org.elasticsearch.search.aggregations.bucket.significant.InternalMappedSignificantTerms Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of org.apache.servicemix.bundles.elasticsearch
Show all versions of org.apache.servicemix.bundles.elasticsearch
This OSGi bundle wraps ${pkgArtifactId} ${pkgVersion} jar file.
/*
* Licensed to Elasticsearch under one or more contributor
* license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright
* ownership. Elasticsearch licenses this file to you under
* the Apache License, Version 2.0 (the "License"); you may
* not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
* KIND, either express or implied. See the License for the
* specific language governing permissions and limitations
* under the License.
*/
package org.elasticsearch.search.aggregations.bucket.significant;
import org.elasticsearch.common.io.stream.StreamInput;
import org.elasticsearch.common.io.stream.StreamOutput;
import org.elasticsearch.common.xcontent.XContentBuilder;
import org.elasticsearch.search.DocValueFormat;
import org.elasticsearch.search.aggregations.bucket.significant.heuristics.SignificanceHeuristic;
import org.elasticsearch.search.aggregations.pipeline.PipelineAggregator;
import java.io.IOException;
import java.util.Iterator;
import java.util.List;
import java.util.Map;
import java.util.Objects;
import java.util.function.Function;
import java.util.stream.Collectors;
public abstract class InternalMappedSignificantTerms<
A extends InternalMappedSignificantTerms,
B extends InternalSignificantTerms.Bucket>
extends InternalSignificantTerms {
protected final DocValueFormat format;
protected final long subsetSize;
protected final long supersetSize;
protected final SignificanceHeuristic significanceHeuristic;
protected final List buckets;
protected Map bucketMap;
protected InternalMappedSignificantTerms(String name, int requiredSize, long minDocCount, List pipelineAggregators,
Map metaData, DocValueFormat format, long subsetSize, long supersetSize,
SignificanceHeuristic significanceHeuristic, List buckets) {
super(name, requiredSize, minDocCount, pipelineAggregators, metaData);
this.format = format;
this.buckets = buckets;
this.subsetSize = subsetSize;
this.supersetSize = supersetSize;
this.significanceHeuristic = significanceHeuristic;
}
protected InternalMappedSignificantTerms(StreamInput in, Bucket.Reader bucketReader) throws IOException {
super(in);
format = in.readNamedWriteable(DocValueFormat.class);
subsetSize = in.readVLong();
supersetSize = in.readVLong();
significanceHeuristic = in.readNamedWriteable(SignificanceHeuristic.class);
buckets = in.readList(stream -> bucketReader.read(stream, subsetSize, supersetSize, format));
}
@Override
protected final void writeTermTypeInfoTo(StreamOutput out) throws IOException {
out.writeNamedWriteable(format);
out.writeVLong(subsetSize);
out.writeVLong(supersetSize);
out.writeNamedWriteable(significanceHeuristic);
out.writeList(buckets);
}
@Override
public Iterator iterator() {
return buckets.stream().map(bucket -> (SignificantTerms.Bucket) bucket).collect(Collectors.toList()).iterator();
}
@Override
public List getBuckets() {
return buckets;
}
@Override
public B getBucketByKey(String term) {
if (bucketMap == null) {
bucketMap = buckets.stream().collect(Collectors.toMap(InternalSignificantTerms.Bucket::getKeyAsString, Function.identity()));
}
return bucketMap.get(term);
}
@Override
protected long getSubsetSize() {
return subsetSize;
}
@Override
protected long getSupersetSize() {
return supersetSize;
}
@Override
protected SignificanceHeuristic getSignificanceHeuristic() {
return significanceHeuristic;
}
@Override
protected boolean doEquals(Object obj) {
InternalMappedSignificantTerms that = (InternalMappedSignificantTerms) obj;
return super.doEquals(obj)
&& Objects.equals(format, that.format)
&& subsetSize == that.subsetSize
&& supersetSize == that.supersetSize
&& Objects.equals(significanceHeuristic, that.significanceHeuristic)
&& Objects.equals(buckets, that.buckets)
&& Objects.equals(bucketMap, that.bucketMap);
}
@Override
protected int doHashCode() {
return Objects.hash(super.doHashCode(), format, subsetSize, supersetSize, significanceHeuristic, buckets, bucketMap);
}
@Override
public XContentBuilder doXContentBody(XContentBuilder builder, Params params) throws IOException {
builder.field(CommonFields.DOC_COUNT.getPreferredName(), subsetSize);
builder.field(BG_COUNT, supersetSize);
builder.startArray(CommonFields.BUCKETS.getPreferredName());
for (Bucket bucket : buckets) {
//There is a condition (presumably when only one shard has a bucket?) where reduce is not called
// and I end up with buckets that contravene the user's min_doc_count criteria in my reducer
if (bucket.subsetDf >= minDocCount) {
bucket.toXContent(builder, params);
}
}
builder.endArray();
return builder;
}
}