
| vila | even with s/competent/knowing that piece of code and all its users/ I have to disagree here :) | 07:46 |
| vila | the closer you get to users, the more time you need to get it right... DWIM being the closest you can get... :) | 07:47 |
| vila | it's more 3.5 years than one hour in that case | 07:48 |
| fullermd | Oh, heck, I don't even worry about _right_; if I can just get it to _work_, I feel good :p | 07:48 |
| vila | ..but you can't know it /works/ until it's /used/... | 07:51 |
| vila | and stop being so modest, thanks for making it happen is my underlying message :) | 07:51 |
| fullermd | Well, in precise terms I guess. But if I can make it do what I ostensibly want and not just blow backtraces and stuff, I call it success. | 07:52 |
| fullermd | Especially in a codebase I don't know in a language and methodology I don't want to know :p | 07:52 |
| vila | that last part is interesting, that part of the code base was (and still is) well tested, so you really experienced the nice aspects of TDD regarding refactoring | 07:54 |
| fullermd | Oh, I don't mean the TDD part. I'm in favor of large test suites (in the abstract anyway; the code I write all day doesn't have any for practical reasons), though not necessarily test-first. | 07:55 |
| fullermd | I mean the whole OO thing. I don't hold with it. | 07:55 |
| vila | gee, so last century :-) Well, I think TDD is more important than OOP, so forget the later but practice the former :) You'll come to OOP eventually... | 07:57 |
| * fullermd proudly wears his Luddite badge. | 07:57 | |
| fullermd | I wish I could write test suites for work stuff. Just a giant pain of ugly problems to get a useful one... | 07:58 |
| vila | http://en.wikipedia.org/wiki/Luddite ? You don't like bzr-loom ? | 07:58 |
| vila | Introducing TDD is a giant pain | 07:59 |
| fullermd | Well, especially with something that reaches so far. Just building a test in general means I have to deal with instantiating and pointing at a properly-populated fresh database, and figuring URL's to access to test the codebase in question, and yada yada. | 08:00 |
| fullermd | Even with the mechanics of testing this vs that worked out, the infrastructure for setting up the environment is daunting. I haven't convinced myself it's reasonably doable given time etc. constraints. | 08:01 |
| vila | yup, infrastructure and test-ready code, if you don't start at the lowest level, you just can't spend the time to write the first useful test | 08:01 |
| vila | These are obviously the main reasons why TDD is not used more broadly | 08:02 |
| fullermd | Yeah. And I need to test the whole stack to catch our common bugs, so I can't even do things like shim database layers or the like. | 08:02 |
| vila | and that's without mentioning setting up even test *hosts* | 08:03 |
| fullermd | Some level of intra-code unit testing may be doable (and I have vague plans to add this on some of our library code), but at the app layer, it's just... | 08:03 |
| fullermd | But hey, I don't need testing. That's what the client is for :p | 08:04 |
| igc | hi vila, fullermd | 08:04 |
| vila | hi Ian | 08:04 |
| * fullermd sails a paper airplane past igc. | 08:04 | |
| vila | testing is one thing, reproducing bugs is another, you can't beat a failing test to speed up debugging... | 08:05 |
| fullermd | It would be _really_ nice for some of the refactoring some projects desperately need :| | 08:06 |
| fullermd | Could spend 3 hours trying to manually test the things a 5 minute change can impact. | 08:06 |
| fullermd | Which means a lot of those 5 minute changes just never get made... | 08:06 |
| vila | That's where TDD is most often badly understood, the time spent in the test infrastructure is the time *not* spent understanding and reproducing bugs... | 08:07 |
| fullermd | Yeah. And trying to tell our clients "OK, here's the hours we'll spend writing your app... and here's 4x as many hours to design and build a testing infrastructure for it"... | 08:09 |
| fullermd | Well, I guess it would be a useful technique to get some free time :p | 08:09 |
| vila | But the point is that if TDD is more expensive than usual methods (including debugging times), then TDD failed and the reasons should be understood | 08:11 |
| vila | But if you have to do a fair try and that implies *ending* with a good infra | 08:12 |
| fullermd | In the long run, having a good test suite would save us time (on the projects that run long, anyway; some are done in a few months, others we keep working on for years) | 08:12 |
| fullermd | But it front-loads the time, even when it's nice over 5 or 6 years. | 08:12 |
| vila | That's not true if you start from scratch, the problem is that you nearly never start from scratch... | 08:12 |
| fullermd | Well, yes and no. There's a lot of codebase-specific stuff that wouldn't be so hard to build in along the way. | 08:14 |
| fullermd | But the general environmental setup stuff would take me a very long time sitting and hammering on to make work well enough. I've thought about it somewhat more than casually, and haven't come up with a scheme I even theoretically feel confident in. | 08:15 |
| vila | IME unless you have the whole infra, you're hitting the wall in unexpected places and that's where you spend a lot of time and need a lot of faith... | 08:16 |
| vila | ...or deep pockets or hiearchy support, whatever | 08:17 |
| fullermd | Yeah. Which is why we don't have test suites :) | 08:17 |
| === _thumper_ is now known as thumper | ||
| james_w | vila: I have one zero-day SRU to shoot for, so is after lunch good with you? | 10:52 |
| vila | james_w: sure | 10:54 |
| awilkins | #ubuntu-release-party is giving me a headache | 10:56 |
| vila | awilkins: you're running the smart server on windows right ? | 11:01 |
| vila | s/the/a/ | 11:02 |
| awilkins | vila: Yes, I think my config is non-optimal but it works | 11:05 |
| vila | awilkins: do you run it as a service or do you use ssh ? | 11:05 |
| awilkins | vila: I'm running it i) in IIS, ii) From the command line on demand... I think I got it working on SSH too | 11:06 |
| vila | awilkins: do you see any problem running it as a service ? | 11:07 |
| awilkins | The SSH was a long time ago ; ICT refused to poke a hole in the firewall for port 22 | 11:07 |
| vila | as in: no ssh setup, no access control | 11:07 |
| awilkins | This is a box exposing IIS to the world... didn't make much sense to me... | 11:07 |
| vila | awilkins: right, but if it was an internal box ? | 11:08 |
| awilkins | vila: I don't see why it would be a problem to run it as a service... there's a little "run program as a service" thingy that you can use | 11:08 |
| vila | awilkins: ok, thanks, that's what I was looking for :) | 11:08 |
| bialix | srvany? | 11:15 |
| vila | hey bialix ! | 11:16 |
| bialix | heya vila! | 11:16 |
| vila | bialix: I didn't know if *you* were runing the smart server or just using windows shares... | 11:16 |
| bialix | I'm running smart server, yes | 11:16 |
| vila | bialix: feel fre to answer my questions above :-) | 11:17 |
| bialix | perhaps I've joined too late | 11:17 |
| vila | ho, you joined just when I asked :-) | 11:17 |
| bialix | the first thing I see in my log is: [13:06]<awilkins>vila: Yes, I think my config is non-optimal but it works | 11:18 |
| vila | do you run it (smart server) as a service or do you use ssh ? | 11:18 |
| vila | bialix: before I asked if he was using the smart server | 11:18 |
| vila | bialix: before, I asked if he was using the smart server | 11:18 |
| bialix | I run bzr serve --alow-writes on internal computer in our intranet network | 11:18 |
| bialix | I run it as windows service via srvany | 11:18 |
| bialix | I've described my setup in Russian if you want to read it | 11:19 |
| bialix | ;-) | 11:19 |
| vila | excellent ! That's exactly the setup I wanted confirmation about ! Unfortunately I don't read russian :-/ | 11:19 |
| vila | bialix: But thanks anyway, I was mostly interested to know if it was possible, no urgency for the details | 11:20 |
| bialix | well, srvany is well documented on microsoft | 11:20 |
| bialix | site | 11:20 |
| bialix | it's a native MS thing | 11:20 |
| bialix | but absense of ACL is pain for me | 11:21 |
| bialix | and plain --allow-writes too | 11:21 |
| bialix | today I've just closed access from outside to my server | 11:21 |
| bialix | but for our small company it's enough | 11:22 |
| vila | bialix: these are the limitations I had in mind, for ACL, I'd say ssh is the way to go (but that's because I know how to set this up in general, not sure how I will proceed on windows...) | 11:22 |
| bialix | yep | 11:22 |
| bialix | btw, hg approach does not require ssh | 11:23 |
| vila | how many users on your server ? | 11:23 |
| bialix | hg beats bzr on this field | 11:23 |
| bialix | 2 full-time programmers | 11:23 |
| vila | ok | 11:23 |
| bialix | vila: http://translate.google.com/translate?prev=hp&hl=ru&js=y&u=http%3A%2F%2Fgroups.google.com%2Fgroup%2Fru_bzr%2Fweb%2Fbzr-serve---windows-2k-xp&sl=ru&tl=en&history_state0= | 11:25 |
| bialix | it's not too bad computer translation | 11:25 |
| bialix | I saw even worse | 11:25 |
| vila | bialix: wow...... | 11:26 |
| * vila reading | 11:26 | |
| vila | so, yes, some translation are strange (and I don't get them) but the overall is excellent, you should really send a submission for inclusion in core doc | 11:30 |
| vila | bialix: or is it already part of the windows survial guide igc is talking about ? | 11:32 |
| bialix | no, I don't think so | 11:33 |
| vila | well, it should, IMHO :) | 11:33 |
| bialix | some russian words in this translation just is not translated, e.g. some slang | 11:33 |
| bialix | I never think it worth | 11:34 |
| vila | Well, all setups are worth documenting IME, it's amazing how you can lose time rediscovering things that can be summarized in one page.... | 11:34 |
| bialix | hm | 11:35 |
| bialix | I may try... but close to december, I have too much urgent work and very critical deadline | 11:36 |
| vila | like, how many times did I spend hours finding the right site to get the right version of some utility... or trying to decide what is the most effective way to achieve X with Y constraints... | 11:36 |
| vila | yea, I know the feeling... | 11:36 |
| vila | bialix: yea, I know the feeling... | 11:36 |
| vila | another good one is seaching the right key in the registry.... | 11:37 |
| bialix | just curious: why you interesting in windows approaches? | 11:37 |
| vila | someone asked offline | 11:37 |
| vila | I'm gate-keeping :) | 11:38 |
| bialix | ok | 11:38 |
| bialix | :-) | 11:38 |
| Mez | bzr: ERROR: Tree transform is malformed [('versioning no contents', 'new-84')] | 11:46 |
| Mez | what does that mean, and how do I fix it? | 11:46 |
| vila | bialix: do you mind if I use the translation to start a wiki page on bazaar-vcs.org ? | 11:46 |
| bialix | vila: feel free | 11:47 |
| vila | Mez: it's a bug | 11:47 |
| Mez | vila: So, workarounds? | 11:47 |
| bialix | vila: just copy code examples from original page, translator slightly broke them. | 11:47 |
| vila | Mez: how did you get there ? | 11:48 |
| Mez | running bzr up | 11:48 |
| vila | Mez: hmm, bad, what does 'bzr st' says ? | 11:48 |
| Mez | http://pastebin.com/m797ef42d | 11:49 |
| bialix | vila: drop me a link then so I will fix some obvious translations errors | 11:49 |
| vila | bialix: asap | 11:50 |
| * bialix -> lunch | 11:50 | |
| vila | Mez: are these local modifications or already brought by the update ? | 11:50 |
| Mez | local mods | 11:51 |
| vila | then do a 'commit --local' first just to avoid too much further problems | 11:51 |
| Mez | bzr: ERROR: Working tree is out of date, please run 'bzr update'. | 11:51 |
| vila | grr | 11:51 |
| vila | ok, 'bzr shelve --all', 'bzr update', 'bzr unshelve' | 11:52 |
| vila | err, bzr unshelve 1 or whatever bzr shelve told you | 11:53 |
| Mez | yeah, I just did that ;) | 11:53 |
| Mez | thought ahead and thought that might work | 11:53 |
| vila | Mez: so you're out of trouble ? | 11:53 |
| Mez | yup | 11:53 |
| Mez | thanks ... kinda annoying that that happened but meh. | 11:53 |
| vila | Mez: can you still reproduce and file a bug ? | 11:53 |
| Mez | no idea. | 11:54 |
| Mez | if I can, I will | 11:56 |
| Mez | but it was someone elses stuff that I got asked how to fux | 11:56 |
| vila | bialix: http://bazaar-vcs.org/SmartServer/AsAserviceOnWindows | 12:07 |
| === mrevell is now known as mrevell-lunch | ||
| === vila is now known as vila-lunch | ||
| === mrevell-lunch is now known as mrevell | ||
| === elmo_ is now known as the_real_elmo | ||
| jam | spiv: your changes seem fine, though I wonder why you feel they are needed, but pqm still merged your earlier patch? | 13:14 |
| IslandUsurper | I'm using Bazaar to work on a project, but the official, public repository is CVS. Has anybody written a post-commit hook to push changes made in bzr to CVS? | 13:22 |
| vila-lunch | morning jam (wow early start ?) | 13:30 |
| vila-lunch | IslandUsurper: Not that I know of | 13:31 |
| === vila-lunch is now known as vila | ||
| spiv | jam: pqm rejected my earlier patch; I decided on reflection that the changes were trivial enough to just send it in rather than keeping the fix hanging around unmerged. | 13:42 |
| * spiv -> zzz | 13:42 | |
| jam | spiv: sleep well | 13:56 |
| jam | vila: sort of. I came around before taking my son to school | 13:56 |
| jam | morning to you | 13:56 |
| jam | sounds like the call went fairly well | 13:56 |
| vila | yes | 13:57 |
| * vila unplug fullermd again just for fun | 14:08 | |
| leo__ | hello | 14:08 |
| vila | jam: err, re-reading Kareem already goes to school ? | 14:08 |
| jam | vila: daycare | 14:09 |
| jam | but we call it school sometimes | 14:09 |
| vila | haa, ok | 14:09 |
| jam | they do teach | 14:09 |
| leo__ | Im looking for some help in order to use bzr-pqm | 14:09 |
| vila | leo__: that's quite the good channel | 14:10 |
| leo__ | got a main bzr repo and I'd like other devs to be able to pqm-submit on it | 14:10 |
| jam | leo__: we can try to help, though bzr-pqm *is* a bit tricky to set up. | 14:10 |
| jam | and we haven't done it in a while, because it tends to be something you do once :) | 14:10 |
| leo__ | I do a branch from main repo | 14:10 |
| leo__ | modify a file & commit | 14:10 |
| leo__ | then try to do bzr pqm-submit | 14:10 |
| leo__ | bzr: ERROR: You must supply a commit message for the pqm to use. | 14:10 |
| vila | bzr pqm-submit -m 'That message will appear in the commit where pqm is the committer' | 14:11 |
| leo__ | I guess I must first setup pqm on the main brach ? | 14:11 |
| vila | err, yes | 14:12 |
| leo__ | bzr pqm-init ? | 14:12 |
| leo__ | hmmmm | 14:13 |
| vila | pqm is kind of a robot that receives mail and do merges/commit/pushes based on that | 14:13 |
| leo__ | ok | 14:13 |
| vila | it's not simply some branch configuration, you need a mail address to send it requests for example | 14:14 |
| leo__ | I get an error now | 14:14 |
| leo__ | $ bzr pqm-submit -m 'essai pqm' | 14:15 |
| leo__ | bzr: ERROR: There is no public branch set for "/home/vincent/ALACARTE_Lagamine/" | 14:15 |
| vila | leo__: You can't send pqm requests before setting up your pqm robot | 14:16 |
| leo__ | that sounds logical :D but how to set it up ? didnt find any documentation | 14:17 |
| vila | pqm-submit is the client part, you want to look at https://edge.launchpad.net/pqm | 14:18 |
| vila | for the "server" part | 14:18 |
| jam | rockstar: by the way, we missed chatting yesterday. Think we'll get a chance today? | 14:19 |
| leo__ | need to build it from source right ? :) | 14:20 |
| vila | It's written in python, what OS are you using ? | 14:20 |
| leo__ | ubuntu | 14:20 |
| vila | leo__: you should be able to install it with synaptic then | 14:20 |
| leo__ | :s | 14:21 |
| vila | err, no ! | 14:21 |
| leo__ | theres bzr-pqm but no pqm | 14:21 |
| vila | only the bzr plugin is available there.. amazing... | 14:21 |
| vila | leo__: so, yes you need to start from the sources | 14:21 |
| vila | leo__: start with 'bzr lp:pqm' and look at the README there | 14:22 |
| leo__ | k tx | 14:22 |
| Tak | hmm...I'm seeing an exception in `bzr log` with bzr branches from svn and git repos - all of the backtrace seems to be in bzrlib core - should I file a bug against bzr, or bzr-svn and bzr-git? | 14:28 |
| metahuman | hi, i have a bzr branch of an svn.repository/trunk. Is it possible to switch to svn.repository/branch? | 14:31 |
| metahuman | bzr switch says i can't be in a branch | 14:31 |
| Tak | doesn't that just mean your local branch has to be bound? | 14:33 |
| metahuman | bzr bind looks like what i need! | 14:35 |
| metahuman | thank you Tak! | 14:35 |
| Tak | \o/ | 14:35 |
| metahuman | YES YOU'RE A GENIUS! | 14:35 |
| metahuman | "checkout of branch" now reflects the branch and it got the changes! | 14:36 |
| james_w | vila: is now a good time? | 14:43 |
| vila | james_w: yes | 14:44 |
| === gioele_ is now known as gioele | ||
| awilkins | metahuman: You may find it better to keep a local repository and track the SVN branches there, and work in your own local bzr branches | 14:47 |
| metahuman | awilkins, what i have been doign is doing bzr branch svn:// | 14:48 |
| metahuman | but i can't switch svn repositories... | 14:48 |
| metahuman | or rather svn branches | 14:48 |
| metahuman | see the problem? | 14:48 |
| awilkins | metahuman: If you do bzr branch, you get a separate local bazaar branch | 14:48 |
| rockstar | jam, yes, we should indeed chat today. | 14:48 |
| awilkins | You can't switch standalone branches as they are not tracking anything | 14:49 |
| awilkins | I do this... | 14:49 |
| metahuman | ok, there's an svn branch/mantis/4453 that i need to switch the bzr working copy to, and commit and pull changes from | 14:49 |
| awilkins | bzr init-repo .repo | 14:49 |
| awilkins | bzr branch svn://branch .repo/branch | 14:49 |
| awilkins | bzr co --lightweight .repo/branch project-name | 14:49 |
| metahuman | hey awilkins | 14:50 |
| metahuman | what would bzr pull svn:///branch/4453 do? | 14:51 |
| awilkins | metahuman: Hi | 14:51 |
| metahuman | ? | 14:51 |
| awilkins | If you're in your local branch, it will pull any revisions into it. | 14:52 |
| metahuman | is there a way to change the parent branch to another svn branch? | 14:52 |
| awilkins | But pnly if it hasn't diverged from /trunk (where I presume you are) | 14:52 |
| awilkins | ie - if trunk has revisions since /4453 was branched, no dice | 14:52 |
| awilkins | metahuman: You could pull --overwrite | 14:53 |
| metahuman | yeah that's what i want | 14:53 |
| awilkins | And that will turn your current branch into a mirror of the remote branch | 14:53 |
| awilkins | If you make commits locally, you'll have to push them for them to be in the remote branch... OR bind your local branch to the remote | 14:53 |
| metahuman | yes i want to push them when i'm ready | 14:54 |
| metahuman | svn is really spotty around here | 14:54 |
| metahuman | down about once a day and horrendously slow | 14:54 |
| metahuman | my parent branch doesn't change | 14:54 |
| metahuman | when i do --overwrite | 14:55 |
| awilkins | The usual SVN working copy is analogous to the bzr "lightweight checkout" | 14:55 |
| awilkins | Hmmph. Add --remember to the pull | 14:55 |
| metahuman | ok!!! | 14:56 |
| metahuman | that's awesome! | 14:56 |
| metahuman | so | 14:56 |
| awilkins | When you push the first time you'll have to specify the target anyway | 14:56 |
| metahuman | to svn switch in bzr, do pull --overwrite --remember and push --remember | 14:56 |
| awilkins | Which will be your branch | 14:56 |
| awilkins | --remember is only need when you want to change the default that it stores first time | 14:57 |
| awilkins | It's not really the same as switch... Bazaar does have "switch" but it only works on bound branches and works best with lightweight checkouts. | 14:58 |
| metahuman | but it is funcitonally equivalent for my working copy | 14:59 |
| awilkins | Which is my practice ; I keep a hidden no-trees repo in the parent folder to my source projects and switch between local and svn-mirrored branches as required. | 14:59 |
| metahuman | the same thing happens to my workign copy as if i was using svn switch | 14:59 |
| awilkins | To the working tree, yes. The repository inside your branch now regards the tip revision to be the HEAD revision of your branch mirrored from SVN | 15:00 |
| awilkins | If you did the same thing to a tree you'd committed changes to locally, you wouldn't LOSE the revisions you'd committed, but you'd have to make some effort to find them | 15:01 |
| metahuman | i already found the workaround for that, tho | 15:02 |
| metahuman | the "lost changes" | 15:02 |
| metahuman | bzr uncommit; bzr shelve; bzr uncommit; bzr shelve; for every local commit | 15:02 |
| metahuman | then bzr shelve; commit again when ready | 15:02 |
| metahuman | or unshelve rather | 15:02 |
| metahuman | thankjs for helping me out, awilkins ; i really appreciate it | 15:03 |
| awilkins | metahuman: That's a rather manual way of doing what the rebase command in the bzr-rewrite plugin does | 15:03 |
| awilkins | metahuman: Only it doesn't smush all your local history into one commit | 15:04 |
| metahuman | maybe bzr rebase is what i should look into? | 15:04 |
| metahuman | nah doesn't look like it | 15:04 |
| awilkins | As I recall, the Bazaar-for-SVN-guys document has some lovely pictures illustrating it | 15:05 |
| awilkins | 'tis a bit slow on the server though what with the Karmic Horde descending | 15:05 |
| * awilkins just happens to have the sources for those locally though | 15:06 | |
| phinze | hmmm what's the status of getting colorized diffs? bzr cdiff i've seen somewhere but doesn't seem to work in 2.0.0 and my inital googles are failing me | 15:16 |
| === Island_Usurper is now known as IslandUsurper | ||
| Tak | cdiff is part of bzrtools, according to `bzr help commands` | 15:26 |
| gioele | phinze: alias color='vim -c ":syntax on" -' | 15:31 |
| gioele | phinze: bzr diff | color | 15:31 |
| Tak | aha, found a dup in bzr bugs | 15:32 |
| phinze | gioele: that's a nice general purpose trick, Tak you're right i just was looking in the wrong place -- simply needed to install bzrtools | 15:36 |
| gioele | phinze: bzrtools brings in too many commands for my taste :) | 15:37 |
| phinze | understandable :) | 15:37 |
| Tak | yeah, damn all those features! | 15:38 |
| === the_real_elmo is now known as elmo | ||
| === loxs_wrk is now known as loxs | ||
| === Island_Usurper is now known as IslandUsurper | ||
| * Ng hrms at the email plugin | 16:06 | |
| Ng | it seems to mostly make bzr explode on me | 16:06 |
| Ng | filed as bug #463428 | 16:10 |
| ubottu | Launchpad bug 463428 in bzr "bzr crash when using the email plugin and smtplib" [Undecided,New] https://launchpad.net/bugs/463428 | 16:10 |
| === beuno is now known as beuno-lunch | ||
| === deryck is now known as deryck[lunch] | ||
| jam | Ng: see bug #338261 | 16:26 |
| ubottu | Launchpad bug 338261 in bzr "Python2.6 hmac.py TypeError: character mapping must return integer, None or unicode"" [Undecided,Fix released] https://launchpad.net/bugs/338261 | 16:26 |
| jam | it seems we fixed that in bzrlib but bzr-email didn't get the fix | 16:27 |
| jam | short answer is to just do "smtp_username = str(smtp_username)" | 16:27 |
| Ng | jam: aha :) | 16:27 |
| maxb | I have a problem with bzr-svn: If I run any bzr command in an unversioned subdirectory of a svn working copy, it climbs up to the working copy and starts processing stuff, sometimes breaking things, always causing a slowdown | 16:45 |
| maxb | This affects me particularly severely since my home directory is a svn working copy | 16:45 |
| maxb | This means I always pay the price of bzr-svn slowdown even when I'm not using it | 16:46 |
| jelmer | maxb: unfortunately that's an issue in bzr itself | 16:49 |
| jelmer | it will just browse up until it finds something it can open | 16:49 |
| maxb | I suppose I could alias sbzr to "BZR_PLUGIN_PATH=something bzr" | 16:50 |
| === beuno-lunch is now known as beuno | ||
| === deryck[lunch] is now known as deryck | ||
| vila | maxb: the idiom is 'BZR_PLUGIN_PATH=-site bzr' to run with only the 'core' plugins (i.e. the ones in bzrlib/plugins) | 17:27 |
| maxb | I guess that doesn't help much if you're using a system-packaged bzr-svn | 17:34 |
| maxb | I might need to switch to a bzr-svn somewhere else | 17:35 |
| hevayo | hi how can I migrate existing cvs code to bzr with revision history | 17:35 |
| === oubiwann__ is now known as oubiwann | ||
| beuno | hevayo, a quick way to do it is import it into Launchpad | 17:36 |
| hevayo | thanks but I am looking for a tool to do that | 17:37 |
| beuno | hevayo, http://bazaar-vcs.org/CVSPSImport | 17:38 |
| hevayo | beuno, thanks for the info I'll check on that | 17:39 |
| rockstar | jam, wanna chat? | 17:44 |
| jam | rockstar: sure. can I get ~ 5 min to finish writing an email? | 17:44 |
| rockstar | jam, yes. | 17:44 |
| maxb | hevayo: Also, cvs2svn has the beginnings of bzr output support | 17:45 |
| rockstar | jam, for context, I have about 45 minutes right now before I get go to shovel 2 ft of snow from the walks at my church. | 17:45 |
| jam | rockstar: ah, you live in *that* part of the country :) | 17:45 |
| jam | do you want to skype or type? | 17:45 |
| rockstar | jam, whichever works. Skype might be better. | 17:46 |
| jam | rockstar: let's just do the call, this email is going to take a while. | 17:47 |
| rockstar | jam, okay. | 17:47 |
| jam | I don't see you online right now | 17:48 |
| rockstar | jam, I'm on now. | 17:48 |
| rockstar | (I have to start Skype with a little pulse misdirection) | 17:48 |
| === rockstar is now known as rockstar-afk | ||
| === asac_ is now known as asac | ||
| Peng | beuno: My shirt got here today. :) | 19:22 |
| beuno | Peng, yay! | 19:23 |
| Peng | Whoever wrote the shipping information on the envelope has nice handwriting. :) | 19:24 |
| === awilkins_ is now known as awilkins | ||
| lifeless | moin | 20:12 |
| RenatoSilva | verterok: hi | 20:43 |
| poolie | hello lifeless | 20:48 |
| verterok | RenatoSilva: hi | 20:51 |
| lifeless | hi poolie | 20:56 |
| === rockstar-afk is now known as rockstar | ||
| vila | g'night folks | 22:02 |
| lifeless | night vila | 22:02 |
| jam | lifeless, poolie: Can one of you guys contact spm or another losa to get 2.0.2 and 2.1.0b2 branches set up? I just realized it is probably better to start the branch tonight, and the filter patches into them. So that we aren't waiting on them over the weekend, etc. | 22:15 |
| jam | If not, I'll try to get a hold of someone tomorrow. | 22:15 |
| jam | I'm off for now. | 22:16 |
| lifeless | spm: ^ | 22:22 |
| igc | morning | 22:33 |
| GaryvdM | Hi igc. | 22:47 |
| igc | hi garyvdm | 22:47 |
| GaryvdM | igc: I just hacked this up: http://garyvdm.googlepages.com/out.ogv | 22:47 |
| igc | garyvdm: I'll take a look | 22:47 |
| GaryvdM | igc: rename from qbrowse (and qcommit) :-) | 22:47 |
| igc | garyvdm: before I forget, I was hoping to ask for some help ... | 22:48 |
| igc | garyvdm: epxlorer on karmic is crashing much more than on jaunty | 22:48 |
| igc | garyvdm: I'm guessing its a segafult | 22:48 |
| igc | maybe related to some data model not quite right? | 22:49 |
| * igc checks the bug list | 22:49 | |
| GaryvdM | igc: If it is a segfault - qbzr had a similar problem | 22:49 |
| GaryvdM | igc: run from the terminal so you can see stdout to check if it is a segfault. | 22:50 |
| igc | garyvdm: I'm suspecting bug 395175 | 22:51 |
| ubottu | Launchpad bug 395175 in bzr-explorer "Refresh twice in a row causes segmentation fault" [High,Confirmed] https://launchpad.net/bugs/395175 | 22:51 |
| igc | garvdm: it's crashing on implicit refresh | 22:51 |
| GaryvdM | igc: oh. | 22:51 |
| igc | garyvdm: is there any chance you could take a look? | 22:51 |
| GaryvdM | igc: ok | 22:51 |
| igc | garyvdm: much apprecated. It's beyond my qt foo right now and other things are higher on my list | 22:52 |
| igc | garyvdm: and I'm fearful that karmic being released with highlight this problem to lots of people in coming days ... | 22:52 |
| GaryvdM | igc: this is the qbzr problem we had: bug 447214 | 22:53 |
| igc | and explorer will get a reputation for unreliability is we don't fix it soon | 22:53 |
| ubottu | Launchpad bug 447214 in qbzr/trunk "Segmentation fault during startup" [Critical,Fix released] https://launchpad.net/bugs/447214 | 22:53 |
| igc | garyvdm: I'll run explorer from a terminal until I have some more data on the problem | 22:53 |
| igc | garyvdm: it definitely triggered by refresh though | 22:54 |
| GaryvdM | igc: That bug is not related to the pyqt karmic upgrade. | 22:54 |
| igc | which happens each time a qbzr dialog closes | 22:54 |
| GaryvdM | igc: so I should be able to reproduce it. (still have jaunty) | 22:54 |
| igc | garyvdm: right. That bug was on jaunty, just 5-10X more common on karmic it feels to me | 22:55 |
| GaryvdM | igc: If it's more - it may be another bug - but I'll have a go at fixing that one first. | 22:55 |
| igc | garyvdm: sounds good | 22:56 |
| GaryvdM | igc: Have you had a look at that vid. How does it look? | 22:56 |
| igc | running it now ... | 22:56 |
| igc | garyvdm: video is sweet - well done!! | 22:58 |
| GaryvdM | :-) | 22:58 |
| igc | garyvdm: do it work on directories and symlinks too? | 22:58 |
| GaryvdM | igc: let me test. | 22:59 |
| GaryvdM | igc: dirs = yes | 22:59 |
| GaryvdM | igc: symlinks = yes :-) | 23:00 |
| igc | garyvdm: now try renaming across types ... empty dir to a file, etc. | 23:01 |
| igc | hmm - probably not easy to even do | 23:01 |
| GaryvdM | igc: I don't understand | 23:01 |
| igc | garyvdm: ignore it - brain fart | 23:02 |
| GaryvdM | igc: how can a rename change a kind | 23:02 |
| GaryvdM | ok | 23:02 |
| igc | garyvdm: it can't | 23:02 |
| igc | garyvdm: just the sort of weird stuff I run into in fast-import streams | 23:02 |
| GaryvdM | igc: mv --after can though I'm sure. | 23:02 |
| spm | lifeless: jam: ok, will do. | 23:04 |
| GaryvdM | igc: what version of qbzr are you running? | 23:10 |
| igc | garyvdm: trunk | 23:11 |
| GaryvdM | :-( | 23:12 |
| GaryvdM | igc: I cant reproduce. I'll have a go when I get karmic. | 23:12 |
| igc | garyvdm: 1026 to be explicit | 23:12 |
| GaryvdM | Which should be on sunday. | 23:12 |
| igc | garyvdm: ok. Thanks | 23:13 |
| GaryvdM | igc: that has the qbzr segfault fix. | 23:13 |
| igc | garyvdm: I'm running explorer rev 299 | 23:13 |
| igc | garyvdm: the problem isn't in qbzr btw - it's definitely in explorer | 23:13 |
| igc | explorer has code which refreshes the view when a qbzr dialog is closed | 23:14 |
| igc | garyvdm: the other important right-click action we're missing is "unversion" | 23:15 |
| GaryvdM | igc: ok - that should be easy to add | 23:16 |
| igc | garyvdm: cool | 23:16 |
| GaryvdM | Good night - bed time. | 23:17 |
| GaryvdM | bye igc | 23:17 |
| igc | garyvdm: night | 23:18 |
| Glenjamin | hey guys, does anyone know a good way of getting an application shortcut to bzr explorer on mac os x? | 23:35 |
| spm | jam: that's done bzr+ssh://bazaar.launchpad.net/~bzr-pqm/bzr/2.0.2 bzr+ssh://bazaar.launchpad.net/~bzr-pqm/bzr/2.1.0b2 | 23:52 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!