<AWT Dev> [9] Review request for 7042645: Numerous api/java_awt jck tests fail - AWT Assertion Failure on fastdebug ri bundles b138 win7 x86

[Semyon Sadetsky]

surely!

webrev: http://cr.openjdk.java.net/~ssadetsky/7042645/webrev.01/

--Semyon

On 4/6/2015 3:33 PM, Alexander Zvegintsev wrote:
> Hello Semyon,
>
> the fix looks good to me,
> but could you please update the copyright year in all files and fix 
> the indentation of if statement in awt_Button.cpp before push?
>
> Thanks,
>
> Alexander.
> On 04/06/2015 03:11 PM, Semyon Sadetsky wrote:
>> Hello,
>>
>> Please review fix for JDK9.
>>
>> webrev: http://cr.openjdk.java.net/~ssadetsky/7042645/webrev.00/
>> bug: https://bugs.openjdk.java.net/browse/JDK-7042645
>>
>> *ROOT CAUSE:
>> The assertion fails message triggered in awt_Button.cpp because a 
>> WinAPI function used to draw button focus rectangle returns zero. 
>> Tests run successfully on the release build just because assertions 
>> are ignored in it but they are in failure as well.
>>
>> MSDN documentation for the function return value:
>>
>> Return value
>> If the function succeeds, the return value is nonzero.
>> If the function fails, the return value is zero.
>>
>> is wrong, because it returns zero when window is not visible (for 
>> example, outside visible screen area).
>> The discrepancy reveals itself when the function returns zero which 
>> means error but subsequent ::GetLastError() call returns error code 0 
>> which means successful execution.
>>
>> The same exists in awt_Checkbox.cpp and awt_Component.cpp
>>
>> *SOLUTION
>> ::GetlastError() returning error code is asserted to be 0 in case if 
>> ::DrawFocusRect() returns zero.
>>
>> *TESTING
>> This is a JCK test failure issue. No extra regression testing needed.
>>
>> --Semyon
>>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openjdk.java.net/pipermail/awt-dev/attachments/20150406/16dadad3/attachment.html>