RFR (S): 6380601: MISC_REGRESSION tests need to be more resilient to ps cmd problems

[Jaroslav Bachorik]

On 13.1.2014 14:57, Yekaterina Kantserova wrote:
> On 01/13/2014 02:29 PM, Alan Bateman wrote:
>> On 13/01/2014 13:17, Yekaterina Kantserova wrote:
>>> Thank you for reviewing this!
>>>
>>> Unfortunantely I can't use Basic.java since we don't use packages for
>>> jtreg tests.
>> It is true that we use the unnamed package in most of the tests but
>> that shouldn't be an issue because the location is unique.
>>
> Sorry, I've not explained in more details. I have sun/tools tests in the
> same project in Eclipse and there I get a compilation error if I have 2
> classes with the same name. I've used this standard for several fixes,
> for example sun/tools/jstatd. Do you think it's OK to keep these names?

Actually, this no-package convention really disturbs all the IDEs making 
them rather unusable when it comes to working with JDK tests. They are 
downgraded to dumb editors with the benefit of receiving of various 
false errors.

-JB-

>>>
>>> Should I use "2005 - 2014" instead of "2005, 2011, 2014"?
>> 2005, 2014.
> Thanks!
>>
>> -Alan.
>
> // Katja