org.apache.flink.formats.hadoop.bulk

接口 HadoopFileCommitter

所有已知实现类:

HadoopRenameFileCommitter

@Internal
public interface HadoopFileCommitter

The committer publishes an intermediate Hadoop file to the target path after it finishes writing.

方法概要

限定符和类型	方法和说明
void	commit() Commits the in-progress file to the target path.
void	commitAfterRecovery() Re-commits the in-progress file to the target path after fail-over.
org.apache.hadoop.fs.Path	getTargetFilePath() Gets the target path to commit to.
org.apache.hadoop.fs.Path	getTempFilePath() Gets the path of the intermediate file to commit.
void	preCommit() Prepares the intermediates file for committing.

方法详细资料

getTargetFilePath

org.apache.hadoop.fs.Path getTargetFilePath()

Gets the target path to commit to.

返回:

The target path to commit to.

getTempFilePath

org.apache.hadoop.fs.Path getTempFilePath()

Gets the path of the intermediate file to commit.

返回:

The path of the intermediate file to commit.

preCommit

void preCommit()
        throws IOException

Prepares the intermediates file for committing.

抛出:

IOException

commit

void commit()
     throws IOException

Commits the in-progress file to the target path.

抛出:

IOException

commitAfterRecovery

void commitAfterRecovery()
                  throws IOException

Re-commits the in-progress file to the target path after fail-over.

抛出:

IOException