[SPARK-20156][CORE][SQL][STREAMING][MLLIB] Java String toLowerCase "Turkish locale bug" causes Spark problems

## What changes were proposed in this pull request?

Add Locale.ROOT to internal calls to String `toLowerCase`, `toUpperCase`, to avoid inadvertent locale-sensitive variation in behavior (aka the "Turkish locale problem").

The change looks large but it is just adding `Locale.ROOT` (the locale with no country or language specified) to every call to these methods.

## How was this patch tested?

Existing tests.

Author: Sean Owen <sowen@cloudera.com>

Closes #17527 from srowen/SPARK-20156.

4 files changed, 12 insertions, 4 deletions

diff --git a/examples/src/main/scala/org/apache/spark/examples/ml/DecisionTreeExample.scala b/examples/src/main/scala/org/apache/spark/examples/ml/DecisionTreeExample.scala
index 1745281c26..f736ceed44 100644
--- a/examples/src/main/scala/org/apache/spark/examples/ml/DecisionTreeExample.scala
+++ b/examples/src/main/scala/org/apache/spark/examples/ml/DecisionTreeExample.scala
@@ -18,6 +18,8 @@
 // scalastyle:off println
 package org.apache.spark.examples.ml
 
+import java.util.Locale
+
 import scala.collection.mutable
 import scala.language.reflectiveCalls
 
@@ -203,7 +205,7 @@ object DecisionTreeExample {
       .getOrCreate()
 
     params.checkpointDir.foreach(spark.sparkContext.setCheckpointDir)
-    val algo = params.algo.toLowerCase
+    val algo = params.algo.toLowerCase(Locale.ROOT)
 
     println(s"DecisionTreeExample with parameters:\n$params")
 
diff --git a/examples/src/main/scala/org/apache/spark/examples/ml/GBTExample.scala b/examples/src/main/scala/org/apache/spark/examples/ml/GBTExample.scala
index db55298d8e..ed598d0d7d 100644
--- a/examples/src/main/scala/org/apache/spark/examples/ml/GBTExample.scala
+++ b/examples/src/main/scala/org/apache/spark/examples/ml/GBTExample.scala
@@ -18,6 +18,8 @@
 // scalastyle:off println
 package org.apache.spark.examples.ml
 
+import java.util.Locale
+
 import scala.collection.mutable
 import scala.language.reflectiveCalls
 
@@ -140,7 +142,7 @@ object GBTExample {
       .getOrCreate()
 
     params.checkpointDir.foreach(spark.sparkContext.setCheckpointDir)
-    val algo = params.algo.toLowerCase
+    val algo = params.algo.toLowerCase(Locale.ROOT)
 
     println(s"GBTExample with parameters:\n$params")
 
diff --git a/examples/src/main/scala/org/apache/spark/examples/ml/RandomForestExample.scala b/examples/src/main/scala/org/apache/spark/examples/ml/RandomForestExample.scala
index a9e07c0705..8fd46c37e2 100644
--- a/examples/src/main/scala/org/apache/spark/examples/ml/RandomForestExample.scala
+++ b/examples/src/main/scala/org/apache/spark/examples/ml/RandomForestExample.scala
@@ -18,6 +18,8 @@
 // scalastyle:off println
 package org.apache.spark.examples.ml
 
+import java.util.Locale
+
 import scala.collection.mutable
 import scala.language.reflectiveCalls
 
@@ -146,7 +148,7 @@ object RandomForestExample {
       .getOrCreate()
 
     params.checkpointDir.foreach(spark.sparkContext.setCheckpointDir)
-    val algo = params.algo.toLowerCase
+    val algo = params.algo.toLowerCase(Locale.ROOT)
 
     println(s"RandomForestExample with parameters:\n$params")
 
diff --git a/examples/src/main/scala/org/apache/spark/examples/mllib/LDAExample.scala b/examples/src/main/scala/org/apache/spark/examples/mllib/LDAExample.scala
index b923e627f2..cd77ecf990 100644
--- a/examples/src/main/scala/org/apache/spark/examples/mllib/LDAExample.scala
+++ b/examples/src/main/scala/org/apache/spark/examples/mllib/LDAExample.scala
@@ -18,6 +18,8 @@
 // scalastyle:off println
 package org.apache.spark.examples.mllib
 
+import java.util.Locale
+
 import org.apache.log4j.{Level, Logger}
 import scopt.OptionParser
 
@@ -131,7 +133,7 @@ object LDAExample {
     // Run LDA.
     val lda = new LDA()
 
-    val optimizer = params.algorithm.toLowerCase match {
+    val optimizer = params.algorithm.toLowerCase(Locale.ROOT) match {
       case "em" => new EMLDAOptimizer
       // add (1.0 / actualCorpusSize) to MiniBatchFraction be more robust on tiny datasets.
       case "online" => new OnlineLDAOptimizer().setMiniBatchFraction(0.05 + 1.0 / actualCorpusSize)