Andreas Ericsson [off-list ref] writes:
Junio C Hamano wrote:
quoted
(this is against 'maint', as I was in the middle of
something else that is based on 'maint' when I started reading
this thread).
That's probably not a bad idea, as it really is a bug, and one
that can cause data-loss at that.
I agree that "git clean" needs to be fixed in 'maint', but the
thing is, setup_standard_excludes() approach would not apply to
anything in 'maint', 'master', nor 'next', and it is more of the
longer term thing to go together with the git-clean in C.
The scripted version needs to be fixed independently in a way
with lessor impact, like the patch Shunichi just posted.