Seth Falcon <seth@xxxxxxxxxxxxxxx> wrote:
> What I would like is a way for the script to determine the appropriate
> tracking branch. So that the usage would look like:
>
> git mark-reviewed someone@xxxxxxxxxxxxxxx
>
> and it would figure out whether it should do trunk.. or release-1.3..,
> etc. Can anyone point me in the right direction?
Something like this, but its uh, ugly due to the use of a network
connection:
branch=$(git symbolic-ref HEAD)
branch=${branch##refs/heads/}
remote=$(git config branch.$branch.remote)
merge=$(git config branch.$branch.merge)
rb=$(git ls-remote $remote $merge | awk '{print $1}')
Then use a filter-branch on "$rb..$branch" as the range.
You may be able to just assume that the remote name is the
refs/remotes prefix and instead do:
branch=$(git symbolic-ref HEAD)
branch=${branch##refs/heads/}
remote=$(git config branch.$branch.remote)
merge=$(git config branch.$branch.merge)
merge=${merge##refs/heads/}
rb=refs/remotes/$remote/$merge
but that's an assumption, and we all know what happens when you
assume things.
Technically you need to look at remote.$remote.fetch lines
in .git/config to figure out the rewriting rules from what
branch.$branch.merge contains to what refs/remotes/* might be,
and doing that is not quite as trivial as either using ls-remote or
assuming your users have the standard layout created by git-clone
and git-remote.
--
Shawn.
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
