[SPARK-18076][CORE][SQL] Fix default Locale used in DateFormat, Numbe…

…rFormat 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 <[email protected]>

Closes apache#15610 from srowen/SPARK-18076.

core/src/main/scala/org/apache/spark/SparkHadoopWriter.scala

@@ -20,7 +20,7 @@ package org.apache.spark
 import java.io.IOException
 import java.text.NumberFormat
 import java.text.SimpleDateFormat
-import java.util.Date
+import java.util.{Date, Locale}
 
 import org.apache.hadoop.fs.FileSystem
 import org.apache.hadoop.fs.Path
@@ -67,12 +67,12 @@ class SparkHadoopWriter(jobConf: JobConf) extends Logging with Serializable {
 
   def setup(jobid: Int, splitid: Int, attemptid: Int) {
     setIDs(jobid, splitid, attemptid)
-    HadoopRDD.addLocalConfiguration(new SimpleDateFormat("yyyyMMddHHmmss").format(now),
+    HadoopRDD.addLocalConfiguration(new SimpleDateFormat("yyyyMMddHHmmss", Locale.US).format(now),
       jobid, splitID, attemptID, conf.value)
   }
 
   def open() {
-    val numfmt = NumberFormat.getInstance()
+    val numfmt = NumberFormat.getInstance(Locale.US)
     numfmt.setMinimumIntegerDigits(5)
     numfmt.setGroupingUsed(false)
 
@@ -162,7 +162,7 @@ class SparkHadoopWriter(jobConf: JobConf) extends Logging with Serializable {
 private[spark]
 object SparkHadoopWriter {
   def createJobID(time: Date, id: Int): JobID = {
-    val formatter = new SimpleDateFormat("yyyyMMddHHmmss")
+    val formatter = new SimpleDateFormat("yyyyMMddHHmmss", Locale.US)
     val jobtrackerID = formatter.format(time)
     new JobID(jobtrackerID, id)
   }

core/src/main/scala/org/apache/spark/deploy/SparkHadoopUtil.scala

@@ -21,7 +21,7 @@ import java.io.IOException
 import java.lang.reflect.Method
 import java.security.PrivilegedExceptionAction
 import java.text.DateFormat
-import java.util.{Arrays, Comparator, Date}
+import java.util.{Arrays, Comparator, Date, Locale}
 
 import scala.collection.JavaConverters._
 import scala.util.control.NonFatal
@@ -357,7 +357,7 @@ class SparkHadoopUtil extends Logging {
    * @return a printable string value.
    */
   private[spark] def tokenToString(token: Token[_ <: TokenIdentifier]): String = {
-    val df = DateFormat.getDateTimeInstance(DateFormat.SHORT, DateFormat.SHORT)
+    val df = DateFormat.getDateTimeInstance(DateFormat.SHORT, DateFormat.SHORT, Locale.US)
     val buffer = new StringBuilder(128)
     buffer.append(token.toString)
     try {

core/src/main/scala/org/apache/spark/deploy/master/Master.scala

@@ -18,7 +18,7 @@
 package org.apache.spark.deploy.master
 
 import java.text.SimpleDateFormat
-import java.util.Date
+import java.util.{Date, Locale}
 import java.util.concurrent.{ScheduledFuture, TimeUnit}
 
 import scala.collection.mutable.{ArrayBuffer, HashMap, HashSet}
@@ -51,7 +51,8 @@ private[deploy] class Master(
 
   private val hadoopConf = SparkHadoopUtil.get.newConfiguration(conf)
 
-  private def createDateFormat = new SimpleDateFormat("yyyyMMddHHmmss") // For application IDs
+  // For application IDs
+  private def createDateFormat = new SimpleDateFormat("yyyyMMddHHmmss", Locale.US)
 
   private val WORKER_TIMEOUT_MS = conf.getLong("spark.worker.timeout", 60) * 1000
   private val RETAINED_APPLICATIONS = conf.getInt("spark.deploy.retainedApplications", 200)

core/src/main/scala/org/apache/spark/deploy/worker/Worker.scala

@@ -20,7 +20,7 @@ package org.apache.spark.deploy.worker
 import java.io.File
 import java.io.IOException
 import java.text.SimpleDateFormat
-import java.util.{Date, UUID}
+import java.util.{Date, Locale, UUID}
 import java.util.concurrent._
 import java.util.concurrent.{Future => JFuture, ScheduledFuture => JScheduledFuture}
 
@@ -68,7 +68,7 @@ private[deploy] class Worker(
     ThreadUtils.newDaemonSingleThreadExecutor("worker-cleanup-thread"))
 
   // For worker and executor IDs
-  private def createDateFormat = new SimpleDateFormat("yyyyMMddHHmmss")
+  private def createDateFormat = new SimpleDateFormat("yyyyMMddHHmmss", Locale.US)
   // Send a heartbeat every (heartbeat timeout) / 4 milliseconds
   private val HEARTBEAT_MILLIS = conf.getLong("spark.worker.timeout", 60) * 1000 / 4
 

core/src/main/scala/org/apache/spark/rdd/HadoopRDD.scala

@@ -19,7 +19,7 @@ package org.apache.spark.rdd
 
 import java.io.IOException
 import java.text.SimpleDateFormat
-import java.util.Date
+import java.util.{Date, Locale}
 
 import scala.collection.immutable.Map
 import scala.reflect.ClassTag
@@ -243,7 +243,8 @@ class HadoopRDD[K, V](
 
       var reader: RecordReader[K, V] = null
       val inputFormat = getInputFormat(jobConf)
-      HadoopRDD.addLocalConfiguration(new SimpleDateFormat("yyyyMMddHHmmss").format(createTime),
+      HadoopRDD.addLocalConfiguration(
+        new SimpleDateFormat("yyyyMMddHHmmss", Locale.US).format(createTime),
         context.stageId, theSplit.index, context.attemptNumber, jobConf)
       reader = inputFormat.getRecordReader(split.inputSplit.value, jobConf, Reporter.NULL)
 

core/src/main/scala/org/apache/spark/rdd/NewHadoopRDD.scala

@@ -19,7 +19,7 @@ package org.apache.spark.rdd
 
 import java.io.IOException
 import java.text.SimpleDateFormat
-import java.util.Date
+import java.util.{Date, Locale}
 
 import scala.reflect.ClassTag
 
@@ -79,7 +79,7 @@ class NewHadoopRDD[K, V](
   // private val serializableConf = new SerializableWritable(_conf)
 
   private val jobTrackerId: String = {
-    val formatter = new SimpleDateFormat("yyyyMMddHHmmss")
+    val formatter = new SimpleDateFormat("yyyyMMddHHmmss", Locale.US)
     formatter.format(new Date())
   }
 

core/src/main/scala/org/apache/spark/rdd/PairRDDFunctions.scala

@@ -19,7 +19,7 @@ package org.apache.spark.rdd
 
 import java.nio.ByteBuffer
 import java.text.SimpleDateFormat
-import java.util.{Date, HashMap => JHashMap}
+import java.util.{Date, HashMap => JHashMap, Locale}
 
 import scala.collection.{mutable, Map}
 import scala.collection.JavaConverters._
@@ -1079,7 +1079,7 @@ class PairRDDFunctions[K, V](self: RDD[(K, V)])
     // Rename this as hadoopConf internally to avoid shadowing (see SPARK-2038).
     val hadoopConf = conf
     val job = NewAPIHadoopJob.getInstance(hadoopConf)
-    val formatter = new SimpleDateFormat("yyyyMMddHHmmss")
+    val formatter = new SimpleDateFormat("yyyyMMddHHmmss", Locale.US)
     val jobtrackerID = formatter.format(new Date())
     val stageId = self.id
     val jobConfiguration = job.getConfiguration

core/src/main/scala/org/apache/spark/status/api/v1/JacksonMessageWriter.scala

@@ -21,7 +21,7 @@ import java.lang.annotation.Annotation
 import java.lang.reflect.Type
 import java.nio.charset.StandardCharsets
 import java.text.SimpleDateFormat
-import java.util.{Calendar, SimpleTimeZone}
+import java.util.{Calendar, Locale, SimpleTimeZone}
 import javax.ws.rs.Produces
 import javax.ws.rs.core.{MediaType, MultivaluedMap}
 import javax.ws.rs.ext.{MessageBodyWriter, Provider}
@@ -86,7 +86,7 @@ private[v1] class JacksonMessageWriter extends MessageBodyWriter[Object]{
 
 private[spark] object JacksonMessageWriter {
   def makeISODateFormat: SimpleDateFormat = {
-    val iso8601 = new SimpleDateFormat("yyyy-MM-dd'T'HH:mm:ss.SSS'GMT'")
+    val iso8601 = new SimpleDateFormat("yyyy-MM-dd'T'HH:mm:ss.SSS'GMT'", Locale.US)
     val cal = Calendar.getInstance(new SimpleTimeZone(0, "GMT"))
     iso8601.setCalendar(cal)
     iso8601

core/src/main/scala/org/apache/spark/status/api/v1/SimpleDateParam.scala

@@ -17,20 +17,20 @@
 package org.apache.spark.status.api.v1
 
 import java.text.{ParseException, SimpleDateFormat}
-import java.util.TimeZone
+import java.util.{Locale, TimeZone}
 import javax.ws.rs.WebApplicationException
 import javax.ws.rs.core.Response
 import javax.ws.rs.core.Response.Status
 
 private[v1] class SimpleDateParam(val originalValue: String) {
 
   val timestamp: Long = {
-    val format = new SimpleDateFormat("yyyy-MM-dd'T'HH:mm:ss.SSSz")
+    val format = new SimpleDateFormat("yyyy-MM-dd'T'HH:mm:ss.SSSz", Locale.US)
     try {
       format.parse(originalValue).getTime()
     } catch {
       case _: ParseException =>
-        val gmtDay = new SimpleDateFormat("yyyy-MM-dd")
+        val gmtDay = new SimpleDateFormat("yyyy-MM-dd", Locale.US)
         gmtDay.setTimeZone(TimeZone.getTimeZone("GMT"))
         try {
           gmtDay.parse(originalValue).getTime()

core/src/main/scala/org/apache/spark/ui/UIUtils.scala

@@ -36,7 +36,8 @@ private[spark] object UIUtils extends Logging {
 
   // SimpleDateFormat is not thread-safe. Don't expose it to avoid improper use.
   private val dateFormat = new ThreadLocal[SimpleDateFormat]() {
-    override def initialValue(): SimpleDateFormat = new SimpleDateFormat("yyyy/MM/dd HH:mm:ss")
+    override def initialValue(): SimpleDateFormat =
+      new SimpleDateFormat("yyyy/MM/dd HH:mm:ss", Locale.US)
   }
 
   def formatDate(date: Date): String = dateFormat.get.format(date)

core/src/main/scala/org/apache/spark/util/logging/RollingPolicy.scala

@@ -18,7 +18,7 @@
 package org.apache.spark.util.logging
 
 import java.text.SimpleDateFormat
-import java.util.Calendar
+import java.util.{Calendar, Locale}
 
 import org.apache.spark.internal.Logging
 
@@ -59,7 +59,7 @@ private[spark] class TimeBasedRollingPolicy(
   }
 
   @volatile private var nextRolloverTime = calculateNextRolloverTime()
-  private val formatter = new SimpleDateFormat(rollingFileSuffixPattern)
+  private val formatter = new SimpleDateFormat(rollingFileSuffixPattern, Locale.US)
 
   /** Should rollover if current time has exceeded next rollover time */
   def shouldRollover(bytesToBeWritten: Long): Boolean = {
@@ -109,7 +109,7 @@ private[spark] class SizeBasedRollingPolicy(
   }
 
   @volatile private var bytesWrittenSinceRollover = 0L
-  val formatter = new SimpleDateFormat("--yyyy-MM-dd--HH-mm-ss--SSSS")
+  val formatter = new SimpleDateFormat("--yyyy-MM-dd--HH-mm-ss--SSSS", Locale.US)
 
   /** Should rollover if the next set of bytes is going to exceed the size limit */
   def shouldRollover(bytesToBeWritten: Long): Boolean = {

core/src/test/scala/org/apache/spark/util/UtilsSuite.scala

@@ -264,7 +264,7 @@ class UtilsSuite extends SparkFunSuite with ResetSystemProperties with Logging {
     val hour = minute * 60
     def str: (Long) => String = Utils.msDurationToString(_)
 
-    val sep = new DecimalFormatSymbols(Locale.getDefault()).getDecimalSeparator()
+    val sep = new DecimalFormatSymbols(Locale.US).getDecimalSeparator
 
     assert(str(123) === "123 ms")
     assert(str(second) === "1" + sep + "0 s")

mesos/src/main/scala/org/apache/spark/deploy/rest/mesos/MesosRestServer.scala

@@ -19,7 +19,7 @@ package org.apache.spark.deploy.rest.mesos
 
 import java.io.File
 import java.text.SimpleDateFormat
-import java.util.Date
+import java.util.{Date, Locale}
 import java.util.concurrent.atomic.AtomicLong
 import javax.servlet.http.HttpServletResponse
 
@@ -62,11 +62,10 @@ private[mesos] class MesosSubmitRequestServlet(
   private val DEFAULT_CORES = 1.0
 
   private val nextDriverNumber = new AtomicLong(0)
-  private def createDateFormat = new SimpleDateFormat("yyyyMMddHHmmss")  // For application IDs
-  private def newDriverId(submitDate: Date): String = {
-    "driver-%s-%04d".format(
-      createDateFormat.format(submitDate), nextDriverNumber.incrementAndGet())
-  }
+  // For application IDs
+  private def createDateFormat = new SimpleDateFormat("yyyyMMddHHmmss", Locale.US)
+  private def newDriverId(submitDate: Date): String =
+    f"driver-${createDateFormat.format(submitDate)}-${nextDriverNumber.incrementAndGet()}%04d"
 
   /**
    * Build a driver description from the fields specified in the submit request.

mllib/src/main/scala/org/apache/spark/mllib/pmml/export/PMMLModelExport.scala

@@ -18,7 +18,7 @@
 package org.apache.spark.mllib.pmml.export
 
 import java.text.SimpleDateFormat
-import java.util.Date
+import java.util.{Date, Locale}
 
 import scala.beans.BeanProperty
 
@@ -34,7 +34,7 @@ private[mllib] trait PMMLModelExport {
     val version = getClass.getPackage.getImplementationVersion
     val app = new Application("Apache Spark MLlib").setVersion(version)
     val timestamp = new Timestamp()
-      .addContent(new SimpleDateFormat("yyyy-MM-dd'T'HH:mm:ss").format(new Date()))
+      .addContent(new SimpleDateFormat("yyyy-MM-dd'T'HH:mm:ss", Locale.US).format(new Date()))
     val header = new Header()
       .setApplication(app)
       .setTimestamp(timestamp)

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/datetimeExpressions.scala

@@ -18,7 +18,7 @@
 package org.apache.spark.sql.catalyst.expressions
 
 import java.text.SimpleDateFormat
-import java.util.{Calendar, TimeZone}
+import java.util.{Calendar, Locale, TimeZone}
 
 import scala.util.Try
 
@@ -331,7 +331,7 @@ case class DateFormatClass(left: Expression, right: Expression) extends BinaryEx
   override def inputTypes: Seq[AbstractDataType] = Seq(TimestampType, StringType)
 
   override protected def nullSafeEval(timestamp: Any, format: Any): Any = {
-    val sdf = new SimpleDateFormat(format.toString)
+    val sdf = new SimpleDateFormat(format.toString, Locale.US)
     UTF8String.fromString(sdf.format(new java.util.Date(timestamp.asInstanceOf[Long] / 1000)))
   }
 
@@ -400,7 +400,7 @@ abstract class UnixTime extends BinaryExpression with ExpectsInputTypes {
 
   private lazy val constFormat: UTF8String = right.eval().asInstanceOf[UTF8String]
   private lazy val formatter: SimpleDateFormat =
-    Try(new SimpleDateFormat(constFormat.toString)).getOrElse(null)
+    Try(new SimpleDateFormat(constFormat.toString, Locale.US)).getOrElse(null)
 
   override def eval(input: InternalRow): Any = {
     val t = left.eval(input)
@@ -425,7 +425,7 @@ abstract class UnixTime extends BinaryExpression with ExpectsInputTypes {
             null
           } else {
             val formatString = f.asInstanceOf[UTF8String].toString
-            Try(new SimpleDateFormat(formatString).parse(
+            Try(new SimpleDateFormat(formatString, Locale.US).parse(
               t.asInstanceOf[UTF8String].toString).getTime / 1000L).getOrElse(null)
           }
       }
@@ -520,7 +520,7 @@ case class FromUnixTime(sec: Expression, format: Expression)
 
   private lazy val constFormat: UTF8String = right.eval().asInstanceOf[UTF8String]
   private lazy val formatter: SimpleDateFormat =
-    Try(new SimpleDateFormat(constFormat.toString)).getOrElse(null)
+    Try(new SimpleDateFormat(constFormat.toString, Locale.US)).getOrElse(null)
 
   override def eval(input: InternalRow): Any = {
     val time = left.eval(input)
@@ -539,9 +539,10 @@ case class FromUnixTime(sec: Expression, format: Expression)
         if (f == null) {
           null
         } else {
-          Try(UTF8String.fromString(new SimpleDateFormat(
-            f.asInstanceOf[UTF8String].toString).format(new java.util.Date(
-              time.asInstanceOf[Long] * 1000L)))).getOrElse(null)
+          Try(
+            UTF8String.fromString(new SimpleDateFormat(f.toString, Locale.US).
+              format(new java.util.Date(time.asInstanceOf[Long] * 1000L)))
+          ).getOrElse(null)
         }
       }
     }

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/stringExpressions.scala

@@ -1415,7 +1415,7 @@ case class Sentences(
       val locale = if (languageStr != null && countryStr != null) {
         new Locale(languageStr.toString, countryStr.toString)
       } else {
-        Locale.getDefault
+        Locale.US
       }
       getSentences(string.asInstanceOf[UTF8String].toString, locale)
     }

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/json/JSONOptions.scala

@@ -17,6 +17,8 @@
 
 package org.apache.spark.sql.catalyst.json
 
+import java.util.Locale
+
 import com.fasterxml.jackson.core.{JsonFactory, JsonParser}
 import org.apache.commons.lang3.time.FastDateFormat
 
@@ -56,11 +58,11 @@ private[sql] class JSONOptions(
 
   // Uses `FastDateFormat` which can be direct replacement for `SimpleDateFormat` and thread-safe.
   val dateFormat: FastDateFormat =
-    FastDateFormat.getInstance(parameters.getOrElse("dateFormat", "yyyy-MM-dd"))
+    FastDateFormat.getInstance(parameters.getOrElse("dateFormat", "yyyy-MM-dd"), Locale.US)
 
   val timestampFormat: FastDateFormat =
     FastDateFormat.getInstance(
-      parameters.getOrElse("timestampFormat", "yyyy-MM-dd'T'HH:mm:ss.SSSZZ"))
+      parameters.getOrElse("timestampFormat", "yyyy-MM-dd'T'HH:mm:ss.SSSZZ"), Locale.US)
 
   // Parse mode flags
   if (!ParseModes.isValidMode(parseMode)) {

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/util/DateTimeUtils.scala

@@ -19,7 +19,7 @@ package org.apache.spark.sql.catalyst.util
 
 import java.sql.{Date, Timestamp}
 import java.text.{DateFormat, SimpleDateFormat}
-import java.util.{Calendar, TimeZone}
+import java.util.{Calendar, Locale, TimeZone}
 import javax.xml.bind.DatatypeConverter
 
 import scala.annotation.tailrec
@@ -79,14 +79,14 @@ object DateTimeUtils {
   // `SimpleDateFormat` is not thread-safe.
   val threadLocalTimestampFormat = new ThreadLocal[DateFormat] {
     override def initialValue(): SimpleDateFormat = {
-      new SimpleDateFormat("yyyy-MM-dd HH:mm:ss")
+      new SimpleDateFormat("yyyy-MM-dd HH:mm:ss", Locale.US)
     }
   }
 
   // `SimpleDateFormat` is not thread-safe.
   private val threadLocalDateFormat = new ThreadLocal[DateFormat] {
     override def initialValue(): SimpleDateFormat = {
-      new SimpleDateFormat("yyyy-MM-dd")
+      new SimpleDateFormat("yyyy-MM-dd", Locale.US)
     }
   }