读代码-InputMapper

package org.apache.mahout.clustering.conversion;
目的：读取输入转换成vector输出

private static final Pattern SPACE = Pattern.compile(" ");
private Constructor<?> constructor;

用反射加载vector类

Configuration conf = context.getConfiguration();
String vectorImplClassName = conf.get("vector.implementation.class.name");
Class<? extends Vector> outputClass = conf.getClassByName(vectorImplClassName).asSubclass(Vector.class);
      constructor = outputClass.getConstructor(int.class);

正则按空格拆开,装入double链表,处理了连续空格情况

String[] numbers = InputMapper.SPACE.split(values.toString());    

Collection<Double> doubles = new ArrayList<Double>();
    for (String value : numbers) {
      if (value.length() > 0) {
        doubles.add(Double.valueOf(value));
      }
    }

建立vector,用writeable包装,key为vector的长度,value为vector写出

Vector result = (Vector) constructor.newInstance(doubles.size());
        int index = 0;
        for (Double d : doubles) {
          result.set(index++, d);
        }
        VectorWritable vectorWritable = new VectorWritable(result);
        context.write(new Text(String.valueOf(index)), vectorWritable);