Search Results

Search found 4953 results on 199 pages for 'git commit'.

Page 14/199 | < Previous Page | 10 11 12 13 14 15 16 17 18 19 20 21  | Next Page >

  • Detach many subdirectories into a new, separate Git repository

    - by prisonerjohn
    This question is based on http://stackoverflow.com/questions/359424/detach-subdirectory-into-separate-git-repository Instead of detaching a single subdirectory, I want to detach a couple. For example, my current directory tree looks like this: /apps /AAA /BBB /CCC /libs /XXX /YYY /ZZZ And I would like this instead: /apps /AAA /libs /XXX The --subdirectory-filter argument to git filter-branch won't work because it gets rid of everything except for the given directory the first time it's run. I thought using the --index-filter argument for all unwanted files would work (albeit tedious), but if I try running it more than once, I get the following message: Cannot create a new backup. A previous backup already exists in refs/original/ Force overwriting the backup with -f Any ideas? TIA

    Read the article

  • Why does the Git community seem to ignore side-by-side diffs

    - by Kyle Heironimus
    I used to use Windows, SVN, Tortoise SVN, and Beyond Compare. It was a great combination for doing code reviews. Now I use OSX and Git. I've managed to kludge together a bash script along with Gitx and DiffMerge to come up with a barely acceptable solution. I've muddled along with this setup, and similar ones, for over a year. I've also tried using the Github diff viewer and the Gitx diff viewer, so it's not like I've not given them a chance. There are so many smart people doing great stuff with Git. Why not the side-by-side diff with the option of seeing the entire file? With people who have used both, I've never heard of anyone that likes the single +/- view better, at least for more than a quick check.

    Read the article

  • git changing head not reflected on co-dev's branch

    - by stevekrzysiak
    Basically, we undid history. I know this is bad, and I am already committed to avoiding this at all costs in the future, but what is done is done. Anyway, I issued a git push origin <1_week_old_sha:master to undo some bad commits. I then deleted a buggered branch called release(which had also received some bad commits) from remote and then branched a new release off master. I pushed this to remote. So basically, remote master & release are clones and just how I want them. The issue is if I clone the repo anew(or work in my current repo) everything looks great....but when my co-devs delete their release branch and create a new one based off the new remote release I created, they still see all the old junk I tried to remove. I feel this has to do with some local .git files mistaking the new branch release for the old release. Any thoughts? Thanks.

    Read the article

  • Forking an open source project using Git

    - by Cromulent
    There is an open source project that I want to fork for my own reasons. It currently has a Git mirror of its main SVN repository which I can use. What are the best techniques for forking a project whilst still maintaining the ability to merge future changes from the original project into your own forked repository using Git? Please note I will not be using Github at all for development so using any features from that is out of the question. The project will be hosted on a private VPS.

    Read the article

  • Git: Stage into Commit, what is the right workflow?

    - by Lukasz Lew
    I just created a big piece of code I want to commit in several separate commits. So I can stage relevant parts, commit, stage, commit, ... and so on until I have all my changes commited. The missing part is how can I test whether I split the commit correcty. I.e. whether the part that is in staging area at least compiles? To do that I must somehow bring my work tree to be in sync with index (staging area) without losing the changes to be committed later. What is the right way to do it? What is the quickest way to do it? Update: How to do it with magit?

    Read the article

  • Git doesn't sync files until committed, even if checked out in a different branch

    - by DertWaiter
    Okay, I have git 1.7.11.1 on Windows and I have a local test repository with 2 branches. One is master with index.php and help.php. I then create another branch called slave :) I run from git bash rm help.php and it disappears from the folder, but I don't stage anything. I switch to checkout master branch and it is supposed to restore file help.php because it is not modified in the master branch, isn't it? And it does not do it. When I go back to the slave branch and commit and then switch to checkout master then help.php appears. Is that the way it is supposed to to work? Why?

    Read the article

  • Can I do a git merge entirely remotely?

    - by CaptainAwesomePants
    My team shares a "work" branch and a "stable" branch. Whenever a particular work branch is approved for further testing/release/etc, we merge it into stable. No code is ever checked directly into the stable branch. Because of this, merge conflicts simply won't happen, and it seems silly to pull down the work branch and the stable branch, merge them, and then push the changes back. Is there a git command to ask a remote git server to commit a merge of two branches that it already knows about?

    Read the article

  • How to regularly merge two git repositories, one with submodules into one without

    - by smoothify
    I maintain a Drupal project in a git repository containing submodules. This works well for me overall, and I like the submodule approach. However, I would like to move my site to a hosting provider that offers deployment via git push but doesn't work with submodules. I would like to keep my current repository intact, and then when I'm ready to deploy, I would like to merge the changes from my repository into the deployment repository, but any submodules need to be exported into the tree. So, it needs to be (semi) automated, so I can just run a command or two and initiate the merge, and then push to the server. Ideally it would keep track of individual commits, but I wouldn't mind if it squashed them into a single commit. How would be the most effective way to achieve this?

    Read the article

  • $HOME git repo (selectively) to github?

    - by user428502
    I keep many files in my home directory under git. Important dotfiles, my thesis, etc. I want to push certain files to github, e.g., my emacs configuration, to share. Obviously, I don't want to push the entire repo. Are submodules the way to go? My first thought is to make a directory ~/github/emacs, and rsync selective files here, then add a submodule under that directory, pointing to github, to push. Is this a good idea, or is there a better way? (I don't want my local git repo storing all files to get muddled up with this stuff, though.)

    Read the article

  • Tracking work history in a git repo

    - by Code-Guru
    Previous related questions: Code bases for desktop and mobile versions of the same app Git branching and tagging best practices Question: I have split my repo into three directories (swing, android, and common) as suggested by @KarlBielefeldt in response to my previous question. Now I am jumping back and forth between developing my Android port and tweaking/adding features to my original Swing app. All of my commits are linear (fast-forward) and only my commit messages give hints indicating whether I'm working on my Swing app or my Android app. Is there a better way to keep track of the work flow in my git repo?

    Read the article

  • Git dont sync files untill committed even if checkout different branch

    - by DertWaiter
    Ok i have git 1.7.11.1 on windows and i have local test reposotory with 2 branches one is master with index.php help.php then i create another branch called slave :) I run from git bash rm help.php and it dissapears from the folder, but i dont stage anything. I switch to checkout master branch and it supposed to restore file help.php because its not modified in master branch isnt it? And it does not do it. When i back to slave branch and commit and then switch to checkout master then help.php appears. is that the way it supposed to be why?

    Read the article

  • How can the Three-Phase Commit Protocol (3PC) guarantee atomicity?

    - by AndiDog
    I'm currently exploring worst case scenarios of atomic commit protocols like 2PC and 3PC and am stuck at the point that I can't find out why 3PC can guarantee atomicity. That is, how does it guarantee that if cohort A commits, cohort B also commits? Here's the simplified 3PC from the Wikipedia article: Now let's assume the following case: Two cohorts participate in the transaction (A and B) Both do their work, then vote for commit Coordinator now sends precommit messages... A receives the precommit message, acknowledges, and then goes offline for a long time B doesn't receive the precommit message (whatever the reason might be) and is thus still in "uncertain" state The results: Coordinator aborts the transaction because not all precommit messages were sent and acknowledged successfully A, who is in precommit state, is still offline, thus times out and commits B aborts in any case: He either stays offline and times out (causes abort) or comes online and receives the abort command from the coordinator And there you have it: One cohort committed, another aborted. The transaction is screwed. So what am I missing here? In my understanding, if the automatic commit on timeout (in precommit state) was replaced by infinitely waiting for a coordinator command, that case should work fine.

    Read the article

  • Can I keep git from pushing the master branch to all remotes by default?

    - by Curtis
    I have a local git repository with two remotes ('origin' is for internal development, and 'other' is for an external contractor to use). The master branch in my local repository tracks the master in 'origin', which is correct. I also have a branch 'external' which tracks the master in 'other'. The problem I have now is that my master brach ALSO wants to push to the master in 'other' as well, which is an issue. Is there any way I can specify that the local master should NOT push to other/master? I've already tried updating my .git/config file to include: [branch "master"] remote = origin merge = refs/heads/master [branch "external"] remote = other merge = refs/heads/master [push] default = upstream But remote show still shows that my master is pushing to both remotes: toko:engine cmlacy$ git remote show origin Password: * remote origin Fetch URL: <REPO LOCATION> Push URL: <REPO LOCATION> HEAD branch: master Remote branches: master tracked refresh-hook tracked Local branch configured for 'git pull': master merges with remote master Local ref configured for 'git push': master pushes to master (up to date) Those are all correct. toko:engine cmlacy$ git remote show other Password: * remote other Fetch URL: <REPO LOCATION> Push URL: <REPO LOCATION> HEAD branch: master Remote branch: master tracked Local branch configured for 'git pull': external merges with remote master Local ref configured for 'git push': master pushes to master (local out of date) That last section is the problem. 'external' should merge with other/master, but master should NEVER push to other/master. It's never gong to work.

    Read the article

  • Git clone/push/pull - where's that username comes from?

    - by Kuroki Kaze
    I've set up gitosis and able to pull/push through ssh. Gitosis is installed on Debian Lenny server, I'm using git from windows machine (msysgit). The strange thing, if I enable loglevel = DEBUG in gitosis.conf, I see something like this when doing any actions with gitosis server: D:\Kaze\source\test-project>git pull origin master DEBUG:gitosis.serve.main:Got command "git-upload-pack 'test_project.git'" DEBUG:gitosis.access.haveAccess:Access check for '[email protected]' as 'writable' on 'test_project.git'... DEBUG:gitosis.access.haveAccess:Stripping .git suffix from 'test_project.git', new value 'test_project' DEBUG:gitosis.group.getMembership:found '[email protected]' in 'test' DEBUG:gitosis.access.haveAccess:Access ok for '[email protected]' as 'writable' on 'test_project' DEBUG:gitosis.access.haveAccess:Using prefix 'repositories' for 'test_project' DEBUG:gitosis.serve.main:Serving git-upload-pack 'repositories/test_project.git' From 192.168.175.128:test_project * branch master -> FETCH_HEAD Already up-to-date. Question is: why am I *[email protected]? This email is in global user.email config variable, too. Yesterday, when the gitosis was installed, it seen me as kaze@KAZE, this is the name under which I was added to gitosis-admin group (and it worked). But today git (or gitosis) started to see me as [email protected]. This is true for all repositories I push or clone. I had to add this address to gitosis.conf directly on server to be able to edit configs again (it worked). There is 2 public keys in keydir: [email protected] and [email protected], their content is identical and they have kaze@KAZE at end. Origin URL looks like git@lennyserver:test_project. Now, the question is - why Git (or gitosis) suddenly decided to call me by email instead of name@machinename? I've changed a couple things trying to set up Gitosis (updated git on server to 1.6.0 for example), but maybe I broke something in my local git installation?

    Read the article

  • fatal: pathspec for removing folder in git

    - by Elnaz Shahmehr
    I want to remove a folder from a Git repository but I get an error. I tried several commands, and here you can see my errors: selnaz:iOS Lnaz$ ls iOS-Tidinfo mockup readme.txt selnaz:iOS Lnaz$ git rm -r -- iOS-Tidinfo/ fatal: pathspec 'tidinfo/iOS/iOS-Tidinfo/' did not match any files selnaz:iOS Lnaz$ git rm -r iOS-Tidinfo/ fatal: pathspec 'tidinfo/iOS/iOS-Tidinfo/' did not match any files selnaz:iOS Lnaz$ git rm -r iOS-Tidinfo/ fatal: pathspec 'tidinfo/iOS/iOS-Tidinfo/' did not match any files selnaz:iOS Lnaz$ git rm -r tidinfo/iOS/iOS-Tidinfo/ fatal: pathspec 'tidinfo/iOS/tidinfo/iOS/iOS-Tidinfo/' did not match any files

    Read the article

  • Git on DreamHost still balking on big files even after I compiled with NO_MMAP=1

    - by fuzzy lollipop
    I compiled Git 1.7.0.3 on DreamHost with the NO_MMAP=1 option, I also supplied that option when I did the "make NO_MMAP=1 install". I have my paths set up correctly, which git reports my ~/bin dir which is correct, git --version returns the correct version. But when I try to do a "git push origin master" with "big" files ~150MB it always fails. Does anyone have an suggestions on how to get DreamHost to accept this "big" files from a git push?

    Read the article

  • Git not work on my Mac OS X .

    - by koko
    I downloaded and install Git from http://code.google.com/p/git-osx-installer/ . After installing I can't use my Git installation. thomas-macbook-:test zozo$ git init -bash: git: command not found What is wrong? Many thanks in advance.

    Read the article

  • Git not work on my MacOSX .

    - by koko
    I downloaded and install Git from http://code.google.com/p/git-osx-installer/ . After installing I can't use my Git installation. thomas-macbook-:test zozo$ git init -bash: git: command not found What is wrong? Many thanks in advance.

    Read the article

  • Uninstall Git completely on Ubuntu?

    - by Millisami
    I installed Git on Ubuntu Lucid (latest) manually as following. cd ~/tmp wget http://kernel.org/pub/software/scm/git/git-1.7.0.6.tar.gz tar -xzvf git-1.7.0.6.tar.gz cd git-1.7.0.6.tar.gz ./configure sudo make sudo make install Now, how can I completely uninstall it?

    Read the article

  • Git multiple config files

    - by tig
    As it is told in git-config manual, I have three files to specify git repo configuration: $GIT_DIR/config, ~/.gitconfig and $(prefix)/etc/gitconfig. Is it possible to make git read another config file (for example .gitconfig in root of the repo) after reading $GIT_DIR/config? I want it as I created git command aliases in .git/config and I would prefer to store them in repo, so I can restore them when cloning and for example.

    Read the article

  • egg git interface for emacs, commit message empty

    - by Gauthier
    I'm using egg (emacs got git) as git interface in emacs. Whenever I try to achieve a commit --amend, I receive a "GIT-COMMIT-AMEND> Aborting commit due to empty commit message". This is what i do: C-u C-x v c Then the commit buffer appears, with the message of my previous commit. Then upon C-c C-c I get the message stated above: empty commit message. I think I've had this behaviour with regular commits (as in not amend) before, but can't remember or find how I solved it. I tried editing the message (adding a space somewhere). No work. I tried saving the buffer before committing, that wouldn't work either (since C-c C-c is not active in another buffer than the commit buffer). Any clue?

    Read the article

  • gitosis and git clone problem

    - by Alexey Poimtsev
    Hi, I have installed gitosis, but i have strange thing when i'm working with repos. In config i have [gitosis] [group gitosis-admin] writable = gitosis-admin members = me@server me@laptop [group prj1] writable = prj1 members = me@laptop and in /home/git/repositories i have created directory prj1.git with empty git repo. i can work with gitosis-admin from server and laptop without problems, but when i'm trying to git clone prj1 on my laptop i see $ git clone git@server:prj1.git Initialized empty Git repository in /Users/alec/temp/prj1/.git/ fatal: no matching remote head ok, lets push prj1 from laptop to server: $ git push origin master:refs/heads/master ERROR:gitosis.serve.main:Repository read access denied fatal: The remote end hung up unexpectedly Whats wrong?

    Read the article

  • Using MobileMe idisk as a git repository

    - by Ben Guest
    I am trying to use git and MobileMe as a version control system for a personal project I am working across several computers. So far i have done the following. Created and empty bare repository on my local computer $ mkdir myproject.git $ cd myproject.git $ git init --bare $ git update-server-info I then copied the myproject.git directory to the mobile me disk, and sync my computer with mobile me. I then switched to the directory where my project was on my local machine, set the remote origin and try to push the local repository to mobile me $ cd myproject $ git remote add origin https://<username>@idisk.me.com/<username>/myproject.git/ $ git push --all Im am then asked for my password twice. The first time is the mobile me password, any other password gets an error. After entering the second password, and believe me i've tried everything, terminal just hangs. So what am I doing wrong? (Besides trying to use mobileme as a git repository) Thanks, Ben.

    Read the article

  • git- how to troubleshoot "cannot find command"

    - by Frank Schwieterman
    I need help getting git extensions to run with msysgit. I have had bad luck with extensions git-tfs and git-fetchall, in both cases it is the same problem. The addon will require a file to be placed where git can find it (git-tfs.exe and git-fetchall.sh). I understand this to mean the files need to be in a directory that is in the 'PATH' environment variable. In both cases I get stuck at this point: $ git-diffall bash: git-diffall: command not found When I run echo %PATH% from a regular command shell, it shows my path variable includes the directories where git-diffall and git-tfs are. How can I debug this, or am I missing something? Is there a way within msysgit to verify the command search path is what I expect?

    Read the article

< Previous Page | 10 11 12 13 14 15 16 17 18 19 20 21  | Next Page >