User-defined Aggregate. This Aggregate will only be executed once and will input a column of type I and output a column of type O. If your aggregator can be recursive (i.e. SUM is recursive while COUNT is not) you should write a RecursiveAggregate

User-defined Aggregate. This Aggregate will only be executed once and will input columns of type I1 and I2 and output a column of type O. By definition, this is not recursive.

User-defined Aggregate. This Aggregate will only be executed once and will input columns of type I1, I2, and I3 and output a column of type O. By definition, this is not recursive.

The Scope representation of the byte[] type TODO: This should inherit from SystemTypeColumnData(of byte[])

The scope representation of a boolean column value

Scope's representation of a column value.

This type contains information about the column exposed in the Schema

This class is used to combine two RowSets matching on their keys.

Scope representation of a DateTime column value

Scope representation of a nullable DateTime column value

Scope's representation of a decimal column value

Scope representation of a decimal? column value

This class is used to extract Rows from text with character delimiters

Default outputter for text output Accepted parameters, for example, DefaultTextOutputter( -d "," ): -d "," use "," to separate columns -e "ASCII" output in ASCII -float cast double columns to float -datetime "pattern" use provided pattern to print DateTime columns Legal encodings: ASCII BigEndianUnicode Default Unicode UTF7 UTF8 UTF32

Scope's representation of a double column value

Scope representation of a double? column value

This class is used to convert arbitrary streams to Row objects. Users need to override the Extract and Produce methods.

Scope representation of a float column value

Scope representation of a float? column value

The Scope representation of an int column value

The Scope representation of an int? column value

The Scope representation of a long column value

The Scope representation of a nullable long column value

This class is used to take Row objects and write to a stream

This class is used to process Row objects. Assume that Rows come into the object in arbitrary order and that a given Processor may see any subset of the data without any guarantees. Users need to override the Process and Produce methods.

User-defined Aggregate will input and output columns of type T. By definition, this aggregate must be able to be executed in arbitrary order and produce the same result. SUM is recursive since SUM(a,b,c) = SUM( SUM(a),SUM(b,c) ) COUNT is not recursive since COUNT(a,b,c) != COUNT( COUNT(a), COUNT(a,b) )

This class is used to process Row objects corresponding to a given key. Scope guarantees that all Rows in the system matching the key will be passed into the Reducer in one call to Reduce. Users need to override the Reduce and Produce methods.

The Row object is the mechanism by which data is passed through Scope processes. It contains ColumnData objects which are containers for string, int, double, float, etc data and automatically serializes to/from binary and text streams

The Schema object defines what column types will be produced or consumed by Scope commands. These are important for compiling valid Scope computations

Base class for all user defined type column data.

Mode of combiner. Enables certain optimizations and special runtime modes