def test_spatial_knn_correctness(self):
polygon_rdd = PolygonRDD(self.sc, input_location, splitter, True)
result_no_index = KNNQuery.SpatialKnnQuery(polygon_rdd,
self.query_point,
self.top_k, False)
polygon_rdd.buildIndex(IndexType.RTREE, False)
result_with_index = KNNQuery.SpatialKnnQuery(polygon_rdd,
self.query_point,
self.top_k, True)
sorted_result_no_index = sorted(
result_no_index,
key=lambda geo_data: distance_sorting_functions(
geo_data, self.query_point))
sorted_result_with_index = sorted(
result_with_index,
key=lambda geo_data: distance_sorting_functions(
geo_data, self.query_point))
difference = 0
for x in range(self.top_k):
difference += sorted_result_no_index[x].geom.distance(
sorted_result_with_index[x].geom)
assert difference == 0
def test_spatial_knn_query_using_index(self):
polygon_rdd = PolygonRDD(self.sc, input_location, splitter, True)
polygon_rdd.buildIndex(IndexType.RTREE, False)
for i in range(self.loop_times):
result = KNNQuery.SpatialKnnQuery(polygon_rdd, self.query_point,
self.top_k, True)
assert result.__len__() > -1
assert result[0].getUserData() is not None
def test_build_index_without_set_grid(self):
spatial_rdd = PolygonRDD(self.sc,
input_location,
FileDataSplitter.CSV,
carryInputData=True,
partitions=num_partitions,
newLevel=StorageLevel.MEMORY_ONLY)
spatial_rdd.analyze()
spatial_rdd.buildIndex(IndexType.RTREE, False)
def test_spatial_range_query_using_index(self):
spatial_rdd = PolygonRDD(self.sc, input_location, splitter, True,
StorageLevel.MEMORY_ONLY)
spatial_rdd.buildIndex(IndexType.RTREE, False)
for i in range(self.loop_times):
result_size = RangeQuery.\
SpatialRangeQuery(spatial_rdd, self.query_envelope, False, False).count()
assert result_size == 704
assert RangeQuery.SpatialRangeQuery(
spatial_rdd, self.query_envelope, False,
False).take(10)[0].getUserData() is not None
def test_empty_constructor(self):
spatial_rdd = PolygonRDD(sparkContext=self.sc,
InputLocation=input_location,
splitter=splitter,
carryInputData=True,
partitions=num_partitions,
newLevel=StorageLevel.MEMORY_ONLY)
spatial_rdd.analyze()
spatial_rdd.spatialPartitioning(grid_type)
spatial_rdd.buildIndex(IndexType.RTREE, True)
spatial_rdd_copy = PolygonRDD()
spatial_rdd_copy.rawSpatialRDD = spatial_rdd
spatial_rdd_copy.analyze()
def test_spatial_join_using_index_on_polygons(self):
query_window = PolygonRDD(self.sc, polygon_rdd_input_location,
polygon_rdd_start_offset,
polygon_rdd_end_offset, polygon_rdd_splitter,
True)
object_rdd = PointRDD(sparkContext=self.sc,
InputLocation=point_rdd_input_location,
Offset=point_rdd_offset,
splitter=point_rdd_splitter,
carryInputData=False)
object_rdd.analyze()
object_rdd.spatialPartitioning(join_query_partitionin_type)
query_window.spatialPartitioning(object_rdd.getPartitioner())
query_window.buildIndex(polygon_rdd_index_type, True)
for i in range(each_query_loop_times):
result_size = JoinQuery.SpatialJoinQuery(object_rdd, query_window,
True, False).count()
def test_polygon_distance_join_with_crs_transformation(self):
query_rdd = PolygonRDD(self.sc, input_location_query_polygon, splitter,
True, num_partitions, StorageLevel.MEMORY_ONLY,
"epsg:4326", "epsg:3857")
window_rdd = CircleRDD(query_rdd, 0.1)
object_rdd = PolygonRDD(self.sc, input_location_query_polygon,
splitter, True, num_partitions,
StorageLevel.MEMORY_ONLY, "epsg:4326",
"epsg:3857")
object_rdd.rawJvmSpatialRDD.jsrdd.repartition(4)
object_rdd.spatialPartitioning(GridType.RTREE)
object_rdd.buildIndex(IndexType.RTREE, True)
window_rdd.spatialPartitioning(object_rdd.grids)
results = JoinQuery.DistanceJoinQuery(object_rdd, window_rdd, True,
False).collect()
assert results.__len__() == 5467
for data in results:
for polygon_data in data[1]:
assert Circle(data[0].geom, 0.1).covers(polygon_data.geom)