org.apache.drill.exec.physical.impl.scan.v3

Class FixedReceiver.Builder

java.lang.Object

org.apache.drill.exec.physical.impl.scan.v3.FixedReceiver.Builder

Enclosing class:

FixedReceiver

public static class FixedReceiver.Builder
extends Object

Constructor Summary

Constructors

Constructor and Description
Builder(SchemaNegotiator negotiator)

Method Summary

Modifier and Type	Method and Description
FixedReceiver	build(TupleMetadata readerSchema) Create a fixed receiver for the provided schema (if any) in the scan plan, and the given reader schema.
StandardConversions.Builder	conversionBuilder() Provides access to the conversion builder to add custom properties.
static TupleMetadata	mergeSchemas(TupleMetadata providedSchema, TupleMetadata readerSchema) Given a desired provided schema and an actual reader schema, create a merged schema that contains the provided column where available, but the reader column otherwise.
FixedReceiver.Builder	schemaIsComplete() Mark that the reader schema provided to build(TupleMetadata) contains all columns that this reader will deliver.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

Builder

public Builder(SchemaNegotiator negotiator)

Method Detail

conversionBuilder

public StandardConversions.Builder conversionBuilder()

Provides access to the conversion builder to add custom properties.

schemaIsComplete

public FixedReceiver.Builder schemaIsComplete()

Mark that the reader schema provided to build(TupleMetadata) contains all columns that this reader will deliver. Allows some optimizations. See SchemaNegotiator.schemaIsComplete(boolean).

build

public FixedReceiver build(TupleMetadata readerSchema)

Create a fixed receiver for the provided schema (if any) in the scan plan, and the given reader schema. Assumes no new columns will be added later in the read.

mergeSchemas

public static TupleMetadata mergeSchemas(TupleMetadata providedSchema,
                                         TupleMetadata readerSchema)

Given a desired provided schema and an actual reader schema, create a merged schema that contains the provided column where available, but the reader column otherwise. Copies provided properties to the output schema.

The result is the schema to use when creating column writers: it reflects the type of the target vector. The reader is responsible for converting from the (possibly different) reader column type to the provided column type.

Note: the provided schema should only contain types that the reader is prepared to offer: there is no requirement that the reader support every possible conversion, only those that make sense for that one reader.

Parameters:

providedSchema - the provided schema from CREATE SCHEMA

readerSchema - the set of column types that the reader can provide "natively"

Returns:

a merged schema to use when creating the ResultSetLoader