NfNitLoop | Kamping_Kaiser: Hrmm. Oh, did you just have local changes in your working copy (but not committed) and 'bzr up' updated them? | 00:11 |
---|---|---|
NfNitLoop | that's pretty standard. | 00:11 |
Kamping_Kaiser | NfNitLoop: its standard to merge upstream changfes without a commit if the working directory has unommited chagnes? | 00:13 |
Kamping_Kaiser | oh, i see what you mean | 00:14 |
Kamping_Kaiser | if you are ahead of trunk, it pulls down chagnes but doens't commit | 00:14 |
marv | how's bzr loom coming along? the documentation (especially http://wiki.bazaar.canonical.com/Documentation/LoomAsSmarterQuilt) make it sound like if i make a wrong move i could screw the whole quilt up | 03:13 |
marv | how well would it work to track say, around 30 separate patches? I'm currently doing it in git and with topic branches for each patch, and a branch where everything is merged together. | 03:18 |
marv | thinking of switching it to bzr, and wondering if i should try loom or keep it as topic branches | 03:19 |
marv | i'm also wondering if you can use loom with bound branches | 03:21 |
fullermd | Well, I don't use loom. But I don't see how it would help you with 30 _separate_ patches. It would fit with a stack of 30 patches, depending on each other in sequence. | 03:24 |
marv | a few of them are related or at least touch the same pieces of code, but a lot of them touch separate pieces of the code and are independent of each other | 03:26 |
fullermd | A loom is basically a single stack. So, if you have 2 or 3 branches that build on each other, a loom could be useful there. But it would be more of a distraction when they're not. | 03:29 |
marv | ok, thanks for the advice | 03:32 |
fullermd | Of course, you can mix&match. If you've got 5 branches that build on each other, that's one loom here. Another 3 that build on each other, a second loom here. A dozen that are independent, there's a dozen branches over there... | 03:33 |
marv | yeah but then i have to teach anyone who works on those about the loom commands, and since they aren't normally required no one will know how to use them, doesn't sound worth the effort | 03:37 |
fullermd | Oh, you have to deal with other _people_. I've located your problem 8-} | 03:38 |
fullermd | A loom with N threads doesn't do anything you can't do with N branches; just automates following some of the steps. So you don't necessarily lose a lot without them. | 03:39 |
marv | well, if i don't deal with other people, then I have to do everything myself. and I don't scale very well. | 03:39 |
fullermd | I understand that if [a subset of] what you want to do fits them, the added convenience can be handy. | 03:39 |
fullermd | Yeah. Just adds fuel to my belief that the universe should be sent back with a failing grade for this beta. Maybe they'll fix that in the final release... | 03:40 |
marv | dealing with other people is a large part of my motivation of trying to switch to bzr from git. have you ever tried to teach people used to svn how to use git? and not just any git, but a git repo with 30+ feature branches, a master that follows upstream, and several branches with everthing merged together? | 03:41 |
Supertanker | Oh hey, using BZR 2.x is way faster than 1.x :P | 04:03 |
* Supertanker is like, super smart for realizing that | 04:03 | |
=== Simon_ is now known as SImonK | ||
=== SImonK is now known as SimonK | ||
=== n0wkul is now known as n3wkul | ||
Solipsist | i got a problem on mac, qbzr cannot find pyqt even though the correct version is installed | 13:01 |
fullermd | Are you sure it's installed for the same version of python as bzr is using? | 13:01 |
Solipsist | i got two instances: /opt/local/var/macports/sources/rsync.macports.org/release/ports/python/py-pyqt4 and /Library/Python/2.6/site-packages/PyQt4 | 13:02 |
Solipsist | the latter is the more recent one | 13:02 |
fullermd | The former sounds like a build location. | 13:03 |
fullermd | See `bzr version`, and look at what it says it's using for the python interpreter and library location. | 13:03 |
Solipsist | http://pastebin.com/N5ttQFSz | 13:04 |
fullermd | Looke like a different python to me. | 13:05 |
Solipsist | yeah it's using the Python i installed from macports | 13:05 |
fullermd | Well, PyQt is. bzr isn't. Or maybe it's vice versa. | 13:06 |
Solipsist | i followed this guide, (installing for 2.6 instead of 2.5 though, rest i followed the guide to the point): http://blog.webgamic.nl/tag/qbzr-mac/ | 13:06 |
Solipsist | how do i fix it? | 13:06 |
fullermd | Dunno. I don't Think Different 8-} | 13:07 |
Solipsist | ill try install pyqt using macports, might sort it | 13:08 |
fullermd | As a guess, I'd try using the stdlib dir bzr is looking at (well, with '/site-packages' appended) for all those -d's, and using its full python path, and see if that works. | 13:08 |
Solipsist | how do i change stdlib dir? | 13:08 |
fullermd | I'd guess that's what the -d's on those command lines refer to. | 13:09 |
Solipsist | $ sudo ln -s /Library/Python/2.6/site-packages/* . | 13:15 |
Solipsist | just symlinking them from the /Library location to the location where macports put Python sorted it | 13:15 |
fullermd | Mmph. That sounds like a nasty ugly hack that'll come back to bite you. | 13:15 |
Solipsist | likely yeah, once i update the macports package | 13:16 |
Solipsist | sorts the problem now, will have to install bzr from source and not use the macports package as its obviously antiquated | 13:17 |
Solipsist | another day, another lesson - lol, thanks for you help | 13:19 |
fullermd | np :) | 13:19 |
Arrrgh | official site claims that there is a Visual Studio plugin for bzr but I can't find any download, who knows where | 15:17 |
Arrrgh | it is? | 15:17 |
jelmer | arrrgh: launchpad.net/bzr-visualstudio IIRC | 15:31 |
jelmer | I think it's unmaintained at this point thouh :_/ | 15:31 |
Arrrgh | I'd like to see even outdated plugin. but there is no any downloadable file | 15:37 |
jelmer | Arrrgh: there's a bzr branch you can download | 16:06 |
Arrrgh | where? | 16:06 |
jelmer | https://code.edge.launchpad.net/~hartke/bzr-visualstudio/banquet.status | 16:07 |
jelmer | (linked from the "branches" page on the bzr-visualstudio project page) | 16:07 |
Arrrgh | I can't branch, it says"bzr: ERROR: Not a branch: "D:/projects/third-party/lp:bzr-visualstudio/"", what's wrong? | 16:26 |
jelmer | what command are you running exactly? | 16:27 |
jelmer | do you have the launchpad plugin installed? | 16:27 |
Arrrgh | so I need a special plugin for launchpad. ok, thanks | 16:29 |
jelmer | Arrrgh: it should be included with Bazaar | 16:32 |
jelmer | Kamping_Kaiser: that's great to hear, thanks :-) | 17:28 |
=== radoe_ is now known as radoe | ||
=== khmarbaise_ is now known as khmarbaise | ||
slestak | Hi guys | 20:37 |
slestak | question, i really want to version some src that lives on an aix box. with the dependency we have on the aix build on a special py build, i cannot install it | 20:41 |
slestak | what i want to know is if it is a bad idea to version an smb share of the items | 20:42 |
slestak | thoughts? | 21:27 |
jelmer | slangasek: using bzr over smb should work ok | 21:29 |
jelmer | s/slangasek/slestak/ | 21:29 |
=== Adys_ is now known as Adys | ||
slestak | k, ive started googling to see if there is prior discussion. | 21:35 |
slestak | one concern i had is the .bzr might be touched by both linux and win32 clients. locally running bzr doesnt do that. that is the potential weak point in doing this | 21:36 |
slangasek | jelmer: but not over nfsv4! :) | 21:44 |
slestak | smb has gotten so pervasive and easy, i dont even consider nfs | 21:45 |
=== Simon_ is now known as SimonK | ||
bronger | I'd like to user "bzr split", however, it always says "To use this feature you must upgrade your branch to a format which supports rich roots". "bzr info" says that I use branch format 7. What should I do? Every "bzr upgrade" I tried ended with "ERROR: The branch format Meta directory format 1 is already at the most recent format". | 22:02 |
Peng | bronger: What version of bzr are you using? What repo format are you using? | 22:07 |
bronger | Repo is version 6. | 22:07 |
Peng | bronger: It's the repo format that matters here, not the branch format. | 22:07 |
Peng | Although that error message is not very obvious. | 22:07 |
bronger | "Packs 6". | 22:08 |
bronger | bzr 2.0.2. | 22:08 |
Peng | bronger: Not Packs 6 Rich Root? | 22:08 |
bronger | No, only "packs 6". | 22:08 |
Peng | Then I guess it's not a rich-root format? | 22:09 |
bronger | I don't know. I do an upgrade in the shared repo directory. | 22:09 |
Peng | Wait wait. bzr 2.0.2? That should upgrade you to 2a by default. | 22:10 |
bronger | So which option should I give? | 22:10 |
Peng | You don't have to give an option. Something is going wrong. Either you're not really using bzr 2.0.2, you already have upgraded to 2a, or you're upgrading the wrong location. | 22:11 |
Peng | What does "bzr info" on the branch say? | 22:11 |
bronger | Format: | 22:11 |
bronger | control: Meta directory format 1 | 22:11 |
bronger | working tree: Working tree format 6 | 22:11 |
bronger | branch: Branch format 7 | 22:11 |
bronger | repository: Repository format 2a - rich roots, group compression and chk inventories | 22:11 |
bronger | (After I did the repo upgrade) | 22:12 |
Peng | OK then. So you should be all set. | 22:12 |
bronger | Okay, thank you! | 22:12 |
Kamping_Kaiser | I just managed to get bzr[ rebase] to explode. http://paste.debian.net/63990/ . should i file a bug on bzr or rebase? the message says bzr, but i'm unsure | 23:46 |
Kamping_Kaiser | s/rebase/rewrite i guess, since i'm uzing the bzr branch | 23:46 |
maxb | Kamping_Kaiser: On bzr-rewrite. The message is generic | 23:49 |
Kamping_Kaiser | maxb: cheers | 23:49 |
Kamping_Kaiser | sigh. launchpad login issues. i'll try again later | 23:57 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!