org.apache.flink.api.java.typeutils.runtime

Class RowSerializer

java.lang.Object

org.apache.flink.api.common.typeutils.TypeSerializer<Row>

org.apache.flink.api.java.typeutils.runtime.RowSerializer

All Implemented Interfaces:

Serializable

@Internal
public final class RowSerializer
extends TypeSerializer<Row>

Serializer for Row.

It uses the following serialization format:

     |bitmask|field|field|....
 

The bitmask serves as a header that consists of ROW_KIND_OFFSET bits for encoding the RowKind and n bits for whether a field is null. For backwards compatibility, those bits can be ignored if serializer runs in legacy mode:

     bitmask with row kind:  |RK RK F1 F2 ... FN|
     bitmask in legacy mode: |F1 F2 ... FN|
 

Field names are an optional part of this serializer. They allow to use rows in named-based field mode. However, the support for name-based rows is limited. Usually, name-based mode should not be used in state but only for in-flight data. For now, names are not part of serializer snapshot or equals/hashCode (similar to RowTypeInfo).

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	RowSerializer.RowSerializerSnapshot A TypeSerializerSnapshot for RowSerializer.

Field Summary

Fields

Modifier and Type	Field and Description
static int	ROW_KIND_OFFSET

Constructor Summary

Constructors

Constructor and Description
RowSerializer(TypeSerializer<?>[] fieldSerializers)
RowSerializer(TypeSerializer<?>[] fieldSerializers, LinkedHashMap<String,Integer> positionByName)
RowSerializer(TypeSerializer<?>[] fieldSerializers, LinkedHashMap<String,Integer> positionByName, boolean supportsRowKind)

Method Summary

Modifier and Type	Method and Description
void	copy(DataInputView source, DataOutputView target) Copies exactly one record from the source input view to the target output view.
Row	copy(Row from) Creates a deep copy of the given element in a new element.
Row	copy(Row from, Row reuse) Creates a copy from the given element.
Row	createInstance() Creates a new instance of the data type.
Row	deserialize(DataInputView source) De-serializes a record from the given source input view.
Row	deserialize(Row reuse, DataInputView source) De-serializes a record from the given source input view into the given reuse record instance if mutable.
TypeSerializer<Row>	duplicate() Creates a deep copy of this serializer if it is necessary, i.e. if it is stateful.
boolean	equals(Object o)
int	getArity()
int	getLength() Gets the length of the data type, if it is a fix length data type.
int	hashCode()
boolean	isImmutableType() Gets whether the type is an immutable type.
void	serialize(Row record, DataOutputView target) Serializes the given record to the given target output view.
TypeSerializerSnapshot<Row>	snapshotConfiguration() Snapshots the configuration of this TypeSerializer.

Methods inherited from class java.lang.Object

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

Field Detail

ROW_KIND_OFFSET

public static final int ROW_KIND_OFFSET

See Also:

Constant Field Values

Constructor Detail

RowSerializer

public RowSerializer(TypeSerializer<?>[] fieldSerializers)

RowSerializer

public RowSerializer(TypeSerializer<?>[] fieldSerializers,
                     @Nullable
                     LinkedHashMap<String,Integer> positionByName)

RowSerializer

public RowSerializer(TypeSerializer<?>[] fieldSerializers,
                     @Nullable
                     LinkedHashMap<String,Integer> positionByName,
                     boolean supportsRowKind)

Method Detail

isImmutableType

public boolean isImmutableType()

Description copied from class: TypeSerializer

Gets whether the type is an immutable type.

Specified by:

isImmutableType in class TypeSerializer<Row>

Returns:

True, if the type is immutable.

duplicate

public TypeSerializer<Row> duplicate()

Description copied from class: TypeSerializer

Creates a deep copy of this serializer if it is necessary, i.e. if it is stateful. This can return itself if the serializer is not stateful.

We need this because Serializers might be used in several threads. Stateless serializers are inherently thread-safe while stateful serializers might not be thread-safe.

Specified by:

duplicate in class TypeSerializer<Row>

createInstance

public Row createInstance()

Description copied from class: TypeSerializer

Creates a new instance of the data type.

Specified by:

createInstance in class TypeSerializer<Row>

Returns:

A new instance of the data type.

copy

public Row copy(Row from)

Description copied from class: TypeSerializer

Creates a deep copy of the given element in a new element.

Specified by:

copy in class TypeSerializer<Row>

Parameters:

from - The element reuse be copied.

Returns:

A deep copy of the element.

copy

public Row copy(Row from,
                Row reuse)

Description copied from class: TypeSerializer

Creates a copy from the given element. The method makes an attempt to store the copy in the given reuse element, if the type is mutable. This is, however, not guaranteed.

Specified by:

copy in class TypeSerializer<Row>

Parameters:

from - The element to be copied.

reuse - The element to be reused. May or may not be used.

Returns:

A deep copy of the element.

getLength

public int getLength()

Description copied from class: TypeSerializer

Gets the length of the data type, if it is a fix length data type.

Specified by:

getLength in class TypeSerializer<Row>

Returns:

The length of the data type, or -1 for variable length data types.

getArity

public int getArity()

serialize

public void serialize(Row record,
                      DataOutputView target)
               throws IOException

Description copied from class: TypeSerializer

Serializes the given record to the given target output view.

Specified by:

serialize in class TypeSerializer<Row>

Parameters:

record - The record to serialize.

target - The output view to write the serialized data to.

Throws:

IOException - Thrown, if the serialization encountered an I/O related error. Typically raised by the output view, which may have an underlying I/O channel to which it delegates.

deserialize

public Row deserialize(DataInputView source)
                throws IOException

Description copied from class: TypeSerializer

De-serializes a record from the given source input view.

Specified by:

deserialize in class TypeSerializer<Row>

Parameters:

source - The input view from which to read the data.

Returns:

The deserialized element.

Throws:

IOException - Thrown, if the de-serialization encountered an I/O related error. Typically raised by the input view, which may have an underlying I/O channel from which it reads.

deserialize

public Row deserialize(Row reuse,
                       DataInputView source)
                throws IOException

Description copied from class: TypeSerializer

De-serializes a record from the given source input view into the given reuse record instance if mutable.

Specified by:

deserialize in class TypeSerializer<Row>

Parameters:

reuse - The record instance into which to de-serialize the data.

source - The input view from which to read the data.

Returns:

The deserialized element.

Throws:

IOException - Thrown, if the de-serialization encountered an I/O related error. Typically raised by the input view, which may have an underlying I/O channel from which it reads.

copy

public void copy(DataInputView source,
                 DataOutputView target)
          throws IOException

Description copied from class: TypeSerializer

Copies exactly one record from the source input view to the target output view. Whether this operation works on binary data or partially de-serializes the record to determine its length (such as for records of variable length) is up to the implementer. Binary copies are typically faster. A copy of a record containing two integer numbers (8 bytes total) is most efficiently implemented as target.write(source, 8);.

Specified by:

copy in class TypeSerializer<Row>

Parameters:

source - The input view from which to read the record.

target - The target output view to which to write the record.

Throws:

IOException - Thrown if any of the two views raises an exception.

equals

public boolean equals(Object o)

Specified by:

equals in class TypeSerializer<Row>

hashCode

public int hashCode()

Specified by:

hashCode in class TypeSerializer<Row>

snapshotConfiguration

public TypeSerializerSnapshot<Row> snapshotConfiguration()

Description copied from class: TypeSerializer

Snapshots the configuration of this TypeSerializer. This method is only relevant if the serializer is used to state stored in checkpoints/savepoints.

The snapshot of the TypeSerializer is supposed to contain all information that affects the serialization format of the serializer. The snapshot serves two purposes: First, to reproduce the serializer when the checkpoint/savepoint is restored, and second, to check whether the serialization format is compatible with the serializer used in the restored program.

IMPORTANT: TypeSerializerSnapshots changed after Flink 1.6. Serializers implemented against Flink versions up to 1.6 should still work, but adjust to new model to enable state evolution and be future-proof. See the class-level comments, section "Upgrading TypeSerializers to the new TypeSerializerSnapshot model" for details.

Specified by:

snapshotConfiguration in class TypeSerializer<Row>

Returns:

snapshot of the serializer's current configuration (cannot be null).

See Also:

TypeSerializerSnapshot.resolveSchemaCompatibility(TypeSerializerSnapshot)