Start line:  
End line:  

Snippet Preview

Snippet HTML Code

Stack Overflow Questions
 /*
  * Licensed to the Apache Software Foundation (ASF) under one
  * or more contributor license agreements.  See the NOTICE file
  * distributed with this work for additional information
  * regarding copyright ownership.  The ASF licenses this file
  * to you under the Apache License, Version 2.0 (the
  * "License"); you may not use this file except in compliance
  * with the License.  You may obtain a copy of the License at
  *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */
package org.apache.pig.backend.hadoop.executionengine.mapReduceLayer;
A special implementation of combiner used only for distinct. This combiner does not even parse out the records. It just throws away duplicate values in the key in order ot minimize the data being sent to the reduce.
public class DistinctCombiner {
    public static class Combine 
        
        private final Log log = LogFactory.getLog(getClass());
        ProgressableReporter pigReporter;
        
        
Configures the reporter
        @Override
        protected void setup(Context contextthrows IOExceptionInterruptedException {
            super.setup(context);
             = new ProgressableReporter();
        }
        
        
The reduce function which removes values.
        @Override
        protected void reduce(PigNullableWritable keyIterable<NullableTupletupIterContext context
                throws IOExceptionInterruptedException {
            
            .setRep(context);
            // Take the first value and the key and collect
            // just that.
            Iterator<NullableTupleiter = tupIter.iterator();
            NullableTuple val = iter.next();
            context.write(keyval);
        }
    }
    
New to GrepCode? Check out our FAQ X