it.unimi.di.mg4j.search
Class AbstractCompositeDocumentIterator

java.lang.Object
  it.unimi.di.mg4j.search.AbstractDocumentIterator
      it.unimi.di.mg4j.search.AbstractIntervalDocumentIterator
          it.unimi.di.mg4j.search.AbstractCompositeDocumentIterator

All Implemented Interfaces:

DocumentIterator

Direct Known Subclasses:

AbstractIntersectionDocumentIterator, AbstractUnionDocumentIterator

public abstract class AbstractCompositeDocumentIterator
extends AbstractIntervalDocumentIterator
implements DocumentIterator

An abstract iterator on documents based on a list of component iterators.

The AbstractCompositeDocumentIterator(Index, Object, DocumentIterator...) caches into documentIterator the component iterators, and sets up a number of protected fields that can be useful to implementors. It also provide abstract member classes that make it easier to implement interval iterators.

Note that this class implements both accept(DocumentIteratorVisitor) and acceptOnTruePaths(DocumentIteratorVisitor) with a series of recursive calls on all component iterator. If you desire a different behaviour for acceptOnTruePaths(DocumentIteratorVisitor) (see, e.g., AbstractUnionDocumentIterator, please override it.

Nested Class Summary

protected static class	AbstractCompositeDocumentIterator.AbstractCompositeIndexIntervalIterator An abstract IndexIterator-based interval iterator.
protected static class	AbstractCompositeDocumentIterator.AbstractCompositeIntervalIterator An abstract interval iterator.

Field Summary

DocumentIterator[]	documentIterator The component document iterators.
IndexIterator[]	indexIterator A cached copy of documentIterator, if all underlying iterators are index iterators; null, otherwise.
protected int	indexIteratorsWithoutPositions If indexIterator is not null, the number of index iterators without positions.
int	n The number of component iterators.

Fields inherited from class it.unimi.di.mg4j.search.AbstractIntervalDocumentIterator

currentIterators, indices, intervalIterators, soleIndex, soleIntervalIterator, unmodifiableCurrentIterators

Fields inherited from class it.unimi.di.mg4j.search.AbstractDocumentIterator

curr, weight

Fields inherited from interface it.unimi.di.mg4j.search.DocumentIterator

END_OF_LIST

Constructor Summary

protected	AbstractCompositeDocumentIterator(DocumentIterator... documentIterator) Creates a new composite document iterator using a given list of component document iterators.
protected	AbstractCompositeDocumentIterator(Index index, Object arg, DocumentIterator... documentIterator) Creates a new composite document iterator using a given list of component document iterators and a specified index.

Method Summary

<T> T	accept(DocumentIteratorVisitor<T> visitor) Accepts a visitor.
<T> T	acceptOnTruePaths(DocumentIteratorVisitor<T> visitor) Accepts a visitor after a call to DocumentIterator.nextDocument(), limiting recursion to true paths.
void	dispose() Disposes this document iterator, releasing all resources.
String	toString()

Methods inherited from class it.unimi.di.mg4j.search.AbstractIntervalDocumentIterator

allIndexIterators, getIntervalIterator, indices, indices

Methods inherited from class it.unimi.di.mg4j.search.AbstractDocumentIterator

document, ensureOnADocument, mayHaveNext, weight, weight

Methods inherited from class java.lang.Object

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

Methods inherited from interface it.unimi.di.mg4j.search.DocumentIterator

document, indices, intervalIterator, intervalIterator, intervalIterators, mayHaveNext, nextDocument, skipTo, weight, weight

Field Detail

n

public final int n

The number of component iterators.

documentIterator

public final DocumentIterator[] documentIterator

The component document iterators.

indexIterator

public final IndexIterator[] indexIterator

A cached copy of documentIterator, if all underlying iterators are index iterators; null, otherwise.

indexIteratorsWithoutPositions

protected int indexIteratorsWithoutPositions

If indexIterator is not null, the number of index iterators without positions.

Constructor Detail

AbstractCompositeDocumentIterator

protected AbstractCompositeDocumentIterator(Index index,
                                            Object arg,
                                            DocumentIterator... documentIterator)

Creates a new composite document iterator using a given list of component document iterators and a specified index.

Parameters:

index - an index that will constitute the only index for which this iterator will return intervals, or null to require the computation of the set of indices as the union of the indices of all component iterators.

arg - an argument that will be passed to AbstractIntervalDocumentIterator.getIntervalIterator(Index, int, boolean, Object).

documentIterator - the component iterators.

AbstractCompositeDocumentIterator

protected AbstractCompositeDocumentIterator(DocumentIterator... documentIterator)

Creates a new composite document iterator using a given list of component document iterators.

Parameters:

documentIterator - the component iterators.

Method Detail

accept

public <T> T accept(DocumentIteratorVisitor<T> visitor)
         throws IOException

Description copied from interface: DocumentIterator

Accepts a visitor.

A document iterator is usually structured as composite, with operators as internal nodes and IndexIterators as leaves. This method implements the visitor pattern.

Specified by:

accept in interface DocumentIterator

Parameters:

visitor - the visitor.

Returns:

an object resulting from the visit, or null if the visit was interrupted.

Throws:

IOException

acceptOnTruePaths

public <T> T acceptOnTruePaths(DocumentIteratorVisitor<T> visitor)
                    throws IOException

Description copied from interface: DocumentIterator

Accepts a visitor after a call to DocumentIterator.nextDocument(), limiting recursion to true paths.

After a call to DocumentIterator.nextDocument(), a document iterator is positioned over a document. This call is equivalent to DocumentIterator.accept(DocumentIteratorVisitor), but visits only along true paths.

We define a true path as a path from the root of the composite that passes only through nodes whose associated subtree is positioned on the same document of the root. Note that OrDocumentIterators detach exhausted iterators from the composite tree, so true paths define the subtree that is causing the current document to satisfy the query represented by this document iterator.

For more elaboration, and the main application of this method, see CounterCollectionVisitor.

Specified by:

acceptOnTruePaths in interface DocumentIterator

Parameters:

visitor - the visitor.

Returns:

an object resulting from the visit, or null if the visit was interrupted.

Throws:

IOException

See Also:

DocumentIterator.accept(DocumentIteratorVisitor), CounterCollectionVisitor

dispose

public void dispose()
             throws IOException

Description copied from interface: DocumentIterator

Disposes this document iterator, releasing all resources.

This method should propagate down to the underlying index iterators, where it should release resources such as open files and network connections. If you're doing your own resource tracking and pooling, then you do not need to call this method.

Specified by:

dispose in interface DocumentIterator

Throws:

IOException

toString

public String toString()

Overrides:

toString in class Object