2010/3/22 Daniel P. Berrange <berrange(a)redhat.com>:
On Sat, Mar 20, 2010 at 06:14:44PM +0100, Matthias Bolte wrote:
> ---
> tests/test-lib.sh | 10 ++++++----
> tests/testutils.c | 2 +-
> 2 files changed, 7 insertions(+), 5 deletions(-)
>
> diff --git a/tests/test-lib.sh b/tests/test-lib.sh
> index 43265f3..57fd438 100644
> --- a/tests/test-lib.sh
> +++ b/tests/test-lib.sh
> @@ -55,10 +55,12 @@ test_final()
>
> if test "$verbose" = "0" ; then
> mod=`eval "expr \( $counter + 1 \) % 40"`
> - for i in `seq $mod 40`
> - do
> - echo -n " "
> - done
> + if test "$mod" != "0" -a "$mod" != "1" ;
then
> + for i in `seq $mod 40`
> + do
> + echo -n " "
> + done
> + fi
> if test "$status" = "0" ; then
> printf " %-3d OK\n" $counter
> else
> diff --git a/tests/testutils.c b/tests/testutils.c
> index 8764673..99bd9df 100644
> --- a/tests/testutils.c
> +++ b/tests/testutils.c
> @@ -531,7 +531,7 @@ cleanup:
> virResetLastError();
> if (!virTestGetVerbose()) {
> int i;
> - for (i = (testCounter % 40) ; i < 40 ; i++)
> + for (i = (testCounter % 40) ; i > 0 && i < 40 ; i++)
> fprintf(stderr, " ");
> fprintf(stderr, " %-3d %s\n", testCounter, ret == 0 ?
"OK" : "FAIL");
> }
I'm not entirely understanding what this is changing ?
This is what I currently see:
$ ./qemuxml2argvtest
TEST: qemuxml2argvtest
........................................ 40
........................................ 80
....... 87 OK
And this change doesn't appear to alter that - what am I missing ?
Regards,
Daniel
This change only affects the output of tests that have an exact
multiple of 40 test cases. For example the domainschematest currently:
TEST: domainschematest
........................................ 40
........................................ 80
........................................ 120
........................................ 160
........................................
200 OK
PASS: domainschematest
It outputs additional 40 spaces on the last line.
The domainschematest output is fixed by the change in test-lib.sh. The
change in testutils.c fixes this for tests written in C. Currently no
C test has an exact multiple of 40 test cases, but I checked it and
the same problem exists there.
This patch stops that in both cases.
Matthias