Re: Measuring Wine's API completeness

[Michael Jung]

Hi Alexandre,

On Sunday 13 November 2005 20:31, Alexandre Julliard wrote:
> If you want to measure the quality of the implementation a much better
> way IMO is to measure regression test coverage. This may also motivate
> people to write more tests, which would do a lot more good than
> spending time adding magic comments all over the source.

That's a great idea (though it's measuring a different kind of quality than 
the one I was thinking of). If I understand correctly, this would mean 
building all of wine with 'make CFLAGS="-fprofile-arcs -ftest-coverage"' (as 
documented by Aaron Arvey), running wine-test and then have some script 
recurse into the dll directories and analyse and build summary information 
out of the '*.c.gcov' files?

Bye,