Package org.apache.cassandra.index.sai.disk.v1.segment

Class SegmentMetadata

java.lang.Object

org.apache.cassandra.index.sai.disk.v1.segment.SegmentMetadata

public class SegmentMetadata
extends Object

Multiple SegmentMetadata are stored in IndexComponent.META file, each corresponds to an on-disk index segment.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	SegmentMetadata.ComponentMetadata
static class	SegmentMetadata.ComponentMetadataMap

Field Summary

Fields

Modifier and Type	Field	Description
final SegmentMetadata.ComponentMetadataMap	componentMetadatas	Root, offset, length for each index structure in the segment.
final PrimaryKey	maxKey
final long	maxSSTableRowId
final ByteBuffer	maxTerm
final PrimaryKey	minKey	Ordered by their token position in current segment
final long	minSSTableRowId	Min and max sstable rowId in current segment.
final ByteBuffer	minTerm	Minimum and maximum indexed column value ordered by its AbstractType.
final long	numRows	number of indexed rows (aka.
final long	rowIdOffset	Used to retrieve sstableRowId which equals to offset plus segmentRowId.

Constructor Summary

Constructors

Constructor	Description
SegmentMetadata(long rowIdOffset, long numRows, long minSSTableRowId, long maxSSTableRowId, PrimaryKey minKey, PrimaryKey maxKey, ByteBuffer minTerm, ByteBuffer maxTerm, SegmentMetadata.ComponentMetadataMap componentMetadatas)

Method Summary

Modifier and Type	Method	Description
static List<SegmentMetadata>	load(MetadataSource source, PrimaryKey.Factory primaryKeyFactory)
int	toSegmentRowId(long sstableRowId)
String	toString()
static void	write(MetadataWriter writer, List<SegmentMetadata> segments)	Writes disk metadata for the given segment list.

Methods inherited from class java.lang.Object

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

Field Details

rowIdOffset

public final long rowIdOffset

Used to retrieve sstableRowId which equals to offset plus segmentRowId.

minSSTableRowId

public final long minSSTableRowId

Min and max sstable rowId in current segment.

For index generated by compaction, minSSTableRowId is the same as segmentRowIdOffset. But for flush, segmentRowIdOffset is taken from previous segment's maxSSTableRowId.

maxSSTableRowId

public final long maxSSTableRowId

numRows

public final long numRows

number of indexed rows (aka. a pair of term and segmentRowId) in the current segment

minKey

public final PrimaryKey minKey

Ordered by their token position in current segment

maxKey

public final PrimaryKey maxKey

minTerm

public final ByteBuffer minTerm

Minimum and maximum indexed column value ordered by its AbstractType.

maxTerm

public final ByteBuffer maxTerm

componentMetadatas

public final SegmentMetadata.ComponentMetadataMap componentMetadatas

Root, offset, length for each index structure in the segment.

Note: postings block offsets are stored in terms dictionary, no need to worry about its root.

Constructor Details

SegmentMetadata

public SegmentMetadata(long rowIdOffset,
 long numRows,
 long minSSTableRowId,
 long maxSSTableRowId,
 PrimaryKey minKey,
 PrimaryKey maxKey,
 ByteBuffer minTerm,
 ByteBuffer maxTerm,
 SegmentMetadata.ComponentMetadataMap componentMetadatas)

Method Details

toSegmentRowId

public int toSegmentRowId(long sstableRowId)

load

public static List<SegmentMetadata> load(MetadataSource source,
 PrimaryKey.Factory primaryKeyFactory)
                                  throws IOException

Throws:

IOException

write

public static void write(MetadataWriter writer,
 List<SegmentMetadata> segments)
                  throws IOException

Writes disk metadata for the given segment list.

Throws:

IOException

toString

public String toString()

Overrides:

toString in class Object