[SPARK-31725][CORE][SQL][TESTS] Set America/Los_Angeles time zone and Locale.US in tests by default

### What changes were proposed in this pull request?
Set default time zone and locale in the default constructor of `SparkFunSuite`:
- Default time zone to `America/Los_Angeles`
- Default locale to `Locale.US`

### Why are the changes needed?
1. To deduplicate code by moving common time zone and locale settings to one place SparkFunSuite
2. To have the same default time zone and locale in all tests. This should prevent errors like #28538

### Does this PR introduce _any_ user-facing change?
No

### How was this patch tested?
by running all affected test suites

Closes #28548 from MaxGekk/timezone-settings-SparkFunSuite.

Authored-by: Max Gekk <[email protected]>
Signed-off-by: Wenchen Fan <[email protected]>

Author: MaxGekk

core/src/test/scala/org/apache/spark/SparkFunSuite.scala

@@ -19,6 +19,7 @@ package org.apache.spark
 
 // scalastyle:off
 import java.io.File
+import java.util.{Locale, TimeZone}
 
 import org.apache.log4j.spi.LoggingEvent
 
@@ -63,6 +64,11 @@ abstract class SparkFunSuite
   with Logging {
 // scalastyle:on
 
+  // Timezone is fixed to America/Los_Angeles for those timezone sensitive tests (timestamp_*)
+  TimeZone.setDefault(TimeZone.getTimeZone("America/Los_Angeles"))
+  // Add Locale setting
+  Locale.setDefault(Locale.US)
+
   protected val enableAutoThreadAudit = true
 
   protected override def beforeAll(): Unit = {

sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/SortOrderExpressionsSuite.scala

@@ -18,7 +18,6 @@
 package org.apache.spark.sql.catalyst.expressions
 
 import java.sql.Timestamp
-import java.util.TimeZone
 
 import org.apache.spark.SparkFunSuite
 import org.apache.spark.sql.types._
@@ -35,15 +34,7 @@ class SortOrderExpressionsSuite extends SparkFunSuite with ExpressionEvalHelper
     val l1 = Literal.create(20132983L, LongType)
     val l2 = Literal.create(-20132983L, LongType)
     val millis = 1524954911000L
-    // Explicitly choose a time zone, since Date objects can create different values depending on
-    // local time zone of the machine on which the test is running
-    val oldDefaultTZ = TimeZone.getDefault
-    val d1 = try {
-      TimeZone.setDefault(TimeZone.getTimeZone("America/Los_Angeles"))
-      Literal.create(new java.sql.Date(millis), DateType)
-    } finally {
-      TimeZone.setDefault(oldDefaultTZ)
-    }
+    val d1 = Literal.create(new java.sql.Date(millis), DateType)
     val t1 = Literal.create(new Timestamp(millis), TimestampType)
     val f1 = Literal.create(0.7788229f, FloatType)
     val f2 = Literal.create(-0.7788229f, FloatType)

sql/core/src/test/scala/org/apache/spark/sql/QueryTest.scala

@@ -17,7 +17,7 @@
 
 package org.apache.spark.sql
 
-import java.util.{Locale, TimeZone}
+import java.util.TimeZone
 
 import scala.collection.JavaConverters._
 
@@ -35,11 +35,6 @@ abstract class QueryTest extends PlanTest {
 
   protected def spark: SparkSession
 
-  // Timezone is fixed to America/Los_Angeles for those timezone sensitive tests (timestamp_*)
-  TimeZone.setDefault(TimeZone.getTimeZone("America/Los_Angeles"))
-  // Add Locale setting
-  Locale.setDefault(Locale.US)
-
   /**
    * Runs the plan and makes sure the answer contains all of the keywords.
    */

sql/core/src/test/scala/org/apache/spark/sql/SQLQueryTestSuite.scala

@@ -18,7 +18,7 @@
 package org.apache.spark.sql
 
 import java.io.File
-import java.util.{Locale, TimeZone}
+import java.util.Locale
 import java.util.regex.Pattern
 
 import scala.collection.mutable.{ArrayBuffer, HashMap}
@@ -672,25 +672,16 @@ class SQLQueryTestSuite extends QueryTest with SharedSparkSession {
     session.sql("DROP TABLE IF EXISTS tenk1")
   }
 
-  private val originalTimeZone = TimeZone.getDefault
-  private val originalLocale = Locale.getDefault
-
   override def beforeAll(): Unit = {
     super.beforeAll()
     createTestTables(spark)
-    // Timezone is fixed to America/Los_Angeles for those timezone sensitive tests (timestamp_*)
-    TimeZone.setDefault(TimeZone.getTimeZone("America/Los_Angeles"))
-    // Add Locale setting
-    Locale.setDefault(Locale.US)
     RuleExecutor.resetMetrics()
     CodeGenerator.resetCompileTime()
     WholeStageCodegenExec.resetCodeGenTime()
   }
 
   override def afterAll(): Unit = {
     try {
-      TimeZone.setDefault(originalTimeZone)
-      Locale.setDefault(originalLocale)
       removeTestTables(spark)
 
       // For debugging dump some statistics about how much time was spent in various optimizer rules

sql/hive/compatibility/src/test/scala/org/apache/spark/sql/hive/execution/HiveCompatibilitySuite.scala

@@ -18,7 +18,6 @@
 package org.apache.spark.sql.hive.execution
 
 import java.io.File
-import java.util.{Locale, TimeZone}
 
 import org.scalatest.BeforeAndAfter
 
@@ -36,8 +35,6 @@ class HiveCompatibilitySuite extends HiveQueryFileTest with BeforeAndAfter {
   private lazy val hiveQueryDir = TestHive.getHiveFile(
     "ql/src/test/queries/clientpositive".split("/").mkString(File.separator))
 
-  private val originalTimeZone = TimeZone.getDefault
-  private val originalLocale = Locale.getDefault
   private val originalColumnBatchSize = TestHive.conf.columnBatchSize
   private val originalInMemoryPartitionPruning = TestHive.conf.inMemoryPartitionPruning
   private val originalCrossJoinEnabled = TestHive.conf.crossJoinEnabled
@@ -51,10 +48,6 @@ class HiveCompatibilitySuite extends HiveQueryFileTest with BeforeAndAfter {
   override def beforeAll(): Unit = {
     super.beforeAll()
     TestHive.setCacheTables(true)
-    // Timezone is fixed to America/Los_Angeles for those timezone sensitive tests (timestamp_*)
-    TimeZone.setDefault(TimeZone.getTimeZone("America/Los_Angeles"))
-    // Add Locale setting
-    Locale.setDefault(Locale.US)
     // Set a relatively small column batch size for testing purposes
     TestHive.setConf(SQLConf.COLUMN_BATCH_SIZE, 5)
     // Enable in-memory partition pruning for testing purposes
@@ -73,8 +66,6 @@ class HiveCompatibilitySuite extends HiveQueryFileTest with BeforeAndAfter {
   override def afterAll(): Unit = {
     try {
       TestHive.setCacheTables(false)
-      TimeZone.setDefault(originalTimeZone)
-      Locale.setDefault(originalLocale)
       TestHive.setConf(SQLConf.COLUMN_BATCH_SIZE, originalColumnBatchSize)
       TestHive.setConf(SQLConf.IN_MEMORY_PARTITION_PRUNING, originalInMemoryPartitionPruning)
       TestHive.setConf(SQLConf.CROSS_JOINS_ENABLED, originalCrossJoinEnabled)

sql/hive/compatibility/src/test/scala/org/apache/spark/sql/hive/execution/HiveWindowFunctionQuerySuite.scala

@@ -18,7 +18,6 @@
 package org.apache.spark.sql.hive.execution
 
 import java.io.File
-import java.util.{Locale, TimeZone}
 
 import org.scalatest.BeforeAndAfter
 
@@ -33,17 +32,11 @@ import org.apache.spark.util.Utils
  * files, every `createQueryTest` calls should explicitly set `reset` to `false`.
  */
 class HiveWindowFunctionQuerySuite extends HiveComparisonTest with BeforeAndAfter {
-  private val originalTimeZone = TimeZone.getDefault
-  private val originalLocale = Locale.getDefault
   private val testTempDir = Utils.createTempDir()
 
   override def beforeAll(): Unit = {
     super.beforeAll()
     TestHive.setCacheTables(true)
-    // Timezone is fixed to America/Los_Angeles for those timezone sensitive tests (timestamp_*)
-    TimeZone.setDefault(TimeZone.getTimeZone("America/Los_Angeles"))
-    // Add Locale setting
-    Locale.setDefault(Locale.US)
 
     // Create the table used in windowing.q
     sql("DROP TABLE IF EXISTS part")
@@ -103,8 +96,6 @@ class HiveWindowFunctionQuerySuite extends HiveComparisonTest with BeforeAndAfte
   override def afterAll(): Unit = {
     try {
       TestHive.setCacheTables(false)
-      TimeZone.setDefault(originalTimeZone)
-      Locale.setDefault(originalLocale)
       TestHive.reset()
     } finally {
       super.afterAll()
@@ -747,17 +738,11 @@ class HiveWindowFunctionQuerySuite extends HiveComparisonTest with BeforeAndAfte
 
 class HiveWindowFunctionQueryFileSuite
   extends HiveCompatibilitySuite with BeforeAndAfter {
-  private val originalTimeZone = TimeZone.getDefault
-  private val originalLocale = Locale.getDefault
   private val testTempDir = Utils.createTempDir()
 
   override def beforeAll(): Unit = {
     super.beforeAll()
     TestHive.setCacheTables(true)
-    // Timezone is fixed to America/Los_Angeles for those timezone sensitive tests (timestamp_*)
-    TimeZone.setDefault(TimeZone.getTimeZone("America/Los_Angeles"))
-    // Add Locale setting
-    Locale.setDefault(Locale.US)
 
     // The following settings are used for generating golden files with Hive.
     // We have to use kryo to correctly let Hive serialize plans with window functions.
@@ -772,8 +757,6 @@ class HiveWindowFunctionQueryFileSuite
   override def afterAll(): Unit = {
     try {
       TestHive.setCacheTables(false)
-      TimeZone.setDefault(originalTimeZone)
-      Locale.setDefault(originalLocale)
       TestHive.reset()
     } finally {
       super.afterAll()

sql/hive/src/test/scala/org/apache/spark/sql/hive/HiveInspectorSuite.scala

@@ -18,7 +18,6 @@
 package org.apache.spark.sql.hive
 
 import java.util
-import java.util.{Locale, TimeZone}
 
 import org.apache.hadoop.hive.ql.udf.UDAFPercentile
 import org.apache.hadoop.hive.serde2.io.DoubleWritable
@@ -74,11 +73,6 @@ class HiveInspectorSuite extends SparkFunSuite with HiveInspectors {
       .get())
   }
 
-  // Timezone is fixed to America/Los_Angeles for those timezone sensitive tests (timestamp_*)
-  TimeZone.setDefault(TimeZone.getTimeZone("America/Los_Angeles"))
-  // Add Locale setting
-  Locale.setDefault(Locale.US)
-
   val data =
     Literal(true) ::
     Literal(null) ::

sql/hive/src/test/scala/org/apache/spark/sql/hive/execution/HiveQuerySuite.scala

@@ -20,7 +20,7 @@ package org.apache.spark.sql.hive.execution
 import java.io.File
 import java.net.URI
 import java.sql.Timestamp
-import java.util.{Locale, TimeZone}
+import java.util.Locale
 
 import scala.util.Try
 
@@ -47,9 +47,6 @@ case class TestData(a: Int, b: String)
  * included in the hive distribution.
  */
 class HiveQuerySuite extends HiveComparisonTest with SQLTestUtils with BeforeAndAfter {
-  private val originalTimeZone = TimeZone.getDefault
-  private val originalLocale = Locale.getDefault
-
   import org.apache.spark.sql.hive.test.TestHive.implicits._
 
   private val originalCrossJoinEnabled = TestHive.conf.crossJoinEnabled
@@ -59,19 +56,13 @@ class HiveQuerySuite extends HiveComparisonTest with SQLTestUtils with BeforeAnd
   override def beforeAll(): Unit = {
     super.beforeAll()
     TestHive.setCacheTables(true)
-    // Timezone is fixed to America/Los_Angeles for those timezone sensitive tests (timestamp_*)
-    TimeZone.setDefault(TimeZone.getTimeZone("America/Los_Angeles"))
-    // Add Locale setting
-    Locale.setDefault(Locale.US)
     // Ensures that cross joins are enabled so that we can test them
     TestHive.setConf(SQLConf.CROSS_JOINS_ENABLED, true)
   }
 
   override def afterAll(): Unit = {
     try {
       TestHive.setCacheTables(false)
-      TimeZone.setDefault(originalTimeZone)
-      Locale.setDefault(originalLocale)
       sql("DROP TEMPORARY FUNCTION IF EXISTS udtf_count2")
       TestHive.setConf(SQLConf.CROSS_JOINS_ENABLED, originalCrossJoinEnabled)
     } finally {