<AWT Dev> [9] Review Request: JDK-8031765 Child process error stream is not empty!

[Anthony Petrov]

+1

--
best regards,
Anthony

On 3/20/2014 3:19 PM, Sergey Bylokhov wrote:
> On 3/20/14 3:14 PM, Petr Pchelko wrote:
>> Hello, Sergey.
>>
>>> The fix looks good in general, but will be good to know that the CR
>>> is filed against jtreg (if the problem in the latest jtreg or
>>> probalbly the problem in java)
>> I think that it’s not the problem in jtreg/java.
>>
>> In version 3.2.3 our code worked because the support classes were
>> copied into the test classes folder.
>> In version 4.1 the support classes are put into their own folder which
>> is the added to the class path.
>>
>> Looks like it was done on purpose to avoid recompilation of the helper
>> classes for every test, so it’s not a bug in jtreg, but a bug in our
>> test helpers.
> Thank for clarification. The fix looks good.
>>
>> With best regards. Petr.
>>
>> 20 марта 2014 г., в 2:51 после полудня, Sergey Bylokhov
>> <Sergey.Bylokhov at oracle.com> написал(а):
>>
>>> Hi, Petr.
>>> The fix looks good in general, but will be good to know that the CR
>>> is filed against jtreg (if the problem in the latest jtreg or
>>> probalbly the problem in java)
>>>
>>> On 3/20/14 2:08 PM, Petr Pchelko wrote:
>>>> Hello, AWT Team.
>>>>
>>>> Please review the fix for the issue:
>>>> https://bugs.openjdk.java.net/browse/JDK-8031765
>>>> The fix is available at:
>>>> http://cr.openjdk.java.net/~pchelko/9/8031765/webrev/
>>>>
>>>> When we are creating a child process on some systems/versions of
>>>> jtreg the class path is not picked up.
>>>> I’ve modified the ProcessCommunicator so that it always explicitly
>>>> sets the class path for the child process.
>>>>
>>>> With best regards. Petr.
>>>
>>> --
>>> Best regards, Sergey.
>>>
>
>