[git-users] Re: Paper on Git to Write, suggestions sought!

2010-12-08 Thread Thomas Ferris Nicolaisen
I wrote a little history/overview here, maybe that helps: http://blog.tfnico.com/2010/06/distributed-source-control-management.html -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to

[git-users] Re: git blame help

2010-12-30 Thread Thomas Ferris Nicolaisen
Just to play the voice of logic here, I think it would be hard to define when a line is introduced vs. changed, as you could say that entering a new line is the same as changing an empty line. An algorithm for finding the difference would have to include some sort of fuzzy consideration of

[git-users] Re: fatal: Unable to create '.../dev/.git/index.lock': File exists.

2011-01-01 Thread Thomas Ferris Nicolaisen
No idea here... This list is mostly focused on introduction-level Git usage though, you're probably better of asking at the regular Git mailing listhttp://vger.kernel.org/vger-lists.html#git, providing them with wich version of Git you are using, OS, and steps to re-create the problem. If the

[git-users] Re: git-svn setup advice for newbie

2011-01-02 Thread Thomas Ferris Nicolaisen
Hi Graham, It sounds like you should just have a go at it :) I would first start off with learning some Git basics. Don't go into git-svn before you know Git a good bit. Quick intro: http://www.spheredev.org/wiki/Git_for_the_lazy Reference: http://progit.org/book/ Now, when you feel you're

Re: [git-users] Re: git-svn setup advice for newbie

2011-01-02 Thread Thomas Ferris Nicolaisen
Hi Graham, I think you're saying, that you don't need to do any git svn dcommit in your git-svn clone. This sounds like the Git repo is a pure read only mirror of the Subversion project. Such a setup is quite straight-forward and won't give any problems. However, as far as I understand, you

[git-users] Re: Blame file execution permissions

2011-01-03 Thread Thomas Ferris Nicolaisen
--summary commit 1c5deadd79556b4cdfc964e826597e60a42f752d Author: Thomas Ferris Nicolaisen tfn...@gmail.com Date: Mon Jan 3 12:44:24 2011 +0100 Set a+x flag. mode change 100644 = 100755 woop Hope this helps :) -- You received this message because you are subscribed to the Google Groups Git

[git-users] Re: Creating a step-by-step tutorial

2011-01-04 Thread Thomas Ferris Nicolaisen
Maybe you could consider documenting each step as a commit instead? I know the author of Test-Driven JavaScript Development did this his example code: http://tddjs.com/code Basically, each tutorial is a git-repository with a number of commits. Each commit is a step in the tutorial, so you can

[git-users] Re: Git Master Reference

2011-01-08 Thread Thomas Ferris Nicolaisen
Impressive! Some tips: Insert some linebreaks in the code sections (they don't wrap automatically) so the page doesn't get so wide. You should also add your resource here, if it isn't already: https://git.wiki.kernel.org/index.php/GitDocumentation And add a link to the dzone page on top of

[git-users] Re: Pulling work from a previous branch into current branch

2011-01-10 Thread Thomas Ferris Nicolaisen
You're better off thinking about the *commits* that you want to pull into the current branch. One way is to cherry-pick the commits where you added the files: git cherry-pick sha This is nice if you want to pull in a single commit or two. Alternatively, if you want to dive into more the more

[git-users] Re: libgit2 usage in git project

2011-01-11 Thread Thomas Ferris Nicolaisen
Hi, Please direct questions related to the development of git at the main mailing list: http://vger.kernel.org/vger-lists.html#git (In other words, I have no idea :)) -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this

[git-users] Re: Problem with branch edits

2011-01-26 Thread Thomas Ferris Nicolaisen
Changes in your working directory are not branch-specific. If you change a file without committing it, you can switch (checkout) branches as much as you want. It's only when you commit that the changes get connected to a branch. If you want to immediately stash away local changes for fixing

[git-users] Drop changes in the tree, keep index

2011-01-27 Thread Thomas Ferris Nicolaisen
Say I have this: # On branch master # Changes to be committed: # (use git reset HEAD file... to unstage) # # modified: foo # # Changed but not updated: # (use git add file... to update what will be committed) # (use git checkout -- file... to discard changes in working directory) # #

Re: [git-users] Drop changes in the tree, keep index

2011-01-27 Thread Thomas Ferris Nicolaisen
Brilliant, that was simpler than I could imagine :) Do you know if the dashes are even necessary? Here it worked without them. -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to

Re: [git-users] Drop changes in the tree, keep index

2011-01-27 Thread Thomas Ferris Nicolaisen
Aha, that clears up a few things. Thanks for the great explanation! -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to git-users@googlegroups.com. To unsubscribe from this group, send email to

[git-users] Re: Git tutorial video

2011-01-29 Thread Thomas Ferris Nicolaisen
Hehe, good one :) Seriously though, the idea of an organization being able to get all their source into Git repositories, and not getting them out again doesn't make sense. First of all, any content in a repository is a replica of code that is on a developer's workstation. You don't lose

[git-users] Re: Something wrong after simple merge

2011-02-03 Thread Thomas Ferris Nicolaisen
The clue to undo a merge is reflog+reset. It's well explained here: http://stackoverflow.com/questions/2389361/git-undo-a-merge Use git reflog (which basically shows a log of commits, merges, checkouts, everything) to find the previous state you want to reset to. If you get a merge conflict,

[git-users] Re: git diff shows some cpp file as binary and core.attributesfile has no effect

2011-02-03 Thread Thomas Ferris Nicolaisen
I just recreated the problem here on OSX with git 1.7.3.3. Maybe it's a bug in git? -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to git-users@googlegroups.com. To unsubscribe from this group, send

[git-users] Re: git diff shows some cpp file as binary and core.attributesfile has no effect

2011-02-04 Thread Thomas Ferris Nicolaisen
On Thursday, February 3, 2011 7:01:27 PM UTC+1, Thomas Ferris Nicolaisen wrote: I just recreated the problem here on OSX with git 1.7.3.3. Maybe it's a bug in git? Just to be clear: I mean I also have the problem with the core.attributesfile configuration seeming to have no effect

[git-users] Re: Any writers out there?

2011-02-07 Thread Thomas Ferris Nicolaisen
Now there's an interesting use-case.. I'm new to git. I'm trying to develop a workflow for my creative writing. I figure that other's must have came before me, so why reinvent the wheel? Note that I don't write poetry myself. I write every day on my worklog, but I just keep that in a text

[git-users] Re: Sporadic failures during git push

2011-02-09 Thread Thomas Ferris Nicolaisen
I see you already posted to the git-list, and I think you'll have higher chances of getting help there. I haven't had the same problem myself. You mention the hook scripts as if you suspect they have something to do with it. Does the problem still occur if you disable them? I'm not on a linux

[git-users] Re: git-svn.perl should check GIT_ASKPASS environment

2011-02-11 Thread Thomas Ferris Nicolaisen
Sounds like something for the main git list: http://vger.kernel.org/vger-lists.html#git -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to git-users@googlegroups.com. To unsubscribe from this group, send

[git-users] Re: Setting up a remote repo when the local is already in course

2011-02-12 Thread Thomas Ferris Nicolaisen
It depends a bit on what you already have on your live site. I think you should make a distinction between working on the source code using git repositories, and doing actual deployment of your site. I think I would do something like this. First you want to set up a reproducable/scriptable

[git-users] Re: Doubts about Wordpress upgrade using git

2011-02-12 Thread Thomas Ferris Nicolaisen
Probably the best thing here is to see how other people do it. I don't know much about Wordpress, but it seems pretty common to deploy it using Git and Capistrano: (Obviously, more recent articles are more likely to be correct today)

Re: [git-users] Re: Doubts about Wordpress upgrade using git

2011-02-14 Thread Thomas Ferris Nicolaisen
Neat solution. Glad you got it working :) -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to git-users@googlegroups.com. To unsubscribe from this group, send email to

[git-users] Revert a single commit in a single file

2011-02-14 Thread Thomas Ferris Nicolaisen
So I have an old file repo/foo.txt foo.txt has undergone many changes over the years. Recently, I changed something in commit Z. Later on, others changed foo.txt again in commits Y, X. Now, I've been put to the task of undoing the changes I made to foo.txt in commit Z. I changed other files

[git-users] Re: Revert a single commit in a single file

2011-02-19 Thread Thomas Ferris Nicolaisen
Hi Stefan, thanks for your answer. But the problem with doing so, is that I lose the changes in Y and X. checkout Z^ checks out the file exactly as it was in the revision before Z, which also means losing subsequent changes. This is what I want to avoid. -- You received this message because

[git-users] Re: Conflicts from stash

2011-02-21 Thread Thomas Ferris Nicolaisen
Try git show stash@{0} It will show you the entire contents of the stash, not just the name of the stash (which could be anything, basically). -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to

[git-users] Re: Conflicts from stash

2011-02-22 Thread Thomas Ferris Nicolaisen
Oops, the correct format is: git stash show stash@{0} The above shows the files and how much they have changed. Too see the diff in patch form: git stash show -p stash@{0} -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to

[git-users] Re: Revert a single commit in a single file

2011-02-25 Thread Thomas Ferris Nicolaisen
Note that my diff solution above is not correct. It should be diff -R Z Z^ (diff between two revisions). Eventually, I took this up in the main git list: http://www.spinics.net/lists/git/msg152112.html Conclusion was that it's a good idea to support git revert taking a path argument. In the

[git-users] Re: git svn branch handling

2011-03-04 Thread Thomas Ferris Nicolaisen
On Friday, March 4, 2011 12:21:03 PM UTC+1, Sabba Hillel wrote: Please respond directly to sabba...@gmail.com as I do not have time to follow this mailing list. Unfortunately, Google Groups camouflages your original email address to prevent spam, or whatever, so I can't CC it in this

[git-users] Re: rebase question

2011-03-12 Thread Thomas Ferris Nicolaisen
Hm, I have to read your question quite a few times before I got it.. It would help if you could illustrate the branches somehow, perhaps by pasting the git log --graph --oneline --all ascii treehttp://stackoverflow.com/questions/1064361/unable-to-show-a-git-tree-in-terminal ? So, you have

[git-users] Re: refnames starting with --

2011-03-13 Thread Thomas Ferris Nicolaisen
I would definitely avoid using -- as a prefix for branch names (I assume you know what the double-dash normally means with git, as you are using it above). I can create branches locally with - and -- prefixes, but I'm unable to check them out. [19:29:08] tfnico@x:~/projects/foo/git branch --

[git-users] Re: refnames starting with --

2011-03-13 Thread Thomas Ferris Nicolaisen
Oh, a quick googling reveals that they're already awarehttp://git.661346.n2.nabble.com/PATCH-disallow-branch-names-that-start-with-a-hyphen-td5449746.htmlof this. -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group,

[git-users] Re: Trying to push a change back to git svn central repository

2011-03-15 Thread Thomas Ferris Nicolaisen
Hi Sabba, The thing is that the repository on *gitserver* has to be a bare repository. Usually all central git repositories are bare. It doesn't make sense for them to have a work-tree, and that's why Git stops you from pushing. I suggest you have a look at my collection of resources and

[git-users] Re: Trying to push a change back to git svn central repository

2011-03-15 Thread Thomas Ferris Nicolaisen
As you may have figured out, the problem is that git svn clone (fetching repo) needs a work-tree to operate. That's why you need to split it into the bare repo and the fetching repo. The former works as a central Git repo, the latter is responsible for fetching stuff from SVN. This is also an

[git-users] Re: Help with git push -u origin master command

2011-03-15 Thread Thomas Ferris Nicolaisen
Hi, Could you perhaps explain your setup a bit more, and show the exact commands and error messages you are getting? -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to git-users@googlegroups.com. To

[git-users] Re: Using branches within a git svn repository

2011-03-16 Thread Thomas Ferris Nicolaisen
Hi Sabba, With the proper configuration, only one push is needed. Check out this more recent post, it covers an example setup for handling branches in a similar setup: http://blog.tfnico.com/2010/11/git-svn-mirror-for-multiple-branches.html As you can see, the magic is in the

[git-users] Re: How does GIT tag works with gmake

2011-03-18 Thread Thomas Ferris Nicolaisen
Hi, Normally, build tools (like gmake) and source code management tools (like git) work adjacent to each other. Simply put: nothing you do with gmake has anything to do with git. If you want more information about git internals, maybe start with a read through this: http://progit.org/book/

[git-users] Re: Odd performance problem with git diff

2011-03-18 Thread Thomas Ferris Nicolaisen
Hi Dave, Just to clarify: The normal way to check for changes in a repository is usually this: cd /path/to/repository git status There are many factors that come into play with git performance, and although it is inherently fast compared to a remote repository system like Subversion, it

[git-users] Re: Running a dcommit for all branches

2011-03-22 Thread Thomas Ferris Nicolaisen
Just to be clear, the upci alias is explained here: http://blog.tfnico.com/2010/11/git-svn-mirror-for-multiple-branches.html It's defined as: upci = !git update-ref refs/remotes/$(git branch | grep '^*' | awk '{print $2}') refs/remotes/origin/$(git branch|grep '^*'|awk '{print $2}') git svn

[git-users] Re: Running a dcommit for all branches

2011-03-22 Thread Thomas Ferris Nicolaisen
Thanks. Unfortunately I do have a lot of branches. Actually, the fetch does pick up all branches and sends them to the bare repository so that every fetch updates every branch. So far it is only the dcommit that is giving the problem. I am trying to avoid having to get everyone use a git svn

[git-users] Re: Running a dcommit for all branches

2011-03-22 Thread Thomas Ferris Nicolaisen
The trick is to first clone the bare repository, and then do a git svn init inside it, so it can be used for dcommitting. There are two ways to set up a repository that you can dcommit from: 1) git svn clone (takes ages with big repositories) 2) git clone existing repository, and then do git

[git-users] Re: Running a dcommit for all branches

2011-03-23 Thread Thomas Ferris Nicolaisen
Almost, this is how: git clone bare-url gitdir cd gitdir git svn init svn-url This is 5. Set up an SVN remote in the developer's repo in this guide: http://blog.tfnico.com/2010/11/git-svn-mirror-for-multiple-branches.html -- You received this message because you are subscribed to the Google

[git-users] Re: Problem Git Push

2011-03-23 Thread Thomas Ferris Nicolaisen
Hard to say, especially without more information. It does seem that others have had the same though: http://support.github.com/discussions/repos/3651-git-push-fails-with-error-unpack-failed-index-pack-abnormal-exit Maybe try some of the workarounds mentioned there? -- You received this

[git-users] Re: Running a dcommit for all branches

2011-03-23 Thread Thomas Ferris Nicolaisen
On Wed, Mar 23, 2011 at 4:41 PM, Hillel (Sabba) Markowitz sabbahil...@gmail.com wrote: On Wed, Mar 23, 2011 at 2:03 AM, Thomas Ferris Nicolaisen tfn...@gmail.com wrote: Almost, this is how: git clone bare-url gitdir cd gitdir git svn init svn-url This is 5. Set up an SVN remote

[git-users] Re: Running a dcommit for all branches

2011-03-23 Thread Thomas Ferris Nicolaisen
No, I haven't heard of any git-svn dealings with svn:externals. That would be beyond my expectations of what git-svn is made to deal with. There are some Google resultshttp://www.google.com/search?q=git+svn+externalsabout it though. Maybe you find something there? -- You received this

[git-users] Re: Odd performance problem with git diff

2011-03-24 Thread Thomas Ferris Nicolaisen
Yes, git log http://www.kernel.org/pub/software/scm/git/docs/git-log.htmlcan take two times as arguments. git log --since two weeks ago --until yesterday You might also want to consider git whatchangedhttp://www.kernel.org/pub/software/scm/git/docs/git-whatchanged.html : git whatchanged

[git-users] Re: Odd performance problem with git diff

2011-03-24 Thread Thomas Ferris Nicolaisen
Actually this was a much better guide to the git date formats: http://www.alexpeattie.com/blog/working-with-dates-in-git/ -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to git-users@googlegroups.com.

[git-users] Re: Running a dcommit for all branches

2011-03-24 Thread Thomas Ferris Nicolaisen
git branch mybranch remotes/origin/mybranch git co mybranch I had created the co alias for checkout and the upci alias. I tried this and modified a file followed by git commit -a -m test file update git upci It gave me an error message unable to determine upstream SVN information from

[git-users] Re: Running a dcommit for all branches

2011-03-25 Thread Thomas Ferris Nicolaisen
OK, glad it worked. What kind of warnings are you getting? On Fri, Mar 25, 2011 at 6:40 PM, Hillel (Sabba) Markowitz sabbahil...@gmail.com wrote: So it's hard to say. I would first step away from the upci alias for a bit, and use the manual update-ref + dcommit commands again, just to be sure

[git-users] Re: Running a dcommit for all branches

2011-03-25 Thread Thomas Ferris Nicolaisen
OK, this is normal. I've always had this and it never gave me any problems. On Fri, Mar 25, 2011 at 8:32 PM, Hillel (Sabba) Markowitz sabbahil...@gmail.com wrote: On Fri, Mar 25, 2011 at 3:29 PM, Thomas Ferris Nicolaisen tfn...@gmail.com wrote: OK, glad it worked. What kind of warnings

[git-users] Re: svn remove branch not passed through to git

2011-03-29 Thread Thomas Ferris Nicolaisen
Yes, this is true. By default, you'll get a lot of stale branches in your fetching repo. For the time being, I've ignored this. I regularly clean it up. Say I have a mirror of a svn repo with trunk, tags/1.0 and a branch branches/1.0.x: Fetching repo: tfnico:~/git-repos/website-fetching/git

[git-users] Re: dcommit with conflict

2011-03-30 Thread Thomas Ferris Nicolaisen
Someone else does a commit to the svn repository of a change to README.txt Is the above commit properly rebased into your local repository before you do the upci? Also, try again to do the manual git update-ref and git svn dcommit instead of using the upci alias. Could be that there is

[git-users] Re: git pull whithout download all file

2011-03-30 Thread Thomas Ferris Nicolaisen
I think the easiest way would be to run instaweb in the git repository, and then get the file you want via http/wget. More info here: https://git.wiki.kernel.org/index.php/Gitweb -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post

[git-users] Re: dcommit with conflict

2011-03-30 Thread Thomas Ferris Nicolaisen
You are right with one clarification: The svn repository must be set up to trigger a rebase in the fetching repository upon committing. I've found the best way to do this is to have a svn post-commit hook that triggers a job that does this (git svn rebase/fetch + push to bare repo). It is also

[git-users] Re: dcommit with conflict

2011-03-31 Thread Thomas Ferris Nicolaisen
No problem, that's what this mailing list is here for. I hope you're successful with your conversion. That's what it's all about in the end! :) -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to

[git-users] Re: Odd performance problem with git diff

2011-04-01 Thread Thomas Ferris Nicolaisen
Yay! These .svn folders are causing havoc to our workspace in general. Amazing how much they slow it down. Another good reason to switch to git :) -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to

[git-users] Re: Git on Android

2011-04-01 Thread Thomas Ferris Nicolaisen
Wow, that sounds ambitious :) Personally, I don't have an Android device, and I imagine if I did, I probably wouldn't use it much for development, hence I wouldn't need Git support on it (I would probably use DropBox for storing files and stuff). Maybe, if I had larger tablet, it would be

Re: [git-users] git read-tree failure

2011-04-05 Thread Thomas Ferris Nicolaisen
I'm not really familiar with the read-tree command, nor have I had any use for it the last year of using git :) I think read-tree is more of a lower level plumbing command that normally we do not have so much need for. Maybe you can explain us what you really want to achieve? -- You received

[git-users] Re: Fwd: Available consultant's for your requirement.

2011-04-07 Thread Thomas Ferris Nicolaisen
And not a single Git consultant. Tsk, tsk! Akshay Korgaonkar wrote: -- Forwarded message -- From: Akshay Korgaonkar akshay.korgaon...@wisdominfotech.us Date: Wed, Apr 6, 2011 at 12:23 AM Subject: Available consultant's for your requirement. To: recruiter.aks...@gmail.com

[git-users] Re: How to manage branches of submodules without publishing those branches to the submodule origins?

2011-04-13 Thread Thomas Ferris Nicolaisen
I'm not sure there are so many Drupal users frequenting this list.. Anyhow, I'm sure there are someone who can suggest something. It's just that your mails are a bit long, and I haven't quite gotten around to read them through yet :) -- You received this message because you are subscribed to

[git-users] Re: Interactive rebase and using date of the most recent commit

2011-04-20 Thread Thomas Ferris Nicolaisen
A commit has a date. When you squash one commit into the previous commit, it keeps the date of the commit you are squashing into. Makes sense. You could maybe reset a previous commit, and then manually amend/edit the changes into the existing commit with the date you want to keep. If you want

[git-users] Re: workflow involving lock

2011-04-20 Thread Thomas Ferris Nicolaisen
Hi Mikael, The short answer is: No, if you want central locking of a resource, Git is not for you. Longer answer: http://stackoverflow.com/questions/119444/locking-binary-files-using-git-version-control-system -- You received this message because you are subscribed to the Google Groups Git

[git-users] Re: Optimal Git setup for multiple websites

2011-04-21 Thread Thomas Ferris Nicolaisen
Yes, if it's two different websites, there's no point in them sharing a repository. Have two repositories: website-us and website-uk. There's nothing wrong with having more machines involved, of course. You can clone around the repository as many times as you want. But before a developer in

Re: [git-users] Re: slowness pulling data to mobile devices

2011-04-21 Thread Thomas Ferris Nicolaisen
If it's only for backup or transport purposes, you could still rsync the repository to your mobile. Making it a bare repository would reduce the size of stuff to be rsynced. I'm not familiar with the internals http://progit.org/book/ch9-6.html of a git push or pull, but there is a lot of

[git-users] Re: Optimal Git setup for multiple websites

2011-04-21 Thread Thomas Ferris Nicolaisen
Oh, I didn't spot the difference there. Hmm. Well, you can have an intermediate intermediate repository if you want. This is a nice place for sharing branches and stuff between developers. If you have it push onward to the webserver automatically or if you have the developers push themselves..

[git-users] Re: Merge went wrong?

2011-04-28 Thread Thomas Ferris Nicolaisen
72065c6 Author: Thomas Ferris Nicolaisen tfn...@gmail.com Date: Thu Apr 28 10:10:19 2011 +0200 Merge branch 'bar' into funk Conflicts: src/business/validation/src/test/java/org/hbird/business/validation/UpperLowerCountTest.java diff --cc src/business/validation/src/test

[git-users] Re: Merge went wrong?

2011-04-28 Thread Thomas Ferris Nicolaisen
When it comes to resolving the problem, you've got three options, I guess: 1) Leave it like it is and keep on working 2) Force a rewrite of history onto the github master branch. This will create problems for anyone who have already cloned with the latest changes 3) Abandon the master branch and

[git-users] Re: Local merges from continuous integration

2011-04-28 Thread Thomas Ferris Nicolaisen
Hi Barry, Do you know exactly how CruiseControl updates the repository (pulls)? Can you paste the .git/config from the CruiseControl repo? -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to

[git-users] Re: Merge went wrong?

2011-04-28 Thread Thomas Ferris Nicolaisen
Hm, that could be. If you can reproduce the problem, maybe this is a bug in EGit that should be reported.. My local EGit installation is a bit outdated so I can't test myself at the moment. -- You received this message because you are subscribed to the Google Groups Git for human beings

[git-users] Aw: which release (non-beta) to download for windows?

2011-05-02 Thread Thomas Ferris Nicolaisen
I think the closest you can get to an official release is the one linked to on http://git-scm.com/ Which at the moment is... Git-1.7.4-preview20110204.exehttp://code.google.com/p/msysgit/downloads/detail?name=Git-1.7.4-preview20110204.execan=3q= - Full installer for official Git 1.7.4 This

[git-users] Re: Local merges from continuous integration

2011-05-03 Thread Thomas Ferris Nicolaisen
I think it sounds a bit unintuitive that you manually created CruiseControl's working copy repository, but at the same time you didn't author how CruiseControl pulls updates. Anyhows, I have little knowledge of how CruiseControl works with Git. Perhaps you should rather try their support

Re: [git-users] Re: error: could not lock config file /path/to/file/.gitconfig

2011-05-05 Thread Thomas Ferris Nicolaisen
Ah, silly me. Maybe also post which version of Git you are using, and whether you're on windows. You can also just jump around the problem and just edit the .gitconfig file manually: [user] name = You Name email = y...@email.com -- You received this message because you are

[git-users] Re: Local site running on webbrick has no data

2011-05-11 Thread Thomas Ferris Nicolaisen
Hi Bill, It's a bit hard to understand what you're problem is. You say it's a DB issue, but this mailing list has little to do with neither databases nor WEBrick. I googled a bit and found that WEBRick is included in the Ruby on Rails project. Perhaps you should try asking for help at their

[git-users] Re: Pushing new VIEW code to production not showing in browser

2011-05-11 Thread Thomas Ferris Nicolaisen
Hi Bill, I'm assuming this is another WEBrick problem. Please see my other mail for advice ;) -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to git-users@googlegroups.com. To unsubscribe from this

[git-users] Re: deleted local changes via checkout

2011-05-17 Thread Thomas Ferris Nicolaisen
Hi Thorolf, I'm afraid you've fallen out of Git's mercy by not adding these changes to the index, or stashing them. Local changes are not specific for any branch, so to speak. They don't stick to the branch you're in before you commit. By adding your files to the index, Git will not overwrite

[git-users] Aw: Re: deleted local changes via checkout

2011-05-17 Thread Thomas Ferris Nicolaisen
Erm, don't forget to commit as well! That's even safer than adding :) -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to git-users@googlegroups.com. To unsubscribe from this group, send email to

[git-users] Re: How to install git-svn from source?

2011-05-20 Thread Thomas Ferris Nicolaisen
You don't need to build git from source. There are plenty of ready binaries around. Just pick your platform in http://git-scm.com/download As far as I know, git-svn are included in most binaries. If you still have problems, let us know which platform you're on (Windows, Linux or OS X). If you

[git-users] Re: How to install git-svn from source?

2011-05-20 Thread Thomas Ferris Nicolaisen
Hm, well this is a bit beyond me, I'm afraid. You could try asking at the main Git list: http://vger.kernel.org/vger-lists.html#git -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to

[git-users] Re: how merge a branch in an empty head

2011-05-23 Thread Thomas Ferris Nicolaisen
Not quite sure what you're trying to do here, but usually there is no point in merging into an empty branch. Rather, just create the branch based on the other branch: git branch master origin/master (sets up a branch called master based on the remote origin/master) -- You received this

Aw: Re: [git-users] Won't push

2011-05-23 Thread Thomas Ferris Nicolaisen
Konstantin beat me to it. I suspect that 'wn' is 'webbynodehttps://github.com/webbynode/webbynode/issues/59 '. git commit -a is synonymous to doing a commit after doing git add -u (adds all changes for files already tracked). -- You received this message because you are subscribed to the

[git-users] Re: -v option for git log --grep

2011-05-24 Thread Thomas Ferris Nicolaisen
If you want to do the equivelant of a svn log -v, the parameter you want is --name-status I use this so often that I have an alias in my .gitconfig: alias.logv=log --name-status If you have any questions to the git developers, the place to ask is the main git mailing list

[git-users] Aw: Submitting into SVN using Git commit entries

2011-05-25 Thread Thomas Ferris Nicolaisen
Hum.. I can certainly understand that this would be a useful feature. It would be kinda like a reverse authormap (git svn fetch -A). Unfortunately, git-svn doesn't support this. I can imagine how it could be implemented, using an array of prepared ~/.subversion/auth files for each author, and

[git-users] Aw: Re: Submitting into SVN using Git commit entries

2011-05-25 Thread Thomas Ferris Nicolaisen
(this is a reply to your second email, my first mail was a reply to your first one) http://blog.tfnico.com/2010/10/gitsvn-5-centralized-git-svn-mirror.htmlI can see how this works and I like the idea, although what would happen if people share code in the Git repository before submitting

[git-users] Re: How can i know who deleted git branch?

2011-05-26 Thread Thomas Ferris Nicolaisen
Hi, It's a bit hard to give an answer without being more concrete.. Have a look at this blog post, it goes through a lot of examples on how to recover a lost branch (or commit): http://programblings.com/2008/06/07/the-illustrated-guide-to-recovering-lost-commits-with-git/ -- You received

[git-users] Re: gitg web interface

2011-05-27 Thread Thomas Ferris Nicolaisen
Hum, well I didn't even notice this one last time I was looking through the alternatives. I was looking for some tool that was able to show side-by-side diffs. In the end we ended up installing Gerrit. We just use normal Gitweb as repo web frontend. I haven't yet found any good reasons to

[git-users] Re: gitg web interface

2011-05-27 Thread Thomas Ferris Nicolaisen
Oh, now I realized that gitg is like a desktop Git client. cgit on the other hand is a repository web frontend. These are two different animals. -- You received this message because you are subscribed to the Google Groups Git for human beings group. To post to this group, send email to

Re: [git-users] Re: gitg web interface

2011-05-27 Thread Thomas Ferris Nicolaisen
I think you've got a user with some bad ideas. Gerrit is for code-reviewing, accepting patches into a given branch. Yes, it can merge patches in, but it can't rebase. I would definitely not recommend Gerrit as a tool for branching, merging and rebasing. Your user just has to learn some Git,

[git-users] Re: 'cannot create thread' doing clone or fetch

2011-06-01 Thread Thomas Ferris Nicolaisen
Hmm.. Is this maybe related? http://comments.gmane.org/gmane.comp.version-control.git/173793 -- You received this message because you are subscribed to the Google Groups Git for human beings group. To view this discussion on the web visit

[git-users] Re: 'cannot create thread' doing clone or fetch

2011-06-01 Thread Thomas Ferris Nicolaisen
Argh, of course it was. Sorry, I didn't see that you had replied to that thread already. Still, I think this subject is beyond the complexity for this newbie list :) -- You received this message because you are subscribed to the Google Groups Git for human beings group. To view this

[git-users] Re: Git on Android

2011-06-13 Thread Thomas Ferris Nicolaisen
Oh, here we go: https://market.android.com/details?id=com.madgag.agit -- You received this message because you are subscribed to the Google Groups Git for human beings group. To view this discussion on the web visit https://groups.google.com/d/msg/git-users/-/FKnL9LSOQ1AJ. To post to this

[git-users] Re: Gerrit can't handle task in the queue and the CPU usage 100%

2011-06-14 Thread Thomas Ferris Nicolaisen
Hi, I think Gerrit is a bit outside the scope of this list. I see that you already tried twice on the Gerrit mailing list (https://groups.google.com/forum/#!forum/repo-discuss), and I doubt you'll get more help here. -- You received this message because you are subscribed to the Google

[git-users] Re: problem with cherry-pick

2011-06-14 Thread Thomas Ferris Nicolaisen
Hi Giuseppe, I find it very hard to understand what you are trying to do. Can you maybe explain in more general terms what you are trying to achieve? Deleting initial commits doesn't make much sense to me, at least. What is the reason you are doing this? -- You received this message because

[git-users] Re: Checkout and reset to a particular commit using git gui

2011-06-15 Thread Thomas Ferris Nicolaisen
Hi, I'm not so familiar with Git GUI, but I see there is this Browse Branch Files command where you can explore the repository in a given branch/revision without doing anything with your working tree. Does that work for you? Also, in gitk (the log visualizer), you can right click commits and

[git-users] Aw: Problem with givt svn clone

2011-06-22 Thread Thomas Ferris Nicolaisen
Hi Thorsten, This is a common problem when there are colliding branch names with branch directories. You can have multiple branches sharing a prefix, and git uses directories to represent this): http://stackoverflow.com/questions/2527355/using-the-slash-character-in-git-branch-name So you

[git-users] Aw: Problem with givt svn clone

2011-06-22 Thread Thomas Ferris Nicolaisen
OK, I've found a working work-around. The git-svn documentation reads: When using multiple --branches or --tags, *git svn* does not automatically handle name collisions (for example, if two branches from different paths have the same name, or if a branch and a tag have the same name). In these

[git-users] Re: Problem with givt svn clone

2011-06-26 Thread Thomas Ferris Nicolaisen
Regarding the missing branches, I don't know how that is. Can you add the full urls to these missing branches the way they look in the SVN repo? Do these branches still exist at HEAD revision? -- You received this message because you are subscribed to the Google Groups Git for human beings

[git-users] Re: Issue with migration from Subversion?

2011-06-26 Thread Thomas Ferris Nicolaisen
Hi Steph, Can you describe the structure of your SVN repository (how the branches are located in particular), and the exact command and parameters you used when starting git-svn? -- You received this message because you are subscribed to the Google Groups Git for human beings group. To view

Re: [git-users] Re: Issue with migration from Subversion?

2011-06-27 Thread Thomas Ferris Nicolaisen
On Monday, June 27, 2011 8:07:04 PM UTC+2, Steph wrote: I've forwarded the error part of this to the larger git mailing list, but I have some PEBKAC issues that I'd like to hash out in a smaller forum... The command I used is this: /usr/local/git/bin/git svn clone https://[server]/svn -A

[git-users] Re: Problem with givt svn clone

2011-06-28 Thread Thomas Ferris Nicolaisen
Hmm, I wouldn't be happy with losing history.. Rather than focusing on the errors that were printed during cloning, if there are no missing branches any more, what is the actual problem after cloning? -- You received this message because you are subscribed to the Google Groups Git for human

  1   2   3   4   5   6   7   8   >