com.bigdata.search

Class ReadIndexTask<V extends Comparable<V>>

java.lang.Object

com.bigdata.search.AbstractIndexTask<V>

com.bigdata.search.ReadIndexTask<V>

Type Parameters:

V - The generic type of the document identifier.

All Implemented Interfaces:

Callable<Object>

public class ReadIndexTask<V extends Comparable<V>>
extends AbstractIndexTask<V>
implements Callable<Object>

Procedure reads on the terms index, aggregating data on a per-Hit basis.

The procedure uses an IRangeQuery.rangeIterator(byte[], byte[]) to perform a key range scan for a specific term. The range iterator will automatically issue queries, obtaining a "chunk" of results at a time. Those results are aggregated on the Hit collection, which is maintained in a thread-safe hash map.

Note: An ISimpleSplitHandler imposes the constraint that index partitions may only fall on a term boundary, hence all tuples for any given term will always be found on the same index partition.

Version:

$Id$

Author:

Bryan Thompson

Field Summary

Fields inherited from class com.bigdata.search.AbstractIndexTask

fromKey, numQueryTerms, queryTerm, queryTermNdx, queryTermWeight, toKey

Constructor Summary

Constructors

Constructor and Description
ReadIndexTask(String termText, int termNdx, int numTerms, boolean prefixMatch, double queryTermWeight, FullTextIndex<V> searchEngine, IHitCollector<V> hits) Setup a task that will perform a range scan for entries matching the search term.

Method Summary

Methods

Modifier and Type	Method and Description
Long	call()

Methods inherited from class java.lang.Object

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

Constructor Detail

ReadIndexTask

public ReadIndexTask(String termText,
             int termNdx,
             int numTerms,
             boolean prefixMatch,
             double queryTermWeight,
             FullTextIndex<V> searchEngine,
             IHitCollector<V> hits)

Setup a task that will perform a range scan for entries matching the search term.

Parameters:

termText - The term text for the search term.

termNdx - The index of this term within the overall search.

numTerms - The overall number of search terms.

prefixMatch - When true any term having termText as a prefix will be matched. Otherwise the term must be an exact match for the termText.

queryTermWeight - The weight for the search term.

searchEngine - The search engine.

hits - The map where the hits are being aggregated.

Method Detail

call

public Long call()
          throws Exception

Specified by:

call in interface Callable<Object>

Returns:

The #of fields with a hit on the search term as a Long.

Throws:

Exception