Package cascading.flow.hadoop

Examples of cascading.flow.hadoop.HadoopFlowProcess

39
40
41
42
43
44
45
   
    @Override
    protected FlowProcess<?> makeFlowProcess() {
        Map<Object, Object> props = new HashMap<Object, Object>();
        TupleSerializationProps.addSerialization(props, BytesSerialization.class.getName());
        return new HadoopFlowProcess(HadoopUtil.createJobConf(props, null));
    }
View Full Code Here
169
170
171
172
173
174
175
176
177
178
179
  private void performMapTest( int numKeys, int listSize, int mapThreshold, int listThreshold, Configuration jobConf )
    {
    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 );
View Full Code Here
111
112
113
114
115
116
117
    }

  @Override
  public FlowProcess getFlowProcess()
    {
    return new HadoopFlowProcess( FlowSession.NULL, (JobConf) getConfiguration(), true );
    }
View Full Code Here
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
    tributaries.remove( this.source ); // we cannot stream and accumulate the same source

    // accumulated paths
    for( Object source : tributaries )
      {
      HadoopFlowProcess hadoopProcess = (HadoopFlowProcess) flowProcess;
      JobConf conf = hadoopProcess.getJobConf();

      // allows client side config to be used cluster side
      String property = conf.getRaw( "cascading.node.accumulated.source.conf." + Tap.id( (Tap) source ) );

      if( property == null )
        throw new IllegalStateException( "accumulated source conf property missing for: " + ( (Tap) source ).getIdentifier() );

      conf = getSourceConf( hadoopProcess, conf, property );
      flowProcess = new HadoopFlowProcess( hadoopProcess, conf );

      handleHead( (Tap) source, flowProcess );
      }
    }
View Full Code Here
58
59
60
61
62
63
64
65
66
67
68
    Hfs temp = new Hfs( new TextLine(), statePath, SinkMode.REPLACE );

    try
      {
      TupleEntryCollector writer = temp.openForWrite( new HadoopFlowProcess( conf ) );

      writer.add( new Tuple( stepState ) );

      writer.close();
      }
View Full Code Here
103
104
105
106
107
108
109
110
111
112
113
    TupleEntryIterator reader = null;

    try
      {
      reader = temp.openForRead( new HadoopFlowProcess( jobConf ) );

      if( !reader.hasNext() )
        throw new FlowException( "step state path is empty: " + temp.getIdentifier() );

      return reader.next().getString( 0 );
View Full Code Here
103
104
105
106
107
108
109
    }

  @Override
  public FlowProcess getFlowProcess()
    {
    return new HadoopFlowProcess( FlowSession.NULL, getConfiguration(), true );
    }
View Full Code Here
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
  public void testAsGroupByValue() throws Exception {
    FileSystem.get(new Configuration()).delete(new Path("/tmp/input"), true);
    FileSystem.get(new Configuration()).delete(new Path("/tmp/output"), true);

    Tap t = new Hfs(new ProtobufScheme("value", Example.Person.class), "/tmp/input");
    TupleEntryCollector tec = t.openForWrite(new HadoopFlowProcess(new JobConf()));

    HashSet<Tuple> expectedTuples = new HashSet<Tuple>(){{
      add(new Tuple(Example.Person.newBuilder().setName("bryan").setId(1).build()));
      add(new Tuple(Example.Person.newBuilder().setName("lucas").setId(2).build()));
    }};

    for (Tuple tuple : expectedTuples) {
      tec.add(tuple);
    }

    tec.close();

    Pipe inPipe = new Pipe("input");
    Pipe injectedPipe = new Each(inPipe, Fields.NONE, new Insert(new Fields("key"), 7), new Fields("key", "value"));
    Pipe groupByPipe = new GroupBy(injectedPipe, new Fields("key"));

    Hfs sink = new Hfs(new ProtobufScheme("value", Example.Person.class), "/tmp/output");
    Map<Object, Object> properties = new HashMap<Object, Object>(){{
      put("io.serializations", new JobConf().get("io.serializations") + "," + ProtobufSerialization.class.getName());
    }};
    new HadoopFlowConnector(properties).connect(t, sink, groupByPipe).complete();

    TupleEntryIterator tei = sink.openForRead(new HadoopFlowProcess(new JobConf()));
    Set<Tuple> tuples = new HashSet<Tuple>();
    while (tei.hasNext()) {
      tuples.add(tei.next().getTupleCopy());
    }
View Full Code Here
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
  public void testAsGroupByKey() throws Exception {
    FileSystem.get(new Configuration()).delete(new Path("/tmp/input"), true);
    FileSystem.get(new Configuration()).delete(new Path("/tmp/output"), true);

    Tap t = new Hfs(new ProtobufScheme("value", Example.Person.class), "/tmp/input");
    TupleEntryCollector tec = t.openForWrite(new HadoopFlowProcess(new JobConf()));

    HashSet<Tuple> expectedTuples = new HashSet<Tuple>(){{
      add(new Tuple(Example.Person.newBuilder().setName("bryan").setId(1).build()));
      add(new Tuple(Example.Person.newBuilder().setName("lucas").setId(2).build()));
    }};

    for (Tuple tuple : expectedTuples) {
      tec.add(tuple);
    }

    tec.close();

    Pipe inPipe = new Pipe("input");
    Pipe groupByPipe = new GroupBy(inPipe, new Fields("value"));

    Hfs sink = new Hfs(new ProtobufScheme("value", Example.Person.class), "/tmp/output");
    Map<Object, Object> properties = new HashMap<Object, Object>(){{
      put("io.serializations",
          new JobConf().get("io.serializations") + "," + ProtobufSerialization.class.getName());
    }};
    new HadoopFlowConnector(properties).connect(t, sink, groupByPipe).complete();

    TupleEntryIterator tei = sink.openForRead(new HadoopFlowProcess(new JobConf()));
    Set<Tuple> tuples = new HashSet<Tuple>();
    while (tei.hasNext()) {
      tuples.add(tei.next().getTupleCopy());
    }
View Full Code Here
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
    expected.add(fixture("bryan", "bryan.duxbury@mail.com", 1));
    expected.add(fixture("lucas", "lucas@mail.com", 2));
    expected.add(fixture("vida", null, 3));

    Tap inputTap = new Hfs(new ProtobufScheme("value", Example.Person.class), "/tmp/input");
    TupleEntryCollector tec = inputTap.openForWrite(new HadoopFlowProcess(), null);

    for (Tuple t : expected) {
      tec.add(new TupleEntry(new Fields("value"), t));
    }
    tec.close();

    // read results back out
    Tap outputTap = new Hfs(new ProtobufScheme("value", Example.Person.class), "/tmp/input");
    TupleEntryIterator iter = outputTap.openForRead(new HadoopFlowProcess(), null);
    List<Tuple> tuples = new ArrayList<Tuple>();
    while (iter.hasNext()) {
      tuples.add(iter.next().getTupleCopy());
    }
View Full Code Here
TOP

Related Classes of cascading.flow.hadoop.HadoopFlowProcess

  • cascading.flow.hadoop.stream.graph.HadoopMapStreamGraph
  • cascading.flow.hadoop.util.HadoopMRUtil
  • cascading.lingual.platform.hadoop.HadoopPlatformBroker
  • cascading.lingual.platform.hadoop2.Hadoop2MR1PlatformBroker
  • cascading.platform.hadoop.HadoopPlatform
  • cascading.platform.hadoop2.Hadoop2MR1Platform
  • cascading.tuple.hadoop.SpillableTupleHadoopTest
  • cascading.tuple.TupleEntryCollector
  • com.etsy.cascading.tap.local.LocalTap
  • com.etsy.cascading.tap.local.LocalTap$LocalScheme
    • 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.