[SPARK-18076][CORE][SQL] Fix default Locale used in DateFormat, NumberFormat to Locale.US

## What changes were proposed in this pull request?

Fix `Locale.US` for all usages of `DateFormat`, `NumberFormat`
## How was this patch tested?

Existing tests.

Author: Sean Owen <sowen@cloudera.com>

Closes #15610 from srowen/SPARK-18076.

3 files changed, 15 insertions, 14 deletions

diff --git a/sql/core/src/test/scala/org/apache/spark/sql/DateFunctionsSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/DateFunctionsSuite.scala
index f7aa3b747a..e05b2252ee 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/DateFunctionsSuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/DateFunctionsSuite.scala
@@ -19,6 +19,7 @@ package org.apache.spark.sql
 
 import java.sql.{Date, Timestamp}
 import java.text.SimpleDateFormat
+import java.util.Locale
 
 import org.apache.spark.sql.catalyst.util.DateTimeUtils
 import org.apache.spark.sql.functions._
@@ -55,8 +56,8 @@ class DateFunctionsSuite extends QueryTest with SharedSQLContext {
     checkAnswer(sql("""SELECT CURRENT_TIMESTAMP() = NOW()"""), Row(true))
   }
 
-  val sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss")
-  val sdfDate = new SimpleDateFormat("yyyy-MM-dd")
+  val sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss", Locale.US)
+  val sdfDate = new SimpleDateFormat("yyyy-MM-dd", Locale.US)
   val d = new Date(sdf.parse("2015-04-08 13:10:15").getTime)
   val ts = new Timestamp(sdf.parse("2013-04-08 13:10:15").getTime)
 
@@ -395,11 +396,11 @@ class DateFunctionsSuite extends QueryTest with SharedSQLContext {
   }
 
   test("from_unixtime") {
-    val sdf1 = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss")
+    val sdf1 = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss", Locale.US)
     val fmt2 = "yyyy-MM-dd HH:mm:ss.SSS"
-    val sdf2 = new SimpleDateFormat(fmt2)
+    val sdf2 = new SimpleDateFormat(fmt2, Locale.US)
     val fmt3 = "yy-MM-dd HH-mm-ss"
-    val sdf3 = new SimpleDateFormat(fmt3)
+    val sdf3 = new SimpleDateFormat(fmt3, Locale.US)
     val df = Seq((1000, "yyyy-MM-dd HH:mm:ss.SSS"), (-1000, "yy-MM-dd HH-mm-ss")).toDF("a", "b")
     checkAnswer(
       df.select(from_unixtime(col("a"))),
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/csv/CSVSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/csv/CSVSuite.scala
index f7c22c6c93..8209b5bd7f 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/csv/CSVSuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/csv/CSVSuite.scala
@@ -21,6 +21,7 @@ import java.io.File
 import java.nio.charset.UnsupportedCharsetException
 import java.sql.{Date, Timestamp}
 import java.text.SimpleDateFormat
+import java.util.Locale
 
 import org.apache.commons.lang3.time.FastDateFormat
 import org.apache.hadoop.io.SequenceFile.CompressionType
@@ -487,7 +488,7 @@ class CSVSuite extends QueryTest with SharedSQLContext with SQLTestUtils {
       .select("date")
       .collect()
 
-    val dateFormat = new SimpleDateFormat("dd/MM/yyyy HH:mm")
+    val dateFormat = new SimpleDateFormat("dd/MM/yyyy HH:mm", Locale.US)
     val expected =
       Seq(Seq(new Timestamp(dateFormat.parse("26/08/2015 18:00").getTime)),
         Seq(new Timestamp(dateFormat.parse("27/10/2014 18:30").getTime)),
@@ -509,7 +510,7 @@ class CSVSuite extends QueryTest with SharedSQLContext with SQLTestUtils {
       .select("date")
       .collect()
 
-    val dateFormat = new SimpleDateFormat("dd/MM/yyyy hh:mm")
+    val dateFormat = new SimpleDateFormat("dd/MM/yyyy hh:mm", Locale.US)
     val expected = Seq(
       new Date(dateFormat.parse("26/08/2015 18:00").getTime),
       new Date(dateFormat.parse("27/10/2014 18:30").getTime),
@@ -728,7 +729,7 @@ class CSVSuite extends QueryTest with SharedSQLContext with SQLTestUtils {
         .option("inferSchema", "false")
         .load(iso8601timestampsPath)
 
-      val iso8501 = FastDateFormat.getInstance("yyyy-MM-dd'T'HH:mm:ss.SSSZZ")
+      val iso8501 = FastDateFormat.getInstance("yyyy-MM-dd'T'HH:mm:ss.SSSZZ", Locale.US)
       val expectedTimestamps = timestamps.collect().map { r =>
         // This should be ISO8601 formatted string.
         Row(iso8501.format(r.toSeq.head))
@@ -761,7 +762,7 @@ class CSVSuite extends QueryTest with SharedSQLContext with SQLTestUtils {
         .option("inferSchema", "false")
         .load(iso8601datesPath)
 
-      val iso8501 = FastDateFormat.getInstance("yyyy-MM-dd")
+      val iso8501 = FastDateFormat.getInstance("yyyy-MM-dd", Locale.US)
       val expectedDates = dates.collect().map { r =>
         // This should be ISO8601 formatted string.
         Row(iso8501.format(r.toSeq.head))
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/csv/CSVTypeCastSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/csv/CSVTypeCastSuite.scala
index 51832a13cf..c74406b9cb 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/csv/CSVTypeCastSuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/csv/CSVTypeCastSuite.scala
@@ -144,13 +144,12 @@ class CSVTypeCastSuite extends SparkFunSuite {
       DateTimeUtils.millisToDays(DateTimeUtils.stringToTime("2015-01-01").getTime))
   }
 
-  test("Float and Double Types are cast correctly with Locale") {
+  test("Float and Double Types are cast without respect to platform default Locale") {
     val originalLocale = Locale.getDefault
     try {
-      val locale : Locale = new Locale("fr", "FR")
-      Locale.setDefault(locale)
-      assert(CSVTypeCast.castTo("1,00", FloatType) == 1.0)
-      assert(CSVTypeCast.castTo("1,00", DoubleType) == 1.0)
+      Locale.setDefault(new Locale("fr", "FR"))
+      assert(CSVTypeCast.castTo("1,00", FloatType) == 100.0) // Would parse as 1.0 in fr-FR
+      assert(CSVTypeCast.castTo("1,00", DoubleType) == 100.0)
     } finally {
       Locale.setDefault(originalLocale)
     }