This is an automated email from the ASF dual-hosted git repository.
uros-b pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/spark.git
The following commit(s) were added to refs/heads/master by this push:
new 7e875f1ee790 [SPARK-58948][SQL] Deduplicate nulls-buffer creation in
columnar decompressors
7e875f1ee790 is described below
commit 7e875f1ee79096a95da97f7f1aa52b538baff57a
Author: Uros <[email protected]>
AuthorDate: Mon Aug 24 18:36:39 2026 +0000
[SPARK-58948][SQL] Deduplicate nulls-buffer creation in columnar
decompressors
### What changes were proposed in this pull request?
Each columnar decompressor repeats the same two-line idiom to build the
"nulls buffer" it reads null-tracking state from:
```scala
val nullsBuffer = buffer.duplicate().order(ByteOrder.nativeOrder())
nullsBuffer.rewind()
```
This adds a `createNullsBuffer(buffer: ByteBuffer)` helper to the
`CompressionScheme` companion object (alongside the existing `columnHeaderSize`
helper) and routes the six call sites through it (`PassThrough`,
`RunLengthEncoding`, `DictionaryEncoding`, `BooleanBitSet`, `IntDelta`,
`LongDelta`). The subsequent `nullCount` / `nextNullIndex` reads stay inline,
since the `nextNullIndex` sentinel legitimately differs (`-1`, and `capacity`
for `PassThrough`).
### Why are the changes needed?
The idiom was duplicated across all six decoders. Naming it once removes
the duplication and gives a single home for the "duplicate, set native order,
rewind" convention. It also normalizes the `LongDelta` site, which called
`rewind` without parentheses.
### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?
Existing columnar compression tests. `createNullsBuffer` returns exactly
the previously-inlined expression, so decompression behavior is unchanged.
### Was this patch authored or co-authored using generative AI tooling?
Generated-by: Claude Code (Opus 4.8)
Closes #58228 from uros-b/helper-compressionschemes-nullindex.
Authored-by: Uros <[email protected]>
Signed-off-by: Uros Bojanic <[email protected]>
---
.../columnar/compression/CompressionScheme.scala | 6 ++++++
.../columnar/compression/compressionSchemes.scala | 19 ++++++-------------
2 files changed, 12 insertions(+), 13 deletions(-)
diff --git
a/sql/core/src/main/scala/org/apache/spark/sql/execution/columnar/compression/CompressionScheme.scala
b/sql/core/src/main/scala/org/apache/spark/sql/execution/columnar/compression/CompressionScheme.scala
index 0fe1fbcf9450..e0f68a70daae 100644
---
a/sql/core/src/main/scala/org/apache/spark/sql/execution/columnar/compression/CompressionScheme.scala
+++
b/sql/core/src/main/scala/org/apache/spark/sql/execution/columnar/compression/CompressionScheme.scala
@@ -84,4 +84,10 @@ private[columnar] object CompressionScheme {
// null count + null positions
4 + 4 * nullCount
}
+
+ def createNullsBuffer(buffer: ByteBuffer): ByteBuffer = {
+ val nullsBuffer = buffer.duplicate().order(ByteOrder.nativeOrder())
+ nullsBuffer.rewind()
+ nullsBuffer
+ }
}
diff --git
a/sql/core/src/main/scala/org/apache/spark/sql/execution/columnar/compression/compressionSchemes.scala
b/sql/core/src/main/scala/org/apache/spark/sql/execution/columnar/compression/compressionSchemes.scala
index 86d76856e12b..72bebb96ed3a 100644
---
a/sql/core/src/main/scala/org/apache/spark/sql/execution/columnar/compression/compressionSchemes.scala
+++
b/sql/core/src/main/scala/org/apache/spark/sql/execution/columnar/compression/compressionSchemes.scala
@@ -18,7 +18,6 @@
package org.apache.spark.sql.execution.columnar.compression
import java.nio.ByteBuffer
-import java.nio.ByteOrder
import scala.collection.mutable
@@ -108,8 +107,7 @@ private[columnar] case object PassThrough extends
CompressionScheme {
capacity: Int,
unitSize: Int,
putFunction: (WritableColumnVector, Int, Int, Int) => Unit): Unit = {
- val nullsBuffer = buffer.duplicate().order(ByteOrder.nativeOrder())
- nullsBuffer.rewind()
+ val nullsBuffer = CompressionScheme.createNullsBuffer(buffer)
val nullCount = ByteBufferHelper.getInt(nullsBuffer)
var nextNullIndex = if (nullCount > 0)
ByteBufferHelper.getInt(nullsBuffer) else capacity
var pos = 0
@@ -307,8 +305,7 @@ private[columnar] case object RunLengthEncoding extends
CompressionScheme {
capacity: Int,
getFunction: (ByteBuffer) => Long,
putFunction: (WritableColumnVector, Int, Long) => Unit): Unit = {
- val nullsBuffer = buffer.duplicate().order(ByteOrder.nativeOrder())
- nullsBuffer.rewind()
+ val nullsBuffer = CompressionScheme.createNullsBuffer(buffer)
val nullCount = ByteBufferHelper.getInt(nullsBuffer)
var nextNullIndex = if (nullCount > 0)
ByteBufferHelper.getInt(nullsBuffer) else -1
var pos = 0
@@ -486,8 +483,7 @@ private[columnar] case object DictionaryEncoding extends
CompressionScheme {
override def hasNext: Boolean = buffer.hasRemaining
override def decompress(columnVector: WritableColumnVector, capacity:
Int): Unit = {
- val nullsBuffer = buffer.duplicate().order(ByteOrder.nativeOrder())
- nullsBuffer.rewind()
+ val nullsBuffer = CompressionScheme.createNullsBuffer(buffer)
val nullCount = ByteBufferHelper.getInt(nullsBuffer)
var nextNullIndex = if (nullCount > 0)
ByteBufferHelper.getInt(nullsBuffer) else -1
var pos = 0
@@ -619,8 +615,7 @@ private[columnar] case object BooleanBitSet extends
CompressionScheme {
override def decompress(columnVector: WritableColumnVector, capacity:
Int): Unit = {
var currentWordLocal: Long = 0
var visitedLocal: Int = 0
- val nullsBuffer = buffer.duplicate().order(ByteOrder.nativeOrder())
- nullsBuffer.rewind()
+ val nullsBuffer = CompressionScheme.createNullsBuffer(buffer)
val nullCount = ByteBufferHelper.getInt(nullsBuffer)
var nextNullIndex = if (nullCount > 0)
ByteBufferHelper.getInt(nullsBuffer) else -1
var pos = 0
@@ -730,8 +725,7 @@ private[columnar] case object IntDelta extends
CompressionScheme {
override def decompress(columnVector: WritableColumnVector, capacity:
Int): Unit = {
var prevLocal: Int = 0
- val nullsBuffer = buffer.duplicate().order(ByteOrder.nativeOrder())
- nullsBuffer.rewind()
+ val nullsBuffer = CompressionScheme.createNullsBuffer(buffer)
val nullCount = ByteBufferHelper.getInt(nullsBuffer)
var nextNullIndex = if (nullCount > 0)
ByteBufferHelper.getInt(nullsBuffer) else -1
var pos = 0
@@ -837,8 +831,7 @@ private[columnar] case object LongDelta extends
CompressionScheme {
override def decompress(columnVector: WritableColumnVector, capacity:
Int): Unit = {
var prevLocal: Long = 0
- val nullsBuffer = buffer.duplicate().order(ByteOrder.nativeOrder())
- nullsBuffer.rewind
+ val nullsBuffer = CompressionScheme.createNullsBuffer(buffer)
val nullCount = ByteBufferHelper.getInt(nullsBuffer)
var nextNullIndex = if (nullCount > 0)
ByteBufferHelper.getInt(nullsBuffer) else -1
var pos = 0
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
