Merge pull request #31 from apache/CRUNCH-693
CRUNCH-693: Make text parsing locale-independent
diff --git a/crunch-contrib/src/main/java/org/apache/crunch/contrib/text/TokenizerFactory.java b/crunch-contrib/src/main/java/org/apache/crunch/contrib/text/TokenizerFactory.java
index f43478d..555ad8c 100644
--- a/crunch-contrib/src/main/java/org/apache/crunch/contrib/text/TokenizerFactory.java
+++ b/crunch-contrib/src/main/java/org/apache/crunch/contrib/text/TokenizerFactory.java
@@ -17,14 +17,14 @@
*/
package org.apache.crunch.contrib.text;
+import com.google.common.base.Preconditions;
+import com.google.common.collect.ImmutableSet;
+
import java.io.Serializable;
import java.util.Locale;
import java.util.Scanner;
import java.util.Set;
-import com.google.common.base.Preconditions;
-import com.google.common.collect.ImmutableSet;
-
/**
* Factory class that constructs {@link Tokenizer} instances for input strings that use a fixed
* set of delimiters, skip patterns, locales, and sets of indices to keep or drop.
@@ -65,6 +65,7 @@
*/
public Tokenizer create(String input) {
Scanner s = new Scanner(input);
+ s.useLocale(Locale.US); // Use period for floating point number formatting
if (delim != null) {
s.useDelimiter(delim);
}
