RFR : 7144488 StackOverflowError occurres on list via Collections.synchronizedList(List)
Seán Coffey
sean.coffey at oracle.com
Thu Feb 23 14:46:50 UTC 2012
ok,
I've updated the changes (and bug synopsis) based on comments.
equals method tweaked for :
Collections.SynchronizedList
Collections.SynchronizedSet
Collections.SynchronizedMap
Collections.UnmodifiableEntry
Testcase updated also. JCK java_util tests run also.
http://cr.openjdk.java.net/~coffeys/webrev.7144488.1/
regards,
Sean.
On 23/02/2012 09:56, David Holmes wrote:
> On 23/02/2012 7:55 PM, Alan Bateman wrote:
>> On 23/02/2012 02:44, David Holmes wrote:
>>>
>>> All of the SynchronizedX.equals methods should be fixed ie
>>> SynchronizedSet, SynchronizedList and SynchronizedMap. It should
>>> always be valid to ask if a.equals(a). The idiomatic form used
>>> elsewhere (CheckedXXX) is:
>>>
>>> public boolean equals(Object o) { return o == this || list.equals(o); }
>>>
>>> I'm not a fan of collections containing themselves, but I think it is
>>> simple to fix contains(o)/contains[Key]|[Value](o) and remove(o) in a
>>> similar way. Though none of the wrapped collections currently handle
>>> that case, so I'm okay with not addressing this part.
>>>
>>> I don't see the recursion potential in hashCode() in the wrappers.
>> The hashCode of a Set requires summing the hash codes of each of the
>> elements so it can't work for the case that a set contains itself.
>
> Right, but that's a general problem with collections, not with the
> wrappers specifically.
>
> Cheers,
> David
>
>> Anyway, I think the best we can do it to have the equals methods use the
>> above form. I wasn't aware of the warning in List that Sean pointed out
>> and maybe there should be similar warnings elsewhere.
>>
>> -Alan
More information about the core-libs-dev
mailing list