# [PATCH] git-commit.sh: Fix usage checks regarding paths given when they do not make sense

2 messages from 2007-11-05 to 2007-11-05. Participants: Björn Steinbrink, Junio C Hamano.
Thread: https://gitlist.dev/t/10673

## Björn Steinbrink, 2007-11-05 19:36

Subject: [PATCH] git-commit.sh: Fix usage checks regarding paths given when they do not make sense
Message-ID: <1194291393-1067-1-git-send-email-B.Steinbrink@gmx.de>
URL: https://gitlist.dev/e/1194291393-1067-1-git-send-email-B.Steinbrink%40gmx.de

```
The checks that looked for paths given to git-commit in addition to
--all or --interactive expected only 3 values, while the case statement
actually provides 4, so the check was never triggered.

The bug was introduced in 6cbf07efc5702351897dee4742525c9b9f7828ac when
the case statement was extended to handle --interactive.

Signed-off-by: Björn Steinbrink <B.Steinbrink@gmx.de>
---
 git-commit.sh |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/git-commit.sh b/git-commit.sh
index fcb8443..d4471ff 100755
--- a/git-commit.sh
+++ b/git-commit.sh
@@ -282,9 +282,9 @@ unset only
 case "$all,$interactive,$also,$#" in
 *t,*t,*)
 	die "Cannot use -a, --interactive or -i at the same time." ;;
-t,,[1-9]*)
+t,,,[1-9]*)
 	die "Paths with -a does not make sense." ;;
-,t,[1-9]*)
+,t,,[1-9]*)
 	die "Paths with --interactive does not make sense." ;;
 ,,t,0)
 	die "No paths with -i does not make sense." ;;
-- 
1.5.3.5.561.g140d-dirty

```

## Junio C Hamano, 2007-11-05 21:39

Subject: Re: [PATCH] git-commit.sh: Fix usage checks regarding paths given when they do not make sense
Message-ID: <7vlk9cnzhb.fsf@gitster.siamese.dyndns.org>
URL: https://gitlist.dev/e/7vlk9cnzhb.fsf%40gitster.siamese.dyndns.org
In-Reply-To: <1194291393-1067-1-git-send-email-B.Steinbrink@gmx.de>

```
Björn Steinbrink <B.Steinbrink@gmx.de> writes:

> The checks that looked for paths given to git-commit in addition to
> --all or --interactive expected only 3 values, while the case statement
> actually provides 4, so the check was never triggered.
>
> The bug was introduced in 6cbf07efc5702351897dee4742525c9b9f7828ac when
> the case statement was extended to handle --interactive.

Gaah, and thanks.

We really should have "negative" tests to catch this kind of
breakage in our testsuite.  People when inventing new features
are eager to write tests to show off the new stuff works, but
not many people are careful enough to add tests to demonstrate
the commands properly catch user errors such as borked command
line options.

```