Package org.apache.hadoop.hdfs.server.namenode

Class FSEditLogOp.Reader

java.lang.Object

org.apache.hadoop.hdfs.server.namenode.FSEditLogOp.Reader

Enclosing class:

FSEditLogOp

public abstract static class FSEditLogOp.Reader
extends java.lang.Object

Class for reading editlog ops from a stream

Method Summary

Modifier and Type	Method	Description
static FSEditLogOp.Reader	create(java.io.DataInputStream in, org.apache.hadoop.hdfs.server.namenode.StreamLimiter limiter, int logVersion)
abstract FSEditLogOp	decodeOp()	Read an opcode from the input stream.
FSEditLogOp	readOp(boolean skipBrokenEdits)	Read an operation from the input stream.
abstract long	scanOp()	Similar to decodeOp(), but we only retrieve the transaction ID of the Op rather than reading it.
void	setMaxOpSize(int maxOpSize)

Methods inherited from class java.lang.Object

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

Method Detail

create

public static FSEditLogOp.Reader create(java.io.DataInputStream in,
                                        org.apache.hadoop.hdfs.server.namenode.StreamLimiter limiter,
                                        int logVersion)

setMaxOpSize

public void setMaxOpSize(int maxOpSize)

readOp

public FSEditLogOp readOp(boolean skipBrokenEdits)
                   throws java.io.IOException

Read an operation from the input stream. Note that the objects returned from this method may be re-used by future calls to the same method.

Parameters:

skipBrokenEdits - If true, attempt to skip over damaged parts of the input stream, rather than throwing an IOException

Returns:

the operation read from the stream, or null at the end of the file

Throws:

java.io.IOException - on error. This function should only throw an exception when skipBrokenEdits is false.

decodeOp

public abstract FSEditLogOp decodeOp()
                              throws java.io.IOException

Read an opcode from the input stream.

Returns:

the opcode, or null on EOF. If an exception is thrown, the stream's mark will be set to the first problematic byte. This usually means the beginning of the opcode.

Throws:

java.io.IOException

scanOp

public abstract long scanOp()
                     throws java.io.IOException

Similar to decodeOp(), but we only retrieve the transaction ID of the Op rather than reading it. If the edit log format supports length prefixing, this can be much faster than full decoding.

Returns:

the last txid of the segment, or INVALID_TXID on EOF.

Throws:

java.io.IOException