org.apache.flink.cdc.runtime.operators.transform

类 PreTransformProcessor

java.lang.Object

org.apache.flink.cdc.runtime.operators.transform.PreTransformProcessor

public class PreTransformProcessor
extends Object

The processor of pre-transform projection in PreTransformOperator.

A pre-transform projection processor handles:

• CreateTableEvent: removes unused (unreferenced) columns from given schema.
• SchemaChangeEvent: update the columns of TransformProjection.
• DataChangeEvent: omits unused columns in data row.

构造器概要

构造器

构造器和说明
PreTransformProcessor(PreTransformChangeInfo tableChangeInfo)

方法概要

限定符和类型	方法和说明
org.apache.flink.cdc.common.event.CreateTableEvent	preTransformCreateTableEvent(org.apache.flink.cdc.common.event.CreateTableEvent createTableEvent) This method analyses (directly and indirectly) referenced columns, and peels unused columns from schema.
org.apache.flink.cdc.common.data.binary.BinaryRecordData	processFillDataField(org.apache.flink.cdc.common.data.binary.BinaryRecordData data)

从类继承的方法 java.lang.Object

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

构造器详细资料

PreTransformProcessor

public PreTransformProcessor(PreTransformChangeInfo tableChangeInfo)

方法详细资料

preTransformCreateTableEvent

public org.apache.flink.cdc.common.event.CreateTableEvent preTransformCreateTableEvent(org.apache.flink.cdc.common.event.CreateTableEvent createTableEvent)

This method analyses (directly and indirectly) referenced columns, and peels unused columns from schema. For example, given original schema with columns (A, B, C, D, E) with projection rule (A, B + 1 as newB) and filtering rule (C > 0), a peeled schema containing (A, B, C) only will be sent to downstream, and (D, E) column along with corresponding data will be trimmed.

processFillDataField

public org.apache.flink.cdc.common.data.binary.BinaryRecordData processFillDataField(org.apache.flink.cdc.common.data.binary.BinaryRecordData data)