Skip to content
Closed
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Original file line number Diff line number Diff line change
Expand Up @@ -92,6 +92,35 @@ abstract class FileCommitProtocol extends Logging {
*/
def newTaskTempFile(taskContext: TaskAttemptContext, dir: Option[String], ext: String): String

/**
* Notifies the commit protocol to add a new file, and gets back the full path that should be
* used. Must be called on the executors when running tasks.
*
* Note that the returned temp file may have an arbitrary path. The commit protocol only
* promises that the file will be at the location specified by the arguments after job commit.
*
* The "dir" parameter specifies the sub-directory within the base path, used to specify
* partitioning. The "spec" parameter specifies the file name. The rest are left to the commit
* protocol implementation to decide.
*
* Important: it is the caller's responsibility to add uniquely identifying content to "spec"
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is not true, commit protocol needs to guarantee name uniqueness, as caller side only gives prefix and suffix

Copy link
Contributor Author

@c21 c21 Jun 23, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The whole sentence is:

it is the caller's responsibility to add uniquely identifying content to "spec"
if a task is going to write out multiple files to the same dir.

I think this refers to the case when in one task, caller calls newTaskTempFile() multiple times with same spec, the caller should not expect commit protocol returning unique different file path every time. The current implementation of HadoopMapReduceCommitProtocol would return same path if ext being same. This is a copy-paste from original comment of newTaskTempFile.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I see!

* if a task is going to write out multiple files to the same dir. The file commit protocol only
* guarantees that files written by different tasks will not conflict.
*
* This API should be implemented and called, instead of
* [[newTaskTempFile(taskContest, dir, ext)]]. Provide a default implementation here to be
* backward compatible with custom [[FileCommitProtocol]] implementations before Spark 3.2.0.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think we should mark this as an API (e.g., @Unstable) ... this is currently internal.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It's a bit weird to say about "backward compatible" here. If this isn't an API, we should explicitly mention that this isn't an API at least here to avoid giving a false impression to dev.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@HyukjinKwon - will it look better as below?

This method should be implemented and called, instead of 
[[newTaskTempFile(taskContest, dir, ext)]]. Provide a default implementation here to be 
compatible with custom [[FileCommitProtocol]] implementations before Spark 3.2.0.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Shall we mark it as @Unstable for now with explaining the context a bit? e.g.) this class is exposed as an API considering the usage of many downstream custom implementations but will be subject to be changed and/or moved.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

cc @cloud-fan to get more feedback

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm fine to mark it Unstable. It doesn't hurt anyway.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sounds good. Let me create a PR now.

*/
def newTaskTempFile(
taskContext: TaskAttemptContext, dir: Option[String], spec: FileNameSpec): String = {
if (spec.prefix.isEmpty) {
newTaskTempFile(taskContext, dir, spec.suffix)
} else {
throw new UnsupportedOperationException(s"${getClass.getSimpleName}.newTaskTempFile does " +
s"not support file name prefix: ${spec.prefix}")
}
}

/**
* Similar to newTaskTempFile(), but allows files to committed to an absolute output location.
* Depending on the implementation, there may be weaker guarantees around adding files this way.
Expand All @@ -103,6 +132,34 @@ abstract class FileCommitProtocol extends Logging {
def newTaskTempFileAbsPath(
taskContext: TaskAttemptContext, absoluteDir: String, ext: String): String

/**
* Similar to newTaskTempFile(), but allows files to committed to an absolute output location.
* Depending on the implementation, there may be weaker guarantees around adding files this way.
*
* The "absoluteDir" parameter specifies the final absolute directory of file. The "spec"
* parameter specifies the file name. The rest are left to the commit protocol implementation to
* decide.
*
* Important: it is the caller's responsibility to add uniquely identifying content to "spec"
* if a task is going to write out multiple files to the same dir. The file commit protocol only
* guarantees that files written by different tasks will not conflict.
*
* This API should be implemented and called, instead of
* [[newTaskTempFileAbsPath(taskContest, absoluteDir, ext)]]. Provide a default implementation
* here to be backward compatible with custom [[FileCommitProtocol]] implementations before
* Spark 3.2.0.
*/
def newTaskTempFileAbsPath(
taskContext: TaskAttemptContext, absoluteDir: String, spec: FileNameSpec): String = {
if (spec.prefix.isEmpty) {
newTaskTempFileAbsPath(taskContext, absoluteDir, spec.suffix)
} else {
throw new UnsupportedOperationException(
s"${getClass.getSimpleName}.newTaskTempFileAbsPath does not support file name prefix: " +
s"${spec.prefix}")
}
}

/**
* Commits a task after the writes succeed. Must be called on the executors when running tasks.
*/
Expand Down Expand Up @@ -174,3 +231,12 @@ object FileCommitProtocol extends Logging {
new Path(path, ".spark-staging-" + jobId)
}
}

/**
* The specification for Spark output file name.
* This is used by [[FileCommitProtocol]] to create full path of file.
*
* @param prefix Prefix of file.
* @param suffix Suffix of file.
*/
final case class FileNameSpec(prefix: String, suffix: String)