htsjdk.samtools

Class SBIIndexWriter

java.lang.Object

htsjdk.samtools.SBIIndexWriter

public final class SBIIndexWriter
extends java.lang.Object

Writes SBI files as understood by SBIIndex.

To use this class, first construct an instance from an output stream, and a desired granularity. Then for each record in the file being indexed, pass the virtual file offset of the record to the processRecord(long) method. The indexer will keep a count of the records passed in an index every nth record. When there are no records left call finish(long, long) to complete writing the index.

Field Summary

Fields

Modifier and Type	Field and Description
static long	DEFAULT_GRANULARITY

Constructor Summary

Constructors

Constructor and Description
SBIIndexWriter(java.io.OutputStream out) Prepare to write an SBI index with the default granularity.
SBIIndexWriter(java.io.OutputStream out, long granularity) Prepare to write an SBI index.

Method Summary

Modifier and Type	Method and Description
void	finish(long finalVirtualOffset, long dataFileLength) Complete the index, and close the output stream.
void	finish(long finalVirtualOffset, long dataFileLength, byte[] md5, byte[] uuid) Complete the index, and close the output stream.
void	processRecord(long virtualOffset) Process a record for the index: the offset of every nth record will be written to the index.

Methods inherited from class java.lang.Object

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

Field Detail

DEFAULT_GRANULARITY

public static final long DEFAULT_GRANULARITY

See Also:

Constant Field Values

Constructor Detail

SBIIndexWriter

public SBIIndexWriter(java.io.OutputStream out)

Prepare to write an SBI index with the default granularity.

Parameters:

out - the stream to write the index to

SBIIndexWriter

public SBIIndexWriter(java.io.OutputStream out,
                      long granularity)

Prepare to write an SBI index.

Parameters:

out - the stream to write the index to

granularity - write the offset of every nth record to the index

Method Detail

processRecord

public void processRecord(long virtualOffset)

Process a record for the index: the offset of every nth record will be written to the index.

Parameters:

virtualOffset - virtual file pointer of the record

finish

public void finish(long finalVirtualOffset,
                   long dataFileLength)

Complete the index, and close the output stream.

Parameters:

finalVirtualOffset - the virtual offset at which the next record would start if it were added to the file

dataFileLength - the length of the data file in bytes

finish

public void finish(long finalVirtualOffset,
                   long dataFileLength,
                   byte[] md5,
                   byte[] uuid)

Complete the index, and close the output stream.

Parameters:

finalVirtualOffset - the virtual offset at which the next record would start if it were added to the file

dataFileLength - the length of the data file in bytes

md5 - the MD5 hash of the data file, or null if not specified

uuid - the UUID for the data file, or null if not specified