org.apache.lucene.search
Class FieldComparator<T>

java.lang.Object
  extended by org.apache.lucene.search.FieldComparator<T>
Direct Known Subclasses:
FieldComparator.ByteComparator, FieldComparator.DocComparator, FieldComparator.DoubleComparator, FieldComparator.FloatComparator, FieldComparator.IntComparator, FieldComparator.LongComparator, FieldComparator.RelevanceComparator, FieldComparator.ShortComparator, FieldComparator.StringComparatorLocale, FieldComparator.StringOrdValComparator, FieldComparator.StringValComparator

public abstract class FieldComparator<T>
extends Object

Expert: a FieldComparator compares hits so as to determine their sort order when collecting the top results with TopFieldCollector. The concrete public FieldComparator classes here correspond to the SortField types.

This API is designed to achieve high performance sorting, by exposing a tight interaction with FieldValueHitQueue as it visits hits. Whenever a hit is competitive, it's enrolled into a virtual slot, which is an int ranging from 0 to numHits-1. The FieldComparator is made aware of segment transitions during searching in case any internal state it's tracking needs to be recomputed during these transitions.

A comparator must define these functions:

WARNING: This API is experimental and might change in incompatible ways in the next release.

Nested Class Summary
static class FieldComparator.ByteComparator
          Parses field's values as byte (using FieldCache.getBytes(org.apache.lucene.index.IndexReader, java.lang.String) and sorts by ascending value
static class FieldComparator.DocComparator
          Sorts by ascending docID
static class FieldComparator.DoubleComparator
          Parses field's values as double (using FieldCache.getDoubles(org.apache.lucene.index.IndexReader, java.lang.String) and sorts by ascending value
static class FieldComparator.FloatComparator
          Parses field's values as float (using FieldCache.getFloats(org.apache.lucene.index.IndexReader, java.lang.String) and sorts by ascending value
static class FieldComparator.IntComparator
          Parses field's values as int (using FieldCache.getInts(org.apache.lucene.index.IndexReader, java.lang.String) and sorts by ascending value
static class FieldComparator.LongComparator
          Parses field's values as long (using FieldCache.getLongs(org.apache.lucene.index.IndexReader, java.lang.String) and sorts by ascending value
static class FieldComparator.RelevanceComparator
          Sorts by descending relevance.
static class FieldComparator.ShortComparator
          Parses field's values as short (using FieldCache.getShorts(org.apache.lucene.index.IndexReader, java.lang.String) and sorts by ascending value
static class FieldComparator.StringComparatorLocale
          Sorts by a field's value using the Collator for a given Locale.
static class FieldComparator.StringOrdValComparator
          Sorts by field's natural String sort order, using ordinals.
static class FieldComparator.StringValComparator
          Sorts by field's natural String sort order.
 
Field Summary
protected  T missingValue
           
 
Constructor Summary
FieldComparator()
           
 
Method Summary
protected static int binarySearch(String[] a, String key)
           
protected static int binarySearch(String[] a, String key, int low, int high)
           
abstract  int compare(int slot1, int slot2)
          Compare hit at slot1 with hit at slot2.
abstract  int compareBottom(int doc)
          Compare the bottom of the queue with doc.
 int compareValues(T first, T second)
          Returns -1 if first is less than second.
abstract  void copy(int slot, int doc)
          This method is called when a new hit is competitive.
abstract  void setBottom(int slot)
          Set the bottom slot, ie the "weakest" (sorted last) entry in the queue.
 FieldComparator<T> setMissingValue(T missingValue)
          Set a default sorting value for documents which lacks one
abstract  void setNextReader(IndexReader reader, int docBase)
          Set a new Reader.
 void setScorer(Scorer scorer)
          Sets the Scorer to use in case a document's score is needed.
abstract  T value(int slot)
          Return the actual value in the slot.
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Field Detail

missingValue

protected T missingValue
Constructor Detail

FieldComparator

public FieldComparator()
Method Detail

setMissingValue

public FieldComparator<T> setMissingValue(T missingValue)
Set a default sorting value for documents which lacks one


compare

public abstract int compare(int slot1,
                            int slot2)
Compare hit at slot1 with hit at slot2.

Parameters:
slot1 - first slot to compare
slot2 - second slot to compare
Returns:
any N < 0 if slot2's value is sorted after slot1, any N > 0 if the slot2's value is sorted before slot1 and 0 if they are equal

setBottom

public abstract void setBottom(int slot)
Set the bottom slot, ie the "weakest" (sorted last) entry in the queue. When compareBottom(int) is called, you should compare against this slot. This will always be called before compareBottom(int).

Parameters:
slot - the currently weakest (sorted last) slot in the queue

compareBottom

public abstract int compareBottom(int doc)
                           throws IOException
Compare the bottom of the queue with doc. This will only invoked after setBottom has been called. This should return the same result as compare(int,int)} as if bottom were slot1 and the new document were slot 2.

For a search that hits many results, this method will be the hotspot (invoked by far the most frequently).

Parameters:
doc - that was hit
Returns:
any N < 0 if the doc's value is sorted after the bottom entry (not competitive), any N > 0 if the doc's value is sorted before the bottom entry and 0 if they are equal.
Throws:
IOException

copy

public abstract void copy(int slot,
                          int doc)
                   throws IOException
This method is called when a new hit is competitive. You should copy any state associated with this document that will be required for future comparisons, into the specified slot.

Parameters:
slot - which slot to copy the hit to
doc - docID relative to current reader
Throws:
IOException

setNextReader

public abstract void setNextReader(IndexReader reader,
                                   int docBase)
                            throws IOException
Set a new Reader. All doc correspond to the current Reader.

Parameters:
reader - current reader
docBase - docBase of this reader
Throws:
IOException
IOException

setScorer

public void setScorer(Scorer scorer)
Sets the Scorer to use in case a document's score is needed.

Parameters:
scorer - Scorer instance that you should use to obtain the current hit's score, if necessary.

value

public abstract T value(int slot)
Return the actual value in the slot.

Parameters:
slot - the value
Returns:
value in this slot

compareValues

public int compareValues(T first,
                         T second)
Returns -1 if first is less than second. Default impl to assume the type implements Comparable and invoke .compareTo; be sure to override this method if your FieldComparator's type isn't a Comparable or if your values may sometimes be null


binarySearch

protected static final int binarySearch(String[] a,
                                        String key)

binarySearch

protected static final int binarySearch(String[] a,
                                        String key,
                                        int low,
                                        int high)


Copyright © 2000-2011 Apache Software Foundation. All Rights Reserved.