A simple way for reducer to get how many records are processed in mapreduce job's mapper

JobClient client = new JobClient(jobConf);
RunningJob mapreduceJob = client.getJob(JobID.forName( jobConf.get("mapred.job.id") ));
long mapProcessedRecordCounter = mapreduceJob .getCounters().findCounter("org.apache.hadoop.mapred.Task$Counter", "MAP_INPUT_RECORDS").getCounter();