From a41ca203753f1c4f533dd78d41cf2f821b332990 Mon Sep 17 00:00:00 2001
From: Matei Zaharia <matei@eecs.berkeley.edu>
Date: Mon, 4 Oct 2010 12:01:05 -0700
Subject: [PATCH] Added splitWords function in Utils
---
src/scala/spark/Utils.scala | 27 ++++++++++++++++++++++++++-
1 file changed, 26 insertions(+), 1 deletion(-)
diff --git a/src/scala/spark/Utils.scala b/src/scala/spark/Utils.scala
index 52bcb89f00..27d73aefbd 100644
--- a/src/scala/spark/Utils.scala
+++ b/src/scala/spark/Utils.scala
@@ -2,7 +2,9 @@ package spark
import java.io._
-private object Utils {
+import scala.collection.mutable.ArrayBuffer
+
+object Utils {
def serialize[T](o: T): Array[Byte] = {
val bos = new ByteArrayOutputStream
val oos = new ObjectOutputStream(bos)
@@ -25,4 +27,27 @@ private object Utils {
}
return ois.readObject.asInstanceOf[T]
}
+
+ def isAlpha(c: Char) = {
+ (c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z')
+ }
+
+ def splitWords(s: String): Seq[String] = {
+ val buf = new ArrayBuffer[String]
+ var i = 0
+ while (i < s.length) {
+ var j = i
+ while (j < s.length && isAlpha(s.charAt(j))) {
+ j += 1
+ }
+ if (j > i) {
+ buf += s.substring(i, j);
+ }
+ i = j
+ while (i < s.length && !isAlpha(s.charAt(i))) {
+ i += 1
+ }
+ }
+ return buf
+ }
}
--
GitLab