Package org.apache.hadoop.hbase.filter

Class FuzzyRowFilter.RowTracker

java.lang.Object

org.apache.hadoop.hbase.filter.FuzzyRowFilter.RowTracker

Enclosing class:

FuzzyRowFilter

private class FuzzyRowFilter.RowTracker
extends Object

If we have multiple fuzzy keys, row tracker should improve overall performance. It calculates all next rows (one per every fuzzy key) and put them (the fuzzy key is bundled) into a priority queue so that the smallest row key always appears at queue head, which helps to decide the "Next Cell Hint". As scanning going on, the number of candidate rows in the RowTracker will remain the size of fuzzy keys until some of the fuzzy keys won't possibly have matches any more.

Field Summary

Fields

Modifier and Type	Field	Description
private boolean	initialized
private final PriorityQueue<Pair<byte[],Pair<byte[],byte[]>>>	nextRows

Constructor Summary

Constructors

Constructor	Description
RowTracker()

Method Summary

Modifier and Type	Method	Description
(package private) boolean	lessThan(Cell currentCell, byte[] nextRowKey)
(package private) byte[]	nextRow()
(package private) boolean	updateTracker(Cell currentCell)
(package private) void	updateWith(Cell currentCell, Pair<byte[],byte[]> fuzzyData)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

nextRows

private final PriorityQueue<Pair<byte[],Pair<byte[],byte[]>>> nextRows

initialized

private boolean initialized

Constructor Details

RowTracker

RowTracker()

Method Details

nextRow

byte[] nextRow()

updateTracker

boolean updateTracker(Cell currentCell)

lessThan

boolean lessThan(Cell currentCell,
 byte[] nextRowKey)

updateWith

void updateWith(Cell currentCell,
 Pair<byte[],byte[]> fuzzyData)