GeoSpark-Scale-Example-0.6.0-later.scala

/*---------------------------- GeoSpark 0.6 (or later) Scala API usage ----------------------------*/

/*
 * If you are writing GeoSpark program in Spark Scala Shell, no need to declare the Spark Context by yourself.
 * If you are writing a self-contained GeoSpark Scala program, please declare the Spark Context as follows and
 * stop it at the end of the entire program.
 */
import org.apache.spark.SparkContext
import org.apache.spark.SparkConf
import org.apache.spark.storage.StorageLevel;


val conf = new SparkConf().setAppName("Simple Application").setMaster("spark://jias-mbp.mobile.asu.edu:7077");
val sc = new SparkContext(conf)

/*---------------------------- Start an example Spatial Range Query without Index ----------------------------*/
import org.datasyslab.geospark.spatialOperator.RangeQuery; 
import org.datasyslab.geospark.spatialRDD.PointRDD;
import com.vividsolutions.jts.geom.Envelope;
import org.datasyslab.geospark.enums.FileDataSplitter;

val queryEnvelope=new Envelope (-113.79,-109.73,32.99,35.08);
/* Range query window format: minX, maxX, minY, maxY*/
val objectRDD = new PointRDD(sc, "/Users/jiayu/VM_Share/resources/arealm.csv", 0, FileDataSplitter.CSV, false, StorageLevel.MEMORY_ONLY); 
/* 
 * 0 is the starting column of spatial data in the input file. 
 * FileDataSplitter.CSV means the data format is CSV. We CSV, TSV, WKT, GeoJSON and self-defined format mapper.
 * false means each spatial object doesn't need to carry the original input tuple with it.
 */
val resultSize = RangeQuery.SpatialRangeQuery(objectRDD, queryEnvelope, false, false).count();
/* 
 * The O means consider a point only if it is fully covered by the query window when doing query.
 * The first false means don't consider objects intersect but not fully covered by the query window.
 * The second false means don't use spatial index.
 */
/*---------------------------- End an example Spatial Range Query without Index ----------------------------*/


/*---------------------------- Start an example Spatial Range Query with Index ----------------------------*/
import org.datasyslab.geospark.spatialOperator.RangeQuery; 
import org.datasyslab.geospark.spatialRDD.PointRDD;
import com.vividsolutions.jts.geom.Envelope;
import org.datasyslab.geospark.enums.FileDataSplitter;
import org.datasyslab.geospark.enums.IndexType;

val queryEnvelope=new Envelope (-113.79,-109.73,32.99,35.08);
/* Range query window format: minX, maxX, minY, maxY*/
val objectRDD = new PointRDD(sc, "/Users/jiayu/VM_Share/resources/arealm.csv", 0, FileDataSplitter.CSV, false, StorageLevel.MEMORY_ONLY); 
/* 
 * 0 is the starting column of spatial data in the input file. 
 * FileDataSplitter.CSV enum means the data format is CSV. We CSV, TSV, WKT, GeoJSON and self-defined format mapper.
 * false means each spatial object doesn't need to carry the original input tuple with it.
 */
objectRDD.buildIndex(IndexType.RTREE,false);
/* 
 * IndexType.RTREE enum means the index type is R-tree. We support R-Tree index and Quad-Tree index.
 * false means just build index on original spatial RDD instead of spatial partitioned RDD. ONLY set true when doing Spatial Join Query.
 */
val resultSize = RangeQuery.SpatialRangeQuery(objectRDD, queryEnvelope, false, true).count();
/* 
 * The O means consider a point only if it is fully covered by the query window when doing query.
 * The first false means don't consider objects intersect but not fully covered by the query window.
 * The true means use spatial index which has been built before.
 */
/*---------------------------- End an example Spatial Range Query with Index ----------------------------*/


/*---------------------------- Start an example Spatial KNN Query without Index ----------------------------*/
import org.datasyslab.geospark.spatialOperator.KNNQuery;
import org.datasyslab.geospark.spatialRDD.PointRDD;
import com.vividsolutions.jts.geom.GeometryFactory;
import com.vividsolutions.jts.geom.Point;
import com.vividsolutions.jts.geom.Coordinate;
import org.datasyslab.geospark.enums.FileDataSplitter;

val fact=new GeometryFactory();
val queryPoint=fact.createPoint(new Coordinate(-109.73, 35.08));
/* Range query window format: X Y */
val objectRDD = new PointRDD(sc, "/Users/jiayu/VM_Share/resources/arealm.csv", 0, FileDataSplitter.CSV, false, StorageLevel.MEMORY_ONLY); 
/* 
 * 0 is the starting column of spatial data in the input file. 
 * FileDataSplitter.CSV means the data format is CSV. We CSV, TSV, WKT, GeoJSON and self-defined format mapper.
 * false means each spatial object doesn't need to carry the original input tuple with it.
 */
val resultSize = KNNQuery.SpatialKnnQuery(objectRDD, queryPoint, 5,false).size();
/* The number 5 means 5 nearest neighbors
 * The false means don't use spatial index.
 */

/*---------------------------- End an example Spatial KNN Query without Index ----------------------------*/


/*---------------------------- Start an example Spatial KNN Query with Index ----------------------------*/
import org.datasyslab.geospark.spatialOperator.KNNQuery;
import org.datasyslab.geospark.spatialRDD.PointRDD;
import com.vividsolutions.jts.geom.GeometryFactory;
import com.vividsolutions.jts.geom.Point;
import com.vividsolutions.jts.geom.Coordinate;
import org.datasyslab.geospark.enums.FileDataSplitter;
import org.datasyslab.geospark.enums.IndexType;

val fact=new GeometryFactory();
val queryPoint=fact.createPoint(new Coordinate(-109.73, 35.08));
/* Range query window format: X Y */
val objectRDD = new PointRDD(sc, "/Users/jiayu/VM_Share/resources/arealm.csv", 0, FileDataSplitter.CSV, false, StorageLevel.MEMORY_ONLY); 
/* 
 * 0 is the starting column of spatial data in the input file. 
 * FileDataSplitter.CSV means the data format is CSV. We CSV, TSV, WKT, GeoJSON and self-defined format mapper.
 * false means each spatial object doesn't need to carry the original input tuple with it.
 */
objectRDD.buildIndex(IndexType.RTREE,false);
/* 
 * IndexType.RTREE enum means the index type is R-tree. We support R-Tree index and Quad-Tree index. But Quad-Tree doesn't support KNN.
 * false means just build index on original spatial RDD instead of spatial partitioned RDD. ONLY set true when doing Spatial Join Query.
 */
val resultSize = KNNQuery.SpatialKnnQuery(objectRDD, queryPoint, 5,true).size();
/* The number 5 means 5 nearest neighbors
 * The true means use spatial index.
 */
/*---------------------------- End an example Spatial KNN Query with Index ----------------------------*/

/*---------------------------- Start an example Spatial Join Query without Index ----------------------------*/
import org.datasyslab.geospark.spatialOperator.JoinQuery;
import org.datasyslab.geospark.spatialRDD.PointRDD;
import org.datasyslab.geospark.spatialRDD.RectangleRDD;
import org.datasyslab.geospark.enums.FileDataSplitter;
import org.datasyslab.geospark.enums.GridType;

val objectRDD = new PointRDD(sc, "/Users/jiayu/VM_Share/resources/arealm.csv", 0, FileDataSplitter.CSV, false, StorageLevel.MEMORY_ONLY); 
/* 
 * 0 is the starting column of spatial data in the input file. 
 * FileDataSplitter.CSV means the data format is CSV. We CSV, TSV, WKT, GeoJSON and self-defined format mapper.
 * false means each spatial object doesn't need to carry the original input tuple with it.
 */
val rectangleRDD = new RectangleRDD(sc, "/Users/jiayu/VM_Share/resources/zcta510.csv", 0, FileDataSplitter.CSV, false, StorageLevel.MEMORY_ONLY); 
/* 
 * 0 is the starting column of spatial data in the input file. 
 * FileDataSplitter.CSV means the data format is CSV. We CSV, TSV, WKT, GeoJSON and self-defined format mapper.
 * false means each spatial object doesn't need to carry the original input tuple with it.
 */
objectRDD.spatialPartitioning(GridType.RTREE);
/*
 * GridType.RTREE means use R-Tree spatial partitioning technique. It will take the leaf node boundaries as parition boundary.
 * We support R-Tree partitioning and Voronoi diagram partitioning.
 */
rectangleRDD.spatialPartitioning(objectRDD.grids);
/* 
 * Use the partition boundary of objectRDD to repartition the query window RDD, This is mandatory.
 */
val resultSize = JoinQuery.SpatialJoinQuery(objectRDD,rectangleRDD,false,false).count();
/*
 * The first false means don't use spatial index.
 * The second false means don't consider objects intersect but not fully covered by the rectangles.
 */
/*---------------------------- End an example Spatial Join Query without Index ----------------------------*/


/*---------------------------- Start an example Spatial Join Query with Index ----------------------------*/
import org.datasyslab.geospark.spatialOperator.JoinQuery;
import org.datasyslab.geospark.spatialRDD.PointRDD;
import org.datasyslab.geospark.spatialRDD.RectangleRDD;
import org.datasyslab.geospark.enums.FileDataSplitter;
import org.datasyslab.geospark.enums.GridType;
import org.datasyslab.geospark.enums.IndexType;

val objectRDD = new PointRDD(sc, "/Users/jiayu/VM_Share/resources/arealm.csv", 0, FileDataSplitter.CSV, false, StorageLevel.MEMORY_ONLY); 
/* 
 * 0 is the starting column of spatial data in the input file. 
 * FileDataSplitter.CSV means the data format is CSV. We CSV, TSV, WKT, GeoJSON and self-defined format mapper.
 * false means each spatial object doesn't need to carry the original input tuple with it.
 */
val rectangleRDD = new RectangleRDD(sc, "/Users/jiayu/VM_Share/resources/zcta510.csv", 0, FileDataSplitter.CSV, false); 
/* 
 * 0 is the starting column of spatial data in the input file. 
 * FileDataSplitter.CSV means the data format is CSV. We CSV, TSV, WKT, GeoJSON and self-defined format mapper.
 * false means each spatial object doesn't need to carry the original input tuple with it.
 */
objectRDD.spatialPartitioning(GridType.RTREE);
/*
 * GridType.RTREE means use R-Tree spatial partitioning technique. It will take the leaf node boundaries as parition boundary.
 * We support R-Tree partitioning and Voronoi diagram partitioning.
 */
objectRDD.buildIndex(IndexType.RTREE,true);
/* 
 * IndexType.RTREE enum means the index type is R-tree. We support R-Tree index and Quad-Tree index. But Quad-Tree doesn't support KNN.
 * True means build index on the spatial partitioned RDD. ONLY set true when doing Spatial Join Query.
 */
rectangleRDD.spatialPartitioning(objectRDD.grids);
/* 
 * Use the partition boundary of objectRDD to repartition the query window RDD, This is mandatory.
 */
val resultSize = JoinQuery.SpatialJoinQuery(objectRDD,rectangleRDD,true, false).count();
/*
 * The first true means use spatial index.
 * The second false means don't consider objects intersect but not fully covered by the rectangles.
 */
/*---------------------------- End an example Spatial Join Query with Index ----------------------------*/


/*---------------------------- Start an example Distance Join Query without Index ----------------------------*/
import org.datasyslab.geospark.spatialOperator.JoinQuery;
import org.datasyslab.geospark.spatialRDD.PointRDD;
import org.datasyslab.geospark.spatialRDD.CircleRDD;
import org.datasyslab.geospark.enums.FileDataSplitter;
import org.datasyslab.geospark.enums.GridType;

val objectRDD = new PointRDD(sc, "/Users/jiayu/VM_Share/resources/arealm.csv", 0, FileDataSplitter.CSV, false, StorageLevel.MEMORY_ONLY); 
/* 
 * 0 is the starting column of spatial data in the input file. 
 * FileDataSplitter.CSV means the data format is CSV. We CSV, TSV, WKT, GeoJSON and self-defined format mapper.
 * false means each spatial object doesn't need to carry the original input tuple with it.
 */
val centerGeometryRDD = new PointRDD(sc, "/Users/jiayu/VM_Share/resources/arealm.csv", 0, FileDataSplitter.CSV, false, StorageLevel.MEMORY_ONLY); 
/* 
 * 0 is the starting column of spatial data in the input file. 
 * FileDataSplitter.CSV means the data format is CSV. We CSV, TSV, WKT, GeoJSON and self-defined format mapper.
 * false means each spatial object doesn't need to carry the original input tuple with it.
 */
val queryRDD = new CircleRDD(centerGeometryRDD,0.1);
/*
 * 0.1 means the distance between two objects from the two Spatial RDDs.
 */
objectRDD.spatialPartitioning(GridType.RTREE);
/*
 * GridType.RTREE means use R-Tree spatial partitioning technique. It will take the leaf node boundaries as parition boundary.
 * We support R-Tree partitioning and Voronoi diagram partitioning.
 */
queryRDD.spatialPartitioning(objectRDD.grids);
/* 
 * Use the partition boundary of objectRDD to repartition the query window RDD, This is mandatory.
 */
val resultSize = JoinQuery.DistanceJoinQuery(objectRDD,queryRDD,false,false).count();
/*
 * The first false means don't use spatial index.
 * The second false means don't consider objects intersect but not fully covered by the rectangles.
 */
/*---------------------------- End an example Distance Join Query without Index ----------------------------*/


/*---------------------------- Start an example Distance Join Query with Index ----------------------------*/
import org.datasyslab.geospark.spatialOperator.JoinQuery;
import org.datasyslab.geospark.spatialRDD.PointRDD;
import org.datasyslab.geospark.spatialRDD.CircleRDD;
import org.datasyslab.geospark.enums.FileDataSplitter;
import org.datasyslab.geospark.enums.GridType;
import org.datasyslab.geospark.enums.IndexType;

val objectRDD = new PointRDD(sc, "/Users/jiayu/VM_Share/resources/arealm.csv", 0, FileDataSplitter.CSV, false, StorageLevel.MEMORY_ONLY); 
/* 
 * 0 is the starting column of spatial data in the input file. 
 * FileDataSplitter.CSV means the data format is CSV. We CSV, TSV, WKT, GeoJSON and self-defined format mapper.
 * false means each spatial object doesn't need to carry the original input tuple with it.
 */
val centerGeometryRDD = new PointRDD(sc, "/Users/jiayu/VM_Share/resources/arealm.csv", 0, FileDataSplitter.CSV, false, StorageLevel.MEMORY_ONLY); 
/* 
 * 0 is the starting column of spatial data in the input file. 
 * FileDataSplitter.CSV means the data format is CSV. We CSV, TSV, WKT, GeoJSON and self-defined format mapper.
 * false means each spatial object doesn't need to carry the original input tuple with it.
 */
val queryRDD = new CircleRDD(centerGeometryRDD,0.1);
/*
 * 0.1 means the distance between two objects from the two Spatial RDDs.
 */
objectRDD.spatialPartitioning(GridType.RTREE);
/*
 * GridType.RTREE means use R-Tree spatial partitioning technique. It will take the leaf node boundaries as parition boundary.
 * We support R-Tree partitioning and Voronoi diagram partitioning.
 */
objectRDD.buildIndex(IndexType.RTREE,true);
/* 
 * IndexType.RTREE enum means the index type is R-tree. We support R-Tree index and Quad-Tree index. But Quad-Tree doesn't support KNN.
 * True means build index on the spatial partitioned RDD. ONLY set true when doing Spatial Join Query.
 */
queryRDD.spatialPartitioning(objectRDD.grids);
/* 
 * Use the partition boundary of objectRDD to repartition the query window RDD, This is mandatory.
 */
val resultSize = JoinQuery.DistanceJoinQuery(objectRDD,queryRDD,true, false).count();
/*
 * The first true means use spatial index.
 * The second false means don't consider objects intersect but not fully covered by the rectangles.
 */
/*---------------------------- End an example Distance Join Query with Index ----------------------------*/