Make thin-pack generation subproject aware.

When a thin pack wants to send a tree object at "sub/dir", and
the commit that is common between the sender and the receiver
that is used as the base object has a subproject at that path,
we should not try to use the data at "sub/dir" of the base tree
as a tree object.  It is not a tree to begin with, and more
importantly, the commit object there does not have to even
exist.

Signed-off-by: Junio C Hamano <gitster@pobox.com>

1 files changed, 52 insertions, 0 deletions

diff --git a/t/t3050-subprojects-fetch.sh b/t/t3050-subprojects-fetch.sh
new file mode 100755
index 000000000..34f26a8d9
--- /dev/null
+++ b/t/t3050-subprojects-fetch.sh
@@ -0,0 +1,52 @@
+#!/bin/sh
+
+test_description='fetching and pushing project with subproject'
+
+. ./test-lib.sh
+
+test_expect_success setup '
+	test_tick &&
+	mkdir -p sub && (
+		cd sub &&
+		git init &&
+		>subfile &&
+		git add subfile
+		git commit -m "subproject commit #1"
+	) &&
+	>mainfile
+	git add sub mainfile &&
+	test_tick &&
+	git commit -m "superproject commit #1"
+'
+
+test_expect_success clone '
+	git clone file://`pwd`/.git cloned &&
+	(git rev-parse HEAD; git ls-files -s) >expected &&
+	(
+		cd cloned &&
+		(git rev-parse HEAD; git ls-files -s) >../actual
+	) &&
+	diff -u expected actual
+'
+
+test_expect_success advance '
+	echo more >mainfile &&
+	git update-index --force-remove sub &&
+	mv sub/.git sub/.git-disabled &&
+	git add sub/subfile mainfile &&
+	mv sub/.git-disabled sub/.git &&
+	test_tick &&
+	git commit -m "superproject commit #2"
+'
+
+test_expect_success fetch '
+	(git rev-parse HEAD; git ls-files -s) >expected &&
+	(
+		cd cloned &&
+		git pull &&
+		(git rev-parse HEAD; git ls-files -s) >../actual
+	) &&
+	diff -u expected actual
+'
+
+test_done