MR- Iterator over Iteration, Can't parse value for output - mapreduce

Question

The raw file for mapreduce is like this (delimiter: Tab)
Apple 11 12 13
Orange 15 26 10
When I try to implement the method to add a new feature and separate the numbers using ",", my expected output should be like this:
Apple 3.0:11,12,13
Orange 3.0:15,26,10
But the final output gets like below:
Apple 3.0:11 12 13
Orange 3.0:15 26 10
I try to print the result for tracing, but it seems next() will skip the parsing and directly jump out the loop. Can anyone help on this?
public static class Mapper1 extends MapReduceBase
implements Mapper<Text, Text, Text, Text> {
#Override
public void map(Text key, Text value, OutputCollector<Text, Text> output,
Reporter reporter) throws IOException {
output.collect(key, value);
}
}
public static class Reducer1 extends MapReduceBase
implements Reducer<Text, Text, Text, Text> {
public void reduce(Text key, Iterator<Text> values,
OutputCollector<Text, Text> output, Reporter reporter)
throws IOException {
String feature = "3.0:";
boolean first = true;
while(values.hasNext()) {
if(!first) {
feature += ",";
}
feature += values.next().toString() ;
System.out.println("count"+feature.length+","+feature);
first = false;
}
output.collect(key, new Text(feature));
}
}

Answer 1

I think it is because your mapper only emits one key value pair for each record, which is not expected. You can check your mapper's output by setting the reducer number to 0 in your driver code:
job.setNumReduceTasks(0);
Mapper input:
Apple 11 12 13
Orange 15 26 10
Actual mapper output: (key, value)
(Apple, 11 12 13)
(Orange, 15 26 10)
Expected mapper output: (key, value)
(Apple, 11)
(Apple, 12)
...
(Orange, 10)
You can either modify your mapper to emit multiple key value pairs for each record, or use split() method of String class to get the substrings from the original one.