----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/6892/#review10980 -----------------------------------------------------------
flume-ng-channels/flume-file-channel/src/main/java/org/apache/flume/channel/file/CheckpointRebuilder.java <https://reviews.apache.org/r/6892/#comment23567> Doesn't this violate the Liskov substitution principal (EJ 2nd ed, p39-40)? If it doesn't make sense to subclass the ComparableFlumeEventPointer class, then can we make the class final? flume-ng-channels/flume-file-channel/src/main/java/org/apache/flume/channel/file/CheckpointRebuilder.java <https://reviews.apache.org/r/6892/#comment23568> Considering the case where pointer can be null, we can use Guava's Objects.equal(..) method: return Objects.equal(this.pointer, ((ComparableFlumeEventPointer)o).pointer); - Will McQueen On Sept. 2, 2012, 6:47 p.m., Hari Shreedharan wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/6892/ > ----------------------------------------------------------- > > (Updated Sept. 2, 2012, 6:47 p.m.) > > > Review request for Flume. > > > Description > ------- > > The equals() method in ComparableFlumeEventPointer was not working correctly, > and hence remove() calls from sets were not removing anything. > > > This addresses bug FLUME-1534. > https://issues.apache.org/jira/browse/FLUME-1534 > > > Diffs > ----- > > > flume-ng-channels/flume-file-channel/src/main/java/org/apache/flume/channel/file/CheckpointRebuilder.java > 4db1b9c > > Diff: https://reviews.apache.org/r/6892/diff/ > > > Testing > ------- > > > Thanks, > > Hari Shreedharan > >
