Re: [abc] test run results (with re-weaving)

Ondrej Lhotak wrote:

>On Thu, Sep 23, 2004 at 06:26:48PM +0200, Aske Simon Christensen wrote:
>
>
>>Quoting Ondrej LHOTAK <olhotak@sable.mcgill.ca>:
>>
>>
>>
>>>Pass: 609 (down 263)
>>>Fail: 282 (up 263)
>>>Skipped: 103 (no change)
>>>
>>>
>>Shouldn't these reweaving run results be relative to the last reweaving run,
>>rather than the normal run?
>>
>>
>
>When I was in Oxford, Ganesh and I agreed that eventually, the reweaving
>run would become the normal run, so that everyone would be discouraged
>from breaking re-weaving in their code (or encouraged to fix their code
>in cases where it breaks re-weaving). I guess that hasn't happened yet.
>
>I'm not sure how the script knows what the previous "official" test run
>result was, and I don't want to mess up the functioning of the
>"official" test runs.
>
>
It takes them from the passed_current.xml, failed_current.xml and
skipped_current.xml, which I think are updated in CVS on the nightly run
after they have been compared to the old version.

- P

>Ondrej
>
>
>
>>-Aske
>>
>>
>>
>>
>
>
>
>
Received on Thu Sep 23 18:09:18 2004