Re: [PATCH v4 00/10] nd/clone-detached

[Nguyen Thai Ngoc Duy <pclouds@gmail.com>]

On Fri, Jan 13, 2012 at 10:53:11PM -0800, Junio C Hamano wrote:
> Nguyen Thai Ngoc Duy <pclouds@gmail.com> writes:
> 
> > 2012/1/14 Junio C Hamano <gitster@pobox.com>:
> >> Thanks, replaced (and updated comment strings read much better).
> >>
> >> There were some conlicts I had to resolve while merging this to 'pu'.
> >> I would appreciate it if you can eyeball it to make sure I didn't make
> >> silly mistakes there.
> >
> > Right, the conflict with nd/clone-single-branch. I kept thinking there
> > would not be conflict because clone-single-branch's big change was in
> > wanted_peer_refs() and missed write_followtags() call. The merge looks
> > good.
> 
> Hmm, 'pu' seems to fail its selftest with this merge present, though.

The commit "refuse to clone if --branch points to bogus ref" from this
series changes clone's behavior that t5500.31, which is added in
nd/clone-single-branch, relies on. This makes "make test" pass for me
on pu:

-- 8< --
diff --git a/t/t5500-fetch-pack.sh b/t/t5500-fetch-pack.sh
index 7e85c71..c4e675f 100755
--- a/t/t5500-fetch-pack.sh
+++ b/t/t5500-fetch-pack.sh
@@ -283,10 +283,7 @@ test_expect_success 'clone shallow object count' '
 '
 
 test_expect_success 'clone shallow with nonexistent --branch' '
-	git clone --depth 1 --branch Z "file://$(pwd)/." shallow4 &&
-	GIT_DIR=shallow4/.git git rev-parse HEAD >actual &&
-	git rev-parse HEAD >expected &&
-	test_cmp expected actual
+	test_must_fail git clone --depth 1 --branch Z "file://$(pwd)/." shallow4
 '
 
 test_expect_success 'clone shallow with detached HEAD' '
-- 8< --

I'd rather remove the test, but removing something in a merge does not
sound wise.

Another cleaner approach is to combine the two clone series into
one. If you want to go this way, pick one as base, I'll rebase the
other on top and resend.
-- 
Duy