In a non-default Git worktree, the .git directory is replaced
by a file containing the path to the real .git directory. This
breaks the version detection logic because it expects .git to
be a directory.
Passing the root of the source tree via "git -C", letting Git
figure out if we're in a repo or not, solves this problem.
Change-Id: I595f1a694258cad490b1a4964f8ae9d51ae76de1
Since we will be transitioning to git, stop considering svn the primary source.
If HEAD is a git-svn revision, return the svn revision number as before
(this means that if you check out old versions they keep their previously
canonical version numbers) but if it's not, then just print the short SHA1.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@31480 a1c6a512-1295-4272-9138-f99709370657
Only I ever used this that I know of, and now we are migrating to git it's
not needed (and won't work correctly, since we won't be svn-derived).
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@31479 a1c6a512-1295-4272-9138-f99709370657
commit ID in the version string, much like is done for repos tracked
with bzr. FS#11297
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@29827 a1c6a512-1295-4272-9138-f99709370657
"git log .. -1" only shows one commit (the last one), so extract the last line (git-svn-id:..) with tail -n 1, instead of the first line (which is the committer's message in case of multi lines body)
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@21188 a1c6a512-1295-4272-9138-f99709370657
