This is an automated email from the ASF dual-hosted git repository.
yangjie01 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 f4d10122fa22 [SPARK-53232][SQL][TESTS] Use Java `Map.copyOf` instead
of `ImmutableMap.copyOf`
f4d10122fa22 is described below
commit f4d10122fa22a9923a619984f12a6149cb52d41e
Author: Dongjoon Hyun <dongj...@apache.org>
AuthorDate: Sun Aug 10 19:40:05 2025 +0800
[SPARK-53232][SQL][TESTS] Use Java `Map.copyOf` instead of
`ImmutableMap.copyOf`
### What changes were proposed in this pull request?
This PR aims to use Java 10+ `Map.copyOf` instead of `ImmutableMap.copyOf`.
In addition, a new Scalastyle rule is added to prevent future regressions.
### Why are the changes needed?
Java native implementation is **significantly faster and simpler** than
`ImmutableMap.copyOf`.
```scala
scala> val m = java.util.Map.of(1, 2, 3, 4, 5, 6, 7, 8, 9, 10)
val m: java.util.Map[Int,Int] = {5=6, 7=8, 9=10, 1=2, 3=4}
scala> spark.time((1 until 100_000_000).foreach(_ =>
com.google.common.collect.ImmutableMap.copyOf(m)))
Time taken: 4404 ms
scala> spark.time((1 until 100_000_000).foreach(_ =>
java.util.Map.copyOf(m)))
Time taken: 223 ms
```
### Does this PR introduce _any_ user-facing change?
No behavior change.
### How was this patch tested?
Pass the CIs.
### Was this patch authored or co-authored using generative AI tooling?
No.
Closes #51958 from dongjoon-hyun/SPARK-53232.
Authored-by: Dongjoon Hyun <dongj...@apache.org>
Signed-off-by: yangjie01 <yangji...@baidu.com>
---
scalastyle-config.xml | 5 +++++
sql/core/src/test/scala/org/apache/spark/sql/FileScanSuite.scala | 7 +++----
2 files changed, 8 insertions(+), 4 deletions(-)
diff --git a/scalastyle-config.xml b/scalastyle-config.xml
index 48362ca6f9a2..cd7c86a8bc1b 100644
--- a/scalastyle-config.xml
+++ b/scalastyle-config.xml
@@ -787,6 +787,11 @@ This file is divided into 3 sections:
<customMessage>Use OutputStream.nullOutputStream instead.</customMessage>
</check>
+ <check customId="ImmutableMapcopyOf" level="error"
class="org.scalastyle.file.RegexChecker" enabled="true">
+ <parameters><parameter
name="regex">\bImmutableMap\.copyOf\b</parameter></parameters>
+ <customMessage>Use Map.copyOf instead.</customMessage>
+ </check>
+
<check customId="maputils" level="error"
class="org.scalastyle.file.RegexChecker" enabled="true">
<parameters><parameter
name="regex">org\.apache\.commons\.collections4\.MapUtils\b</parameter></parameters>
<customMessage>Use org.apache.spark.util.collection.Utils
instead.</customMessage>
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/FileScanSuite.scala
b/sql/core/src/test/scala/org/apache/spark/sql/FileScanSuite.scala
index f8a3cfc50a2e..c7ea8eca75ea 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/FileScanSuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/FileScanSuite.scala
@@ -21,7 +21,6 @@ import java.util.{Map => JMap}
import scala.collection.mutable
-import com.google.common.collect.ImmutableMap
import org.apache.hadoop.fs.{FileStatus, Path}
import org.apache.spark.sql.catalyst.dsl.expressions._
@@ -84,9 +83,9 @@ trait FileScanSuiteBase extends SharedSparkSession {
val pushedFiltersNotEqual =
Array[Filter](sources.And(sources.IsNull("data"),
sources.LessThan("data", 1)))
val optionsMap = JMap.of("key", "value")
- val options = new CaseInsensitiveStringMap(ImmutableMap.copyOf(optionsMap))
+ val options = new CaseInsensitiveStringMap(JMap.copyOf(optionsMap))
val optionsNotEqual =
- new CaseInsensitiveStringMap(ImmutableMap.copyOf(JMap.of("key2",
"value2")))
+ new CaseInsensitiveStringMap(JMap.copyOf(JMap.of("key2", "value2")))
val partitionFilters = Seq(And(IsNull($"data".int), LessThan($"data".int,
0)))
val partitionFiltersNotEqual = Seq(And(IsNull($"data".int),
LessThan($"data".int, 1)))
@@ -115,7 +114,7 @@ trait FileScanSuiteBase extends SharedSparkSession {
readDataSchema.copy(),
readPartitionSchema.copy(),
pushedFilters.clone(),
- new CaseInsensitiveStringMap(ImmutableMap.copyOf(optionsMap)),
+ new CaseInsensitiveStringMap(JMap.copyOf(optionsMap)),
Seq(partitionFilters: _*),
Seq(dataFilters: _*))
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org
For additional commands, e-mail: commits-h...@spark.apache.org
