> On June 24, 2012, 10:09 a.m., Juhani Connolly wrote: > > If there's anything to comment on, it's that relying on the order of the > > methods in the file for the exception test to run last is a bit dubious. > > >From the faq http://junit.sourceforge.net/doc/faq/faq.htm#tests_2 > > "The ordering of test-method invocations is not guaranteed, so > > testOneItemCollection() might be executed before testEmptyCollection(). But > > it doesn't matter, because each method gets its own instance of the > > collection." > > > > Also, I wasn't aware that HBaseTestingUtility.createTable overwrites > > existing ones, but since the tests work I assume it does(or did I miss > > something)? > > > > I did however confirm that all the tests ran fine. So if we're ok with > > that, it's good to go as far as I'm concerned
Juhani, Thanks for the review. You are right, there is no guarantee that the order is the same as in the file, which is why I mentioned that we can comment out the failure test before committing this, and it can be used during development only. I know that it is isn't exactly a great thing to do, but this does test one more scenario. Also, that test does take very long too, so I wouldn't mind disabling it by default and enabling it only when needed. Regarding the createTable() method, the table is actually deleted at the end of every test(except the last one - which is one more reason to keep it disabled by default). - Hari ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/5405/#review8516 ----------------------------------------------------------- On June 18, 2012, 10:49 p.m., Hari Shreedharan wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/5405/ > ----------------------------------------------------------- > > (Updated June 18, 2012, 10:49 p.m.) > > > Review request for Flume. > > > Description > ------- > > Make sure table and column family exists in start(). > > > This addresses bug FLUME-1288. > https://issues.apache.org/jira/browse/FLUME-1288 > > > Diffs > ----- > > > flume-ng-sinks/flume-ng-hbase-sink/src/main/java/org/apache/flume/sink/hbase/AsyncHBaseSink.java > d08d7a7 > > flume-ng-sinks/flume-ng-hbase-sink/src/test/java/org/apache/flume/sink/hbase/TestAsyncHBaseSink.java > 3de651f > > Diff: https://reviews.apache.org/r/5405/diff/ > > > Testing > ------- > > Added several more unit tests. > > > Thanks, > > Hari Shreedharan > >
