[SPARK-37004][PYTHON] Upgrade to Py4J 0.10.9.3

### What changes were proposed in this pull request?

This PR upgrades Py4J from 0.10.9.2 to 0.10.9.3 which contains the bug fix (py4j/py4j#440) that directly affected us.

For example, once you cancel a cell in Jupyter, all following cells simply fail. This PR fixes the bug by upgrading Py4J.

### Why are the changes needed?

To fix a regression in Spark 3.2.0 in notebooks like Jupyter.

### Does this PR introduce _any_ user-facing change?

Fixes a regression described in SPARK-37004

### How was this patch tested?

Manually tested the fix when I land py4j/py4j#440 to Py4J.

Closes apache#34814 from HyukjinKwon/SPARK-37004.

Authored-by: Hyukjin Kwon <gurwls223@apache.org>
Signed-off-by: Hyukjin Kwon <gurwls223@apache.org>

bin/pyspark

@@ -50,7 +50,7 @@ export PYSPARK_DRIVER_PYTHON_OPTS
 
 # Add the PySpark classes to the Python path:
 export PYTHONPATH="${SPARK_HOME}/python/:$PYTHONPATH"
-export PYTHONPATH="${SPARK_HOME}/python/lib/py4j-0.10.9.2-src.zip:$PYTHONPATH"
+export PYTHONPATH="${SPARK_HOME}/python/lib/py4j-0.10.9.3-src.zip:$PYTHONPATH"
 
 # Load the PySpark shell.py script when ./pyspark is used interactively:
 export OLD_PYTHONSTARTUP="$PYTHONSTARTUP"

bin/pyspark2.cmd

@@ -30,7 +30,7 @@ if "x%PYSPARK_DRIVER_PYTHON%"=="x" (
 )
 
 set PYTHONPATH=%SPARK_HOME%\python;%PYTHONPATH%
-set PYTHONPATH=%SPARK_HOME%\python\lib\py4j-0.10.9.2-src.zip;%PYTHONPATH%
+set PYTHONPATH=%SPARK_HOME%\python\lib\py4j-0.10.9.3-src.zip;%PYTHONPATH%
 
 set OLD_PYTHONSTARTUP=%PYTHONSTARTUP%
 set PYTHONSTARTUP=%SPARK_HOME%\python\pyspark\shell.py

core/pom.xml

@@ -427,7 +427,7 @@
     <dependency>
       <groupId>net.sf.py4j</groupId>
       <artifactId>py4j</artifactId>
-      <version>0.10.9.2</version>
+      <version>0.10.9.3</version>
     </dependency>
     <dependency>
       <groupId>org.apache.spark</groupId>

core/src/main/scala/org/apache/spark/api/python/PythonUtils.scala

@@ -27,7 +27,7 @@ import org.apache.spark.SparkContext
 import org.apache.spark.api.java.{JavaRDD, JavaSparkContext}
 
 private[spark] object PythonUtils {
-  val PY4J_ZIP_NAME = "py4j-0.10.9.2-src.zip"
+  val PY4J_ZIP_NAME = "py4j-0.10.9.3-src.zip"
 
   /** Get the PYTHONPATH for PySpark, either from SPARK_HOME, if it is set, or from our JAR */
   def sparkPythonPath: String = {

dev/deps/spark-deps-hadoop-2.7-hive-2.3

@@ -216,7 +216,7 @@ parquet-hadoop/1.12.2//parquet-hadoop-1.12.2.jar
 parquet-jackson/1.12.2//parquet-jackson-1.12.2.jar
 pickle/1.2//pickle-1.2.jar
 protobuf-java/2.5.0//protobuf-java-2.5.0.jar
-py4j/0.10.9.2//py4j-0.10.9.2.jar
+py4j/0.10.9.3//py4j-0.10.9.3.jar
 remotetea-oncrpc/1.1.2//remotetea-oncrpc-1.1.2.jar
 rocksdbjni/6.20.3//rocksdbjni-6.20.3.jar
 scala-collection-compat_2.12/2.1.1//scala-collection-compat_2.12-2.1.1.jar

dev/deps/spark-deps-hadoop-3.2-hive-2.3

@@ -203,7 +203,7 @@ parquet-hadoop/1.12.2//parquet-hadoop-1.12.2.jar
 parquet-jackson/1.12.2//parquet-jackson-1.12.2.jar
 pickle/1.2//pickle-1.2.jar
 protobuf-java/2.5.0//protobuf-java-2.5.0.jar
-py4j/0.10.9.2//py4j-0.10.9.2.jar
+py4j/0.10.9.3//py4j-0.10.9.3.jar
 remotetea-oncrpc/1.1.2//remotetea-oncrpc-1.1.2.jar
 rocksdbjni/6.20.3//rocksdbjni-6.20.3.jar
 scala-collection-compat_2.12/2.1.1//scala-collection-compat_2.12-2.1.1.jar

python/docs/Makefile

@@ -21,7 +21,7 @@ SPHINXBUILD   ?= sphinx-build
 SOURCEDIR     ?= source
 BUILDDIR      ?= build
 
-export PYTHONPATH=$(realpath ..):$(realpath ../lib/py4j-0.10.9.2-src.zip)
+export PYTHONPATH=$(realpath ..):$(realpath ../lib/py4j-0.10.9.3-src.zip)
 
 # Put it first so that "make" without argument is like "make help".
 help:

python/docs/make2.bat

@@ -25,7 +25,7 @@ if "%SPHINXBUILD%" == "" (
 set SOURCEDIR=source
 set BUILDDIR=build
 
-set PYTHONPATH=..;..\lib\py4j-0.10.9.2-src.zip
+set PYTHONPATH=..;..\lib\py4j-0.10.9.3-src.zip
 
 if "%1" == "" goto help
 

python/docs/source/getting_started/install.rst

@@ -157,7 +157,7 @@ Package       Minimum supported version Note
 `pandas`      1.0.5                     Optional for Spark SQL
 `NumPy`       1.7                       Required for MLlib DataFrame-based API
 `pyarrow`     1.0.0                     Optional for Spark SQL
-`Py4J`        0.10.9.2                  Required
+`Py4J`        0.10.9.3                  Required
 `pandas`      1.0.5                     Required for pandas API on Spark
 `pyarrow`     1.0.0                     Required for pandas API on Spark
 `Numpy`       1.14                      Required for pandas API on Spark

python/setup.py

@@ -258,7 +258,7 @@ def run(self):
         license='http://www.apache.org/licenses/LICENSE-2.0',
         # Don't forget to update python/docs/source/getting_started/install.rst
         # if you're updating the versions or dependencies.
-        install_requires=['py4j==0.10.9.2'],
+        install_requires=['py4j==0.10.9.3'],
         extras_require={
             'ml': ['numpy>=1.7'],
             'mllib': ['numpy>=1.7'],

sbin/spark-config.sh

@@ -28,6 +28,6 @@ export SPARK_CONF_DIR="${SPARK_CONF_DIR:-"${SPARK_HOME}/conf"}"
 # Add the PySpark classes to the PYTHONPATH:
 if [ -z "${PYSPARK_PYTHONPATH_SET}" ]; then
   export PYTHONPATH="${SPARK_HOME}/python:${PYTHONPATH}"
-  export PYTHONPATH="${SPARK_HOME}/python/lib/py4j-0.10.9.2-src.zip:${PYTHONPATH}"
+  export PYTHONPATH="${SPARK_HOME}/python/lib/py4j-0.10.9.3-src.zip:${PYTHONPATH}"
   export PYSPARK_PYTHONPATH_SET=1
 fi