[NOTE]
And those "too deep" descriptions are often marked as Note.
+[NOTE]
+If you are already familiar with another version control system,
+like CVS, you may want to take a look at
+link:everyday.html[Everyday GIT in 20 commands or so] first
+before reading this.
+
Creating a git repository
-------------------------
HTTP(S)::
`http://remote.machine/path/to/repo.git/`
+
-HTTP and HTTPS transport are used only for downloading. They
-first obtain the topmost commit object name from the remote site
-by looking at `repo.git/info/refs` file, tries to obtain the
+Downloader from http and https URL
+first obtains the topmost commit object name from the remote site
+by looking at the specified refname under `repo.git/refs/` directory,
+and then tries to obtain the
commit object by downloading from `repo.git/objects/xx/xxx\...`
using the object name of that commit object. Then it reads the
commit object to find out its parent commits and the associate
The 'commit walkers' are sometimes also called 'dumb
transports', because they do not require any git aware smart
server like git Native transport does. Any stock HTTP server
-would suffice.
+that does not even support directory index would suffice. But
+you must prepare your repository with `git-update-server-info`
+to help dumb transport downloaders.
+
There are (confusingly enough) `git-ssh-fetch` and `git-ssh-upload`
programs, which are 'commit walkers'; they outlived their
2. Prepare a public repository accessible to others.
+
If other people are pulling from your repository over dumb
-transport protocols, you need to keep this repository 'dumb
-transport friendly'. After `git init-db`,
+transport protocols (HTTP), you need to keep this repository
+'dumb transport friendly'. After `git init-db`,
`$GIT_DIR/hooks/post-update` copied from the standard templates
would contain a call to `git-update-server-info` but the
`post-update` hook itself is disabled by default -- enable it
-with `chmod +x post-update`.
+with `chmod +x post-update`. This makes sure `git-update-server-info`
+keeps the necessary files up-to-date.
3. Push into the public repository from your primary
repository.
For this, set up a public repository on a machine that is
reachable via SSH by people with "commit privileges". Put the
committers in the same user group and make the repository
-writable by that group.
+writable by that group. Make sure their umasks are set up to
+allow group members to write into directories other members
+have created.
You, as an individual committer, then:
and put them in the same group. Make sure that the repository
shared among these developers is writable by that group.
+. Initializing the shared repository with `git-init-db --shared`
+helps somewhat.
+
+. Run the following in the shared repository:
++
+------------
+$ chgrp -R $group repo.git
+$ find repo.git -type d -print | xargs chmod ug+rwx,g+s
+$ GIT_DIR=repo.git git repo-config core.sharedrepository true
+------------
+
+The above measures make sure that directories lazily created in
+`$GIT_DIR` are writable by group members. You, as the
+repository administrator, are still responsible to make sure
+your developers belong to that shared repository group and set
+their umask to a value no stricter than 027 (i.e. at least allow
+reading and searching by group members).
+
You can implement finer grained branch policies using update
hooks. There is a document ("control access to branches") in
Documentation/howto by Carl Baldwin and JC outlining how to (1)
projects / git.git / blobdiff