Module datarush.matching
Package com.pervasive.datarush.matching.block

Class GroupPairsSortedRows

  • All Implemented Interfaces:
    LogicalOperator
    public class GroupPairsSortedRows
extends ExecutableOperator
    Finds key groupings within the input key fields and, for each key group, generates all pairwise combinations of distinct rows in that group.

    The output is similar to an inner join of the data against itself, except that only distinct combinations are generated. These combinations are useful for comparing rows that may be duplicates, and for mining association rules.

    For example, given three rows A, B, C in a key group, a join would generate all nine of the combinations shown in the matrix below. For efficiency, this operator generates only 4, 7, and 8: just the "strictly upper triangular" entries marked with a * in the matrix: 

     1  A with A  4* B with A  7* C with A
 2  A with B  5  B with B  8* C with B
 3  A with C  6  B with C  9  C with C

    Combinations 1, 5, and 9 are omitted because they join a row with itself, and are thus not useful candidates when looking for duplicate rows or mining association rules.

    Combinations 2, 3, and 6 are omitted because they are the same as combinations 4, 7, and 8 respectively, with field order reversed.

    As with GroupSortedRows and JoinSortedRows, the input for this operator must be sorted so that values in the same key group are consecutive, and the output is sorted by the same key.

    • Field Detail

      • input

        protected final RecordPort input
        The input control port. Contains the an identifier that is unique per key group. A change in the input value signifies the end of a group.

      • keys

        protected String[] keys

    • Constructor Detail

      • GroupPairsSortedRows

        public GroupPairsSortedRows()
        Block records from a single source. Records from the source will be blocked based on key sets and record pairs generated from these groups. Key fields must be identified before execution.
        See Also:
        setKeys(String[])

    • Method Detail

      • getOutput

        public RecordPort getOutput()
        Gets the record port providing the results of the pair generation.
        Returns:
        the output port for the operation

      • getLeftFieldPattern

        public String getLeftFieldPattern()
        Gets the output naming pattern for fields on the left hand side of the pair.
        Returns:
        the pattern for the left hand side field names in output.

      • setLeftFieldPattern

        public void setLeftFieldPattern​(String pattern)
        Sets the output naming pattern for fields on the left hand side of pairs. This is used to ensure distinct names in the output pairs.
        Parameters:
        pattern - name pattern for the left hand side field names

      • getRightFieldPattern

        public String getRightFieldPattern()
        Gets the output naming pattern for fields on the right hand side of the pair.
        Returns:
        the pattern for the right hand side field names in output.

      • setRightFieldPattern

        public void setRightFieldPattern​(String pattern)
        Sets the output naming pattern for fields on the right hand side of pairs. This is used to ensure distinct names in the output pairs.
        Parameters:
        pattern - name pattern for the eight hand side field names

      • computeMetadata

        protected void computeMetadata​(StreamingMetadataContext ctx)
        Description copied from class: StreamingOperator
        Implementations must adhere to the following contracts

        General

        Regardless of input ports/output port types, all implementations must do the following:

        1. Validation. Validation of configuration should always be performed first.
        2. Declare parallelizability.. Implementations must declare parallelizability by calling StreamingMetadataContext.parallelize(ParallelismStrategy).

        Input record ports

        Implementations with input record ports must declare the following:
        1. Required data ordering:
          2. Implementations that have data ordering requirements must declare them by calling RecordPort#setRequiredDataOrdering, otherwise data may arrive in any order.
        2. Required data distribution (only applies to parallelizable operators):
          3. Implementations that have data distribution requirements must declare them by calling RecordPort#setRequiredDataDistribution, otherwise data will arrive in an unspecified partial distribution.
        Note that if the upstream operator's output distribution/ordering is compatible with those required, we avoid a re-sort/re-distribution which is generally a very large savings from a performance standpoint. In addition, some operators may chose to query the upstream output distribution/ordering by calling RecordPort#getSourceDataDistribution and RecordPort#getSourceDataOrdering. These should be viewed as a hints to help chose a more efficient algorithm. In such cases, though, operators must still declare data ordering and data distribution requirements; otherwise there is no guarantee that data will arrive sorted/distributed as required.

        Output record ports

        Implementations with output record ports must declare the following:
        1. Type: Implementations must declare their output type by calling RecordPort#setType.
        Implementations with output record ports may declare the following:
        1. Output data ordering: Implementations that can make guarantees as to their output ordering may do so by calling RecordPort#setOutputDataOrdering
        2. Output data distribution (only applies to parallelizable operators): Implementations that can make guarantees as to their output distribution may do so by calling RecordPort#setOutputDataDistribution
        Note that both of these properties are optional; if unspecified, performance may suffer since the framework may unnecessarily re-sort/re-distributed the data.

        Input model ports

        In general, there is nothing special to declare for input model ports. Models are implicitly duplicated to all partitions when going from non-parallel to parallel operators. The case of a model going from a parallel to a non-parallel node is a special case of a "model reducer" operator. In the case of a model reducer, the downstream operator, must declare the following:
        1. Merge handler: Model reducers must declare a merge handler by calling AbstractModelPort#setMergeHandler.
        Note that MergeModel is a convenient, re-usable model reducer, parameterized with a merge-handler.

        Output model ports

        SimpleModelPort's have no associated metadata and therefore there is never any output metadata to declare. PMMLPort's, on the other hand, do have associated metadata. For all PMMLPorts, implementations must declare the following:
        1. pmmlModelSpec: Implementations must declare the PMML model spec by calling PMMLPort.setPMMLModelSpec.
        Parameters:
        ctx - the context

      • execute

        protected void execute​(ExecutionContext ctx)
        Description copied from class: ExecutableOperator
        Executes the operator. Implementations should adhere to the following contracts:
        1. Following execution, all input ports must be at end-of-data.
        2. Following execution, all output ports must be at end-of-data.
        Parameters:
        ctx - context in which to lookup physical ports bound to logical ports

      • handleRow

        protected void handleRow​(boolean endOfGroup)
        Called once per input row. Aggregators must step to the next row of their inputs and handle group boundaries indicated by endOfGroup.
        Parameters:
        endOfGroup - true iff the input row is the last in the key group

      • endOfData

        protected void endOfData​(boolean emptyInput)
        Called once at the end of run. Aggregators must step to the end of their inputs and push EOD on their outputs.
        Parameters:
        emptyInput - true iff handleRow was called zero times (no input rows to aggregate)

      • getKeys

        public String[] getKeys()

      • setKeys

        public void setKeys​(String[] keys)

      • getNumInputCopies

        protected int getNumInputCopies​(LogicalPort inputPort)
        Description copied from class: ExecutableOperator
        May be overridden to specify that multiple input copies are needed for a given input port. By default this is one. This can be used in rare cases when we must examine multiple positions in the same input stream.
        Overrides:
        getNumInputCopies in class ExecutableOperator
        Parameters:
        inputPort - the port
        Returns:
        the number of input copies for the port