Do you think we should tackle this now or in a future release? I'd be inclined
to leave things as-is, hide the methods from the Javadocs, and focus on
completing what's there in terms of tests and docs. What do you think?
On 03 Jun 2010, at 17:08, Chris Dennis wrote:
> Yes... that would be good!
>
> On Jun 3, 2010, at 11:08 AM, Geert Bevin wrote:
>
>> I wasn't aware of this, imho we should go the other way and make the
>> orphaned values work without having to iterate through the
>> constituent maps.
>>
>> On 03 Jun 2010, at 17:03, Chris Dennis wrote:
>>
>>> Currently if you want to get the keys for orphaned values on a CDM
>>> you
>>> have to do:
>>>
>>> Set<K> allKeys = new HashSet<K>();
>>> for (Map<K,V> m : map.getConstituentMaps()) {
>>> allKeys.add(clusterInfo.getKeysForOrphanedValues(m));
>>> }
>>>
>>> I was just suggesting that one thing we could do is get rid of TCMap,
>>> and make users do this for the local keys method too.
--
Geert Bevin
Terracotta - http://www.terracotta.org
_______________________________________________
tc-dev mailing list
tc-dev@lists.terracotta.org
http://lists.terracotta.org/mailman/listinfo/tc-dev
