User: mnf999 Date: 01/11/13 15:41:20 Modified: src/main/org/jboss/ejb CacheKey.java Log: The performance of the copy semantics is just too expensive, if we want to enforce the copy semantics in the future (commented out here) then we should put that in some cache class that can read a configuration field. This is intended to speed up the finder (large ones) and 2.0 CMR that makes extensive use of these. Revision Changes Path 1.17 +11 -1 jboss/src/main/org/jboss/ejb/CacheKey.java Index: CacheKey.java =================================================================== RCS file: /cvsroot/jboss/jboss/src/main/org/jboss/ejb/CacheKey.java,v retrieving revision 1.16 retrieving revision 1.17 diff -u -r1.16 -r1.17 --- CacheKey.java 2001/08/03 17:15:43 1.16 +++ CacheKey.java 2001/11/13 23:41:20 1.17 @@ -29,7 +29,7 @@ * * @author <a href="mailto:[EMAIL PROTECTED]">Marc Fleury</a> * @author <a href="[EMAIL PROTECTED]">Bill Burke</a> - * @version $Revision: 1.16 $ + * @version $Revision: 1.17 $ */ public class CacheKey implements Externalizable @@ -78,11 +78,21 @@ this.id = null; try { + + + /* + * FIXME MARCF: The reuse of the primary key is an "exception" and this fix makes + everyone pay an hefty price. If we really want this behavior we can put it in + the cache. Is there a test for this? + // Equals rely on the MarshalledObject itself mo = new MarshalledObject(id); // Make a copy of the id to enforce copy semantics and // allow reuse of the original primary key this.id = mo.get(); + + */ + // Precompute the hashCode (speed) hashCode = mo.hashCode(); }
_______________________________________________ Jboss-development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development