Package org.apache.crunch.impl.mr

Examples of org.apache.crunch.impl.mr.MRPipeline

260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
  public void testHFileUsesFamilyConfig() throws IOException {
    DataBlockEncoding newBlockEncoding = DataBlockEncoding.PREFIX;
    assertNotSame(newBlockEncoding, DataBlockEncoding.valueOf(HColumnDescriptor.DEFAULT_DATA_BLOCK_ENCODING));

    Configuration conf = HBASE_TEST_UTILITY.getConfiguration();
    Pipeline pipeline = new MRPipeline(HFileTargetIT.class, conf);
    Path inputPath = copyResourceFileToHDFS("shakes.txt");
    Path outputPath = getTempPathOnHDFS("out");
    HColumnDescriptor hcol = new HColumnDescriptor(TEST_FAMILY);
    hcol.setDataBlockEncoding(newBlockEncoding);
    HTable testTable = createTable(10, hcol);

    PCollection<String> shakespeare = pipeline.read(At.textFile(inputPath, Writables.strings()));
    PCollection<String> words = split(shakespeare, "\\s+");
    PTable<String,Long> wordCounts = words.count();
    PCollection<Put> wordCountPuts = convertToPuts(wordCounts);
    HFileUtils.writePutsToHFilesForIncrementalLoad(
        wordCountPuts,
        testTable,
        outputPath);

    PipelineResult result = pipeline.run();
    assertTrue(result.succeeded());

    int hfilesCount = 0;
    FileSystem fs = outputPath.getFileSystem(conf);
    for (FileStatus e : fs.listStatus(new Path(outputPath, Bytes.toString(TEST_FAMILY)))) {
View Full Code Here
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
   * files in that path. The method return back a {@link PObject} containing a
   * {@link Map} having file names as keys and filters as values
   */
  public static PObject<Map<String, BloomFilter>> createFilter(Path inputPath, BloomFilterFn<String> filterFn)
      throws IOException {
    MRPipeline pipeline = new MRPipeline(BloomFilterFactory.class);
    FileStatus[] listStatus = FileSystem.get(pipeline.getConfiguration()).listStatus(inputPath);
    PTable<String, BloomFilter> filterTable = null;
    for (FileStatus fileStatus : listStatus) {
      Path path = fileStatus.getPath();
      PCollection<String> readTextFile = pipeline.readTextFile(path.toString());
      pipeline.getConfiguration().set(BloomFilterFn.CRUNCH_FILTER_NAME, path.getName());
      PTable<String, BloomFilter> currentTable = createFilterTable(readTextFile, filterFn);
      if (filterTable != null) {
        filterTable = filterTable.union(currentTable);
      } else {
        filterTable = currentTable;
View Full Code Here
54
55
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
    server.stop();
  }

  @Test
  public void testReadFromSource() throws Exception {
    Pipeline pipeline = new MRPipeline(DataBaseSourceIT.class);
    DataBaseSource<IdentifiableName> dbsrc = new DataBaseSource.Builder<IdentifiableName>(IdentifiableName.class)
        .setDriverClass(org.h2.Driver.class)
        .setUrl("jdbc:h2:tcp://localhost/~/test").setUsername("sa").setPassword("")
        .selectSQLQuery("SELECT ID, NAME FROM TEST").countSQLQuery("select count(*) from Test").build();

    PCollection<IdentifiableName> cdidata = pipeline.read(dbsrc);
    PCollection<String> names = cdidata.parallelDo(new DoFn<IdentifiableName, String>() {

      @Override
      public void process(IdentifiableName input, Emitter<String> emitter) {
        emitter.emit(input.name.toString());
      }

    }, Writables.strings());

    List<String> nameList = Lists.newArrayList(names.materialize());
    pipeline.done();

    assertEquals(2, nameList.size());
    assertEquals(Sets.newHashSet("Hello", "World"), Sets.newHashSet(nameList));

  }
View Full Code Here
55
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
      System.err.println();
      GenericOptionsParser.printGenericCommandUsage(System.err);
      return 1;
    }
    // Create an object to coordinate pipeline creation and execution.
    Pipeline pipeline = new MRPipeline(AverageBytesByIP.class, getConf());
    // Reference a given text file as a collection of Strings.
    PCollection<String> lines = pipeline.readTextFile(args[0]);

    // Combiner used for summing up response size and count
    CombineFn<String, Pair<Long, Long>> stringPairOfLongsSumCombiner = CombineFn.pairAggregator(CombineFn.SUM_LONGS,
        CombineFn.SUM_LONGS);

    // Table of (ip, sum(response size), count)
    PTable<String, Pair<Long, Long>> remoteAddrResponseSize = lines
        .parallelDo(extractResponseSize,
            Writables.tableOf(Writables.strings(), Writables.pairs(Writables.longs(), Writables.longs()))).groupByKey()
        .combineValues(stringPairOfLongsSumCombiner);

    // Calculate average response size by ip address
    PTable<String, Double> avgs = remoteAddrResponseSize.parallelDo(calulateAverage,
        Writables.tableOf(Writables.strings(), Writables.doubles()));

    // write the result to a text file
    pipeline.writeTextFile(avgs, args[1]);
    // Execute the pipeline as a MapReduce.
    PipelineResult result = pipeline.done();

    return result.succeeded() ? 0 : 1;
  }
View Full Code Here
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
      System.err.println();
      GenericOptionsParser.printGenericCommandUsage(System.err);
      return 1;
    }
    // Create an object to coordinate pipeline creation and execution.
    Pipeline pipeline = new MRPipeline(TotalBytesByIP.class, getConf());
    // Reference a given text file as a collection of Strings.
    PCollection<String> lines = pipeline.readTextFile(args[0]);

    // Combiner used for summing up response size
    CombineFn<String, Long> longSumCombiner = CombineFn.SUM_LONGS();

    // Table of (ip, sum(response size))
    PTable<String, Long> ipAddrResponseSize = lines
        .parallelDo(extractIPResponseSize, Writables.tableOf(Writables.strings(), Writables.longs())).groupByKey()
        .combineValues(longSumCombiner);

    pipeline.writeTextFile(ipAddrResponseSize, args[1]);
    // Execute the pipeline as a MapReduce.
    PipelineResult result = pipeline.done();

    return result.succeeded() ? 0 : 1;
  }
View Full Code Here
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
    createTable(configuration, TABLE_TARGET, Bytes.toString(COLUMN_FAMILY_TARGET));

    putInHbase(putList, configuration);

    // We create the pipeline which will handle most of the job.
    Pipeline pipeline = new MRPipeline(WordAggregationHBase.class, HBaseConfiguration.create());

    // The scan which will retrieve the data from the source in hbase.
    Scan scan = new Scan();
    scan.addColumn(COLUMN_FAMILY_SOURCE, COLUMN_QUALIFIER_SOURCE_PLAY);
    scan.addColumn(COLUMN_FAMILY_SOURCE, COLUMN_QUALIFIER_SOURCE_QUOTE);

    // Our hbase source
    HBaseSourceTarget source = new HBaseSourceTarget(TABLE_SOURCE, scan);

    // Our source, in a format which can be use by crunch
    PTable<ImmutableBytesWritable, Result> rawText = pipeline.read(source);

    // We process the data from the source HTable then concatenate all data
    // with the same rowkey
    PTable<String, String> textExtracted = extractText(rawText);
    CombineFn<String, String> stringConcatCombine = CombineFn.STRING_CONCAT(" ", true);
    PTable<String, String> result = textExtracted.groupByKey().combineValues(stringConcatCombine);

    // We create the collection of puts from the concatenated datas
    PCollection<Put> resultPut = createPut(result);

    // We write the puts in hbase, in the target table
    pipeline.write(resultPut, new HBaseTarget(TABLE_TARGET));

    pipeline.done();
    return 0;
  }
View Full Code Here
58
59
60
61
62
63
64
65
  @Rule
  public transient TemporaryPath tmpDir = TemporaryPaths.create();

  @Test
  public void testWritableSortAsc() throws Exception {
    runSingle(new MRPipeline(SortIT.class, tmpDir.getDefaultConfiguration()), WritableTypeFamily.getInstance(), Order.ASCENDING,
        "A\tand this text as well");
  }
View Full Code Here
64
65
66
67
68
69
70
71
        "A\tand this text as well");
  }

  @Test
  public void testWritableSortDesc() throws Exception {
    runSingle(new MRPipeline(SortIT.class, tmpDir.getDefaultConfiguration()), WritableTypeFamily.getInstance(), Order.DESCENDING,
        "B\tthis doc has some text");
  }
View Full Code Here
70
71
72
73
74
75
76
77
        "B\tthis doc has some text");
  }

  @Test
  public void testWritableSortAscDesc() throws Exception {
    runPair(new MRPipeline(SortIT.class, tmpDir.getDefaultConfiguration()), WritableTypeFamily.getInstance(), by(1, ASCENDING), by(2, DESCENDING), "A",
        "this doc has this text");
  }
View Full Code Here
76
77
78
79
80
81
82
83
        "this doc has this text");
  }

  @Test
  public void testWritableSortSecondDescFirstAsc() throws Exception {
    runPair(new MRPipeline(SortIT.class, tmpDir.getDefaultConfiguration()), WritableTypeFamily.getInstance(), by(2, DESCENDING), by(1, ASCENDING), "A",
        "this doc has this text");
  }
View Full Code Here
TOP

Related Classes of org.apache.crunch.impl.mr.MRPipeline

  • com.cloudera.cdk.data.crunch.TestCrunchDatasets
  • org.apache.bigtop.bigpetstore.etl.CrunchETL
  • org.apache.crunch.Breakpoint2IT
  • org.apache.crunch.BreakpointIT
  • org.apache.crunch.CancelJobsIT
  • org.apache.crunch.CheckpointIT
  • org.apache.crunch.CleanTextIT
  • org.apache.crunch.CollectionPObjectIT
  • org.apache.crunch.CollectionsIT
  • org.apache.crunch.CollectionsLengthIT
    • 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.