Package com.cloudera.cdk.data

Examples of com.cloudera.cdk.data.DatasetRepository

27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
public class CreateStagedDataset extends Configured implements Tool {

  @Override
  public int run(String[] args) throws Exception {
    DatasetRepository repo = DatasetRepositories.open("repo:file:/tmp/data");

    // where the schema is stored
    URI schemaURI = URI.create("resource:simple-log.avsc");

    // create a Parquet dataset for long-term storage
    repo.create("logs", new DatasetDescriptor.Builder()
        .format(Formats.PARQUET)
        .schemaUri(schemaURI)
        .partitionStrategy(new PartitionStrategy.Builder()
            .year("timestamp", "year")
            .month("timestamp", "month")
            .day("timestamp", "day")
            .build())
        .build());

    // create an Avro dataset to temporarily hold data
    repo.create("logs-staging", new DatasetDescriptor.Builder()
        .format(Formats.AVRO)
        .schemaUri(schemaURI)
        .partitionStrategy(new PartitionStrategy.Builder()
            .day("timestamp", "day")
            .build())
View Full Code Here
50
51
52
53
54
55
56
57
58
59
60
61
62
63
  public int run(String[] args) throws Exception {
    // going to generate a lot of random log messages
    final Random rand = new Random();

    // open the repository
    final DatasetRepository repo = DatasetRepositories.open("repo:file:/tmp/data");

    // data is written to the staging dataset
    final Dataset<GenericRecord> staging = repo.load("logs-staging");
    final DatasetWriter<GenericRecord> writer = staging.newWriter();

    // this is going to build our simple log records
    final GenericRecordBuilder builder = new GenericRecordBuilder(
        staging.getDescriptor().getSchema());
View Full Code Here
31
32
33
34
35
36
37
38
39
40
41
42
43
44
  @Override
  public int run(String[] args) throws Exception {

    // Construct a local filesystem dataset repository rooted at /tmp/data
    DatasetRepository repo = DatasetRepositories.open("repo:hdfs:/tmp/data");

    // Load the events dataset
    Dataset<GenericRecord> events = repo.load("events");

    // Get a reader for the dataset and read all the events
    DatasetReader<GenericRecord> reader = events.newReader();
    try {
      reader.open();
View Full Code Here
35
36
37
38
39
40
41
42
43
44
45
46
    // Get a log4j logger
    Logger logger = Logger.getLogger(App.class);

    // Find the schema from the repository
    DatasetRepository repo = DatasetRepositories.open("repo:hdfs:/tmp/data");
    Schema schema = repo.load("events").getDescriptor().getSchema();

    // Build some events using the generic Avro API and log them using log4j
    GenericRecordBuilder builder = new GenericRecordBuilder(schema);
    for (long i = 0; i < 10; i++) {
      GenericRecord event = builder.set("id", i)
View Full Code Here
28
29
30
31
32
33
34
35
36
37
38
39
  @Override
  public int run(String[] args) throws Exception {

    // Construct a local filesystem dataset repository rooted at /tmp/data
    DatasetRepository repo = DatasetRepositories.open("repo:hdfs:/tmp/data");

    // Drop the events dataset
    boolean success = repo.delete("events");

    return success ? 0 : 1;
  }
View Full Code Here
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
  @Override
  public int run(String[] args) throws Exception {

    // Construct a local filesystem dataset repository rooted at /tmp/data
    DatasetRepository repo = DatasetRepositories.open("repo:hdfs:/tmp/data");

    // Create a dataset of events with the Avro schema in the repository
    DatasetDescriptor descriptor = new DatasetDescriptor.Builder()
        .schemaUri("resource:event.avsc")
        .build();
    repo.create("events", descriptor);

    return 0;
  }
View Full Code Here
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
public class DescribeDatasets extends Configured implements Tool {

  @Override
  public int run(String[] args) throws Exception {
    DatasetRepository repo = DatasetRepositories.open("repo:hdfs://localhost:8020/user/cloudera");

    Schema ratingSchema = SchemaBuilder.record("Rating")
        .fields()
        .name("userId").type().intType().noDefault()
        .name("movieId").type().intType().noDefault()
        .name("rating").type().intType().noDefault()
        .name("timeInSeconds").type().intType().noDefault()
        .endRecord();

    // create
    repo.create("ratings", new DatasetDescriptor.Builder()
//        .location("hdfs:u.data")
        .format(Formats.CSV)
        .property("cdk.csv.delimiter", "\t")
        .schema(ratingSchema)
        .build());

//    movie id | movie title | release date | video release date |
//    IMDb URL | unknown | Action | Adventure | Animation |
//        Children's | Comedy | Crime | Documentary | Drama | Fantasy |
//    Film-Noir | Horror | Musical | Mystery | Romance | Sci-Fi |
//        Thriller | War | Western |
    Schema movieSchema = SchemaBuilder.record("Movie")
        .fields()
        .name("movieId").type().intType().noDefault()
        .name("title").type().stringType().noDefault()
        .name("releaseDate").type().stringType().noDefault()
        .name("videoReleaseDate").type().stringType().noDefault()
        .name("imdbURL").type().stringType().noDefault()
        // ignore genre fields for now
        .endRecord();

    repo.create("movies", new DatasetDescriptor.Builder()
//        .location("hdfs:u.item")
        .format(Formats.CSV)
        .property("cdk.csv.delimiter", "|")
        .schema(movieSchema)
        .build());
View Full Code Here
15
16
17
18
19
20
21
22
23
24
25
26
27
public class ReadMovies extends Configured implements Tool {

  @Override
  public int run(String[] args) throws Exception {
    DatasetRepository repo = DatasetRepositories.open("repo:hdfs://localhost:8020/user/cloudera");

    Dataset movies = repo.load("movies");
    DatasetReader reader = movies.newReader();
    try {
      reader.open();
      for (Object rec : reader) {
        System.err.println("Movie: " + rec);
View Full Code Here
41
42
43
44
45
46
47
48
49
50
51
52
53
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
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
  @Override
  public int run(String[] args) throws Exception {

    // Construct a local filesystem dataset repository rooted at /tmp/data
    DatasetRepository fsRepo = DatasetRepositories.open("repo:hdfs:/tmp/data");


    // Construct an HCatalog dataset repository using external Hive tables
    DatasetRepository hcatRepo = DatasetRepositories.open("repo:hive:/tmp/data");

    // Turn debug on while in development.
    getPipeline().enableDebug();
    getPipeline().getConfiguration().set("crunch.log.job.progress", "true");

    // Load the events dataset and get the correct partition to sessionize
    Dataset<StandardEvent> eventsDataset = fsRepo.load("events");
    Dataset<StandardEvent> partition;
    if (args.length == 0 || (args.length == 1 && args[0].equals("LATEST"))) {
      partition = getLatestPartition(eventsDataset);
    } else {
      partition = getPartitionForURI(eventsDataset, args[0]);
    }

    // Create a parallel collection from the working partition
    PCollection<StandardEvent> events = read(
        CrunchDatasets.asSource(partition, StandardEvent.class));

    // Process the events into sessions, using a combiner
    PCollection<Session> sessions = events
      .parallelDo(new DoFn<StandardEvent, Session>() {
        @Override
        public void process(StandardEvent event, Emitter<Session> emitter) {
          emitter.emit(Session.newBuilder()
              .setUserId(event.getUserId())
              .setSessionId(event.getSessionId())
              .setIp(event.getIp())
              .setStartTimestamp(event.getTimestamp())
              .setDuration(0)
              .setSessionEventCount(1)
              .build());
        }
      }, Avros.specifics(Session.class))
      .by(new MapFn<Session, Pair<Long, String>>() {
        @Override
        public Pair<Long, String> map(Session session) {
          return Pair.of(session.getUserId(), session.getSessionId());
        }
      }, Avros.pairs(Avros.longs(), Avros.strings()))
      .groupByKey()
      .combineValues(new CombineFn<Pair<Long, String>, Session>() {
        @Override
        public void process(Pair<Pair<Long, String>, Iterable<Session>> pairIterable,
            Emitter<Pair<Pair<Long, String>, Session>> emitter) {
          String ip = null;
          long startTimestamp = Long.MAX_VALUE;
          long endTimestamp = Long.MIN_VALUE;
          int sessionEventCount = 0;
          for (Session s : pairIterable.second()) {
            ip = s.getIp();
            startTimestamp = Math.min(startTimestamp, s.getStartTimestamp());
            endTimestamp = Math.max(endTimestamp, s.getStartTimestamp() + s.getDuration());
            sessionEventCount += s.getSessionEventCount();
          }
          emitter.emit(Pair.of(pairIterable.first(), Session.newBuilder()
              .setUserId(pairIterable.first().first())
              .setSessionId(pairIterable.first().second())
              .setIp(ip)
              .setStartTimestamp(startTimestamp)
              .setDuration(endTimestamp - startTimestamp)
              .setSessionEventCount(sessionEventCount)
              .build()));
        }
      })
      .parallelDo(new DoFn<Pair<Pair<Long, String>, Session>, Session>() {
        @Override
        public void process(Pair<Pair<Long, String>, Session> pairSession,
            Emitter<Session> emitter) {
          emitter.emit(pairSession.second());
        }
      }, Avros.specifics(Session.class));

    // Write the sessions to the "sessions" Dataset
    getPipeline().write(sessions, CrunchDatasets.asTarget(hcatRepo.load("sessions")),
        Target.WriteMode.APPEND);

    return run().succeeded() ? 0 : 1;
  }
View Full Code Here
TOP

Related Classes of com.cloudera.cdk.data.DatasetRepository

  • com.cloudera.cdk.data.filesystem.TestFileSystemURIs
  • com.cloudera.cdk.data.flume.Log4jAppender
  • com.cloudera.cdk.data.hcatalog.TestHiveURIs
  • com.cloudera.cdk.examples.data.CreateHCatalogUserDatasetGeneric
  • com.cloudera.cdk.examples.data.CreateProductDatasetPojo
  • com.cloudera.cdk.examples.data.CreateUserDatasetGeneric
  • com.cloudera.cdk.examples.data.CreateUserDatasetGenericParquet
  • com.cloudera.cdk.examples.data.CreateUserDatasetGenericPartitioned
  • com.cloudera.cdk.examples.data.DeleteHCatalogUserDataset
  • com.cloudera.cdk.examples.data.DeleteProductDataset
    • 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.