From: Junio C Hamano <junkio@cox.net>
Date: Sat, 17 Sep 2005 08:50:37 GMT
Subject: Help cloning over http.
Message-ID: <7vvf10axia.fsf_-_@assigned-by-dhcp.cox.net>
In-Reply-To: <7vacicccxl.fsf@assigned-by-dhcp.cox.net>

Even when repository is packed, the new "we trust only the refs
we already have" code wasted too much time verifying what are
already in packs, walking down to the root commits.  Since we
slurp all the necessary packs upfront and run verify pack on
them, temporarily register their edge commits as "known to be
good" before invoking git-http-fetch.

I am not proud of this workaround -- this would probably break
in a pathological case where the remote repository has packs
that depend on ancient unpacked objects, but people should not
be making such packs anyway (git repack would not create such
packs).

Signed-off-by: Junio C Hamano <junkio@cox.net>

---

 git-clone.sh |   19 +++++++++++++++++++
 1 files changed, 19 insertions(+), 0 deletions(-)

34414e4170ab8d1d2a9ef582256530dc3b42d3e2
diff --git a/git-clone.sh b/git-clone.sh
--- a/git-clone.sh
+++ b/git-clone.sh
@@ -50,11 +50,30 @@ Perhaps git-update-server-info needs to 
 		git-verify-pack ".git/objects/pack/$idx" || exit 1
 	done <"$clone_tmp/packs"
 
+	while read type idx sha1 commit
+	do
+		case "$type,$commit" in
+		T,commit) ;;
+		*) continue ;;
+		esac &&
+		echo "$sha1" >.git/refs/tags/clone-tmp-$sha1
+	done <"$clone_tmp/packs"
+
 	while read sha1 refname
 	do
 		name=`expr "$refname" : 'refs/\(.*\)'` &&
 		git-http-fetch -v -a -w "$name" "$name" "$1/" || exit 1
 	done <"$clone_tmp/refs"
+
+	while read type idx sha1 commit
+	do
+		case "$type,$commit" in
+		T,commit) ;;
+		*) continue ;;
+		esac &&
+		rm -f .git/refs/tags/clone-tmp-$sha1
+	done <"$clone_tmp/packs"
+
 	rm -fr "$clone_tmp"
 }