Re: [PATCH] repack: allow simultaneous packing and pruning

From

Junio C Hamano <junkio@cox.net>

Date

Oct 10, 2006, 20:24 UTC

Message-ID

<7vk637zzpk.fsf@assigned-by-dhcp.cox.net>

In-Reply-To

<Pine.LNX.4.64.0610100800490.3952@g5.osdl.org>

Linus Torvalds <torvalds@osdl.org> writes:

> On Tue, 10 Oct 2006, Sam Vilain wrote:
>>
>> If using git-repack -a, unreferenced objects are kept behind in the
>> pack.  This might be the best default, but there are no good ways
>> to clean up the packfiles if a lot of rebasing is happening, or
>> branches have been deleted.
>
> Don't do this.
>
> I understand why you want to do it, but the fact is, it's dangerous.

@@ -32,6 +33,11 @@ case ",$all_into_one," in
 ,,)
 	rev_list='--unpacked'
 	pack_objects='--incremental'
+	if [ -n "$prune" ]
+	then
+	    # prune junk first
+	    git-prune
+	fi
 	;;
 ,t,)
 	rev_list=

This shouldn't make any difference if the repository is
quiescent (and is dangerous if it isn't).  pack-objects will
not get fed things that are not reachable.

@@ -40,8 +46,14 @@ case ",$all_into_one," in
 	# Redundancy check in all-into-one case is trivial.
 	existing=`cd "$PACKDIR" && \
 	    find . -type f \( -name '*.pack' -o -name '*.idx' \) -print`
+
+	if [ -n "$prune" ]
+	then
+	    rev_list=`cd "$GIT_DIR" && find refs -type f -print`
+	fi
 	;;
 esac
+

We give --all to rev-list so this should not have any effect
either; other than that the code introduced by this hunk is
broken with packed-refs.

Isn't "repack -a -d" what Sam wants?