New wintest executable is there (but there are some issues due to introduction of skip)

[Paul Vriens]

Hi,

as of today there's a new winetest executable available at the usual place:

http://www.astro.gla.ac.uk/users/paulm/WRT/CrossBuilt/

There is however an issue with the generated reports. It looks like this is 
introduced when the 'skip' thingy was added to the test infrastructure. I'm 
currently going through the perl scripts but have no fix yet. Most likely we 
have to patch the 'dissect' script at line 158:

      } elsif (s/^($test: (\d+) tests executed, (\d+) marked as todo, (\d+) 
failures?\.)\r?$/$1/) {

The previous lines were like:

advapi32:crypt start dlls/advapi32/tests/crypt.c 1.31
crypt: 69 tests executed, 0 marked as todo, 0 failures.
advapi32:crypt done (0)

now it's:

advapi32:crypt start dlls/advapi32/tests/crypt.c 1.31
crypt: 69 tests executed (0 marked as todo, 0 failures), 0 skipped.
advapi32:crypt done (0)

Anybody?

Cheers,

Paul.

[Paul Vriens]

Paul Vriens wrote:
> Hi,
>
> as of today there's a new winetest executable available at the usual place:
>
> http://www.astro.gla.ac.uk/users/paulm/WRT/CrossBuilt/
>
> There is however an issue with the generated reports. It looks like this 
> is introduced when the 'skip' thingy was added to the test 
> infrastructure. I'm currently going through the perl scripts but have no 
> fix yet. Most likely we have to patch the 'dissect' script at line 158:
>
> } elsif (s/^($test: (\d+) tests executed, (\d+) marked as todo, 
> (\d+) failures?\.)\r?$/$1/) {
>
> The previous lines were like:
>
> advapi32:crypt start dlls/advapi32/tests/crypt.c 1.31
> crypt: 69 tests executed, 0 marked as todo, 0 failures.
> advapi32:crypt done (0)
>
> now it's:
>
> advapi32:crypt start dlls/advapi32/tests/crypt.c 1.31
> crypt: 69 tests executed (0 marked as todo, 0 failures), 0 skipped.
> advapi32:crypt done (0)
>
> Anybody?
>
> Cheers,
>
> Paul.

Hi,

the attached patch should fix the 'dissect' script and retains backwards
compatibility (if needed).

It doesn't make use of the skip-count as that will mean fixing up a lot more. If
somebody could have a look at the attached patch?

Cheers,

Paul.

[Felix Nawothnig]

And while we're at it, there seems to be some misconfiguration of the 
httpd (for quite some time), as the following URL generates an 500 
Internal Server Error:

http://test.winehq.org/data/200702241000/2003_W2K3-SE-IE7/ntdll:error.txt

New wintest executable is there (but there are some issues due to introduction of skip)

[Wagner Ferenc]

Paul Vriens <paul.vriens.wine(a)gmail.com> writes:
> It doesn't make use of the skip-count as that will mean fixing up a
> lot more. If somebody could have a look at the attached patch?

Looks good.  The already submitted report files could be moved from
the data dir back into the queue on the web server as
http://cvs.winehq.org/cvsweb/tools/winetest/README tries to explain.
(The original directories can be removed afterwards.)
-- 
Regards,
Feri.