Question

我想在空间上索引形状以支持基于多边形和点的查询。但是，我很难创建所需的lucene文档。我们的想法是使用递归前缀树基于geohash快速过滤形状，然后通过检查形状的序列化来查看查询多边形是否确实匹配（explained here）来进行粒度过滤。

我遇到的第一个问题是无法创建要添加到空间文档的可索引字段：

Document doc = new Document();
Field[] fields = rptStrategy.createIndexableFields(spatial4jShape.getBoundingBox());
for (Field f : fields) {
  doc.add(f);
}

我遇到的第二个问题是无法使用Jts形状进行序列化部分：

fields = dvStrategy.createIndexableFields((Shape)spatial4jShape);

抛出IllegalArgumentException java.lang.IllegalArgumentException: Unsupported shape class com.spatial4j.core.shape.jts.JtsGeometry

我现在的问题是

我在使用递归前缀树做错了什么？
如何使用序列化策略索引多边形本身？

完整代码：

import com.spatial4j.core.context.SpatialContext;
import com.spatial4j.core.context.jts.JtsSpatialContext;
import com.spatial4j.core.io.jts.JtsWKTReader;
import com.spatial4j.core.shape.Shape;
import org.apache.lucene.document.Document;
import org.apache.lucene.document.Field;
import org.apache.lucene.index.IndexableField;
import org.apache.lucene.spatial.prefix.RecursivePrefixTreeStrategy;
import org.apache.lucene.spatial.prefix.tree.GeohashPrefixTree;
import org.apache.lucene.spatial.serialized.SerializedDVStrategy;

import java.text.ParseException;

public class TestLuceneSpatial {
  public static void main(String[] args) throws ParseException {
    String wkt = "POLYGON((-122.4604 37.7818,-122.4707 37.7645,-122.4659 37.7509,-122.4337 37.7476,-122.4192 37.7856,-122.4604 37.7818))";
    SpatialContext ctx = SpatialContext.GEO;
    GeohashPrefixTree grid = new GeohashPrefixTree(ctx, 6);
    RecursivePrefixTreeStrategy rptStrategy = new RecursivePrefixTreeStrategy(grid, "rptBBox");
    rptStrategy.setPrefixGridScanLevel(grid.getMaxLevels() - 1);
    SerializedDVStrategy dvStrategy = new SerializedDVStrategy(ctx, "polygon");

    JtsWKTReader wktReader = new JtsWKTReader(JtsSpatialContext.GEO, null);
    Shape spatial4jShape = wktReader.parse(wkt);

    Document doc = new Document();
    Field[] fields = rptStrategy.createIndexableFields(spatial4jShape.getBoundingBox());
    for (Field f : fields) {
      doc.add(f);
    }

    fields = dvStrategy.createIndexableFields((Shape)spatial4jShape);
    for (Field f : fields) {
      doc.add(f);
    }

    for (IndexableField f : doc.getFields()) {
      System.out.printf("%s => %s\n", f.name(), f.binaryValue());
    }

    System.out.println(doc);

  }
}

Answer 1

为时已晚，但可能会有所帮助

 private val ctx = new JtsSpatialContextFactory().newSpatialContext

并直接使用形状

Field[] fields = rptStrategy.createIndexableFields(spatial4jShape);

Lucene 5.5.3使用递归前缀树策略进行空间索引

完整代码：

1 个答案: