[SPARK-41108][SPARK-41005][CONNECT][FOLLOW-UP] Deduplicate ArrowConverters codes #38618
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
cloud-fan
approved these changes
Nov 11, 2022
HyukjinKwon
force-pushed
the
SPARK-41108-followup-1
branch
from
22ecb3f to
cf2de76
Compare
zhengruifeng
approved these changes
Nov 11, 2022
Member
Author
|
Merged to master. |
hvanhovell
reviewed
Nov 16, 2022
| extends ArrowBatchIterator( | ||
| rowIter, schema, maxRecordsPerBatch, timeZoneId, context) { | ||
|
|
||
| private val arrowSchemaSize = SizeEstimator.estimate(arrowSchema) |
Contributor
There was a problem hiding this comment.
What is this supposed to achieve? This uses a lot of reflective code to figure out the size of the schema object. How is this related to the size of the batch?
Member
Author
There was a problem hiding this comment.
@hvanhovell, this PR is virtually pure refactoring except the couple of points I mentioned in the PR description. For the question, it came from #38612 to estimate the size of the batch before creating an Arrow batch.
hvanhovell
reviewed
Nov 16, 2022
| rowIter, schema, maxRecordsPerBatch, timeZoneId, context) { | ||
|
|
||
| private val arrowSchemaSize = SizeEstimator.estimate(arrowSchema) | ||
| var rowCountInLastBatch: Long = 0 |
Contributor
There was a problem hiding this comment.
A couple of questions. Why do we need the rowcount? It is already encoded in the batch itself. If we do need the rowcount, please make the iterator return it in the next call instead relying on a side effect.
Member
Author
There was a problem hiding this comment.
This logic is also from #38468, and this PR is a followup. The return type here is Array[Byte] that is raw binary record batch. So we cannot get the count from that unless we define other case classes to keep the row count. This class is private that is only used in the specific case.
SandishKumarHN
pushed a commit
to SandishKumarHN/spark
that referenced
this pull request
Dec 12, 2022
…rters codes ### What changes were proposed in this pull request? This PR is a followup of both apache#38468 and apache#38612 that proposes to deduplicate codes in `ArrowConverters` by creating two classes `ArrowBatchIterator` and `ArrowBatchWithSchemaIterator`. In addition, we reuse `ArrowBatchWithSchemaIterator` when creating an empty Arrow batch at `createEmptyArrowBatch`. While I am here, - I addressed my own comment at apache#38612 (comment) - Kept the support of both max rows and size. Max row size check was removed in apache#38612 ### Why are the changes needed? For better readability and maintenance. ### Does this PR introduce _any_ user-facing change? No, both codes are not related. ### How was this patch tested? This is refactoring so existing CI should cover. Closes apache#38618 from HyukjinKwon/SPARK-41108-followup-1. Authored-by: Hyukjin Kwon <[email protected]> Signed-off-by: Hyukjin Kwon <[email protected]>
3 tasks
pan3793
added a commit
to apache/kyuubi
that referenced
this pull request
Apr 23, 2023
…tead of `ArrowConveters#toBatchIterator` ### _Why are the changes needed?_ to adapt Spark 3.4 the signature of function `ArrowConveters#toBatchIterator` is changed in apache/spark#38618 (since Spark 3.4) Before Spark 3.4: ``` private[sql] def toBatchIterator( rowIter: Iterator[InternalRow], schema: StructType, maxRecordsPerBatch: Int, timeZoneId: String, context: TaskContext): Iterator[Array[Byte]] ``` Spark 3.4 ``` private[sql] def toBatchIterator( rowIter: Iterator[InternalRow], schema: StructType, maxRecordsPerBatch: Long, timeZoneId: String, context: TaskContext): ArrowBatchIterator ``` the return type is changed from `Iterator[Array[Byte]]` to `ArrowBatchIterator` ### _How was this patch tested?_ - [ ] Add some test cases that check the changes thoroughly including negative and positive cases if possible - [ ] Add screenshots for manual tests if appropriate - [x] [Run test](https://kyuubi.readthedocs.io/en/master/develop_tools/testing.html#running-tests) locally before make a pull request Closes #4754 from cfmcgrady/arrow-spark34. Closes #4754 a3c58d0 [Fu Chen] fix ci 32704c5 [Fu Chen] Revert "fix ci" e32311a [Fu Chen] fix ci a76af62 [Cheng Pan] Update externals/kyuubi-spark-sql-engine/src/main/scala/org/apache/spark/sql/kyuubi/SparkDatasetHelper.scala 453b6a6 [Cheng Pan] Update externals/kyuubi-spark-sql-engine/src/main/scala/org/apache/spark/sql/kyuubi/SparkDatasetHelper.scala 74a9f7a [Cheng Pan] Update externals/kyuubi-spark-sql-engine/src/main/scala/org/apache/spark/sql/kyuubi/SparkDatasetHelper.scala 4ce5844 [Fu Chen] adapt Spark 3.4 Lead-authored-by: Fu Chen <[email protected]> Co-authored-by: Cheng Pan <[email protected]> Signed-off-by: Cheng Pan <[email protected]>
pan3793
added a commit
to apache/kyuubi
that referenced
this pull request
Apr 23, 2023
…tead of `ArrowConveters#toBatchIterator` ### _Why are the changes needed?_ to adapt Spark 3.4 the signature of function `ArrowConveters#toBatchIterator` is changed in apache/spark#38618 (since Spark 3.4) Before Spark 3.4: ``` private[sql] def toBatchIterator( rowIter: Iterator[InternalRow], schema: StructType, maxRecordsPerBatch: Int, timeZoneId: String, context: TaskContext): Iterator[Array[Byte]] ``` Spark 3.4 ``` private[sql] def toBatchIterator( rowIter: Iterator[InternalRow], schema: StructType, maxRecordsPerBatch: Long, timeZoneId: String, context: TaskContext): ArrowBatchIterator ``` the return type is changed from `Iterator[Array[Byte]]` to `ArrowBatchIterator` ### _How was this patch tested?_ - [ ] Add some test cases that check the changes thoroughly including negative and positive cases if possible - [ ] Add screenshots for manual tests if appropriate - [x] [Run test](https://kyuubi.readthedocs.io/en/master/develop_tools/testing.html#running-tests) locally before make a pull request Closes #4754 from cfmcgrady/arrow-spark34. Closes #4754 a3c58d0 [Fu Chen] fix ci 32704c5 [Fu Chen] Revert "fix ci" e32311a [Fu Chen] fix ci a76af62 [Cheng Pan] Update externals/kyuubi-spark-sql-engine/src/main/scala/org/apache/spark/sql/kyuubi/SparkDatasetHelper.scala 453b6a6 [Cheng Pan] Update externals/kyuubi-spark-sql-engine/src/main/scala/org/apache/spark/sql/kyuubi/SparkDatasetHelper.scala 74a9f7a [Cheng Pan] Update externals/kyuubi-spark-sql-engine/src/main/scala/org/apache/spark/sql/kyuubi/SparkDatasetHelper.scala 4ce5844 [Fu Chen] adapt Spark 3.4 Lead-authored-by: Fu Chen <[email protected]> Co-authored-by: Cheng Pan <[email protected]> Signed-off-by: Cheng Pan <[email protected]> (cherry picked from commit d0a7ca4) Signed-off-by: Cheng Pan <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What changes were proposed in this pull request?
This PR is a followup of both #38468 and #38612 that proposes to deduplicate codes in
ArrowConvertersby creating two classesArrowBatchIteratorandArrowBatchWithSchemaIterator. In addition, we reuseArrowBatchWithSchemaIteratorwhen creating an empty Arrow batch atcreateEmptyArrowBatch.While I am here,
Why are the changes needed?
For better readability and maintenance.
Does this PR introduce any user-facing change?
No, both codes are not related.
How was this patch tested?
This is refactoring so existing CI should cover.