org.neo4j.kernel.api.impl.index

Class AbstractLuceneIndex

java.lang.Object

org.neo4j.kernel.api.impl.index.AbstractLuceneIndex

public abstract class AbstractLuceneIndex
extends Object

Abstract implementation of a partitioned index. Such index may consist of one or multiple separate Lucene indexes that are represented as independent partitions. Class and it's subclasses should not be directly used, instead please use corresponding writable or read only wrapper.

See Also:

WritableAbstractDatabaseIndex, ReadOnlyAbstractDatabaseIndex

Constructor Summary

Constructors

Constructor and Description
AbstractLuceneIndex(PartitionedIndexStorage indexStorage, IndexPartitionFactory partitionFactory)

Method Summary

Modifier and Type	Method and Description
LuceneAllDocumentsReader	allDocumentsReader() Creates an iterable over all documents in all partitions.
void	close()
void	create() Creates new index.
void	drop() Close index and deletes all it's partitions.
boolean	exists() Check lucene index existence within all allocated partitions.
void	flush(boolean merge) Commits all index partitions.
AbstractIndexPartition	getFirstPartition(List<AbstractIndexPartition> partitions)
List<AbstractIndexPartition>	getPartitions()
boolean	hasSinglePartition(List<AbstractIndexPartition> partitions)
boolean	isOpen()
boolean	isValid() Verify state of the index.
void	maybeRefreshBlocking() Refresh all partitions to make newly inserted data visible for readers.
void	open() Open index with all allocated partitions.
org.neo4j.graphdb.ResourceIterator<File>	snapshot() Snapshot of all file in all index partitions.

Methods inherited from class java.lang.Object

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

Constructor Detail

AbstractLuceneIndex

public AbstractLuceneIndex(PartitionedIndexStorage indexStorage,
                           IndexPartitionFactory partitionFactory)

Method Detail

create

public void create()
            throws IOException

Creates new index. As part of creation process index will allocate all required folders, index failure storage and will create its first partition.

Index creation do not automatically open it. To be able to use index please open it first.

Throws:

IOException

open

public void open()
          throws IOException

Open index with all allocated partitions.

Throws:

IOException

isOpen

public boolean isOpen()

exists

public boolean exists()
               throws IOException

Check lucene index existence within all allocated partitions.

Returns:

true if index exist in all partitions, false when index is empty or does not exist

Throws:

IOException

isValid

public boolean isValid()

Verify state of the index. If index is already open and in use method assume that index is valid since lucene already operating with it, otherwise necessary checks perform.

Returns:

true if lucene confirm that index is in valid clean state or index is already open.

drop

public void drop()
          throws IOException

Close index and deletes all it's partitions.

Throws:

IOException

flush

public void flush(boolean merge)
           throws IOException

Commits all index partitions.

Parameters:

merge - also merge all segments together. This should be done before reading term frequencies.

Throws:

IOException - on Lucene I/O error.

close

public void close()
           throws IOException

Throws:

IOException

allDocumentsReader

public LuceneAllDocumentsReader allDocumentsReader()

Creates an iterable over all documents in all partitions.

Returns:

LuceneAllDocumentsReader over all documents

snapshot

public org.neo4j.graphdb.ResourceIterator<File> snapshot()
                                                  throws IOException

Snapshot of all file in all index partitions.

Returns:

iterator over all index files.

Throws:

IOException

See Also:

WritableIndexSnapshotFileIterator

maybeRefreshBlocking

public void maybeRefreshBlocking()
                          throws IOException

Refresh all partitions to make newly inserted data visible for readers.

Throws:

IOException

getPartitions

public List<AbstractIndexPartition> getPartitions()

hasSinglePartition

public boolean hasSinglePartition(List<AbstractIndexPartition> partitions)

getFirstPartition

public AbstractIndexPartition getFirstPartition(List<AbstractIndexPartition> partitions)