Examples with HadoopDruidIndexerJob io.druid.indexer.HadoopDruidIndexerJob used on opensource projects

Example 1 with HadoopDruidIndexerJob

use of io.druid.indexer.HadoopDruidIndexerJob in project druid by druid-io.

the class HadoopConverterJobTest method setUp.

@Before
public void setUp() throws Exception {
    final MetadataStorageUpdaterJobSpec metadataStorageUpdaterJobSpec = new MetadataStorageUpdaterJobSpec() {

        @Override
        public String getSegmentTable() {
            return derbyConnectorRule.metadataTablesConfigSupplier().get().getSegmentsTable();
        }

        @Override
        public MetadataStorageConnectorConfig get() {
            return derbyConnectorRule.getMetadataConnectorConfig();
        }
    };
    final File scratchFileDir = temporaryFolder.newFolder();
    storageLocProperty = System.getProperty(STORAGE_PROPERTY_KEY);
    tmpSegmentDir = temporaryFolder.newFolder();
    System.setProperty(STORAGE_PROPERTY_KEY, tmpSegmentDir.getAbsolutePath());
    final URL url = Preconditions.checkNotNull(Query.class.getClassLoader().getResource("druid.sample.tsv"));
    final File tmpInputFile = temporaryFolder.newFile();
    FileUtils.retryCopy(new ByteSource() {

        @Override
        public InputStream openStream() throws IOException {
            return url.openStream();
        }
    }, tmpInputFile, FileUtils.IS_EXCEPTION, 3);
    final HadoopDruidIndexerConfig hadoopDruidIndexerConfig = new HadoopDruidIndexerConfig(new HadoopIngestionSpec(new DataSchema(DATASOURCE, HadoopDruidIndexerConfig.JSON_MAPPER.convertValue(new StringInputRowParser(new DelimitedParseSpec(new TimestampSpec("ts", "iso", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(Arrays.asList(TestIndex.DIMENSIONS)), null, null), "\t", "
", Arrays.asList(TestIndex.COLUMNS)), null), Map.class), new AggregatorFactory[] { new DoubleSumAggregatorFactory(TestIndex.METRICS[0], TestIndex.METRICS[0]), new HyperUniquesAggregatorFactory("quality_uniques", "quality") }, new UniformGranularitySpec(Granularities.MONTH, Granularities.DAY, ImmutableList.<Interval>of(interval)), HadoopDruidIndexerConfig.JSON_MAPPER), new HadoopIOConfig(ImmutableMap.<String, Object>of("type", "static", "paths", tmpInputFile.getAbsolutePath()), metadataStorageUpdaterJobSpec, tmpSegmentDir.getAbsolutePath()), new HadoopTuningConfig(scratchFileDir.getAbsolutePath(), null, null, null, null, null, false, false, false, false, null, false, false, null, null, null, false, false)));
    metadataStorageTablesConfigSupplier = derbyConnectorRule.metadataTablesConfigSupplier();
    connector = derbyConnectorRule.getConnector();
    try {
        connector.getDBI().withHandle(new HandleCallback<Void>() {

            @Override
            public Void withHandle(Handle handle) throws Exception {
                handle.execute("DROP TABLE druid_segments");
                return null;
            }
        });
    } catch (CallbackFailedException e) {
    // Who cares
    }
    List<Jobby> jobs = ImmutableList.of(new Jobby() {

        @Override
        public boolean run() {
            connector.createSegmentTable(metadataStorageUpdaterJobSpec.getSegmentTable());
            return true;
        }
    }, new HadoopDruidDetermineConfigurationJob(hadoopDruidIndexerConfig), new HadoopDruidIndexerJob(hadoopDruidIndexerConfig, new SQLMetadataStorageUpdaterJobHandler(connector)));
    JobHelper.runJobs(jobs, hadoopDruidIndexerConfig);
}

Example 2 with HadoopDruidIndexerJob

use of io.druid.indexer.HadoopDruidIndexerJob in project druid by druid-io.

the class CliInternalHadoopIndexer method run.

@Override
public void run() {
    try {
        Injector injector = makeInjector();
        config = getHadoopDruidIndexerConfig();
        MetadataStorageUpdaterJobSpec metadataSpec = config.getSchema().getIOConfig().getMetadataUpdateSpec();
        // override metadata storage type based on HadoopIOConfig
        Preconditions.checkNotNull(metadataSpec.getType(), "type in metadataUpdateSpec must not be null");
        injector.getInstance(Properties.class).setProperty("druid.metadata.storage.type", metadataSpec.getType());
        config = HadoopDruidIndexerConfig.fromSpec(HadoopIngestionSpec.updateSegmentListIfDatasourcePathSpecIsUsed(config.getSchema(), HadoopDruidIndexerConfig.JSON_MAPPER, new MetadataStoreBasedUsedSegmentLister(injector.getInstance(IndexerMetadataStorageCoordinator.class))));
        List<Jobby> jobs = Lists.newArrayList();
        jobs.add(new HadoopDruidDetermineConfigurationJob(config));
        jobs.add(new HadoopDruidIndexerJob(config, injector.getInstance(MetadataStorageUpdaterJobHandler.class)));
        JobHelper.runJobs(jobs, config);
    } catch (Exception e) {
        throw Throwables.propagate(e);
    }
}