Merge pull request csuldw#4 from Canappeco/patch-1

csuldw · web-flow · commit 67ac02f7895a · 2017-09-14T19:00:45.000+08:00
Update invertedIndex.scala
diff --git a/spark-demo/invertedIndex/src/main/scala/invertedIndex.scala b/spark-demo/invertedIndex/src/main/scala/invertedIndex.scala
@@ -18,10 +18,11 @@ object InvertedIndex{
     val cfg = new HierConf(args(0))
     val inputfile = cfg.getString("inputfile")
     val result = sc.textFile(inputfile)
-      .map(x => x.split("\t"))
-      .map(x => (x(0), x(1)))
-      .map(x => x._2.split(" ").map(y => (y, x._1)))
-      .flatMap(x => x)
+      .flatMap(x => {
+          val arr = x.split("\t")
+          val arr_words = arr(1).split(" ")
+          arr_words.map(y => (y, arr(0)))
+        })
       .reduceByKey( (x, y) => x + "|" + y)
     result.collect.foreach(println)
     sc.stop()
