Reputation: 413
i'd like to just checkout the files without the .git files and the whole repository. It's because i'd like to manage a website (php & html) with git and i'm looking for an easy way to update the files in the htdocs folder from the repository, without having the repository public. (now it's in the home-dir and is accessed via ssh, but i have to put the new files to htdocs manually.
Upvotes: 41
Views: 33356
Reputation: 17524
Git is much easier than Subversion for this, as the whole repository is held in a single directory. You only need to delete the hidden ".git" folder in the root of the project to create a production-ready copy of your site.
In Linux/OSX this would be:
mkdir dist && cd dist
git checkout --depth=1 http://path/to/your/project.git
rm -rf .git
Upvotes: 12
Reputation: 3335
You'd probably benefit from a bare repo plus post-receive hook.
This method is useful if you desire to push changes to your server but don't want the htdocs themselves under version control.
We'll make a folder for the bare git repo. I like to put it in the same parent folder as htdocs:
mkdir barerepo.git
cd barerepo.git
git --bare init
Then create a post-receive hook file, make it executable:
touch hooks/post-receive
chmod ug+x hooks/post-receive
Edit post-receive in your favorite text editor:
GIT_WORK_TREE=/full/path/to/htdocs git checkout -f
# optional stuff:
cd down/to/some/directory
[do some stuff]
Now every time you push to this bare repo it will checkout the working tree to htdocs. But htdocs itself is not under version control; running git status
in htdocs will give the error fatal: Not a git repository (or any parent up to mount point /data)
. It's just plain files.
Note that you must always push from master for this to work otherwise git checkout
won't know what branch to checkout.
Upvotes: 1
Reputation: 609
git clone --depth 1
does exactly what you want. I have a git repository with 41,000 commits and this is vastly faster. See this answer or this more general response for a more detailed explanation.
If you want to specify the branch to use (instead of the default "master" branch) use this syntax (requires Git 1.9+):
git clone -b <remoteBranch> --single-branch --depth 1 ssh://[email protected]:serverport/PathToProject <FolderName>
Upvotes: 13
Reputation: 3664
Expanding on seanhodges answer, you could just rsync the project's files to wherever you need to put them and use --exclude to avoid the .git directory.
rsync -avz --exclude=.git /source/directory/ [email protected]:/target-directory
I didn't test that command so feel free to edit it, but you get the jist.
A useful option is to add --dry-run to the end of the command, that will show you what files will be sent without actually making any changes.
Upvotes: 2
Reputation: 323932
The git-archive manpage contains the following example:
git archive --format=tar --prefix=junk/ HEAD | (cd /var/tmp/ && tar xf -)
Create a tar archive that contains the contents of the latest commit on the current branch, and extract it in the '/var/tmp/junk' directory.
Or you can use low level git-checkout-index, which manpage contains the following example:
Using git checkout-index to "export an entire tree"
The prefix ability basically makes it trivial to use '
git checkout-index
' as an "export as tree" function. Just read the desired tree into the index, and do$ git checkout-index --prefix=git-export-dir/ -a
git checkout-index
will "export" the index into the specified directory.The final "/" is important. The exported name is literally just prefixed with the specified string.
Or you can try to use --work-tree
option to git wrapper, or GIT_WORK_TREE environment variable, e.g. by using "git --work-tree=/somwehere/else checkout -- .
".
Upvotes: 37
Reputation: 602635
You could create a tar archive in a git working directory and copy the tar to the server:
git archive -o foo.tar HEAD
scp foo.tar server:
Upvotes: 6