> On April 12, 2016, 3:07 p.m., Ken Howe wrote: > > geode-core/src/test/java/com/gemstone/gemfire/internal/cache/SearchLoadAndWriteProcessorTest.java, > > lines 59-62 > > <https://reviews.apache.org/r/46060/diff/1/?file=1340084#file1340084line59> > > > > I think it be good to verify that EntryEvetnImpl.release() is called in > > addition to verifying that retain and release are called the correct number > > of times? Can you spy EntryEventImpl to accomplish this.
No, the EntryEventImpl that the test knows about is not released by this code. A copy of it is made internally and it is that one that is released. The identity of that one is not visible to this test. All I care about in this test is that we don't leak the StoredObject. - Darrel ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/46060/#review128546 ----------------------------------------------------------- On April 11, 2016, 4:05 p.m., Darrel Schneider wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/46060/ > ----------------------------------------------------------- > > (Updated April 11, 2016, 4:05 p.m.) > > > Review request for geode, Scott Jewell, Ken Howe, and Sai Boorlagadda. > > > Bugs: GEODE-1199 > https://issues.apache.org/jira/browse/GEODE-1199 > > > Repository: geode > > > Description > ------- > > doNetWrite now calls release when it creates a new EntryEventImpl. > Added a unit test for doNetWrite. > > > Diffs > ----- > > > geode-core/src/main/java/com/gemstone/gemfire/internal/cache/SearchLoadAndWriteProcessor.java > 8224fc2826cb7650a812cef6e04f399000211593 > > geode-core/src/test/java/com/gemstone/gemfire/internal/cache/SearchLoadAndWriteProcessorTest.java > PRE-CREATION > > Diff: https://reviews.apache.org/r/46060/diff/ > > > Testing > ------- > > precheckin > > > Thanks, > > Darrel Schneider > >
