[PATCH] Fix "git submodule status" output when "git describe" is not possible

[PATCH] Fix "git submodule status" output when "git describe" is not possible

[Alberto Bertogli]

When a submodule cannot be described, doing a "git submodule status" on
the supermodule produces unnecesary and scary output, for example:

    fatal: cannot describe 'cac575f1bccf0c6942d11962cf4d44ba974d3e8e'
     cac575f1bccf0c6942d11962cf4d44ba974d3e8e megaco/megaco_tipc

Where it should just show:

     cac575f1bccf0c6942d11962cf4d44ba974d3e8e megaco/megaco_tipc

The culprit (after manual inspection of git log) seems to be commit
a2cf9f4.

This patch fixes this by making git submodule aware of that behaviour
change.

Signed-off-by: Alberto Bertogli <albertito@gmail.com>
---
 git-submodule.sh |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/git-submodule.sh b/git-submodule.sh
index a6aaf40..5e8b84f 100755
--- a/git-submodule.sh
+++ b/git-submodule.sh
@@ -315,7 +315,7 @@ set_name_rev () {
 		cd "$1" && {
 			git describe "$2" 2>/dev/null ||
 			git describe --tags "$2" 2>/dev/null ||
-			git describe --contains --tags "$2"
+			git describe --contains --tags "$2" 2>/dev/null
 		}
 	) )
 	test -z "$revname" || revname=" ($revname)"
-- 
1.5.4.1.1279.g0d205f-dirty

Re: [PATCH] Fix "git submodule status" output when "git describe" is not possible

[Junio C Hamano]

Alberto Bertogli <albertito@gmail.com> writes:

> When a submodule cannot be described, doing a "git submodule status" on
> the supermodule produces unnecesary and scary output,...

I had an impression that describe got a new feature for a case
like this recently.

Re: [PATCH] Fix "git submodule status" output when "git describe" is not possible

[Alberto Bertogli]

On Tue, Feb 12, 2008 at 05:01:52PM -0800, Junio C Hamano wrote:
> Alberto Bertogli <albertito@gmail.com> writes:
> 
> > When a submodule cannot be described, doing a "git submodule status" on
> > the supermodule produces unnecesary and scary output,...
> 
> I had an impression that describe got a new feature for a case
> like this recently.

I'm sorry but I've read the docs and tried with a few options and I have
no idea what that one may be.

The bug is still present on 'next' as of an hour ago.

Thanks,
		Alberto