[2/3] gdiplus: fix GdipPathIterNextMarker behaviour on path without markers. fix tests.

Nikolay Sivov bunglehead at gmail.com
Sun Jul 13 03:42:00 CDT 2008


Alexandre Julliard wrote:
> Nikolay Sivov <bunglehead at gmail.com> writes:
>
>   
>> I don't think so. If the call fails testing return value doesn't make 
>> any sense. In this case 'result' is uninitialized and remains 
>> uninitialized after a call if it fails (here I mean a native too). So 
>> why should we check something which has unpredictable value?
>> By the way first time I saw this here 
>> 8fd6f0e26ae28f2ba4938e2fbcc4851f47baa53c..
>>     
>
> The difference is that this was a todo_wine, so it's expected to fail
> and we don't want to test uninitialized data since that can cause random
> test failures.
>
> In your case the test is expected to succeed, so the only way you'll get
> uninitialized data is when the first test fails, and then it doesn't
> matter if we get a random second failure, the test failed already.
>
>   
I see. I've just reposted it with 'try3' label, there's no linking based 
on return value now. Now I'm explicitly initialize output before each 
call to be sure that values changed or not.

What do you think about matrix comparison or in general float comparison 
implementation? What is the best way to implement this. We have two 
cases now:

1. GdipIsMatrixEqual which compares two matrices (with simple bitwise 
operation now)
2. GdipIsMatrixInvertible which contains a check for 'not above zero' 
determinant.

I think first of all we should choose an appropriate method to do so and 
only after that try to compare results with native. What do you think of 
that?





More information about the wine-devel mailing list