> On Aug. 11, 2014, 4:05 p.m., Brock Noland wrote: > > Thank you very much! Two comments below.
Thanks for reviewing this quickly - Suma ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/24497/#review50182 ----------------------------------------------------------- On Aug. 12, 2014, 6:28 a.m., Suma Shivaprasad wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/24497/ > ----------------------------------------------------------- > > (Updated Aug. 12, 2014, 6:28 a.m.) > > > Review request for hive. > > > Bugs: HIVE-7629 > https://issues.apache.org/jira/browse/HIVE-7629 > > > Repository: hive-git > > > Description > ------- > > Map Joins between 2 parquet tables are failing since the Mapper is trying to > access the columns of the first table(bigger table) while trying to load the > second table(smaller map join table). Fixed this by adding a guard on the > column indexes passed by hive > > > Diffs > ----- > > > ql/src/java/org/apache/hadoop/hive/ql/io/parquet/read/DataWritableReadSupport.java > d6be4bd > ql/src/test/queries/clientpositive/parquet_join.q PRE-CREATION > > Diff: https://reviews.apache.org/r/24497/diff/ > > > Testing > ------- > > parquet_join.q covers most types of joins between 2 parquet tables - Normal, > Map join, SMB join > > > Thanks, > > Suma Shivaprasad > >
