Bug 38840 - Adding coverage analysis to unit tests

[Stephan Bergmann]

On 08/24/2012 12:41 PM, John Smith wrote:
>>> I did have one test fail. See the 'make check failure' thread:
>>> http://nabble.documentfoundation.org/make-check-failure-td4002979.html
>>> Could that explain the omission ?
>>
>> of course it could, i wasn't connecting those threads :)
>>
> Thats good. Well then at least that makes me fairly confident that the
> results are correct, but others are still invited to take a look
> though.

 From a very quick first glance, results do look plausible.  -- Though 
the 75% threshold for yellow is probably a bit high for us ;)  Great to 
have this set up, indeed.

>> ok then i guess the test wasn't run because of the previous failure; if
>> that failure happens every time for you, then you can disable the test
>> locally by commenting out the line sc.ScAccessiblePageHeaderArea in
>> sc/qa/unoapi/sc.sce and try again.
>>
> Thanks, but since the failure of a single test doesnt stop it from
> executing any further tests.

It does, unless one calls make with "-k" (and I'm not sure this would 
already be propagated reliably to the relevant places, with our somewhat 
complex Makefile, Makefile.top, etc. setup).

Stephan