comp

Type Members

1. trait CollectFunctions extends AnyRef

   

   This functions can be used to filter or collect specific nodes in collections

2. case class Combine(in: CompNode, dofn: DoFunction, wfk: WireReaderWriter, wfv: WireReaderWriter, wfu: WireReaderWriter, nodeSinks: Seq[Sink] = Seq(), bridgeStoreId: String = randomUUID.toString) extends ProcessNodeImpl with Product with Serializable

   

   The Combine node type specifies the building of a CompNode as a result of applying an associative function to the values of an existing key-values CompNode

3. trait CompNodes extends GraphNodes

   

   General methods for navigating a graph of CompNodes

4. trait GraphNodes extends AttributionCore

   

   generic functions for a nodes graph

5. case class GroupByKey(in: CompNode, wfk: WireReaderWriter, gpk: KeyGrouping, wfv: WireReaderWriter, nodeSinks: Seq[Sink] = Seq(), bridgeStoreId: String = randomUUID.toString) extends ProcessNodeImpl with Product with Serializable

   

   The GroupByKey node type specifies the building of a CompNode as a result of partitioning an exiting key-value CompNode by key

6. case class Load(source: Source, wf: WireReaderWriter, sinks: Seq[Sink] = Seq()) extends ValueNodeImpl with Product with Serializable

   

   The Load node type specifies the creation of a CompNode from some source other than another CompNode.

   The Load node type specifies the creation of a CompNode from some source other than another CompNode. A DataSource object specifies how the loading is performed

7. case class Materialise(in: ProcessNode, wf: WireReaderWriter, sinks: Seq[Sink] = Seq()) extends ValueNodeImpl with Product with Serializable

   

8. case class Op(in1: CompNode, in2: CompNode, f: (Any, Any) ⇒ Any, wf: WireReaderWriter, sinks: Seq[Sink] = Seq()) extends ValueNodeImpl with Product with Serializable

   

   The Op node type specifies the building of Exp CompNode by applying a function to the values of two other CompNode nodes

9. trait Optimiser extends CompNodes with MemoRewriter

   

   Optimiser for the CompNode graph

   Optimiser for the CompNode graph

   It uses the [Kiama](http://code.google.com/p/kiama) rewriting library by defining Strategies for traversing the graph and rules to rewrite it. Usually the rules are applied in a top-down fashion at every node where they can be applied (using the everywhere strategy).

10. case class ParallelDo(ins: Seq[CompNode], env: ValueNode, dofn: DoFunction, wfa: WireReaderWriter, wfb: WireReaderWriter, nodeSinks: Seq[Sink] = Seq(), bridgeStoreId: String = randomUUID.toString) extends ProcessNodeImpl with Product with Serializable

    

    The ParallelDo node type specifies the building of a CompNode as a result of applying a function to all elements of an existing CompNode and concatenating the results

11. trait ProcessNodeImpl extends ProcessNode

    

    Processing node in the computation graph.

    Processing node in the computation graph.

    It has a unique id, a BridgeStore for its outputs and some possible additional sinks.

12. case class Return(rt: ReturnedValue, wf: WireReaderWriter, sinks: Seq[Sink] = Seq()) extends ValueNodeImpl with Product with Serializable

    

    The Return node type specifies the building of a Exp CompNode from an "ordinary" value.

13. case class ReturnSC(in: (ScoobiConfiguration) ⇒ Seq[Any], wf: WireReaderWriter = UnitFmt, sinks: Seq[Sink] = Seq()) extends ValueNodeImpl with Product with Serializable

    

14. class ReturnedValue extends AnyRef

    

15. case class Root(ins: Seq[CompNode], sinks: Seq[Sink] = Seq()) extends ValueNodeImpl with Product with Serializable

    

16. trait ShowNode extends CompNodes

    

    This trait contains some display functions for computation graphs

17. trait ValueNodeImpl extends ValueNode with WithEnvironment

    

    Value node to either load or materialise a value

18. trait WithEnvironment extends AnyRef

    

    Value nodes have environments which are determined by the job configuration because they are effectively files which are distributed via the distributed cache