Package org.apache.flink.api.java.record.operators

Examples of org.apache.flink.api.java.record.operators.FileDataSource

55
56
57
58
59
60
61
62
63
64
65
66
  @SuppressWarnings("unchecked")
  @Test
  public void testUnionPropertyOldApiPropagation() {
    // construct the plan

    FileDataSource sourceA = new FileDataSource(new DummyInputFormat(), IN_FILE);
    FileDataSource sourceB = new FileDataSource(new DummyInputFormat(), IN_FILE);
   
    ReduceOperator redA = ReduceOperator.builder(new IdentityReduce(), IntValue.class, 0)
      .input(sourceA)
      .build();
    ReduceOperator redB = ReduceOperator.builder(new IdentityReduce(), IntValue.class, 0)
View Full Code Here
54
55
56
57
58
59
60
61
62
63
64
65
    final String edgeInput = args.length > 1 ? args[1] : "";
    final String output    = args.length > 2 ? args[2] : "";
    final char delimiter   = args.length > 3 ? (char) Integer.parseInt(args[3]) : ',';
   

    FileDataSource edges = new FileDataSource(new EdgeInputFormat(), edgeInput, "Input Edges");
    edges.setParameter(EdgeInputFormat.ID_DELIMITER_CHAR, delimiter);

    // =========================== Vertex Degree ============================
   
    MapOperator projectEdge = MapOperator.builder(new ProjectEdge())
        .input(edges).name("Project Edge").build();
View Full Code Here
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
      this.outputPath = args[7];
    }
   
    /* Create the 6 data sources: */
    /* part: (partkey | name, mfgr, brand, type, size, container, retailprice, comment) */
    FileDataSource partInput = new FileDataSource(
      new IntTupleDataInFormat(), this.partInputPath, "\"part\" source");
    //partInput.setOutputContract(UniqueKey.class);
//    partInput.getCompilerHints().setAvgNumValuesPerKey(1);

    /* partsupp: (partkey | suppkey, availqty, supplycost, comment) */
    FileDataSource partSuppInput = new FileDataSource(
      new IntTupleDataInFormat(), this.partSuppInputPath, "\"partsupp\" source");

    /* orders: (orderkey | custkey, orderstatus, totalprice, orderdate, orderpriority, clerk, shippriority, comment) */
    FileDataSource ordersInput = new FileDataSource(
      new IntTupleDataInFormat(), this.ordersInputPath, "\"orders\" source");
    //ordersInput.setOutputContract(UniqueKey.class);
//    ordersInput.getCompilerHints().setAvgNumValuesPerKey(1);

    /* lineitem: (orderkey | partkey, suppkey, linenumber, quantity, extendedprice, discount, tax, ...) */
    FileDataSource lineItemInput = new FileDataSource(
      new IntTupleDataInFormat(), this.lineItemInputPath, "\"lineitem\" source");

    /* supplier: (suppkey | name, address, nationkey, phone, acctbal, comment) */
    FileDataSource supplierInput = new FileDataSource(
      new IntTupleDataInFormat(), this.supplierInputPath, "\"supplier\" source");
    //supplierInput.setOutputContract(UniqueKey.class);
//    supplierInput.getCompilerHints().setAvgNumValuesPerKey(1);

    /* nation: (nationkey | name, regionkey, comment) */
    FileDataSource nationInput = new FileDataSource(
      new IntTupleDataInFormat(), this.nationInputPath, "\"nation\" source");
    //nationInput.setOutputContract(UniqueKey.class);
//    nationInput.getCompilerHints().setAvgNumValuesPerKey(1);

    /* Filter on part's name, project values to NULL: */
 
View Full Code Here
252
253
254
255
256
257
258
259
260
261
262
    // parse job parameters
    int numSubTasks   = (args.length > 0 ? Integer.parseInt(args[0]) : 1);
    String edgeInput = (args.length > 1 ? args[1] : "");
    String output    = (args.length > 2 ? args[2] : "");

    FileDataSource edges = new FileDataSource(new EdgeInFormat(), edgeInput, "BTC Edges");
   
    ReduceOperator buildTriads = ReduceOperator.builder(new BuildTriads(), StringValue.class, 0)
      .name("Build Triads")
      .build();
View Full Code Here
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
    int numSubTasks   = (args.length > 0 ? Integer.parseInt(args[0]) : 1);
    String paths     = (args.length > 1 ? args[1] : "");
    String output    = (args.length > 2 ? args[2] : "");
    boolean rdfInput = (args.length > 3 ? Boolean.parseBoolean(args[3]) : false);

    FileDataSource pathsInput;
   
    if(rdfInput) {
      pathsInput = new FileDataSource(new RDFTripleInFormat(), paths, "RDF Triples");
    } else {
      pathsInput = new FileDataSource(new PathInFormat(), paths, "Paths");
    }
    pathsInput.setDegreeOfParallelism(numSubTasks);

    JoinOperator concatPaths =
        JoinOperator.builder(new ConcatPaths(), StringValue.class, 0, 1)
      .name("Concat Paths")
      .build();
View Full Code Here
106
107
108
109
110
111
112
113
114
115
116
  public Plan getPlan(int numSubTasks, String dataInput, String output) {


    // input is {word, count} pair
    FileDataSource source = new FileDataSource(new TextInputFormat(), dataInput, "Input Lines");

    //do a selection using cached file
    MapOperator mapper = MapOperator.builder(new TokenizeLine())
      .input(source)
      .name("Tokenize Lines")
View Full Code Here
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
    int numSubtasksInput2 = (args.length > 4 ? Integer.parseInt(args[4]) : 1);

    // create DataSourceContract for Orders input
    @SuppressWarnings("unchecked")
    CsvInputFormat format1 = new CsvInputFormat('|', IntValue.class, IntValue.class);
    FileDataSource input1 = new FileDataSource(format1, input1Path, "Input 1");
   
    ReduceOperator aggInput1 = ReduceOperator.builder(DummyReduce.class, IntValue.class, 0)
      .input(input1)
      .name("AggOrders")
      .build();

   
    // create DataSourceContract for Orders input
    @SuppressWarnings("unchecked")
    CsvInputFormat format2 = new CsvInputFormat('|', IntValue.class, IntValue.class);
    FileDataSource input2 = new FileDataSource(format2, input2Path, "Input 2");
    input2.setDegreeOfParallelism(numSubtasksInput2);

    ReduceOperator aggInput2 = ReduceOperator.builder(DummyReduce.class, IntValue.class, 0)
      .input(input2)
      .name("AggLines")
      .build();
View Full Code Here
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
    final String dependencySetInput = (args.length > 3 ? args[3] : "");
    final String output = (args.length > 4 ? args[4] : "");
    final int maxIterations = (args.length > 5 ? Integer.parseInt(args[5]) : 1);
   
    // create DataSourceContract for the initalSolutionSet
    FileDataSource initialSolutionSet = new FileDataSource(new CsvInputFormat(' ', LongValue.class, DoubleValue.class), solutionSetInput, "Initial Solution Set");

    // create DataSourceContract for the initalDeltaSet
    FileDataSource initialDeltaSet = new FileDataSource(new CsvInputFormat(' ', LongValue.class, DoubleValue.class), deltasInput, "Initial DeltaSet");
       
    // create DataSourceContract for the edges
    FileDataSource dependencySet = new FileDataSource(new CsvInputFormat(' ', LongValue.class, LongValue.class, LongValue.class), dependencySetInput, "Dependency Set");
   
    DeltaIteration iteration = new DeltaIteration(0, "Delta PageRank");
    iteration.setInitialSolutionSet(initialSolutionSet);
    iteration.setInitialWorkset(initialDeltaSet);
    iteration.setMaximumNumberOfIterations(maxIterations);
View Full Code Here
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
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
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
   
    String lineitemsPath = (args.length > 5 ? args[5] : "");
    String output        = (args.length > 6 ? args[6] : "");

    // create DataSourceContract for Orders input
    FileDataSource orders1 = new FileDataSource(new CsvInputFormat(), orders1Path, "Orders 1");
    CsvInputFormat.configureRecordFormat(orders1)
      .recordDelimiter('\n')
      .fieldDelimiter('|')
      .field(LongValue.class, 0)    // order id
      .field(IntValue.class, 7)     // ship prio
      .field(StringValue.class, 2, 2)  // order status
      .field(StringValue.class, 4, 10)  // order date
      .field(StringValue.class, 5, 8);  // order prio
   
    FileDataSource orders2 = new FileDataSource(new CsvInputFormat(), orders2Path, "Orders 2");
    CsvInputFormat.configureRecordFormat(orders2)
      .recordDelimiter('\n')
      .fieldDelimiter('|')
      .field(LongValue.class, 0)    // order id
      .field(IntValue.class, 7)     // ship prio
      .field(StringValue.class, 2, 2)  // order status
      .field(StringValue.class, 4, 10)  // order date
      .field(StringValue.class, 5, 8);  // order prio
   
    // create DataSourceContract for LineItems input
    FileDataSource lineitems = new FileDataSource(new CsvInputFormat(), lineitemsPath, "LineItems");
    CsvInputFormat.configureRecordFormat(lineitems)
      .recordDelimiter('\n')
      .fieldDelimiter('|')
      .field(LongValue.class, 0)
      .field(DoubleValue.class, 5);

    // create MapOperator for filtering Orders tuples
    MapOperator filterO1 = MapOperator.builder(new FilterO())
      .name("FilterO")
      .input(orders1)
      .build();
    // filter configuration
    filterO1.setParameter(TPCHQuery3.YEAR_FILTER, 1993);
    filterO1.setParameter(TPCHQuery3.PRIO_FILTER, "5");
    filterO1.getCompilerHints().setFilterFactor(0.05f);
   
    // create MapOperator for filtering Orders tuples
    MapOperator filterO2 = MapOperator.builder(new FilterO())
      .name("FilterO")
      .input(orders2)
      .build();
    // filter configuration
    filterO2.setParameter(TPCHQuery3.YEAR_FILTER, 1993);
    filterO2.setParameter(TPCHQuery3.PRIO_FILTER, "5");

    // create JoinOperator for joining Orders and LineItems
    @SuppressWarnings("unchecked")
    JoinOperator joinLiO = JoinOperator.builder(new JoinLiO(), LongValue.class, 0, 0)
      .input1(filterO2, filterO1)
      .input2(lineitems)
      .name("JoinLiO")
      .build();
   
    FileDataSource partJoin1 = new FileDataSource(new CsvInputFormat(), partJoin1Path, "Part Join 1");
    CsvInputFormat.configureRecordFormat(partJoin1)
      .recordDelimiter('\n')
      .fieldDelimiter('|')
      .field(LongValue.class, 0)
      .field(IntValue.class, 1)
      .field(DoubleValue.class, 2);
   
    FileDataSource partJoin2 = new FileDataSource(new CsvInputFormat(), partJoin2Path, "Part Join 2");
    CsvInputFormat.configureRecordFormat(partJoin2)
      .recordDelimiter('\n')
      .fieldDelimiter('|')
      .field(LongValue.class, 0)
      .field(IntValue.class, 1)
View Full Code Here
82
83
84
85
86
87
88
89
90
91
92
    String dataInput = (args.length > 1 ? args[1] : "");
    String output = (args.length > 2 ? args[2] : "");

    @SuppressWarnings("unchecked")
    CsvInputFormat format = new CsvInputFormat(' ', IntValue.class, IntValue.class);
    FileDataSource input = new FileDataSource(format, dataInput, "Input");
   
    // create the reduce contract and sets the key to the first field
    ReduceOperator sorter = ReduceOperator.builder(new IdentityReducer(), IntValue.class, 0)
      .input(input)
      .name("Reducer")
View Full Code Here
TOP

Related Classes of org.apache.flink.api.java.record.operators.FileDataSource

  • org.apache.flink.compiler.AdditionalOperatorsTest
  • org.apache.flink.compiler.BranchingPlansCompilerTest
  • org.apache.flink.compiler.DOPChangeTest
  • org.apache.flink.compiler.GroupOrderTest
  • org.apache.flink.compiler.HardPlansCompilationTest
  • org.apache.flink.compiler.ReduceAllTest
  • org.apache.flink.compiler.UnionPropertyPropagationTest
  • org.apache.flink.compiler.WorksetIterationsRecordApiCompilerTest
  • org.apache.flink.test.accumulators.AccumulatorIterativeITCase
  • org.apache.flink.test.broadcastvars.BroadcastBranchingITCase
    • 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.