JDK 9 RFR of JDK-8036842: Remove unneeded/obsolete -source/-target options in javac tests, part 2

Jonathan Gibbons jonathan.gibbons at oracle.com
Fri Mar 7 21:13:48 UTC 2014


Joe,

If you remove the numerically named duplicate test for "enum as 
identifier", I approve the changes.

-- Jon

On 03/07/2014 09:58 AM, Joe Darcy wrote:
> PS Revised webrev at
>
>     http://cr.openjdk.java.net/~darcy/8036842.1/
>
> -Joe
>
> On 03/07/2014 09:25 AM, Joe Darcy wrote:
>> Hi Jon,
>>
>> On 03/07/2014 07:58 AM, Jonathan Gibbons wrote:
>>> On 03/06/2014 11:40 PM, Joe Darcy wrote:
>>>> Hello,
>>>>
>>>> Please review the webrev of changes for
>>>>
>>>>     JDK-8036842: Remove unneeded/obsolete -source/-target options 
>>>> in javac tests, part 2
>>>>     http://cr.openjdk.java.net/~darcy/8036842.0/
>>>>
>>>> The langtools regression tests suite still passes with these 
>>>> changes in place.
>>>>
>>>> Thanks,
>>>>
>>>> -Joe
>>>
>>> Joe,
>>>
>>> I think the tests for 6464451 should be deleted as no longer 
>>> interesting.
>>
>> Removed.
>>
>>>
>>> ClassLit.java:
>>> T6266772.java:
>>> you can remove the @compile/@run main and use the default action of 
>>> the test
>>
>> Done.
>>
>>>
>>> T6265137.java
>>> do you need any -target
>>
>> The bug being tested seemed to be "if you use -target, you get a 
>> crash" so I left the use of -target in.
>>
>>>
>>> ClassVersionChecker.java
>>> The set of jdk versions should auto adapt to the current 
>>> specification version
>>> The comment of line 50 is out of date.
>>
>> I changed the comment to say: " * 1. If both empty, version num is 
>> for the current release"
>>
>> Sonali, as the author of this test, please file an RFE to address 
>> Jon's comment to make the test adapt to the JDK's version.
>>
>>>
>>> T6384542.java
>>> The first @compile is redundant
>>
>> Do you mean a different file? T6384542.java has been deleted.
>>
>>>
>>> EnumAsIdentifier.java,.out
>>> The test is redundant (equivalent to T6384542.java)
>>>
>>>
>>
>> Keeping one of the files is probably sufficient :-)
>>
>> Thanks,
>>
>> -Joe
>>
>



More information about the compiler-dev mailing list