Lucene 自定义距离排序

<div class="iteye-blog-content-contain" style="font-size: 14px">

 

 

 

测试类

package sort.distanceSort;

import java.io.IOException;

import org.apache.lucene.analysis.standard.StandardAnalyzer;
import org.apache.lucene.document.Document;
import org.apache.lucene.document.Field.Store;
import org.apache.lucene.document.IntField;
import org.apache.lucene.document.StringField;
import org.apache.lucene.index.DirectoryReader;
import org.apache.lucene.index.IndexWriter;
import org.apache.lucene.index.IndexWriterConfig;
import org.apache.lucene.index.Term;
import org.apache.lucene.search.IndexSearcher;
import org.apache.lucene.search.Query;
import org.apache.lucene.search.ScoreDoc;
import org.apache.lucene.search.Sort;
import org.apache.lucene.search.SortField;
import org.apache.lucene.search.TermQuery;
import org.apache.lucene.search.TopDocs;
import org.apache.lucene.store.RAMDirectory;
import org.apache.lucene.util.Version;

/**
 * @author
 * @Description TODO
 */
public class DistanceSortingTest{
    private RAMDirectory directory;
    private IndexSearcher searcher;
    private Query query;
    public void setUp() throws IOException{
        directory=new RAMDirectory();
        IndexWriter writer=new IndexWriter(directory,
                new IndexWriterConfig(Version.LUCENE_43,
                        new StandardAnalyzer(Version.LUCENE_43)));
        addPoint(writer, "El", "Charro", 1, 2);
        addPoint(writer, "Cafe Poca Cosa", "restaurant", 5, 9);
        addPoint(writer, "Los Betos", "restaurant", 9, 6);
        addPoint(writer, "Nico's Taco Shop", "restaurant", 3, 8);
        addPoint(writer, "Nico's Taco Shopa", "restaurant", 1, 8);
        writer.close();
        searcher=new IndexSearcher(DirectoryReader.open(directory));
        query=new TermQuery(new Term("type","restaurant"));
        System.out.println(query);
    }
   
    private void addPoint(IndexWriter writer,String name,String type,int x,int y )
        throws IOException{
        Document doc=new Document();
        Store yes=Store.YES;
        doc.add(new StringField("name", name, yes));
        doc.add(new StringField("type", type, yes));
        doc.add(new IntField("x", x, yes));
        doc.add(new IntField("y", y, yes));
        writer.addDocument(doc);
    }
    public void testNearestRestaurantToHome() throws IOException{
        Sort sort=new Sort(new SortField("location", new DistanceComparatorSource(0, 0)));
        TopDocs hits=searcher.search(query,10,sort);
        ScoreDoc[] sds=hits.scoreDocs;
        for(ScoreDoc sd:sds){
            System.out.println( searcher.doc(sd.doc).get("name"));
        }
    }
    public static void main(String[] args) throws IOException {
        DistanceSortingTest d=new DistanceSortingTest();
        d.setUp();
        d.testNearestRestaurantToHome();
    }
}

自定义排序类

 

package sort.distanceSort;

import java.io.IOException;

import org.apache.lucene.index.AtomicReaderContext;
import org.apache.lucene.search.FieldCache;
import org.apache.lucene.search.FieldCache.Ints;
import org.apache.lucene.search.FieldComparator;
import org.apache.lucene.search.FieldComparatorSource;

/**
 * @author
 * @Description TODO
 */
public class DistanceComparatorSource extends FieldComparatorSource{
    private int x;
    private int y;
    public DistanceComparatorSource(int x,int y){
        this.x=x;
        this.y=y;
    }
    /* (non-Javadoc)
     * @see org.apache.lucene.search.FieldComparatorSource#newComparator(java.lang.String, int, int, boolean)
     */
    @Override
    public FieldComparator<?> newComparator(String fieldName, int numHits,
            int sortPos, boolean reversed) throws IOException {
        // TODO Auto-generated method stub
        return new DistanceScoreDocLookupComparator(fieldName, numHits);
    }
   
    private class DistanceScoreDocLookupComparator extends FieldComparator{
        private Ints xDoc,yDoc;
        private float[] value;
        private float bottom;
        String fieldName;
       
        public DistanceScoreDocLookupComparator(String fidldName,int numHits){
            value=new float[numHits];
            this.fieldName=fidldName;
        }
       
        /* (non-Javadoc)
         * 传入新的AtomicReaderContext 所有的后续的docIDs都会关联到当前的reader（
         *如果您需要将其映射到一个顶级docID您必须添加docBase）
         *context 当前reader的上下文环境
         *Returns  比较器需要用到这个实例，大多数比较器都是返回当前实例，夸segment的重用当前实例
         *
         */
        @Override
        public FieldComparator setNextReader(AtomicReaderContext context)
                throws IOException {
            xDoc=FieldCache.DEFAULT.getInts(context.reader(), "x", false);
            yDoc=FieldCache.DEFAULT.getInts(context.reader(), "y", false);
            return this;
        }
   
       
        @Override
        public int compare(int slot1, int slot2) {
            if(value[slot1]<value[slot2])return -1;
            if(value[slot1]>value[slot2])return 1;
            return 0;
        }

        /* (non-Javadoc)
         * 设置 最底位置，即 进到对比队列中的最小值（排在末尾的）；
         */
        @Override
        public void setBottom(int slot) {
            bottom=value[slot];
        }

        /* (non-Javadoc)
         * 与文档进行底部比对，只有在setBottom 之后才会调用，与 compare（int,int）功能相同
         * 对于一次命中相对较多的搜索中，这个方法会成为一个调用热点。（是到目前为止调用最为频繁的一个方法）
         */
        @Override
        public int compareBottom(int doc) throws IOException {
            float docDistance=getDistance(doc);
            if(bottom<docDistance)return -1;
            if(bottom>docDistance)return 1;
            return 0;
        }

        /* (non-Javadoc)
         * 当一个命中结果具有争议性的时候会调用这个方法，你需要复制任何一个与文档相关的状态
         * 在之后会再次比对，并放入指定的位置
         */
        @Override
        public void copy(int slot, int doc) throws IOException {
            value[slot]=getDistance(doc);
        }

        public float getDistance(int doc){
            int deltax=xDoc.get(doc)-x;
            int deltay=yDoc.get(doc)-y;
            return (float)Math.sqrt(deltax*deltax+deltay*deltay);
        }
       
        /* (non-Javadoc)
         * 返回当前位置的实际值
         */
        @Override
        public Object value(int slot) {
            return Float.valueOf(value[slot]);
        }

        /* (non-Javadoc)
         * 如果 doc的值小于给定值返回负数
         */
        @Override
        public int compareDocToValue(int doc, Object value) throws IOException {
            return 0;
        }
       
    }
}

 

</div>