Package org.apache.lucene.codecs

Class DocValuesFormat

java.lang.Object

org.apache.lucene.codecs.DocValuesFormat

All Implemented Interfaces:

NamedSPILoader.NamedSPI

Direct Known Subclasses:

Lucene40DocValuesFormat, Lucene42DocValuesFormat, Lucene45DocValuesFormat, PerFieldDocValuesFormat

public abstract class DocValuesFormat
extends Object
implements NamedSPILoader.NamedSPI

Encodes/decodes per-document values.

Note, when extending this class, the name (getName()) may written into the index in certain configurations. In order for the segment to be read, the name must resolve to your implementation via forName(String). This method uses Java's Service Provider Interface (SPI) to resolve format names.

If you implement your own format, make sure that it has a no-arg constructor so SPI can load it.

See Also:

ServiceLoader

Constructor Summary

Constructors

Modifier	Constructor	Description
protected	DocValuesFormat(String name)	Creates a new docvalues format.

Method Summary

Modifier and Type	Method	Description
static Set<String>	availableDocValuesFormats()	returns a list of all available format names
abstract DocValuesConsumer	fieldsConsumer(SegmentWriteState state)	Returns a DocValuesConsumer to write docvalues to the index.
abstract DocValuesProducer	fieldsProducer(SegmentReadState state)	Returns a DocValuesProducer to read docvalues from the index.
static DocValuesFormat	forName(String name)	looks up a format by name
String	getName()
static void	reloadDocValuesFormats(ClassLoader classloader)	Reloads the DocValues format list from the given ClassLoader.
String	toString()

Methods inherited from class java.lang.Object

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

Constructor Detail

DocValuesFormat

protected DocValuesFormat(String name)

Creates a new docvalues format.

The provided name will be written into the index segment in some configurations (such as when using PerFieldDocValuesFormat): in such configurations, for the segment to be read this class should be registered with Java's SPI mechanism (registered in META-INF/ of your jar file, etc).

Parameters:

name - must be all ascii alphanumeric, and less than 128 characters in length.

Method Detail

fieldsConsumer

public abstract DocValuesConsumer fieldsConsumer(SegmentWriteState state)
                                          throws IOException

Returns a DocValuesConsumer to write docvalues to the index.

Throws:

IOException

fieldsProducer

public abstract DocValuesProducer fieldsProducer(SegmentReadState state)
                                          throws IOException

Returns a DocValuesProducer to read docvalues from the index.

NOTE: by the time this call returns, it must hold open any files it will need to use; else, those files may be deleted. Additionally, required files may be deleted during the execution of this call before there is a chance to open them. Under these circumstances an IOException should be thrown by the implementation. IOExceptions are expected and will automatically cause a retry of the segment opening logic with the newly revised segments.

Throws:

IOException

getName

public final String getName()

Specified by:

getName in interface NamedSPILoader.NamedSPI

toString

public String toString()

Overrides:

toString in class Object

forName

public static DocValuesFormat forName(String name)

looks up a format by name

availableDocValuesFormats

public static Set<String> availableDocValuesFormats()

returns a list of all available format names

reloadDocValuesFormats

public static void reloadDocValuesFormats(ClassLoader classloader)

Reloads the DocValues format list from the given ClassLoader. Changes to the docvalues formats are visible after the method ends, all iterators (availableDocValuesFormats(),...) stay consistent.

NOTE: Only new docvalues formats are added, existing ones are never removed or replaced.

This method is expensive and should only be called for discovery of new docvalues formats on the given classpath/classloader!