case class SpacePartitioner[K](bounds: Bounds[K])(implicit evidence$1: Boundable[K], evidence$2: ClassTag[K], index: PartitionerIndex[K]) extends Partitioner with Product with Serializable
Linear Supertypes
Ordering
- Alphabetic
- By Inheritance
Inherited
- SpacePartitioner
- Product
- Equals
- Partitioner
- Serializable
- Serializable
- AnyRef
- Any
- Hide All
- Show All
Visibility
- Public
- All
Instance Constructors
- new SpacePartitioner(bounds: Bounds[K])(implicit arg0: Boundable[K], arg1: ClassTag[K], index: PartitionerIndex[K])
Value Members
-
def
apply[V, M](rdd: RDD[(K, V)] with Metadata[M])(implicit arg0: ClassTag[V], arg1: GetComponent[M, Bounds[K]]): RDD[(K, V)] with Metadata[Bounds[K]]
Use this partitioner as a partitioner for rdd.
Use this partitioner as a partitioner for rdd. The rdd may have a SpacePartitioner already. If it is in sync with Bounds in the Metadata and PartitionIndex we assume it to be valid . Otherwise we assume it has degraded to be a hash partitioner and we must perform a shuffle.
- val bounds: Bounds[K]
- def containsKey(key: Any): Boolean
-
def
getPartition(key: Any): Int
- Definition Classes
- SpacePartitioner → Partitioner
-
def
hasSameIndex(other: SpacePartitioner[K]): Boolean
Is another space partitioner compatible in the sense of key to index mapping?
- implicit val index: PartitionerIndex[K]
-
def
numPartitions: Int
- Definition Classes
- SpacePartitioner → Partitioner
-
def
regionIndex(region: BigInt): Option[Int]
Map given spatial region index to offset in region array (aka partition id)
- val regions: Array[BigInt]