Team Chat Logs

March 10, 2010

2010 2
Mo Tu We Th Fr Sa Su
1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30 31        
<< <   > >>

[00:09:16]*cxreg!count@62.f9.1243.static.theplanet.com sees a list mail mentioning "log --pretty=raw" and his mind goes in the gutter
[00:35:02]<VxJasonxV>I'm unable to git svn rebase, the migration (to v2, apparently?) is failing. I'm looking at the release notes for 1.7.0 and beyond, but I don't see anything about a git-svn change. I'm trudging through the git ML, but any one have a link handy already?
[01:12:33]<crankharder>i git svn clone'd a subversion repository and it tracked the svn repo pretty well (git br -av shows all of my svn remote branches)
[01:12:57]<crankharder>however, for some reason 'git svn rebase' isn't pulling in the latest changes
[01:13:31]<crankharder>any guesses as to why?
[01:16:04]<jaysern>how do you fork in git ?
[01:20:44]<crab>fork what?
[01:22:33]<joec4i>jaysern, you mean branching?
[01:22:41]<jaysern>a repo
[01:22:53]<jaysern>i'm confused with the terminology used
[01:23:33]<joec4i>so what do you want with the repo?
[01:23:45]<jaysern>i have a repo with this project, i have a few branches in it
[01:24:07]<jaysern>and now i decided that I want to take one of the branches in a completely different direction
[01:24:22]<jaysern>so i would like this branch to become a separate repo
[01:24:57]<jaysern>because the 2 repos are going to be 2 different projects
[01:26:04]<joec4i>you can use another branch for it or just clone the repo to something like repo2, and do whatever you want with repo2 .
[01:26:35]<jaysern>so clone the repo
[01:26:43]<jaysern>how do i clone a repo?
[01:27:04]<jaysern>is that what the "fork" button does on github?>
[01:27:14]<joec4i>i guess you can start with man gittutorial
[01:27:14]<Gitbot>joec4i: the gittutorial manpage can be found at http://git.or.cz/man/gittutorial
[01:31:05]<cancuengt>hello everybody
[01:32:18]<cancuengt>I'm new with git, and follow instructions to commiitting but didn't worked. I have used SVN if any can help
[01:34:28]<bob2>what did you do, and what happened?
[01:35:38]<cancuengt>bob2 thanks
[01:36:35]<cancuengt>i'm using a github account and this is my first time with this. I "cloned" the repo and modiy the source
[01:37:00]<cancuengt>then I "# git add .
[01:37:22]<cancuengt>and then git commit -m "message"
[01:37:39]<cancuengt>and then went to the github and nothing happened
[01:38:20]<cancuengt>nothing has changed
[01:38:39]<cancuengt>I have read/write access and the repo was created by other person
[01:39:38]<jaysern>you didn't push
[01:39:40]<jaysern>git push
[01:40:54]<jaysern>i think of it as "pushing" the code from your local machine, to the git server (github).
[01:41:53]<davr>cancuengt: i recommend http://progit.org/book/ -- I came from SVN, and this guide explained Git to me very well
[01:43:18]<wereHamster>cancuengt: only two git commands actually transfer commits between repositories (push and fetch). Everything else is strictly local
[02:03:22]<fynn>Hey, I'm trying to push this emergency patch and git thinks that's the best time to run auto gc that can take 30 minutes
[02:03:29]<fynn>How do I push regardless?
[02:03:40]<fynn>i.e. without the auto gc completing?
[02:04:34]<fynn>(this is running locally)
[02:04:45]<wereHamster>isn't gc run after the push updates the refs in the remote repo?
[02:04:58]<wereHamster>oh, is gc running locally?
[02:05:25]<fynn>wereHamster: yeah, it's a pull --rebase
[02:05:34]<fynn>that's when the gc happens.
[02:05:38]<wereHamster>abort the gc and set gc.auto to zero
[02:06:09]<wereHamster>(that's my guess what could avoid the gc)
[02:06:19]<fynn>wereHamster: thanks
[02:06:48]<wereHamster>btw, does the gc run after or before the pull finishes?
[02:07:01]<fynn>wereHamster: looks like after.
[02:07:09]<wereHamster>then you can safely abort it
[02:07:21]<fynn>wereHamster: interesting, what's the difference?
[02:07:39]<wereHamster>operations will be a bit slower than usual
[02:08:09]<wereHamster>git gc optimizes the repository, that's about all it does
[02:08:22]<fynn>wereHamster: I mean, why is a post-pull gc abortion safer?
[02:09:29]<wereHamster>if it ran in the middle of the pull, I wouldn't abort it. But if it is running after a successful pull, then I don't think anything bad can happen
[02:10:04]<fynn>makes sense, thanks!
[02:36:16]<EugenMayer>hello guys, i have an old commit and i want to start a branch out of this commit
[02:36:31]<EugenMayer>i have checked out those commit, but how to start a branch from this commit now?
[02:38:48]<thiago>git branch <branchname> <thatcommit>
[02:38:56]<thiago>note that this doesn't affect your checkout
[02:39:08]<thiago>to create a branch from the commit you're currently in and check it out:
[02:39:12]<thiago>git checkout -b <branchname>
[02:41:05]<EugenMayer>thats clear
[02:41:07]<EugenMayer>thank you
[04:13:47]<beni->wereHamster: I found the reason for my problem we discussed yesterday (there were modifications right after the clone): I am using Mac with HFS+ formatted drive with case-*in*sensitivity. in kernel, there are files which are case-sensitive (e.g. there is a file ABC.c and Abc.c in the same directory). this caused the problem.
[04:17:43]<wereHamster>beni-: I saw the emails. And didn't I suggest that it might be a problem with case insensitivity?
[04:18:03]<beni->wereHamster: hm, did not see that
[04:56:18]<disappearedng_>Hey how do I update all my remote branches on my local comp
[04:56:54]<bob2>how would git know where all your repositories are?
[04:57:17]<disappearedng_>as in on origin
[04:57:25]<disappearedng_>I want to update all my origin's branches
[04:57:52]<maiOo>disappearedng_: git fetch
[04:57:54]<bob2>git remote update? git fetc?
[04:58:02]<j416>disappearedng_: git fetch
[04:58:54]<disappearedng_>oh
[04:58:58]<j416>disappearedng_: if you have more than one remote: git fetch --all
[04:59:33]<disappearedng_>ok cool
[04:59:44]<disappearedng_>so after that if I want to check out a branch that was on remote
[05:00:10]<disappearedng_>git checkout -b remoteNAME:localNAME right
[05:00:15]<ohan_magopian>hi there, anyone could recommend me a web frontend for git?
[05:00:27]<j416>ohan_magopian: gitweb?
[05:00:44]<rvsjoen>trac had a nice one last I checked as well
[05:00:46]<ohan_magopian>j416, thanks
[05:00:53]<j416>ohan_magopian: man git-instaweb
[05:00:53]<Gitbot>ohan_magopian: the git-instaweb manpage can be found at http://git.or.cz/man/git-instaweb
[05:01:12]<ohan_magopian>thanks :D
[05:03:24]<disappearedng_>oh didn't know you are here too bob2
[05:05:45]<myst>anyone knows how to add a file in edit to the index in Emacs?
[05:06:19]<hendry>i have an open repo on my remoteserver:/srv/git/repo. what is the typical hook i use to make sure the open contents update on remote commit?
[05:06:19]<myst>C-v i says the file already registered
[05:08:53]<bob2>myst: C-x v v?
[05:09:03]<bob2>myst: btw magit is pretty awesome
[05:14:45]<tschundeee>hey I want to git-commit my rails apps like a pro! so whenever I push to my repo my server shall execute a git-hook that deploys my railsapp... anybody has a good tutorial on this? I googled a bit and found this one here: http://blog.divoxx.com/2010/02/14/rails-application-deploy-using-git/ BUT I use nginx + passenger and so this is not fitting for me...
[05:18:28]<_rane>why would you do such hook?
[05:19:26]<_rane>I don't see how pushing should be tied to deployment
[05:20:20]<tschundeee>_rane: hmm because I want my server to behave a bit like linode...
[05:21:19]<_rane>maybe if you have more than one remote
[05:21:50]<shruggar>deploying on push is very common
[05:22:21]<shruggar>and "more than one remote" is not at all a requirement
[05:22:35]<shruggar>just a specific ref within your remote
[05:23:07]<shruggar>though specifics about how to set things up to work with rails, I do not know
[05:34:34]<mawe>hi. i'm using git 1.6.4.2 and subversion 1.6.6. ssl client certificates are used to authenticate with the svn repository. passphrase for certificate is stored in gnome-keyring. everythink works fine with plain svn, but with git-svn I get a prompt for the gnome keyring every time I run a command. is there a way to cache the password?
[05:35:14]<TenJack>hey I just did a merge and I would like to undo it, how do I do that?
[05:35:32]<Ilari>TenJack: git reset with suitable parameters.
[05:35:56]<TenJack>Ilari: like what parameters?
[05:39:27]<gehdan>Hi, i'm working with git on Windoes (XP, msysgit
[05:39:34]<gehdan>version 1.6.5.1.1367.gcd48)
[05:40:03]<ljungk>is there a way to share a repository with someone else without a server? Say A initially has the project, then B wants to clone the repository. They both do some changes and want to send them to each other in some way. How can this be done without a server?
[05:40:05]<j416>TenJack: man git-reset
[05:40:05]<Gitbot>TenJack: the git-reset manpage can be found at http://git.or.cz/man/git-reset
[05:40:08]<Ilari>TenJack: --merge HEAD^ or --hard ... (depending on wheither merge has already been committed, and that latter is more dangerous as it removes all uncommitted changes)
[05:40:28]<gehdan>didn't have any problems up until now: gitk and git gui show a wrong diff, not representing the changes that actually are in the file in question
[05:41:02]<gehdan>the diff does show changes that once were in that file, though. Any ideas why that is?
[05:41:24]<TenJack>Ilari: so i did 'git reset --hard' and it is not back to the version before the merge. i committed after the merge...ugh
[05:41:42]<j416>TenJack: did you forget HEAD^ ?
[05:41:49]<TenJack>is there a way to just go back versions?
[05:42:14]<j416>TenJack: if you do "git log" you will see sha1s for all commits
[05:42:14]<TenJack>I just did "git reset --hard"
[05:42:25]<j416>you can do git reset <sha1> where <sha1> is one of those
[05:42:35]<j416>adding --hard will reset your work tree as well
[05:42:44]<locklace>argh. any idea why a push would fail with this error:
[05:42:49]<locklace>"error: dst refspec foo matches more than one."
[05:43:17]<locklace>foo is a unique branch name on the remote