O
- The "outer" (user-facing) Java typeI
- The "inner" (Cassandra-facing, intermediary) Java typepublic abstract class MappingCodec<O,I> extends TypeCodec<O>
TypeCodec
that maps a Java Pojo to another Java object
that can in turn be serialized into a CQL type.
This can serve as a base for libraries dealing with Pojo mappings.
This codec can be seen as a convenience base class for libraries
dealing with Pojo-to-Pojo mappings, but it comes
with a performance penalty: each Java object is serialized
in two steps: first to an intermediate object, and then to a ByteBuffer
,
which means that each serialization actually incurs in two potentially
expensive operations being carried.
If such operations are really expensive, and your mapping library is capable
of serializing objects directly to ByteBuffer
s,
consider writing your own codec instead of using this one.TypeCodec.AbstractCollectionCodec<E,C extends Collection<E>>, TypeCodec.AbstractMapCodec<K,V>, TypeCodec.AbstractTupleCodec<T>, TypeCodec.AbstractUDTCodec<T>, TypeCodec.PrimitiveBooleanCodec, TypeCodec.PrimitiveByteCodec, TypeCodec.PrimitiveDoubleCodec, TypeCodec.PrimitiveFloatCodec, TypeCodec.PrimitiveIntCodec, TypeCodec.PrimitiveLongCodec, TypeCodec.PrimitiveShortCodec
Modifier and Type | Field and Description |
---|---|
protected TypeCodec<I> |
innerCodec |
Constructor and Description |
---|
MappingCodec(TypeCodec<I> innerCodec,
Class<O> javaType) |
MappingCodec(TypeCodec<I> innerCodec,
com.google.common.reflect.TypeToken<O> javaType) |
Modifier and Type | Method and Description |
---|---|
O |
deserialize(ByteBuffer bytes,
ProtocolVersion protocolVersion)
Deserialize the given
ByteBuffer instance according to the CQL type
handled by this codec. |
protected abstract O |
deserialize(I value) |
String |
format(O value)
Format the given value as a valid CQL literal according
to the CQL type handled by this codec.
|
O |
parse(String value)
Parse the given CQL literal into an instance of the Java type
handled by this codec.
|
protected abstract I |
serialize(O value) |
ByteBuffer |
serialize(O value,
ProtocolVersion protocolVersion)
Serialize the given value according to the CQL type
handled by this codec.
|
accepts, accepts, accepts, accepts, ascii, bigint, blob, cboolean, cdouble, cfloat, cint, counter, custom, date, decimal, getCqlType, getJavaType, inet, list, map, set, smallInt, time, timestamp, timeUUID, tinyInt, toString, tuple, userType, uuid, varchar, varint
public ByteBuffer serialize(O value, ProtocolVersion protocolVersion) throws InvalidTypeException
TypeCodec
null
input as
the equivalent of an empty collection.serialize
in class TypeCodec<O>
value
- An instance of T; may be null
.protocolVersion
- the protocol version to use when serializing
bytes
. In most cases, the proper value to provide for this argument
is the value returned by ProtocolOptions.getProtocolVersion()
(which
is the protocol version in use by the driver).ByteBuffer
instance containing the serialized form of TInvalidTypeException
- if the given value does not have the expected typepublic O deserialize(ByteBuffer bytes, ProtocolVersion protocolVersion) throws InvalidTypeException
TypeCodec
ByteBuffer
instance according to the CQL type
handled by this codec.
Implementation notes:
null
or a default value for the corresponding Java type, if applicable;null
;
they should return empty collections instead (the driver's default collection codecs all comply with this rule).ByteBuffer
should never be consumed by read operations that
modify its current position; if necessary,
ByteBuffer.duplicate()
duplicate} it before consuming.deserialize
in class TypeCodec<O>
bytes
- A ByteBuffer
instance containing the serialized form of T;
may be null
or empty.protocolVersion
- the protocol version to use when serializing
bytes
. In most cases, the proper value to provide for this argument
is the value returned by ProtocolOptions.getProtocolVersion()
(which
is the protocol version in use by the driver).InvalidTypeException
- if the given ByteBuffer
instance cannot be deserializedpublic O parse(String value) throws InvalidTypeException
TypeCodec
"NULL"
;
in most cases, implementations should interpret these inputs has equivalent to a null
reference.
Implementing this method is not strictly mandatory: internally, the driver only uses it to
parse the INITCOND when building the metadata of an aggregate function (and in most cases it
will use a built-in codec, unless the INITCOND has a custom type).parse
in class TypeCodec<O>
value
- The CQL string to parse, may be null
or empty.null
on a null input
.InvalidTypeException
- if the given value cannot be parsed into the expected typepublic String format(O value) throws InvalidTypeException
TypeCodec
"NULL"
for null
inputs.
Implementing this method is not strictly mandatory. It is used:
BuiltStatement
for a detailed
explanation of when this happens);QueryLogger
, if parameter logging is enabled;AggregateMetadata.asCQLQuery(boolean)
;toString()
implementation of some objects (UDTValue
,
TupleValue
, and the internal representation of a ROWS
response),
which may appear in driver logs.format
in class TypeCodec<O>
value
- An instance of T; may be null
.InvalidTypeException
- if the given value does not have the expected type