Package org.apache.lucene.index.sorter

Class Sorter

java.lang.Object

org.apache.lucene.index.sorter.Sorter

Direct Known Subclasses:

BlockJoinSorter, NumericDocValuesSorter

public abstract class Sorter
extends Object

Sorts documents of a given index by returning a permutation on the document IDs.

NOTE: A Sorter implementation can be easily written from a document comparator by using the sort(int, DocComparator) helper method. This is especially useful when documents are directly comparable by their field values.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	Sorter.DocComparator	A comparator of doc IDs.
static class	Sorter.DocMap	A permutation of doc IDs.

Field Summary

Fields

Modifier and Type	Field	Description
static final Sorter.DocComparator	INDEX_ORDER_COMPARATOR	A comparator that keeps documents in index order.
static final Sorter	REVERSE_DOCS	Sorts documents in reverse order.

Constructor Summary

Constructors

Constructor	Description
Sorter()

Method Summary

Modifier and Type	Method	Description
abstract String	getID()	Returns the identifier of this Sorter.
abstract Sorter.DocMap	sort(AtomicReader reader)	Returns a mapping from the old document ID to its new location in the sorted index.
String	toString()

Methods inherited from class java.lang.Object

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

Field Details

INDEX_ORDER_COMPARATOR

public static final Sorter.DocComparator INDEX_ORDER_COMPARATOR

A comparator that keeps documents in index order.

REVERSE_DOCS

public static final Sorter REVERSE_DOCS

Sorts documents in reverse order. NOTE: This Sorter is not idempotent. Sorting an AtomicReader once or twice will return two different AtomicReader views. This Sorter should not be used with SortingMergePolicy.

Constructor Details

Sorter

public Sorter()

Method Details

sort

public abstract Sorter.DocMap sort(AtomicReader reader)
                            throws IOException

Returns a mapping from the old document ID to its new location in the sorted index. Implementations can use the auxiliary sort(int, DocComparator) to compute the old-to-new permutation given a list of documents and their corresponding values.

A return value of null is allowed and means that reader is already sorted.

NOTE: deleted documents are expected to appear in the mapping as well, they will however be marked as deleted in the sorted view.

Throws:

IOException

getID

public abstract String getID()

Returns the identifier of this Sorter.

This identifier is similar to Object.hashCode() and should be chosen so that two instances of this class that sort documents likewise will have the same identifier. On the contrary, this identifier should be different on different sorters.

toString

public String toString()

Overrides:

toString in class Object