RFR: JDK-8022554 -- Fix Warnings in sun.invoke.anon Package

Dan Xu dan.xu at oracle.com
Wed Aug 7 19:16:04 UTC 2013


Thank you for your review!

-Dan

On 08/07/2013 12:08 PM, Remi Forax wrote:
> On 08/07/2013 08:59 PM, Joe Darcy wrote:
>> Amended version approved to go back.
>>
>> Thanks,
>>
>> -Joe
>
> As one of the guys involved in the design of this API :)
> I'm Ok with this change.
>
> Rémi
>
>>
>> On 08/07/2013 11:54 AM, Dan Xu wrote:
>>> I see, Thanks! I have updated my changeto 
>>> http://cr.openjdk.java.net/~dxu/8022554/webrev1/.
>>>
>>> -Dan
>>>
>>> On 08/07/2013 11:46 AM, Joe Darcy wrote:
>>>> Hi Dan,
>>>>
>>>> Even if the compiler does not complain, using "Class" or "Class[]" 
>>>> is using a raw type and raw types should generally be viewed as 
>>>> unacceptable in modern code.
>>>>
>>>> Cheers,
>>>>
>>>> -Joe
>>>>
>>>> On 08/07/2013 11:36 AM, Dan Xu wrote:
>>>>> Thanks for your review!
>>>>>
>>>>> I was thinking of that. But without Class<?>[] on the left, the 
>>>>> compiler just worked fine.
>>>>>
>>>>> Here is a simple example,
>>>>>
>>>>> //Main.java
>>>>>
>>>>> import java.util.*;
>>>>>
>>>>> public class Main {
>>>>>     public static final Class[] TEST_CLASS = new Class<?>[16];
>>>>>     public static final List[] TEST_MAP = new ArrayList<?>[16];
>>>>> }
>>>>>
>>>>> After compiling with javac Xlint:all Main.java, no warnings are 
>>>>> printed out. Is it a compiler issue? Thanks!
>>>>>
>>>>> -Dan
>>>>>
>>>>>
>>>>>
>>>>> On 08/07/2013 11:11 AM, Joe Darcy wrote:
>>>>>> I agree with Mike; Class<?>[] should be used on both sides.
>>>>>>
>>>>>> Cheers,
>>>>>>
>>>>>> -Joe
>>>>>>
>>>>>> On 08/07/2013 10:53 AM, Mike Duigou wrote:
>>>>>>> Why not have Class<?>[] on the left side as well?
>>>>>>>
>>>>>>> Mike
>>>>>>>
>>>>>>> On Aug 7 2013, at 10:49 , Dan Xu wrote:
>>>>>>>
>>>>>>>> Hi All,
>>>>>>>>
>>>>>>>> Please review the simple warning fix in 
>>>>>>>> src/share/classes/sun/invoke/anon/ConstantPoolPatch.java.
>>>>>>>>
>>>>>>>> webrev: http://cr.openjdk.java.net/~dxu/8022554/webrev/
>>>>>>>>
>>>>>>>> Thanks,
>>>>>>>>
>>>>>>>> -Dan
>>>>>>
>>>>>
>>>>
>>>
>>
>




More information about the core-libs-dev mailing list