Search in sources :

Example 1 with LocalDataSegmentPusher

use of io.druid.segment.loading.LocalDataSegmentPusher in project hive by apache.

the class TestDruidRecordWriter method testWrite.

// This test need this patch https://github.com/druid-io/druid/pull/3483
@Ignore
@Test
public void testWrite() throws IOException, SegmentLoadingException {
    final String dataSourceName = "testDataSource";
    final File segmentOutputDir = temporaryFolder.newFolder();
    final File workingDir = temporaryFolder.newFolder();
    Configuration config = new Configuration();
    final InputRowParser inputRowParser = new MapInputRowParser(new TimeAndDimsParseSpec(new TimestampSpec(DruidTable.DEFAULT_TIMESTAMP_COLUMN, "auto", null), new DimensionsSpec(ImmutableList.<DimensionSchema>of(new StringDimensionSchema("host")), null, null)));
    final Map<String, Object> parserMap = objectMapper.convertValue(inputRowParser, Map.class);
    DataSchema dataSchema = new DataSchema(dataSourceName, parserMap, new AggregatorFactory[] { new LongSumAggregatorFactory("visited_sum", "visited_sum"), new HyperUniquesAggregatorFactory("unique_hosts", "unique_hosts") }, new UniformGranularitySpec(Granularity.DAY, QueryGranularities.NONE, ImmutableList.of(INTERVAL_FULL)), objectMapper);
    RealtimeTuningConfig tuningConfig = RealtimeTuningConfig.makeDefaultTuningConfig(temporaryFolder.newFolder());
    LocalFileSystem localFileSystem = FileSystem.getLocal(config);
    DataSegmentPusher dataSegmentPusher = new LocalDataSegmentPusher(new LocalDataSegmentPusherConfig() {

        @Override
        public File getStorageDirectory() {
            return segmentOutputDir;
        }
    }, objectMapper);
    Path segmentDescriptroPath = new Path(workingDir.getAbsolutePath(), DruidStorageHandler.SEGMENTS_DESCRIPTOR_DIR_NAME);
    druidRecordWriter = new DruidRecordWriter(dataSchema, tuningConfig, dataSegmentPusher, 20, segmentDescriptroPath, localFileSystem);
    List<DruidWritable> druidWritables = Lists.transform(expectedRows, new Function<ImmutableMap<String, Object>, DruidWritable>() {

        @Nullable
        @Override
        public DruidWritable apply(@Nullable ImmutableMap<String, Object> input) {
            return new DruidWritable(ImmutableMap.<String, Object>builder().putAll(input).put(Constants.DRUID_TIMESTAMP_GRANULARITY_COL_NAME, Granularity.DAY.truncate(new DateTime((long) input.get(DruidTable.DEFAULT_TIMESTAMP_COLUMN))).getMillis()).build());
        }
    });
    for (DruidWritable druidWritable : druidWritables) {
        druidRecordWriter.write(druidWritable);
    }
    druidRecordWriter.close(false);
    List<DataSegment> dataSegmentList = DruidStorageHandlerUtils.getPublishedSegments(segmentDescriptroPath, config);
    Assert.assertEquals(1, dataSegmentList.size());
    File tmpUnzippedSegmentDir = temporaryFolder.newFolder();
    new LocalDataSegmentPuller().getSegmentFiles(dataSegmentList.get(0), tmpUnzippedSegmentDir);
    final QueryableIndex queryableIndex = DruidStorageHandlerUtils.INDEX_IO.loadIndex(tmpUnzippedSegmentDir);
    QueryableIndexStorageAdapter adapter = new QueryableIndexStorageAdapter(queryableIndex);
    Firehose firehose = new IngestSegmentFirehose(ImmutableList.of(new WindowedStorageAdapter(adapter, adapter.getInterval())), ImmutableList.of("host"), ImmutableList.of("visited_sum", "unique_hosts"), null, QueryGranularities.NONE);
    List<InputRow> rows = Lists.newArrayList();
    while (firehose.hasMore()) {
        rows.add(firehose.nextRow());
    }
    verifyRows(expectedRows, rows);
}
Also used : IngestSegmentFirehose(io.druid.segment.realtime.firehose.IngestSegmentFirehose) LocalDataSegmentPusher(io.druid.segment.loading.LocalDataSegmentPusher) DataSegmentPusher(io.druid.segment.loading.DataSegmentPusher) Configuration(org.apache.hadoop.conf.Configuration) MapInputRowParser(io.druid.data.input.impl.MapInputRowParser) LongSumAggregatorFactory(io.druid.query.aggregation.LongSumAggregatorFactory) DataSegment(io.druid.timeline.DataSegment) DateTime(org.joda.time.DateTime) TimeAndDimsParseSpec(io.druid.data.input.impl.TimeAndDimsParseSpec) UniformGranularitySpec(io.druid.segment.indexing.granularity.UniformGranularitySpec) LocalDataSegmentPuller(io.druid.segment.loading.LocalDataSegmentPuller) TimestampSpec(io.druid.data.input.impl.TimestampSpec) WindowedStorageAdapter(io.druid.segment.realtime.firehose.WindowedStorageAdapter) Path(org.apache.hadoop.fs.Path) Firehose(io.druid.data.input.Firehose) IngestSegmentFirehose(io.druid.segment.realtime.firehose.IngestSegmentFirehose) LocalDataSegmentPusherConfig(io.druid.segment.loading.LocalDataSegmentPusherConfig) QueryableIndexStorageAdapter(io.druid.segment.QueryableIndexStorageAdapter) RealtimeTuningConfig(io.druid.segment.indexing.RealtimeTuningConfig) LocalDataSegmentPusher(io.druid.segment.loading.LocalDataSegmentPusher) ImmutableMap(com.google.common.collect.ImmutableMap) StringDimensionSchema(io.druid.data.input.impl.StringDimensionSchema) DataSchema(io.druid.segment.indexing.DataSchema) DruidWritable(org.apache.hadoop.hive.druid.serde.DruidWritable) LocalFileSystem(org.apache.hadoop.fs.LocalFileSystem) QueryableIndex(io.druid.segment.QueryableIndex) HyperUniquesAggregatorFactory(io.druid.query.aggregation.hyperloglog.HyperUniquesAggregatorFactory) InputRow(io.druid.data.input.InputRow) DimensionsSpec(io.druid.data.input.impl.DimensionsSpec) MapInputRowParser(io.druid.data.input.impl.MapInputRowParser) InputRowParser(io.druid.data.input.impl.InputRowParser) File(java.io.File) DruidRecordWriter(org.apache.hadoop.hive.druid.io.DruidRecordWriter) Nullable(javax.annotation.Nullable) Ignore(org.junit.Ignore) Test(org.junit.Test)

Example 2 with LocalDataSegmentPusher

use of io.druid.segment.loading.LocalDataSegmentPusher in project druid by druid-io.

the class KafkaIndexTaskTest method makeToolboxFactory.

private void makeToolboxFactory() throws IOException {
    directory = tempFolder.newFolder();
    final TestUtils testUtils = new TestUtils();
    final ObjectMapper objectMapper = testUtils.getTestObjectMapper();
    for (Module module : new KafkaIndexTaskModule().getJacksonModules()) {
        objectMapper.registerModule(module);
    }
    final TaskConfig taskConfig = new TaskConfig(new File(directory, "taskBaseDir").getPath(), null, null, 50000, null, false, null, null);
    final TestDerbyConnector derbyConnector = derby.getConnector();
    derbyConnector.createDataSourceTable();
    derbyConnector.createPendingSegmentsTable();
    derbyConnector.createSegmentTable();
    derbyConnector.createRulesTable();
    derbyConnector.createConfigTable();
    derbyConnector.createTaskTables();
    derbyConnector.createAuditTable();
    taskStorage = new MetadataTaskStorage(derbyConnector, new TaskStorageConfig(null), new SQLMetadataStorageActionHandlerFactory(derbyConnector, derby.metadataTablesConfigSupplier().get(), objectMapper));
    metadataStorageCoordinator = new IndexerSQLMetadataStorageCoordinator(testUtils.getTestObjectMapper(), derby.metadataTablesConfigSupplier().get(), derbyConnector);
    taskLockbox = new TaskLockbox(taskStorage);
    final TaskActionToolbox taskActionToolbox = new TaskActionToolbox(taskLockbox, metadataStorageCoordinator, emitter, new SupervisorManager(null));
    final TaskActionClientFactory taskActionClientFactory = new LocalTaskActionClientFactory(taskStorage, taskActionToolbox);
    final SegmentHandoffNotifierFactory handoffNotifierFactory = new SegmentHandoffNotifierFactory() {

        @Override
        public SegmentHandoffNotifier createSegmentHandoffNotifier(String dataSource) {
            return new SegmentHandoffNotifier() {

                @Override
                public boolean registerSegmentHandoffCallback(SegmentDescriptor descriptor, Executor exec, Runnable handOffRunnable) {
                    if (doHandoff) {
                        // Simulate immediate handoff
                        exec.execute(handOffRunnable);
                    }
                    return true;
                }

                @Override
                public void start() {
                //Noop
                }

                @Override
                public void close() {
                //Noop
                }
            };
        }
    };
    final LocalDataSegmentPusherConfig dataSegmentPusherConfig = new LocalDataSegmentPusherConfig();
    dataSegmentPusherConfig.storageDirectory = getSegmentDirectory();
    final DataSegmentPusher dataSegmentPusher = new LocalDataSegmentPusher(dataSegmentPusherConfig, objectMapper);
    toolboxFactory = new TaskToolboxFactory(taskConfig, taskActionClientFactory, emitter, dataSegmentPusher, new TestDataSegmentKiller(), // DataSegmentMover
    null, // DataSegmentArchiver
    null, new TestDataSegmentAnnouncer(), handoffNotifierFactory, makeTimeseriesOnlyConglomerate(), // queryExecutorService
    MoreExecutors.sameThreadExecutor(), EasyMock.createMock(MonitorScheduler.class), new SegmentLoaderFactory(new SegmentLoaderLocalCacheManager(null, new SegmentLoaderConfig() {

        @Override
        public List<StorageLocationConfig> getLocations() {
            return Lists.newArrayList();
        }
    }, testUtils.getTestObjectMapper())), testUtils.getTestObjectMapper(), testUtils.getTestIndexMerger(), testUtils.getTestIndexIO(), MapCache.create(1024), new CacheConfig(), testUtils.getTestIndexMergerV9());
}
Also used : DataSegmentPusher(io.druid.segment.loading.DataSegmentPusher) LocalDataSegmentPusher(io.druid.segment.loading.LocalDataSegmentPusher) LocalTaskActionClientFactory(io.druid.indexing.common.actions.LocalTaskActionClientFactory) TaskActionClientFactory(io.druid.indexing.common.actions.TaskActionClientFactory) TestDataSegmentAnnouncer(io.druid.indexing.test.TestDataSegmentAnnouncer) TaskConfig(io.druid.indexing.common.config.TaskConfig) TestUtils(io.druid.indexing.common.TestUtils) Executor(java.util.concurrent.Executor) TaskToolboxFactory(io.druid.indexing.common.TaskToolboxFactory) SegmentDescriptor(io.druid.query.SegmentDescriptor) TaskActionToolbox(io.druid.indexing.common.actions.TaskActionToolbox) LocalTaskActionClientFactory(io.druid.indexing.common.actions.LocalTaskActionClientFactory) SegmentLoaderConfig(io.druid.segment.loading.SegmentLoaderConfig) SegmentLoaderFactory(io.druid.indexing.common.SegmentLoaderFactory) SegmentLoaderLocalCacheManager(io.druid.segment.loading.SegmentLoaderLocalCacheManager) CacheConfig(io.druid.client.cache.CacheConfig) DefaultObjectMapper(io.druid.jackson.DefaultObjectMapper) ObjectMapper(com.fasterxml.jackson.databind.ObjectMapper) IndexerSQLMetadataStorageCoordinator(io.druid.metadata.IndexerSQLMetadataStorageCoordinator) StorageLocationConfig(io.druid.segment.loading.StorageLocationConfig) TaskStorageConfig(io.druid.indexing.common.config.TaskStorageConfig) LocalDataSegmentPusherConfig(io.druid.segment.loading.LocalDataSegmentPusherConfig) SegmentHandoffNotifier(io.druid.segment.realtime.plumber.SegmentHandoffNotifier) TestDerbyConnector(io.druid.metadata.TestDerbyConnector) LocalDataSegmentPusher(io.druid.segment.loading.LocalDataSegmentPusher) TestDataSegmentKiller(io.druid.indexing.test.TestDataSegmentKiller) SegmentHandoffNotifierFactory(io.druid.segment.realtime.plumber.SegmentHandoffNotifierFactory) SupervisorManager(io.druid.indexing.overlord.supervisor.SupervisorManager) SQLMetadataStorageActionHandlerFactory(io.druid.metadata.SQLMetadataStorageActionHandlerFactory) TaskLockbox(io.druid.indexing.overlord.TaskLockbox) Module(com.fasterxml.jackson.databind.Module) MetadataTaskStorage(io.druid.indexing.overlord.MetadataTaskStorage) File(java.io.File)

Aggregations

DataSegmentPusher (io.druid.segment.loading.DataSegmentPusher)2 LocalDataSegmentPusher (io.druid.segment.loading.LocalDataSegmentPusher)2 LocalDataSegmentPusherConfig (io.druid.segment.loading.LocalDataSegmentPusherConfig)2 File (java.io.File)2 Module (com.fasterxml.jackson.databind.Module)1 ObjectMapper (com.fasterxml.jackson.databind.ObjectMapper)1 ImmutableMap (com.google.common.collect.ImmutableMap)1 CacheConfig (io.druid.client.cache.CacheConfig)1 Firehose (io.druid.data.input.Firehose)1 InputRow (io.druid.data.input.InputRow)1 DimensionsSpec (io.druid.data.input.impl.DimensionsSpec)1 InputRowParser (io.druid.data.input.impl.InputRowParser)1 MapInputRowParser (io.druid.data.input.impl.MapInputRowParser)1 StringDimensionSchema (io.druid.data.input.impl.StringDimensionSchema)1 TimeAndDimsParseSpec (io.druid.data.input.impl.TimeAndDimsParseSpec)1 TimestampSpec (io.druid.data.input.impl.TimestampSpec)1 SegmentLoaderFactory (io.druid.indexing.common.SegmentLoaderFactory)1 TaskToolboxFactory (io.druid.indexing.common.TaskToolboxFactory)1 TestUtils (io.druid.indexing.common.TestUtils)1 LocalTaskActionClientFactory (io.druid.indexing.common.actions.LocalTaskActionClientFactory)1