Added TopTitle implementation xldrx#1.

andresgomezfrr · Sep 11, 2015 · 140b27b · 140b27b
1 parent a5439e6
commit 140b27b
Showing 1 changed file with 42 additions and 7 deletions.
diff --git a/TopTitles.java b/TopTitles.java
@@ -126,20 +126,31 @@ protected void setup(Context context) throws IOException,InterruptedException {
 
         @Override
         public void map(Object key, Text value, Context context) throws IOException, InterruptedException {
-        // TODO
+           String line = value.toString();
+           StringTokenizer tokenizer = new StringTokenizer(line, this.delimiters);
+           while (tokenizer.hasMoreTokens()) {
+               String nextToken = tokenizer.nextToken().trim().toLowerCase();
+               if (!this.stopWords.contains(nextToken)) {
+                  context.write(new Text(nextToken), new IntWritable(1)); 
+               } 
+           }  
         }
     }
 
     public static class TitleCountReduce extends Reducer<Text, IntWritable, Text, IntWritable> {
         @Override
         public void reduce(Text key, Iterable<IntWritable> values, Context context) throws IOException, InterruptedException {
-            // TODO
+            int count = 0;
+            for (IntWritable v : values) {
+                count += v.get();
+            }
+            context.write(key, new IntWritable(count));
         }
     }
 
     public static class TopTitlesMap extends Mapper<Text, Text, NullWritable, TextArrayWritable> {
         Integer N;
-        // TODO
+        private TreeSet<Pair<Integer, String>> countToWordMap = new TreeSet<Pair<Integer, String>>();
 
         @Override
         protected void setup(Context context) throws IOException,InterruptedException {
@@ -149,18 +160,27 @@ protected void setup(Context context) throws IOException,InterruptedException {
 
         @Override
         public void map(Text key, Text value, Context context) throws IOException, InterruptedException {
-            // TODO
+            Integer count = Integer.parseInt(value.toString());
+            String word = key.toString();
+            countToWordMap.add(new Pair<Integer, String>(count, word));
+            if (countToWordMap.size() > this.N) {
+                countToWordMap.remove(countToWordMap.first());
+            }
         }
 
         @Override
         protected void cleanup(Context context) throws IOException, InterruptedException {
-            // TODO
+            for (Pair<Integer, String> item : countToWordMap) {
+                String[] strings = {item.second, item.first.toString()};
+                TextArrayWritable val = new TextArrayWritable(strings);
+                context.write(NullWritable.get(), val);
+            }
         }
     }
 
     public static class TopTitlesReduce extends Reducer<NullWritable, TextArrayWritable, Text, IntWritable> {
         Integer N;
-        // TODO
+        private TreeSet<Pair<Integer, String>> countToWordMap = new TreeSet<Pair<Integer, String>>();
 
         @Override
         protected void setup(Context context) throws IOException,InterruptedException {
@@ -170,7 +190,22 @@ protected void setup(Context context) throws IOException,InterruptedException {
 
         @Override
         public void reduce(NullWritable key, Iterable<TextArrayWritable> values, Context context) throws IOException, InterruptedException {
-            // TODO
+            for (TextArrayWritable val: values) {
+                Text[] pair= (Text[]) val.toArray();
+                String word = pair[0].toString();
+                Integer count = Integer.parseInt(pair[1].toString());
+                countToWordMap.add(new Pair<Integer, String>(count, word));
+
+                if (countToWordMap.size() > this.N) {
+                    countToWordMap.remove(countToWordMap.first());
+                }
+            }
+
+            for (Pair<Integer, String> item: countToWordMap) {
+                Text word = new Text(item.second);
+                IntWritable value = new IntWritable(item.first);
+                context.write(word, value);
+            }
         }
     }