-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://review.cloudera.org/r/1208/
-----------------------------------------------------------
Review request for hbase, stack and Kannan Muthukkaruppan.
Summary
-------
Adds a new filter, KeyOnlyFilter. The idea is that this will make it so only
the key portion of all the KVs are returned. Could imagine a few use cases
where you just need the keys/index not the values. We have one where we have
giant rows with big values and want to just get the qualifiers/versions w/o
values.
Adds a new method in KeyValue, convertToKeyOnly(). From javadoc:
/**
* Converts this KeyValue to only contain the key portion (the value is
* changed to be null). This method does a full copy of the backing byte
* array and does not modify the original byte array of this KeyValue.
* <p>
* This method is used by {...@link KeyOnlyFilter} and is an advanced feature
of
* KeyValue, proceed with caution.
*/
This addresses bug HBASE-3211.
http://issues.apache.org/jira/browse/HBASE-3211
Diffs
-----
trunk/src/main/java/org/apache/hadoop/hbase/KeyValue.java 1033617
trunk/src/main/java/org/apache/hadoop/hbase/filter/KeyOnlyFilter.java
PRE-CREATION
trunk/src/main/java/org/apache/hadoop/hbase/io/HbaseObjectWritable.java
1033617
trunk/src/test/java/org/apache/hadoop/hbase/TestKeyValue.java 1033617
trunk/src/test/java/org/apache/hadoop/hbase/filter/TestFilter.java 1033617
Diff: http://review.cloudera.org/r/1208/diff
Testing
-------
Test of the KV method added to TestKeyValue. Test of KeyOnlyFilter added to
TestFilter. Both passing.
Thanks,
Jonathan
