[HIVE-5601] NPE in ORC's PPD when using select * from table with where predicate

• Type: Bug
• Status: Resolved
• Priority: Critical
• Resolution: Fixed
• Affects Version/s: 0.12.0
• Fix Version/s: 0.12.1, 0.13.0
• Labels: ORC
• Assignee: Prasanth Jayachandran
• Reporter: Prasanth Jayachandran
• Created: 21/Oct/13 21:26
• Updated: 16/Nov/13 07:27
• Resolved: 14/Nov/13 19:08

Description

ORCInputFormat has a method findIncludedColumns() which returns boolean array of included columns. In case of the following query

select * from qlog_orc where id<1000 limit 10;

where all columns are selected the findIncludedColumns() returns null. This will result in a NPE when PPD is enabled. Following is the stack trace

Caused by: java.lang.NullPointerException
	at org.apache.hadoop.hive.ql.io.orc.RecordReaderImpl.planReadPartialDataStreams(RecordReaderImpl.java:2387)
	at org.apache.hadoop.hive.ql.io.orc.RecordReaderImpl.readPartialDataStreams(RecordReaderImpl.java:2543)
	at org.apache.hadoop.hive.ql.io.orc.RecordReaderImpl.readStripe(RecordReaderImpl.java:2200)
	at org.apache.hadoop.hive.ql.io.orc.RecordReaderImpl.advanceStripe(RecordReaderImpl.java:2573)
	at org.apache.hadoop.hive.ql.io.orc.RecordReaderImpl.advanceToNextRow(RecordReaderImpl.java:2615)
	at org.apache.hadoop.hive.ql.io.orc.RecordReaderImpl.<init>(RecordReaderImpl.java:132)
	at org.apache.hadoop.hive.ql.io.orc.ReaderImpl.rows(ReaderImpl.java:348)
	at org.apache.hadoop.hive.ql.io.orc.OrcInputFormat$OrcRecordReader.<init>(OrcInputFormat.java:99)
	at org.apache.hadoop.hive.ql.io.orc.OrcInputFormat.getRecordReader(OrcInputFormat.java:241)
	at org.apache.hadoop.hive.ql.io.HiveInputFormat.getRecordReader(HiveInputFormat.java:237)
	... 8 more