HHH-10023 Make hibernate-infinispan compiled with Infinispan 7.x but … #1045
Conversation
|
Pushed again, those two failures masked that I haven't fixed two keySet() usages |
…runnable with Infinispan 8.x * workaround for ISPN-5676 * fix for ClassCastException in ClusteredTimestampsRegionImpl * minor fixes in the testsuite
|
I've found ISPN-5676 - the PR now contains a workaround for that (+ few other fixes). |
| @@ -291,6 +291,10 @@ public static void removeAll(AdvancedCache cache) { | |||
| } | |||
|
|
|||
| public static CollectableCloseableIterable keys(AdvancedCache cache) { | |||
| if (cache.getCacheConfiguration().transaction().transactionMode().isTransactional()) { | |||
| // Dummy read to enlist the LocalTransaction as workaround for ISPN-5676 | |||
| cache.containsKey(false); | |||
There was a problem hiding this comment.
Wondering maybe you want to do a withFlags(CACHE_MODE_LOCAL) ? So this way if it is distributed you won't incur the remote overhead.
There was a problem hiding this comment.
Good point. @rvansa I'll integrate this PR asap but can you make sure this gets addressed later on?
There was a problem hiding this comment.
Also this will be fixed in ISPN 7.2.5 which will render this change unneeded. But I am not for sure when that will be released.
There was a problem hiding this comment.
I think that in all places where I use this, local cache is passed to the method. I don't think that the additional flag should be passed at this place (which is some kind of generic helper on Cache API).
…runnable with Infinispan 8.x
https://hibernate.atlassian.net/browse/HHH-10023
There are two tests failing yet because of https://issues.jboss.org/browse/ISPN-5605 - that should be fixed in Infinispan upstream.