org.apache.lucene.index
Class SegmentReadState

java.lang.Object
  org.apache.lucene.index.SegmentReadState

public class SegmentReadState
extends Object

Holder class for common parameters used during read.

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

Field Summary

IOContext	context IOContext to pass to Directory.openInput(String,IOContext).
Directory	directory Directory where this segment is read from.
FieldInfos	fieldInfos FieldInfos describing all fields in this segment.
SegmentInfo	segmentInfo SegmentInfo describing this segment.
String	segmentSuffix Unique suffix for any postings files read for this segment.
int	termsIndexDivisor The termInfosIndexDivisor to use, if appropriate (not all PostingsFormats support it; in particular the current default does not).

Constructor Summary

SegmentReadState(Directory dir, SegmentInfo info, FieldInfos fieldInfos, IOContext context, int termsIndexDivisor)
Create a SegmentReadState.

SegmentReadState(Directory dir, SegmentInfo info, FieldInfos fieldInfos, IOContext context, int termsIndexDivisor, String segmentSuffix)
Create a SegmentReadState.

SegmentReadState(SegmentReadState other, String newSegmentSuffix)
Create a SegmentReadState.

Method Summary

Methods inherited from class java.lang.Object

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

Field Detail

directory

public final Directory directory

Directory where this segment is read from.

segmentInfo

public final SegmentInfo segmentInfo

SegmentInfo describing this segment.

fieldInfos

public final FieldInfos fieldInfos

FieldInfos describing all fields in this segment.

context

public final IOContext context

IOContext to pass to Directory.openInput(String,IOContext).

termsIndexDivisor

public int termsIndexDivisor

The termInfosIndexDivisor to use, if appropriate (not all PostingsFormats support it; in particular the current default does not).

NOTE: if this is < 0, that means "defer terms index load until needed". But if the codec must load the terms index on init (preflex is the only once currently that must do so), then it should negate this value to get the app's terms divisor

segmentSuffix

public final String segmentSuffix

Unique suffix for any postings files read for this segment. PerFieldPostingsFormat sets this for each of the postings formats it wraps. If you create a new PostingsFormat then any files you write/read must be derived using this suffix (use IndexFileNames.segmentFileName(String,String,String)).

Constructor Detail

SegmentReadState

public SegmentReadState(Directory dir,
                        SegmentInfo info,
                        FieldInfos fieldInfos,
                        IOContext context,
                        int termsIndexDivisor)

Create a SegmentReadState.

SegmentReadState

public SegmentReadState(Directory dir,
                        SegmentInfo info,
                        FieldInfos fieldInfos,
                        IOContext context,
                        int termsIndexDivisor,
                        String segmentSuffix)

Create a SegmentReadState.

SegmentReadState

public SegmentReadState(SegmentReadState other,
                        String newSegmentSuffix)

Create a SegmentReadState.