openide.util/src/org/openide/util/lookup/ProxyLookup.java
Tue, 05 Feb 2008 19:43:00 +0100 An attempt to make the result hold the data weak, but it does not work because of GC of template's class
Tue, 19 Feb 2008 07:23:16 +0100 Getting rid of Xlint warning
Tue, 19 Feb 2008 07:16:25 +0100 Giving up, returning back the finalizer to make testGarbageCollect pass
Mon, 18 Feb 2008 20:45:21 +0100 Reorganized, support for cleanup of templates, but the classloader is still not GCed
Mon, 18 Feb 2008 20:36:46 +0100 Making ProxyLookup with one result smaller by 1/3 of original size
Mon, 18 Feb 2008 20:20:20 +0100 Less generics
Mon, 18 Feb 2008 20:01:01 +0100 Empty ProxyLookup smaller to 16B again, but the filled one still occupies quite a big amount of memory
Mon, 18 Feb 2008 19:51:41 +0100 Amount of needed memory for empty lookup got down to 32, which is still twice as it used to
Mon, 18 Feb 2008 19:08:55 +0100 Almost works, except it is still too big
Mon, 18 Feb 2008 18:47:05 +0100 Works with recurrent calls to ProxyLookup.setLookups
Mon, 18 Feb 2008 16:10:20 +0100 If setLookups is called on the same data, it does nothing internally
Mon, 18 Feb 2008 15:37:51 +0100 Too much asserts, removing one that I have no clue what it is supposed to test
Wed, 06 Feb 2008 18:50:03 +0100 So finally I know who exits from the critical section into foreing code. It is res.lookup(template) in lookupChange. I guess I start from scratch and fix this differently.
Wed, 06 Feb 2008 16:09:50 +0100 Tests are OK and R is now static
Tue, 05 Feb 2008 17:53:22 +0100 tests are passing again
Tue, 05 Feb 2008 16:37:24 +0100 rewritten to have unmodifiable data inside ProxyLookup
Wed, 16 Jan 2008 08:46:13 +0000 #123679: Make setLookups really reentrant
Thu, 01 Nov 2007 09:30:18 +0000 Fixing compilation warning
Tue, 30 Oct 2007 17:51:00 +0000 #119292: Making setLookups re-entrant from add/remove LookupListener
Tue, 30 Oct 2007 17:33:24 +0000 #119292: Lookup.equals could be used to execute setLookups re-entrantly and get IndexOutOfBounds
Wed, 26 Sep 2007 20:53:05 +0000 I love free software very very very very very very very very very very much
Wed, 15 Aug 2007 12:13:23 +0000 #112574: Under concurent access it is not possible to guarantee a state of cache after exiting the synchronized block - let's remove the assert, it was apperently wrong
Tue, 12 Sep 2006 14:46:50 +0000 A bit more generification
Fri, 04 Aug 2006 21:27:32 +0000 Some varargs in method signatures where it makes sense.
Tue, 11 Jul 2006 18:07:00 +0000 Trivial Javadoc simplification; removing uninformational sentence.
Sat, 01 Jul 2006 09:07:16 +0000 Final touches for CDDL upgrade in module openide
Tue, 27 Jun 2006 12:23:26 +0000 Remove unnecessary casts.
Tue, 21 Mar 2006 08:19:37 +0000 #67888 and #73573: Generified
Tue, 07 Mar 2006 06:12:43 +0000 #73263: Using WeakResult the same way SimpleProxyLookup does. Fixes Nejedlak's memory test.
Fri, 02 Dec 2005 10:45:40 +0000 #69633: All changes in results in org.openide.util.lookup are first processed and only then fired to LookupListeners registered from outside of the package. This prevents the situation when one listens on interface A and then queries interface B, while the result for the interface B is not yet updated.
Fri, 11 Nov 2005 23:14:02 +0000 "Lookup" is a backformation from "[to] look [something] up".
Fri, 11 Nov 2005 16:49:03 +0000 #68031: Making the AbstractLookup and especially ProxyLookup much, much improved. The results are now really unmodifiable, events are fired only if something really changes, etc. Small step for me, but big step forward for the mankind. I am just curious how much unit tests will fail on monday...
Sun, 12 Jun 2005 15:44:26 +0000 #58941: The array passed to setLookups was thought as not modifiable and if someone modified it and setLookups once again things were obviously broken. Making clone of the array in setLookups and also creating clone in getLookups. Testing both cases.
Thu, 21 Apr 2005 20:12:08 +0000 #57941: Physically moving openide sources into their logical locations