On Tue, Nov 24, 2009 at 09:13:39AM +0800, bill lam wrote:
> On Mon, 23 Nov 2009, Nicolas Pitre wrote:
> > Then, during the pack-objects process, there are 3 phases: counting
> > objects, compressing objects, and writing objects. However in the fetch
>
> during git-gc it shows yet another progress
>
> Removing duplicate objects: 100% (256/256), done.
Thanks, this doesn't seem to have been guarded at all (but since it is
on a 2-second delay, you have to have quite a lot of loose objects or a
slow disk to trigger it).
We should apply the patch below to keep things consistent.
I also checked every other call to start_progress; everything else seems
to be guarded. Most of them were easy to trace to an isatty check,
though the one in unpack-trees is influenced by o->verbose_update. That
in turn usually corresponds to a quiet option, though merge does seem to
use it unconditionally. Maybe that should be tweaked, too?
-- >8 --
Subject: [PATCH] prune-packed: only show progress when stderr is a tty
This matches the behavior of other git programs, and helps
keep cruft out of things like cron job output.
Signed-off-by: Jeff King <peff@xxxxxxxx>
---
builtin-prune-packed.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/builtin-prune-packed.c b/builtin-prune-packed.c
index be99eb0..f9463de 100644
--- a/builtin-prune-packed.c
+++ b/builtin-prune-packed.c
@@ -71,7 +71,7 @@ void prune_packed_objects(int opts)
int cmd_prune_packed(int argc, const char **argv, const char *prefix)
{
- int opts = VERBOSE;
+ int opts = isatty(2) ? VERBOSE : 0;
const struct option prune_packed_options[] = {
OPT_BIT('n', "dry-run", &opts, "dry run", DRY_RUN),
OPT_NEGBIT('q', "quiet", &opts, "be quiet", VERBOSE),
--
1.6.6.rc0.249.g9b4cf.dirty
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html