The problem is that the semantics of calling some of the hooks is not as
simple as just calling them all in a row. How do you split up the input
going to the hooks? How do you combine the output coming from the hooks?
What is the resulting exit code? If one hook fails, do we indicate
failure? Or if one hook succeeds, do we indicate success?
So before any such code could go into git proper, there would have to be
agreement on how those issues are resolved. In the meantime, it probably
makes more sense to implement a "master" post-commit hook that uses the
semantics that you find useful.
-Peff
--