Package cascading.tuple.hadoop.collect

Examples of cascading.tuple.hadoop.collect.HadoopSpillableTupleMap


    {
    jobConf.set( "io.serializations", TestSerialization.class.getName() + "," + WritableSerialization.class.getName() ); // disable/replace WritableSerialization class
    jobConf.set( "cascading.serialization.tokens", "1000=" + BooleanWritable.class.getName() + ",10001=" + Text.class.getName() ); // not using Text, just testing parsing

    HadoopFlowProcess flowProcess = new HadoopFlowProcess( jobConf );
    HadoopSpillableTupleMap map = new HadoopSpillableTupleMap( SpillableProps.defaultMapInitialCapacity, SpillableProps.defaultMapLoadFactor, mapThreshold, listThreshold, flowProcess );

    Set<Integer> keySet = new HashSet<Integer>();
    Random gen = new Random( 1 );

    for( int i = 0; i < listSize * numKeys; i++ )
      {
      String aString = "string number " + i;
      double random = Math.random();

      double keys = numKeys / 3.0;
      int key = (int) ( gen.nextDouble() * keys + gen.nextDouble() * keys + gen.nextDouble() * keys );

      Tuple tuple = new Tuple( i, aString, random, new Text( aString ), new TestText( aString ), new Tuple( "inner tuple", new BytesWritable( aString.getBytes() ) ) );

      map.get( new Tuple( key ) ).add( tuple );

      keySet.add( key );
      }

    // the list test above verifies the contents are being serialized, the Map is just a container of lists.
    assertEquals( "not equal: map.size();", keySet.size(), map.size() );
    }
View Full Code Here

TOP

Related Classes of cascading.tuple.hadoop.collect.HadoopSpillableTupleMap

Copyright © 2018 www.massapicom. All rights reserved.
All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.