Source code

001/*
002 * Licensed to the Apache Software Foundation (ASF) under one or more
003 * contributor license agreements.  See the NOTICE file distributed with
004 * this work for additional information regarding copyright ownership.
005 * The ASF licenses this file to You under the Apache License, Version 2.0
006 * (the "License"); you may not use this file except in compliance with
007 * the License.  You may obtain a copy of the License at
008 *
009 *     http://www.apache.org/licenses/LICENSE-2.0
010 *
011 * Unless required by applicable law or agreed to in writing, software
012 * distributed under the License is distributed on an "AS IS" BASIS,
013 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
014 * See the License for the specific language governing permissions and
015 * limitations under the License.
016 */
017package org.apache.lucene.demo.facet;
018
019import java.io.Closeable;
020import java.io.IOException;
021import java.text.ParseException;
022
023import org.apache.lucene.analysis.core.WhitespaceAnalyzer;
024import org.apache.lucene.document.Document;
025import org.apache.lucene.document.DoublePoint;
026import org.apache.lucene.document.NumericDocValuesField;
027import org.apache.lucene.expressions.Expression;
028import org.apache.lucene.expressions.SimpleBindings;
029import org.apache.lucene.expressions.js.JavascriptCompiler;
030import org.apache.lucene.facet.DrillDownQuery;
031import org.apache.lucene.facet.DrillSideways;
032import org.apache.lucene.facet.FacetResult;
033import org.apache.lucene.facet.Facets;
034import org.apache.lucene.facet.FacetsCollector;
035import org.apache.lucene.facet.FacetsConfig;
036import org.apache.lucene.facet.range.DoubleRange;
037import org.apache.lucene.facet.range.DoubleRangeFacetCounts;
038import org.apache.lucene.facet.taxonomy.TaxonomyReader;
039import org.apache.lucene.index.DirectoryReader;
040import org.apache.lucene.index.IndexWriter;
041import org.apache.lucene.index.IndexWriterConfig;
042import org.apache.lucene.index.IndexWriterConfig.OpenMode;
043import org.apache.lucene.search.BooleanClause;
044import org.apache.lucene.search.BooleanQuery;
045import org.apache.lucene.search.DoubleValuesSource;
046import org.apache.lucene.search.IndexSearcher;
047import org.apache.lucene.search.MatchAllDocsQuery;
048import org.apache.lucene.search.Query;
049import org.apache.lucene.search.TopDocs;
050import org.apache.lucene.store.Directory;
051import org.apache.lucene.store.RAMDirectory;
052import org.apache.lucene.util.SloppyMath;
053
054/** Shows simple usage of dynamic range faceting, using the
055 *  expressions module to calculate distance. */
056public class DistanceFacetsExample implements Closeable {
057
058  final DoubleRange ONE_KM = new DoubleRange("< 1 km", 0.0, true, 1.0, false);
059  final DoubleRange TWO_KM = new DoubleRange("< 2 km", 0.0, true, 2.0, false);
060  final DoubleRange FIVE_KM = new DoubleRange("< 5 km", 0.0, true, 5.0, false);
061  final DoubleRange TEN_KM = new DoubleRange("< 10 km", 0.0, true, 10.0, false);
062
063  private final Directory indexDir = new RAMDirectory();
064  private IndexSearcher searcher;
065  private final FacetsConfig config = new FacetsConfig();
066
067  /** The "home" latitude. */
068  public final static double ORIGIN_LATITUDE = 40.7143528;
069
070  /** The "home" longitude. */
071  public final static double ORIGIN_LONGITUDE = -74.0059731;
072
073  /** Mean radius of the Earth in KM
074   *
075   * NOTE: this is approximate, because the earth is a bit
076   * wider at the equator than the poles.  See
077   * http://en.wikipedia.org/wiki/Earth_radius */
078  // see http://earth-info.nga.mil/GandG/publications/tr8350.2/wgs84fin.pdf
079  public final static double EARTH_RADIUS_KM = 6_371.0087714;
080
081  /** Empty constructor */
082  public DistanceFacetsExample() {}
083  
084  /** Build the example index. */
085  public void index() throws IOException {
086    IndexWriter writer = new IndexWriter(indexDir, new IndexWriterConfig(
087        new WhitespaceAnalyzer()).setOpenMode(OpenMode.CREATE));
088
089    // TODO: we could index in radians instead ... saves all the conversions in getBoundingBoxFilter
090
091    // Add documents with latitude/longitude location:
092    // we index these both as DoublePoints (for bounding box/ranges) and as NumericDocValuesFields (for scoring)
093    Document doc = new Document();
094    doc.add(new DoublePoint("latitude", 40.759011));
095    doc.add(new NumericDocValuesField("latitude", Double.doubleToRawLongBits(40.759011)));
096    doc.add(new DoublePoint("longitude", -73.9844722));
097    doc.add(new NumericDocValuesField("longitude", Double.doubleToRawLongBits(-73.9844722)));
098    writer.addDocument(doc);
099    
100    doc = new Document();
101    doc.add(new DoublePoint("latitude", 40.718266));
102    doc.add(new NumericDocValuesField("latitude", Double.doubleToRawLongBits(40.718266)));
103    doc.add(new DoublePoint("longitude", -74.007819));
104    doc.add(new NumericDocValuesField("longitude", Double.doubleToRawLongBits(-74.007819)));
105    writer.addDocument(doc);
106    
107    doc = new Document();
108    doc.add(new DoublePoint("latitude", 40.7051157));
109    doc.add(new NumericDocValuesField("latitude", Double.doubleToRawLongBits(40.7051157)));
110    doc.add(new DoublePoint("longitude", -74.0088305));
111    doc.add(new NumericDocValuesField("longitude", Double.doubleToRawLongBits(-74.0088305)));
112    writer.addDocument(doc);
113
114    // Open near-real-time searcher
115    searcher = new IndexSearcher(DirectoryReader.open(writer));
116    writer.close();
117  }
118
119  private DoubleValuesSource getDistanceValueSource() {
120    Expression distance;
121    try {
122      distance = JavascriptCompiler.compile(
123                  "haversin(" + ORIGIN_LATITUDE + "," + ORIGIN_LONGITUDE + ",latitude,longitude)");
124    } catch (ParseException pe) {
125      // Should not happen
126      throw new RuntimeException(pe);
127    }
128    SimpleBindings bindings = new SimpleBindings();
129    bindings.add("latitude", DoubleValuesSource.fromDoubleField("latitude"));
130    bindings.add("longitude", DoubleValuesSource.fromDoubleField("longitude"));
131
132    return distance.getDoubleValuesSource(bindings);
133  }
134
135  /** Given a latitude and longitude (in degrees) and the
136   *  maximum great circle (surface of the earth) distance,
137   *  returns a simple Filter bounding box to "fast match"
138   *  candidates. */
139  public static Query getBoundingBoxQuery(double originLat, double originLng, double maxDistanceKM) {
140
141    // Basic bounding box geo math from
142    // http://JanMatuschek.de/LatitudeLongitudeBoundingCoordinates,
143    // licensed under creative commons 3.0:
144    // http://creativecommons.org/licenses/by/3.0
145
146    // TODO: maybe switch to recursive prefix tree instead
147    // (in lucene/spatial)?  It should be more efficient
148    // since it's a 2D trie...
149
150    // Degrees -> Radians:
151    double originLatRadians = SloppyMath.toRadians(originLat);
152    double originLngRadians = SloppyMath.toRadians(originLng);
153
154    double angle = maxDistanceKM / EARTH_RADIUS_KM;
155
156    double minLat = originLatRadians - angle;
157    double maxLat = originLatRadians + angle;
158
159    double minLng;
160    double maxLng;
161    if (minLat > SloppyMath.toRadians(-90) && maxLat < SloppyMath.toRadians(90)) {
162      double delta = Math.asin(Math.sin(angle)/Math.cos(originLatRadians));
163      minLng = originLngRadians - delta;
164      if (minLng < SloppyMath.toRadians(-180)) {
165        minLng += 2 * Math.PI;
166      }
167      maxLng = originLngRadians + delta;
168      if (maxLng > SloppyMath.toRadians(180)) {
169        maxLng -= 2 * Math.PI;
170      }
171    } else {
172      // The query includes a pole!
173      minLat = Math.max(minLat, SloppyMath.toRadians(-90));
174      maxLat = Math.min(maxLat, SloppyMath.toRadians(90));
175      minLng = SloppyMath.toRadians(-180);
176      maxLng = SloppyMath.toRadians(180);
177    }
178
179    BooleanQuery.Builder f = new BooleanQuery.Builder();
180
181    // Add latitude range filter:
182    f.add(DoublePoint.newRangeQuery("latitude", SloppyMath.toDegrees(minLat), SloppyMath.toDegrees(maxLat)),
183          BooleanClause.Occur.FILTER);
184
185    // Add longitude range filter:
186    if (minLng > maxLng) {
187      // The bounding box crosses the international date
188      // line:
189      BooleanQuery.Builder lonF = new BooleanQuery.Builder();
190      lonF.add(DoublePoint.newRangeQuery("longitude", SloppyMath.toDegrees(minLng), Double.POSITIVE_INFINITY),
191               BooleanClause.Occur.SHOULD);
192      lonF.add(DoublePoint.newRangeQuery("longitude", Double.NEGATIVE_INFINITY, SloppyMath.toDegrees(maxLng)),
193               BooleanClause.Occur.SHOULD);
194      f.add(lonF.build(), BooleanClause.Occur.MUST);
195    } else {
196      f.add(DoublePoint.newRangeQuery("longitude", SloppyMath.toDegrees(minLng), SloppyMath.toDegrees(maxLng)),
197            BooleanClause.Occur.FILTER);
198    }
199
200    return f.build();
201  }
202
203  /** User runs a query and counts facets. */
204  public FacetResult search() throws IOException {
205
206    FacetsCollector fc = new FacetsCollector();
207
208    searcher.search(new MatchAllDocsQuery(), fc);
209
210    Facets facets = new DoubleRangeFacetCounts("field", getDistanceValueSource(), fc,
211                                               getBoundingBoxQuery(ORIGIN_LATITUDE, ORIGIN_LONGITUDE, 10.0),
212                                               ONE_KM,
213                                               TWO_KM,
214                                               FIVE_KM,
215                                               TEN_KM);
216
217    return facets.getTopChildren(10, "field");
218  }
219
220  /** User drills down on the specified range. */
221  public TopDocs drillDown(DoubleRange range) throws IOException {
222
223    // Passing no baseQuery means we drill down on all
224    // documents ("browse only"):
225    DrillDownQuery q = new DrillDownQuery(null);
226    final DoubleValuesSource vs = getDistanceValueSource();
227    q.add("field", range.getQuery(getBoundingBoxQuery(ORIGIN_LATITUDE, ORIGIN_LONGITUDE, range.max), vs));
228    DrillSideways ds = new DrillSideways(searcher, config, (TaxonomyReader) null) {
229        @Override
230        protected Facets buildFacetsResult(FacetsCollector drillDowns, FacetsCollector[] drillSideways, String[] drillSidewaysDims) throws IOException {        
231          assert drillSideways.length == 1;
232          return new DoubleRangeFacetCounts("field", vs, drillSideways[0], ONE_KM, TWO_KM, FIVE_KM, TEN_KM);
233        }
234      };
235    return ds.search(q, 10).hits;
236  }
237
238  @Override
239  public void close() throws IOException {
240    searcher.getIndexReader().close();
241    indexDir.close();
242  }
243
244  /** Runs the search and drill-down examples and prints the results. */
245  public static void main(String[] args) throws Exception {
246    DistanceFacetsExample example = new DistanceFacetsExample();
247    example.index();
248
249    System.out.println("Distance facet counting example:");
250    System.out.println("-----------------------");
251    System.out.println(example.search());
252
253    System.out.println("Distance facet drill-down example (field/< 2 km):");
254    System.out.println("---------------------------------------------");
255    TopDocs hits = example.drillDown(example.TWO_KM);
256    System.out.println(hits.totalHits + " totalHits");
257
258    example.close();
259  }
260}