Alternative versions of word count mappers that are closer to pseudo-…

…code.
lintool · Jan 10, 2017 · 0bdde0a · 0bdde0a
1 parent d8b1198
commit 0bdde0a
Showing 1 changed file with 42 additions and 0 deletions.
diff --git a/src/main/scala/io/bespin/scala/mapreduce/wordcount/WordCount.scala b/src/main/scala/io/bespin/scala/mapreduce/wordcount/WordCount.scala
@@ -53,6 +53,16 @@ object WordCount extends Configured with Tool with WritableConversions with Toke
     }
   }
 
+  // Same as above, except using explicit loops to look more like pseudo-code
+  class AlternativeMapper extends Mapper[LongWritable, Text, Text, IntWritable] {
+    override def map(key: LongWritable, value: Text,
+                     context: Mapper[LongWritable, Text, Text, IntWritable]#Context) = {
+      for (word <- tokenize(value)) {
+        context.write(word, 1)
+      }
+    }
+  }
+
   class MyMapperHistogram extends Mapper[LongWritable, Text, Text, IntWritable] {
     override def map(key: LongWritable, value: Text,
                      context: Mapper[LongWritable, Text, Text, IntWritable]#Context) = {
@@ -62,6 +72,20 @@ object WordCount extends Configured with Tool with WritableConversions with Toke
     }
   }
 
+  // Same as above, except using explicit loops to look more like pseudo-code
+  class AlternativeMapperHistogram extends Mapper[LongWritable, Text, Text, IntWritable] {
+    override def map(key: LongWritable, value: Text,
+                     context: Mapper[LongWritable, Text, Text, IntWritable]#Context) = {
+      val counts = new HashMap[String, Int]() { override def default(key: String) = 0 }
+      for (word <- tokenize(value)) {
+        counts(word) += 1
+      }
+      for ((k, v) <- counts) {
+        context.write(k, v)
+      }
+    }
+  }
+
   class MyMapperIMC extends Mapper[LongWritable, Text, Text, IntWritable] {
     val counts = new HashMap[String, Int]().withDefaultValue(0)
 
@@ -75,6 +99,24 @@ object WordCount extends Configured with Tool with WritableConversions with Toke
     }
   }
 
+  // Same as above, except using explicit loops to look more like pseudo-code
+  class AlternativeMapperIMC extends Mapper[LongWritable, Text, Text, IntWritable] {
+    val counts = new HashMap[String, Int]().withDefaultValue(0)
+
+    override def map(key: LongWritable, value: Text,
+                     context: Mapper[LongWritable, Text, Text, IntWritable]#Context) = {
+      for (word <- tokenize(value)) {
+        counts(word) += 1
+      }
+    }
+
+    override def cleanup(context: Mapper[LongWritable, Text, Text, IntWritable]#Context) = {
+      for ((k, v) <- counts) {
+        context.write(k, v)
+      }
+    }
+  }
+
   class MyReducer extends Reducer[Text, IntWritable, Text, IntWritable] {
     override def reduce(key: Text, values: java.lang.Iterable[IntWritable],
                         context: Reducer[Text, IntWritable, Text, IntWritable]#Context) = {