Hadoop 将自定义对象传递给reducer，完全获取null

我是mapreduce的新手，我的映射器输出了一个DbWritable对象，在reducer过程中，我无法从传递的对象中获得任何值，也许它根本就没有传递？ 这是我的密码

可写数据库

public   class StWritable implements DBWritable,Writable { 
    private String stockName; 
    private double q1,q2,q3,q4; // quarter  of year 
    ... 
    @Override
    public void readFields(DataInput input) throws IOException {

    }
    @Override
    public void write(DataOutput output) throws IOException {

    }
    @Override
    public void readFields(ResultSet rs) throws SQLException {
       stockName = rs.getString("STOCK_NAME");
      // year      = rs.getInt("RECORDING_YEAR");
       q1        = rs.getDouble("Q1");
       q2        = rs.getDouble("Q2");
       q3        = rs.getDouble("Q3");
       q4        = rs.getDouble("Q4");   
    }
    @Override
    public void write(PreparedStatement pstmt) throws SQLException {
       pstmt.setString(1, stockName);
       pstmt.setDouble(2, q1); 
       pstmt.setDouble(3, q2); 
       pstmt.setDouble(4, q3); 
       pstmt.setDouble(5, q4);  
    }
    @Override
    public String toString() {
        return Double.toString(q1)+","+Double.toString(q2)+","+Double.toString(q3)+","+Double.toString(q4);
    }    
}

制图员

public static class StockMapper extends Mapper<NullWritable, StWritable, Text , StWritable> {

        private Text stock = new Text(); 
        private Text value = new Text(); 
        private StWritable  stockq=  new StWritable() ;
        public static final Log log = LogFactory.getLog(StockMapper.class);

        @Override
        protected void map(NullWritable key, StWritable stockqWritable, Context context) throws IOException, InterruptedException {

          final String stockName = stockqWritable.getStockName();          
          final Double q1        = stockqWritable.getQ1();
          final Double q2        = stockqWritable.getQ2();
          final Double q3        = stockqWritable.getQ3();
          final Double q4        = stockqWritable.getQ4();

           stock.set(stockName);
           stockq.setStockName(stockName);
           stockq.setQ1(q1);
           stockq.setQ2(q2);
           stockq.setQ3(q3);
           stockq.setQ4(q4);
          // value.set(stockq.toString());
           log.info("map the stockq value is  "+ stockqWritable.toString());
           context.write(stock,stockq);
        }

    }

一些日志信息

INFO StockMr$StockMapper: map the stockq value is  86.29,86.58,81.9,83.8
INFO StockMr$StockMapper: map the stockq value is  199.27,200.26,192.55,194.84

INFO StockMr$StockReducer: reduce  the stockq value is  4.9E-324,4.9E-324,4.9E-324,4.9E-324
INFO StockMr$StockReducer: reduce  the stockq value is  4.9E-324,4.9E-324,4.9E-324,4.9E-324

谢谢

参考文献

新编辑

我有两种解决问题的方法：一种是映射程序将复合值作为文本发送到reducer，然后将Text.toString（）.split（“，”）发送到reducer； 另一种方法是映射程序将MapWritable发送到reducer。 我不知道哪个更好，或者两者都不好

configuration.set("hbase.zookeeper.quorum", "backup103");
final String selectQuery = "SELECT STOCK_NAME,Q1,Q2,Q3,Q4 FROM STOCKQ"; 
final Job job = Job.getInstance(configuration, "phoenix-mr-job");    
job.setJarByClass(StockMr.class);
PhoenixMapReduceUtil.setInput(job, StWritable.class, "STOCKQ",  selectQuery);
PhoenixMapReduceUtil.setOutput(job, "STOCKQ_SUM", "STOCK_NAME,Q1,Q2,Q3,Q4");
job.setMapperClass(StockMapper.class);
//job.setCombinerClass(StockCombiner.class);
job.setReducerClass(StockReducer.class); 
job.setOutputFormatClass(PhoenixOutputFormat.class);
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(StWritable.class);
job.setOutputKeyClass(NullWritable.class);
job.setOutputValueClass(StWritable.class); 
job.setNumReduceTasks(1);
TableMapReduceUtil.addDependencyJars(job);
job.waitForCompletion(true); 

INFO StockMr$StockMapper: map the stockq value is  86.29,86.58,81.9,83.8
INFO StockMr$StockMapper: map the stockq value is  199.27,200.26,192.55,194.84

INFO StockMr$StockReducer: reduce  the stockq value is  4.9E-324,4.9E-324,4.9E-324,4.9E-324
INFO StockMr$StockReducer: reduce  the stockq value is  4.9E-324,4.9E-324,4.9E-324,4.9E-324