Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
Hive >> mail # dev >> Review Request 15097: Remove .array() assumptions from ORC RecordReaderImpl and InStream


Copy link to this message
-
Review Request 15097: Remove .array() assumptions from ORC RecordReaderImpl and InStream

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/15097/
-----------------------------------------------------------

Review request for hive and Owen O'Malley.
Bugs: HIVE-5663
    https://issues.apache.org/jira/browse/HIVE-5663
Repository: hive-git
Description
-------

This patch scrubs all ByteBuffer.wrap assumptions from the RecordReaderImpl and InStream classes.

The code still uses wrapped bytebuffers in the read path, but accesses all data using the ByteBuffer methods without resorting to .array() data.
Diffs
-----

  ql/src/java/org/apache/hadoop/hive/ql/io/orc/InStream.java 3164a19
  ql/src/java/org/apache/hadoop/hive/ql/io/orc/RecordReaderImpl.java 71484a3

Diff: https://reviews.apache.org/r/15097/diff/
Testing
-------

tested with TestORCFile, -Dtestcase=TestCliDriver -Dqfile_regex=.*orc.*
Thanks,

Gopal V