RFR: JDK-8219483: j.l.c.ClassDesc::nested(String, String...) doesn't throw NPE if any arg is null

Joseph D. Darcy joe.darcy at oracle.com
Tue Apr 30 22:00:26 UTC 2019


Hi Vicente,

CSR reviewed.

I suggesting adding a test case for

     cd.nested("good", null)

Thanks,

-Joe

On 4/29/2019 2:41 PM, Vicente Romero wrote:
> Hi Joe,
>
> Thanks for the review. I have modified the patch, please see [1] . I 
> still need a reviewer for the CSR [2],
>
> Vicente
>
> [1] http://cr.openjdk.java.net/~vromero/8219483/webrev.01/
> [2] https://bugs.openjdk.java.net/browse/JDK-8223034
>
>
> On 4/26/19 9:32 PM, Joe Darcy wrote:
>> Hi Vicente,
>>
>> For purposes of a better exception message, do you want to explicitly 
>> check moreNestedNames for null in some way before accessing its 
>> contents? Also, I'd commend the spec be updated slightly to
>>
>>     @throws NullPointerException if any argument or its contents is 
>> {@code null}
>>
>> assuming the desired behavior is a NPE if an element of 
>> moreNestedNames is null as opposed to ust moreNestedNames itself.
>>
>> Thanks,
>>
>> -Joe
>>
>> On 4/26/2019 9:33 AM, Vicente Romero wrote:
>>> Hi,
>>>
>>> Please review fix [1] and CSR [2] for [3]. The API for method 
>>> j.l.c.ClassDesc::nested(String, String...) states that it should 
>>> throw NPE if any of the arguments is null. The implementation is not 
>>> in sync with the API and should be corrected,
>>>
>>> Thanks,
>>> Vicente
>>>
>>> [1] http://cr.openjdk.java.net/~vromero/8219483/webrev.00/
>>> [2] https://bugs.openjdk.java.net/browse/JDK-8223034
>>> [3] https://bugs.openjdk.java.net/browse/JDK-8219483
>>>
>



More information about the core-libs-dev mailing list