Package org.elasticsearch.search.aggregations.bucket

Class BestBucketsDeferringCollector

java.lang.Object

org.elasticsearch.search.aggregations.BucketCollector

org.elasticsearch.search.aggregations.bucket.DeferringBucketCollector

org.elasticsearch.search.aggregations.bucket.BestBucketsDeferringCollector

All Implemented Interfaces:

org.apache.lucene.search.Collector

public class BestBucketsDeferringCollector
extends DeferringBucketCollector

A specialization of DeferringBucketCollector that collects all matches and then is able to replay a given subset of buckets which represent the survivors from a pruning process performed by the aggregator that owns this collector.

Nested Class Summary

Nested classes/interfaces inherited from class org.elasticsearch.search.aggregations.bucket.DeferringBucketCollector

DeferringBucketCollector.WrappedAggregator

Field Summary

Fields inherited from class org.elasticsearch.search.aggregations.BucketCollector

NO_OP_COLLECTOR

Constructor Summary

Constructors

Constructor	Description
BestBucketsDeferringCollector(org.apache.lucene.search.Query topLevelQuery, org.apache.lucene.search.IndexSearcher searcher, boolean isGlobal)	Sole constructor.

Method Summary

Modifier and Type	Method	Description
LeafBucketCollector	getLeafCollector(org.apache.lucene.index.LeafReaderContext ctx)
void	postCollection()	Post-collection callback.
void	preCollection()	Pre collection callback.
void	prepareSelectedBuckets(long... selectedBuckets)	Replay the wrapped collector, but only on a selection of buckets.
void	rewriteBuckets(LongUnaryOperator howToRewrite)	Merge or prune the selected buckets.
org.apache.lucene.search.ScoreMode	scoreMode()
void	setDeferredCollector(Iterable<BucketCollector> deferredCollectors)	Set the deferred collectors.
Aggregator	wrap(Aggregator in)	Wrap the provided aggregator so that it behaves (almost) as if it had been collected directly.

Methods inherited from class java.lang.Object

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

Constructor Details

BestBucketsDeferringCollector

public BestBucketsDeferringCollector(org.apache.lucene.search.Query topLevelQuery,
 org.apache.lucene.search.IndexSearcher searcher,
 boolean isGlobal)

Sole constructor.

Parameters:

isGlobal - Whether this collector visits all documents (global context)

Method Details

scoreMode

public org.apache.lucene.search.ScoreMode scoreMode()

setDeferredCollector

public void setDeferredCollector(Iterable<BucketCollector> deferredCollectors)

Set the deferred collectors.

Specified by:

setDeferredCollector in class DeferringBucketCollector

getLeafCollector

public LeafBucketCollector getLeafCollector(org.apache.lucene.index.LeafReaderContext ctx)
                                     throws IOException

Specified by:

getLeafCollector in interface org.apache.lucene.search.Collector

Specified by:

getLeafCollector in class BucketCollector

Throws:

IOException

preCollection

public void preCollection()
                   throws IOException

Description copied from class: BucketCollector

Pre collection callback.

Specified by:

preCollection in class BucketCollector

Throws:

IOException

postCollection

public void postCollection()
                    throws IOException

Description copied from class: BucketCollector

Post-collection callback.

Specified by:

postCollection in class BucketCollector

Throws:

IOException

prepareSelectedBuckets

public void prepareSelectedBuckets(long... selectedBuckets)
                            throws IOException

Replay the wrapped collector, but only on a selection of buckets.

Specified by:

prepareSelectedBuckets in class DeferringBucketCollector

Throws:

IOException

wrap

public Aggregator wrap(Aggregator in)

Wrap the provided aggregator so that it behaves (almost) as if it had been collected directly.

Overrides:

wrap in class DeferringBucketCollector

rewriteBuckets

public void rewriteBuckets(LongUnaryOperator howToRewrite)

Merge or prune the selected buckets.

This process rebuilds some packed structures and is O(number_of_collected_docs) so do your best to skip calling it unless you need it.

Parameters:

howToRewrite - a unary operator which maps a bucket's ordinal to the ordinal it has after this process. If a bucket's ordinal is mapped to -1 then the bucket is removed entirely.