Skip to content

[SPARK-21440][SQL][PYSPARK] Refactor ArrowConverters and add ArrayType and StructType support. #18655

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
ueshin wants to merge 17 commits into from
Closed

[SPARK-21440][SQL][PYSPARK] Refactor ArrowConverters and add ArrayType and StructType support. #18655

Changes from 1 commit
Commits
Show all changes
17 commits
Select commit Hold shift + click to select a range
1439fe6
Introduce ArrowWriter and ArrowColumnVector.
ueshin Jul 13, 2017
6fcf700
Use ArrowWriter for ArrowConverters.
ueshin Jul 14, 2017
579def2
Refactor ArrowConverters.
ueshin Jul 14, 2017
58cd465
Move releasing memory into task completion listener.
ueshin Jul 17, 2017
8ffedda
Revert ArrowColumnVector related implementations to split a pr into m…
ueshin Jul 18, 2017
e3a4fc0
Use rowCount instead of rowId.
ueshin Jul 19, 2017
b5988f9
Move files back to arrow package.
ueshin Jul 19, 2017
a50a271
Modify ArrowUtils to avoid deprecated APIs.
ueshin Jul 19, 2017
7084b38
Modify to close resources also immediately after row iterator is cons…
ueshin Jul 20, 2017
53fba41
Merge branch 'master' into issues/SPARK-21440
ueshin Jul 20, 2017
6fc4da0
Add ArrowWriterSuite.
ueshin Jul 20, 2017
926a624
Merge branch 'master' into issues/SPARK-21440
ueshin Jul 26, 2017
5bbb46f
Remove DecimalType support for now.
ueshin Jul 26, 2017
beff6ef
Fix skip() for StructType.
ueshin Jul 26, 2017
19f3973
Address comments.
ueshin Jul 26, 2017
0bac10d
Modify skip semantic.
ueshin Jul 27, 2017
b85dc23
Inline writeNull().
ueshin Jul 27, 2017
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
Prev Previous commit
Next Next commit
Modify to close resources also immediately after row iterator is cons… 
…umed.
  • Loading branch information
@ueshin
ueshin committed Jul 20, 2017
commit 7084b388d87c8347b79898827658d7827bf5649d
15 changes: 12 additions & 3 deletions sql/core/src/main/scala/org/apache/spark/sql/execution/arrow/ArrowConverters.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -69,14 +69,23 @@ private[sql] object ArrowConverters {
val root = VectorSchemaRoot.create(arrowSchema, allocator)
val arrowWriter = ArrowWriter.create(root)

var closed = false

context.addTaskCompletionListener { _ =>
root.close()
allocator.close()
if (!closed) {
Copy link
Contributor

@cloud-fan cloud-fan Jul 26, 2017

Choose a reason for hiding this comment

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

do we really need this? I think it's ok to close twice?

Copy link
Member Author

@ueshin ueshin Jul 26, 2017

Choose a reason for hiding this comment

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

The allocator can be closed twice, but the root throws an exception after allocator is closed.

Copy link
Contributor

@cloud-fan cloud-fan Jul 26, 2017

Choose a reason for hiding this comment

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

is this a bug in arrow? cc @BryanCutler

Copy link
Member

@BryanCutler BryanCutler Jul 26, 2017

Choose a reason for hiding this comment

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

The root just releases the buffers from the FieldVectors, so I would think it should be able to handle being closed twice. I'll check tomorrow if seems reasonable.

Copy link
Member

@BryanCutler BryanCutler Jul 27, 2017

Choose a reason for hiding this comment

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

I filed https://issues.apache.org/jira/browse/ARROW-1283 to fix this. For now, it looks like we need this.

root.close()
allocator.close()
}
}

new Iterator[ArrowPayload] {

override def hasNext: Boolean = rowIter.hasNext
override def hasNext: Boolean = rowIter.hasNext || {
root.close()
allocator.close()
closed = true
false
}

override def next(): ArrowPayload = {
val out = new ByteArrayOutputStream()
Expand Down