Package org.apache.jackrabbit.core.query.lucene

Class CachingMultiIndexReader

java.lang.Object

org.apache.lucene.index.IndexReader

org.apache.lucene.index.MultiReader

org.apache.jackrabbit.core.query.lucene.CachingMultiIndexReader

All Implemented Interfaces:

Closeable, AutoCloseable, Cloneable, HierarchyResolver, MultiIndexReader, ReleaseableIndexReader

public final class CachingMultiIndexReader
extends MultiReader
implements HierarchyResolver, MultiIndexReader

Extends a MultiReader with support for cached TermDocs on FieldNames.UUID field.

Nested Class Summary

Nested classes/interfaces inherited from class org.apache.lucene.index.IndexReader

IndexReader.ReaderClosedListener

Field Summary

Fields inherited from class org.apache.lucene.index.MultiReader

starts

Fields inherited from class org.apache.lucene.index.IndexReader

hasChanges

Constructor Summary

Constructors

Constructor	Description
CachingMultiIndexReader(org.apache.jackrabbit.core.query.lucene.ReadOnlyIndexReader[] subReaders, org.apache.jackrabbit.core.query.lucene.DocNumberCache cache)	Creates a new CachingMultiIndexReader based on sub readers.

Method Summary

Modifier and Type	Method	Description
org.apache.jackrabbit.core.query.lucene.ForeignSegmentDocId	createDocId(NodeId id)	Creates a document id for the given node identifier.
protected void	doClose()
int	getDocumentNumber(org.apache.jackrabbit.core.query.lucene.ForeignSegmentDocId docId)	Returns the document number for the passed docId.
IndexReader[]	getIndexReaders()
org.apache.jackrabbit.core.query.lucene.DocId	getParentDocId(int n)	Returns the DocId of the parent of n or DocId.NULL if n does not have a parent (n is the root node).
int[]	getParents(int n, int[] docNumbers)	Returns the document number of the parent of n or an empty array if n does not have a parent (n is the root node).
void	release()	Releases this index reader and potentially frees resources.
TermDocs	termDocs(Term term)

Methods inherited from class org.apache.lucene.index.MultiReader

clone, clone, docFreq, doCommit, document, doDelete, doOpenIfChanged, doOpenIfChanged, doSetNorm, doUndeleteAll, getFieldInfos, getSequentialSubReaders, getTermFreqVector, getTermFreqVector, getTermFreqVector, getTermFreqVectors, getVersion, hasDeletions, hasNorms, isCurrent, isDeleted, isOptimized, maxDoc, norms, norms, numDocs, readerIndex, termDocs, termPositions, terms, terms

Methods inherited from class org.apache.lucene.index.IndexReader

acquireWriteLock, addReaderClosedListener, close, commit, commit, decRef, deleteDocument, deleteDocuments, directory, document, doOpenIfChanged, doOpenIfChanged, ensureOpen, flush, flush, getCommitUserData, getCommitUserData, getCoreCacheKey, getCurrentVersion, getDeletesCacheKey, getIndexCommit, getRefCount, getTermInfosIndexDivisor, getUniqueTermCount, incRef, indexExists, lastModified, listCommits, numDeletedDocs, open, open, open, open, open, open, open, open, open, open, open, openIfChanged, openIfChanged, openIfChanged, openIfChanged, removeReaderClosedListener, reopen, reopen, reopen, reopen, setNorm, setNorm, termPositions, toString, tryIncRef, undeleteAll

Methods inherited from class java.lang.Object

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

Constructor Detail

CachingMultiIndexReader

public CachingMultiIndexReader(org.apache.jackrabbit.core.query.lucene.ReadOnlyIndexReader[] subReaders,
                               org.apache.jackrabbit.core.query.lucene.DocNumberCache cache)

Creates a new CachingMultiIndexReader based on sub readers.

Parameters:

subReaders - the sub readers.

cache - the document number cache.

Method Detail

getParents

public int[] getParents(int n,
                        int[] docNumbers)
                 throws IOException

Returns the document number of the parent of n or an empty array if n does not have a parent (n is the root node).

Specified by:

getParents in interface HierarchyResolver

Parameters:

n - the document number.

docNumbers - an array for reuse. An implementation should use the passed array as a container for the return value, unless the length of the returned array is different from docNumbers. In which case an implementation will create a new array with an appropriate size.

Returns:

the document number of n's parent.

Throws:

IOException - if an error occurs while reading from the index.

getParentDocId

public org.apache.jackrabbit.core.query.lucene.DocId getParentDocId(int n)
                                                             throws IOException

Returns the DocId of the parent of n or DocId.NULL if n does not have a parent (n is the root node).

Parameters:

n - the document number.

Returns:

the DocId of n's parent.

Throws:

IOException - if an error occurs while reading from the index.

termDocs

public TermDocs termDocs(Term term)
                  throws IOException

Overrides:

termDocs in class MultiReader

Throws:

IOException

release

public final void release()
                   throws IOException

Releases this index reader and potentially frees resources. In contrast to IndexReader.close() this method does not necessarily close the index reader, but gives the implementation the opportunity to do reference counting.

Specified by:

release in interface ReleaseableIndexReader

Throws:

IOException - if an error occurs while releasing the index reader.

doClose

protected void doClose()
                throws IOException

Overrides:

doClose in class MultiReader

Throws:

IOException

getIndexReaders

public IndexReader[] getIndexReaders()

Specified by:

getIndexReaders in interface MultiIndexReader

Returns:

the IndexReaders that are contained in this MultiIndexReader.

createDocId

public org.apache.jackrabbit.core.query.lucene.ForeignSegmentDocId createDocId(NodeId id)
                                                                        throws IOException

Creates a document id for the given node identifier.

Specified by:

createDocId in interface MultiIndexReader

Parameters:

id - the id of the node.

Returns:

a foreign segment doc id or null if there is no node with the given id.

Throws:

IOException - if an error occurs while reading from the index.

getDocumentNumber

public int getDocumentNumber(org.apache.jackrabbit.core.query.lucene.ForeignSegmentDocId docId)

Returns the document number for the passed docId. If the id is invalid -1 is returned.

Specified by:

getDocumentNumber in interface MultiIndexReader

Parameters:

docId - the document id to resolve.

Returns:

the document number or -1 if it is invalid (e.g. does not exist).