001 package org.apache.lucene.demo.facet;
002
003 import java.io.IOException;
004 import java.text.ParseException;
005
006 import org.apache.lucene.analysis.core.WhitespaceAnalyzer;
007 import org.apache.lucene.document.Document;
008 import org.apache.lucene.document.Field.Store;
009 import org.apache.lucene.document.NumericDocValuesField;
010 import org.apache.lucene.document.TextField;
011 import org.apache.lucene.expressions.Expression;
012 import org.apache.lucene.expressions.SimpleBindings;
013 import org.apache.lucene.expressions.js.JavascriptCompiler;
014 import org.apache.lucene.facet.FacetField;
015 import org.apache.lucene.facet.FacetResult;
016 import org.apache.lucene.facet.Facets;
017 import org.apache.lucene.facet.FacetsCollector;
018 import org.apache.lucene.facet.FacetsConfig;
019 import org.apache.lucene.facet.taxonomy.TaxonomyFacetSumValueSource;
020 import org.apache.lucene.facet.taxonomy.TaxonomyReader;
021 import org.apache.lucene.facet.taxonomy.directory.DirectoryTaxonomyReader;
022 import org.apache.lucene.facet.taxonomy.directory.DirectoryTaxonomyWriter;
023 import org.apache.lucene.index.DirectoryReader;
024 import org.apache.lucene.index.IndexWriter;
025 import org.apache.lucene.index.IndexWriterConfig;
026 import org.apache.lucene.search.IndexSearcher;
027 import org.apache.lucene.search.MatchAllDocsQuery;
028 import org.apache.lucene.search.SortField;
029 import org.apache.lucene.store.Directory;
030 import org.apache.lucene.store.RAMDirectory;
031
032 /*
033 * Licensed to the Apache Software Foundation (ASF) under one or more
034 * contributor license agreements. See the NOTICE file distributed with
035 * this work for additional information regarding copyright ownership.
036 * The ASF licenses this file to You under the Apache License, Version 2.0
037 * (the "License"); you may not use this file except in compliance with
038 * the License. You may obtain a copy of the License at
039 *
040 * http://www.apache.org/licenses/LICENSE-2.0
041 *
042 * Unless required by applicable law or agreed to in writing, software
043 * distributed under the License is distributed on an "AS IS" BASIS,
044 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
045 * See the License for the specific language governing permissions and
046 * limitations under the License.
047 */
048
049 /** Shows facets aggregation by an expression. */
050 public class ExpressionAggregationFacetsExample {
051
052 private final Directory indexDir = new RAMDirectory();
053 private final Directory taxoDir = new RAMDirectory();
054 private final FacetsConfig config = new FacetsConfig();
055
056 /** Empty constructor */
057 public ExpressionAggregationFacetsExample() {}
058
059 /** Build the example index. */
060 private void index() throws IOException {
061 IndexWriter indexWriter = new IndexWriter(indexDir, new IndexWriterConfig(FacetExamples.EXAMPLES_VER,
062 new WhitespaceAnalyzer(FacetExamples.EXAMPLES_VER)));
063
064 // Writes facet ords to a separate directory from the main index
065 DirectoryTaxonomyWriter taxoWriter = new DirectoryTaxonomyWriter(taxoDir);
066
067 Document doc = new Document();
068 doc.add(new TextField("c", "foo bar", Store.NO));
069 doc.add(new NumericDocValuesField("popularity", 5L));
070 doc.add(new FacetField("A", "B"));
071 indexWriter.addDocument(config.build(taxoWriter, doc));
072
073 doc = new Document();
074 doc.add(new TextField("c", "foo foo bar", Store.NO));
075 doc.add(new NumericDocValuesField("popularity", 3L));
076 doc.add(new FacetField("A", "C"));
077 indexWriter.addDocument(config.build(taxoWriter, doc));
078
079 indexWriter.close();
080 taxoWriter.close();
081 }
082
083 /** User runs a query and aggregates facets. */
084 private FacetResult search() throws IOException, ParseException {
085 DirectoryReader indexReader = DirectoryReader.open(indexDir);
086 IndexSearcher searcher = new IndexSearcher(indexReader);
087 TaxonomyReader taxoReader = new DirectoryTaxonomyReader(taxoDir);
088
089 // Aggregate categories by an expression that combines the document's score
090 // and its popularity field
091 Expression expr = JavascriptCompiler.compile("_score * sqrt(popularity)");
092 SimpleBindings bindings = new SimpleBindings();
093 bindings.add(new SortField("_score", SortField.Type.SCORE)); // the score of the document
094 bindings.add(new SortField("popularity", SortField.Type.LONG)); // the value of the 'popularity' field
095
096 // Aggregates the facet values
097 FacetsCollector fc = new FacetsCollector(true);
098
099 // MatchAllDocsQuery is for "browsing" (counts facets
100 // for all non-deleted docs in the index); normally
101 // you'd use a "normal" query:
102 FacetsCollector.search(searcher, new MatchAllDocsQuery(), 10, fc);
103
104 // Retrieve results
105 Facets facets = new TaxonomyFacetSumValueSource(taxoReader, config, fc, expr.getValueSource(bindings));
106 FacetResult result = facets.getTopChildren(10, "A");
107
108 indexReader.close();
109 taxoReader.close();
110
111 return result;
112 }
113
114 /** Runs the search example. */
115 public FacetResult runSearch() throws IOException, ParseException {
116 index();
117 return search();
118 }
119
120 /** Runs the search and drill-down examples and prints the results. */
121 public static void main(String[] args) throws Exception {
122 System.out.println("Facet counting example:");
123 System.out.println("-----------------------");
124 FacetResult result = new ExpressionAggregationFacetsExample().runSearch();
125 System.out.println(result);
126 }
127 }