summaryrefslogtreecommitdiff
path: root/core/src/main/java/org/elasticsearch/search/aggregations/bucket/terms/TermsAggregatorFactory.java
blob: f668e3ad9972ecfcb1e4888e829e7f272eaa8162 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
/*
 * 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.terms;

import org.apache.lucene.search.IndexSearcher;
import org.elasticsearch.common.ParseField;
import org.elasticsearch.common.ParseFieldMatcher;
import org.elasticsearch.search.aggregations.AggregationExecutionException;
import org.elasticsearch.search.aggregations.Aggregator;
import org.elasticsearch.search.aggregations.Aggregator.SubAggCollectionMode;
import org.elasticsearch.search.aggregations.AggregatorFactories;
import org.elasticsearch.search.aggregations.InternalAggregation;
import org.elasticsearch.search.aggregations.NonCollectingAggregator;
import org.elasticsearch.search.aggregations.bucket.terms.support.IncludeExclude;
import org.elasticsearch.search.aggregations.pipeline.PipelineAggregator;
import org.elasticsearch.search.aggregations.support.AggregationContext;
import org.elasticsearch.search.aggregations.support.ValuesSource;
import org.elasticsearch.search.aggregations.support.ValuesSourceAggregatorFactory;
import org.elasticsearch.search.aggregations.support.ValuesSourceParser;

import java.io.IOException;
import java.util.List;
import java.util.Map;

/**
 *
 */
public class TermsAggregatorFactory extends ValuesSourceAggregatorFactory<ValuesSource> {

    public enum ExecutionMode {

        MAP(new ParseField("map")) {

            @Override
            Aggregator create(String name, AggregatorFactories factories, ValuesSource valuesSource, Terms.Order order,
                    TermsAggregator.BucketCountThresholds bucketCountThresholds, IncludeExclude includeExclude,
                    AggregationContext aggregationContext, Aggregator parent, SubAggCollectionMode subAggCollectMode,
                    boolean showTermDocCountError, List<PipelineAggregator> pipelineAggregators, Map<String, Object> metaData)
                    throws IOException {
                final IncludeExclude.StringFilter filter = includeExclude == null ? null : includeExclude.convertToStringFilter();
                return new StringTermsAggregator(name, factories, valuesSource, order, bucketCountThresholds, filter, aggregationContext,
                        parent, subAggCollectMode, showTermDocCountError, pipelineAggregators, metaData);
            }

            @Override
            boolean needsGlobalOrdinals() {
                return false;
            }

        },
        GLOBAL_ORDINALS(new ParseField("global_ordinals")) {

            @Override
            Aggregator create(String name, AggregatorFactories factories, ValuesSource valuesSource, Terms.Order order,
                    TermsAggregator.BucketCountThresholds bucketCountThresholds, IncludeExclude includeExclude,
                    AggregationContext aggregationContext, Aggregator parent, SubAggCollectionMode subAggCollectMode,
                    boolean showTermDocCountError, List<PipelineAggregator> pipelineAggregators, Map<String, Object> metaData)
                    throws IOException {
                final IncludeExclude.OrdinalsFilter filter = includeExclude == null ? null : includeExclude.convertToOrdinalsFilter();
                return new GlobalOrdinalsStringTermsAggregator(name, factories, (ValuesSource.Bytes.WithOrdinals) valuesSource, order,
                        bucketCountThresholds, filter, aggregationContext, parent, subAggCollectMode, showTermDocCountError,
                        pipelineAggregators, metaData);
            }

            @Override
            boolean needsGlobalOrdinals() {
                return true;
            }

        },
        GLOBAL_ORDINALS_HASH(new ParseField("global_ordinals_hash")) {

            @Override
            Aggregator create(String name, AggregatorFactories factories, ValuesSource valuesSource, Terms.Order order,
                    TermsAggregator.BucketCountThresholds bucketCountThresholds, IncludeExclude includeExclude,
                    AggregationContext aggregationContext, Aggregator parent, SubAggCollectionMode subAggCollectMode,
                    boolean showTermDocCountError, List<PipelineAggregator> pipelineAggregators, Map<String, Object> metaData)
                    throws IOException {
                final IncludeExclude.OrdinalsFilter filter = includeExclude == null ? null : includeExclude.convertToOrdinalsFilter();
                return new GlobalOrdinalsStringTermsAggregator.WithHash(name, factories,
                        (ValuesSource.Bytes.WithOrdinals.FieldData) valuesSource, order, bucketCountThresholds, filter, aggregationContext,
                        parent, subAggCollectMode, showTermDocCountError, pipelineAggregators, metaData);
            }

            @Override
            boolean needsGlobalOrdinals() {
                return true;
            }
        },
        GLOBAL_ORDINALS_LOW_CARDINALITY(new ParseField("global_ordinals_low_cardinality")) {

            @Override
            Aggregator create(String name, AggregatorFactories factories, ValuesSource valuesSource, Terms.Order order,
                    TermsAggregator.BucketCountThresholds bucketCountThresholds, IncludeExclude includeExclude,
                    AggregationContext aggregationContext, Aggregator parent, SubAggCollectionMode subAggCollectMode,
                    boolean showTermDocCountError, List<PipelineAggregator> pipelineAggregators, Map<String, Object> metaData)
                    throws IOException {
                if (includeExclude != null || factories.count() > 0) {
                    return GLOBAL_ORDINALS.create(name, factories, valuesSource, order, bucketCountThresholds, includeExclude,
                            aggregationContext, parent, subAggCollectMode, showTermDocCountError, pipelineAggregators, metaData);
                }
                return new GlobalOrdinalsStringTermsAggregator.LowCardinality(name, factories,
                        (ValuesSource.Bytes.WithOrdinals) valuesSource, order, bucketCountThresholds, aggregationContext, parent,
                        subAggCollectMode, showTermDocCountError, pipelineAggregators, metaData);
            }

            @Override
            boolean needsGlobalOrdinals() {
                return true;
            }
        };

        public static ExecutionMode fromString(String value, ParseFieldMatcher parseFieldMatcher) {
            for (ExecutionMode mode : values()) {
                if (parseFieldMatcher.match(value, mode.parseField)) {
                    return mode;
                }
            }
            throw new IllegalArgumentException("Unknown `execution_hint`: [" + value + "], expected any of " + values());
        }

        private final ParseField parseField;

        ExecutionMode(ParseField parseField) {
            this.parseField = parseField;
        }

        abstract Aggregator create(String name, AggregatorFactories factories, ValuesSource valuesSource, Terms.Order order,
                TermsAggregator.BucketCountThresholds bucketCountThresholds, IncludeExclude includeExclude,
                AggregationContext aggregationContext, Aggregator parent, SubAggCollectionMode subAggCollectMode,
                boolean showTermDocCountError, List<PipelineAggregator> pipelineAggregators, Map<String, Object> metaData)
                throws IOException;

        abstract boolean needsGlobalOrdinals();

        @Override
        public String toString() {
            return parseField.getPreferredName();
        }
    }

    private final Terms.Order order;
    private final IncludeExclude includeExclude;
    private final String executionHint;
    private final SubAggCollectionMode collectMode;
    private final TermsAggregator.BucketCountThresholds bucketCountThresholds;
    private final boolean showTermDocCountError;

    public TermsAggregatorFactory(String name, ValuesSourceParser.Input input, Terms.Order order,
            TermsAggregator.BucketCountThresholds bucketCountThresholds, IncludeExclude includeExclude, String executionHint,
            SubAggCollectionMode executionMode, boolean showTermDocCountError) {
        super(name, StringTerms.TYPE, input);
        this.order = order;
        this.includeExclude = includeExclude;
        this.executionHint = executionHint;
        this.bucketCountThresholds = bucketCountThresholds;
        this.collectMode = executionMode;
        this.showTermDocCountError = showTermDocCountError;
    }

    @Override
    protected Aggregator createUnmapped(AggregationContext aggregationContext, Aggregator parent,
            List<PipelineAggregator> pipelineAggregators, Map<String, Object> metaData) throws IOException {
        final InternalAggregation aggregation = new UnmappedTerms(name, order, bucketCountThresholds.getRequiredSize(),
                bucketCountThresholds.getShardSize(), bucketCountThresholds.getMinDocCount(), pipelineAggregators, metaData);
        return new NonCollectingAggregator(name, aggregationContext, parent, factories, pipelineAggregators, metaData) {
            {
                // even in the case of an unmapped aggregator, validate the order
                InternalOrder.validate(order, this);
            }

            @Override
            public InternalAggregation buildEmptyAggregation() {
                return aggregation;
            }
        };
    }

    @Override
    protected Aggregator doCreateInternal(ValuesSource valuesSource, AggregationContext aggregationContext, Aggregator parent,
            boolean collectsFromSingleBucket, List<PipelineAggregator> pipelineAggregators, Map<String, Object> metaData)
            throws IOException {
        if (collectsFromSingleBucket == false) {
            return asMultiBucketAggregator(this, aggregationContext, parent);
        }
        if (valuesSource instanceof ValuesSource.Bytes) {
            ExecutionMode execution = null;
            if (executionHint != null) {
                execution = ExecutionMode.fromString(executionHint, aggregationContext.searchContext().parseFieldMatcher());
            }

            // In some cases, using ordinals is just not supported: override it
            if (!(valuesSource instanceof ValuesSource.Bytes.WithOrdinals)) {
                execution = ExecutionMode.MAP;
            }

            final long maxOrd;
            final double ratio;
            if (execution == null || execution.needsGlobalOrdinals()) {
                ValuesSource.Bytes.WithOrdinals valueSourceWithOrdinals = (ValuesSource.Bytes.WithOrdinals) valuesSource;
                IndexSearcher indexSearcher = aggregationContext.searchContext().searcher();
                maxOrd = valueSourceWithOrdinals.globalMaxOrd(indexSearcher);
                ratio = maxOrd / ((double) indexSearcher.getIndexReader().numDocs());
            } else {
                maxOrd = -1;
                ratio = -1;
            }

            // Let's try to use a good default
            if (execution == null) {
                // if there is a parent bucket aggregator the number of instances of this aggregator is going
                // to be unbounded and most instances may only aggregate few documents, so use hashed based
                // global ordinals to keep the bucket ords dense.
                if (Aggregator.descendsFromBucketAggregator(parent)) {
                    execution = ExecutionMode.GLOBAL_ORDINALS_HASH;
                } else {
                    if (factories == AggregatorFactories.EMPTY) {
                        if (ratio <= 0.5 && maxOrd <= 2048) {
                            // 0.5: At least we need reduce the number of global ordinals look-ups by half
                            // 2048: GLOBAL_ORDINALS_LOW_CARDINALITY has additional memory usage, which directly linked to maxOrd, so we need to limit.
                            execution = ExecutionMode.GLOBAL_ORDINALS_LOW_CARDINALITY;
                        } else {
                            execution = ExecutionMode.GLOBAL_ORDINALS;
                        }
                    } else {
                        execution = ExecutionMode.GLOBAL_ORDINALS;
                    }
                }
            }

            return execution.create(name, factories, valuesSource, order, bucketCountThresholds, includeExclude, aggregationContext,
                    parent, collectMode, showTermDocCountError, pipelineAggregators, metaData);
        }

        if ((includeExclude != null) && (includeExclude.isRegexBased())) {
            throw new AggregationExecutionException(
                    "Aggregation ["
                            + name
                            + "] cannot support regular expression style include/exclude "
                            + "settings as they can only be applied to string fields. Use an array of numeric values for include/exclude clauses used to filter numeric fields");
        }

        if (valuesSource instanceof ValuesSource.Numeric) {
            IncludeExclude.LongFilter longFilter = null;
            if (((ValuesSource.Numeric) valuesSource).isFloatingPoint()) {
                if (includeExclude != null) {
                    longFilter = includeExclude.convertToDoubleFilter();
                }
                return new DoubleTermsAggregator(name, factories, (ValuesSource.Numeric) valuesSource, config.format(), order,
                        bucketCountThresholds, aggregationContext, parent, collectMode, showTermDocCountError, longFilter,
                        pipelineAggregators, metaData);
            }
            if (includeExclude != null) {
                longFilter = includeExclude.convertToLongFilter();
            }
            return new LongTermsAggregator(name, factories, (ValuesSource.Numeric) valuesSource, config.format(), order,
                    bucketCountThresholds, aggregationContext, parent, collectMode, showTermDocCountError, longFilter, pipelineAggregators,
                    metaData);
        }

        throw new AggregationExecutionException("terms aggregation cannot be applied to field [" + config.fieldContext().field()
                + "]. It can only be applied to numeric or string fields.");
    }

}