java/pig-galago/src/com/yahoo/pig/GroupFunc.java

   1 /*
   2  * Copyright (c) 2007 Yahoo! Inc.  All rights reserved.
   3  * See accompanying LICENSE file.
   4  */
   5 package com.yahoo.pig;
   6
   7 import java.io.IOException;
   8
   9 import com.yahoo.pig.data.Datum;
  10 import com.yahoo.pig.data.Tuple;
  11 import com.yahoo.pig.impl.logicalLayer.schema.SchemaItem;
  12
  13 /**
  14  * The class is used to implement functions that generate an array of Datums to represent
  15  * the group of a given tuple. The group can be thought of as the "key", in Map-Reduce
  16  * terminology.
  17  *
  18  * The programmer should not make assumptions about state maintained
  19  * between invocations of the invoke() method since the Pig runtime
  20  * will schedule and localize invocations based on information provided
  21  * at runtime.
  22  *
  23  * @author database-systems@yahoo.research
  24  *
  25  */
  26 public abstract class GroupFunc extends InstantiatedFunc {
  27
  28         /**
  29          * This callback method must be implemented by all subclasses. This method
  30          * defines the group that will be used by the grouping operator.
  31          * Since the dataset may be divided up in a variety of ways the programmer
  32          * should not make assumptions about state that is maintained between
  33          * invocations of this method.
  34          *
  35          * @param input the tuple to be processed.
  36          * @return the group that the specified tuple belongs to.
  37          * @throws IOException
  38          */
  39         abstract public Datum[] exec(Tuple input);
  40
  41         public String name() { return this.getClass().getName();}
  42     /**
  43      * Returns the String representation of the function. It will
  44      * have a function syntax: name(ColumnList).
  45      */
  46         public String toString() {
  47                 return this.name();
  48         }
  49
  50         public SchemaItem outputSchema(SchemaItem input){
  51                 return input;
  52         }
  53 }