Examples with CSVParseSpec io.druid.data.input.impl.CSVParseSpec used on opensource projects

Example 1 with CSVParseSpec

use of io.druid.data.input.impl.CSVParseSpec in project druid by druid-io.

the class IndexGeneratorJobTest method constructFeed.

@Parameterized.Parameters(name = "useCombiner={0}, partitionType={1}, interval={2}, shardInfoForEachSegment={3}, " + "data={4}, inputFormatName={5}, inputRowParser={6}, maxRowsInMemory={7}, " + "aggs={8}, datasourceName={9}, forceExtendableShardSpecs={10}, buildV9Directly={11}")
public static Collection<Object[]> constructFeed() {
    final List<Object[]> baseConstructors = Arrays.asList(new Object[][] { { false, "single", "2014-10-22T00:00:00Z/P2D", new String[][][] { { { null, "c.example.com" }, { "c.example.com", "e.example.com" }, { "e.example.com", "g.example.com" }, { "g.example.com", "i.example.com" }, { "i.example.com", null } }, { { null, "c.example.com" }, { "c.example.com", "e.example.com" }, { "e.example.com", "g.example.com" }, { "g.example.com", "i.example.com" }, { "i.example.com", null } } }, ImmutableList.of("2014102200,a.example.com,100", "2014102200,b.exmaple.com,50", "2014102200,c.example.com,200", "2014102200,d.example.com,250", "2014102200,e.example.com,123", "2014102200,f.example.com,567", "2014102200,g.example.com,11", "2014102200,h.example.com,251", "2014102200,i.example.com,963", "2014102200,j.example.com,333", "2014102300,a.example.com,100", "2014102300,b.exmaple.com,50", "2014102300,c.example.com,200", "2014102300,d.example.com,250", "2014102300,e.example.com,123", "2014102300,f.example.com,567", "2014102300,g.example.com,11", "2014102300,h.example.com,251", "2014102300,i.example.com,963", "2014102300,j.example.com,333"), null, new StringInputRowParser(new CSVParseSpec(new TimestampSpec("timestamp", "yyyyMMddHH", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(ImmutableList.of("host")), null, null), null, ImmutableList.of("timestamp", "host", "visited_num")), null), null, aggs1, "website" }, { false, "hashed", "2014-10-22T00:00:00Z/P1D", new Integer[][][] { { { 0, 4 }, { 1, 4 }, { 2, 4 }, { 3, 4 } } }, ImmutableList.of("2014102200,a.example.com,100", "2014102201,b.exmaple.com,50", "2014102202,c.example.com,200", "2014102203,d.example.com,250", "2014102204,e.example.com,123", "2014102205,f.example.com,567", "2014102206,g.example.com,11", "2014102207,h.example.com,251", "2014102208,i.example.com,963", "2014102209,j.example.com,333", "2014102210,k.example.com,253", "2014102211,l.example.com,321", "2014102212,m.example.com,3125", "2014102213,n.example.com,234", "2014102214,o.example.com,325", "2014102215,p.example.com,3533", "2014102216,q.example.com,500", "2014102216,q.example.com,87"), null, new HadoopyStringInputRowParser(new CSVParseSpec(new TimestampSpec("timestamp", "yyyyMMddHH", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(ImmutableList.of("host")), null, null), null, ImmutableList.of("timestamp", "host", "visited_num"))), null, aggs1, "website" }, { true, "hashed", "2014-10-22T00:00:00Z/P1D", new Integer[][][] { { { 0, 4 }, { 1, 4 }, { 2, 4 }, { 3, 4 } } }, ImmutableList.of("2014102200,a.example.com,100", "2014102201,b.exmaple.com,50", "2014102202,c.example.com,200", "2014102203,d.example.com,250", "2014102204,e.example.com,123", "2014102205,f.example.com,567", "2014102206,g.example.com,11", "2014102207,h.example.com,251", "2014102208,i.example.com,963", "2014102209,j.example.com,333", "2014102210,k.example.com,253", "2014102211,l.example.com,321", "2014102212,m.example.com,3125", "2014102213,n.example.com,234", "2014102214,o.example.com,325", "2014102215,p.example.com,3533", "2014102216,q.example.com,500", "2014102216,q.example.com,87"), null, new StringInputRowParser(new CSVParseSpec(new TimestampSpec("timestamp", "yyyyMMddHH", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(ImmutableList.of("host")), null, null), null, ImmutableList.of("timestamp", "host", "visited_num")), null), null, aggs1, "website" }, { false, "single", "2014-10-22T00:00:00Z/P2D", new String[][][] { { { null, "c.example.com" }, { "c.example.com", "e.example.com" }, { "e.example.com", "g.example.com" }, { "g.example.com", "i.example.com" }, { "i.example.com", null } }, { { null, "c.example.com" }, { "c.example.com", "e.example.com" }, { "e.example.com", "g.example.com" }, { "g.example.com", "i.example.com" }, { "i.example.com", null } } }, ImmutableList.of("2014102200,a.example.com,100", "2014102200,b.exmaple.com,50", "2014102200,c.example.com,200", "2014102200,d.example.com,250", "2014102200,e.example.com,123", "2014102200,f.example.com,567", "2014102200,g.example.com,11", "2014102200,h.example.com,251", "2014102200,i.example.com,963", "2014102200,j.example.com,333", "2014102300,a.example.com,100", "2014102300,b.exmaple.com,50", "2014102300,c.example.com,200", "2014102300,d.example.com,250", "2014102300,e.example.com,123", "2014102300,f.example.com,567", "2014102300,g.example.com,11", "2014102300,h.example.com,251", "2014102300,i.example.com,963", "2014102300,j.example.com,333"), SequenceFileInputFormat.class.getName(), new HadoopyStringInputRowParser(new CSVParseSpec(new TimestampSpec("timestamp", "yyyyMMddHH", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(ImmutableList.of("host")), null, null), null, ImmutableList.of("timestamp", "host", "visited_num"))), null, aggs1, "website" }, { // Tests that new indexes inherit the dimension order from previous index
    false, "hashed", "2014-10-22T00:00:00Z/P1D", new Integer[][][] { { // use a single partition, dimension order inheritance is not supported across partitions
    { 0, 1 } } }, ImmutableList.of("{\"ts\":\"2014102200\", \"X\":\"x.example.com\"}", "{\"ts\":\"2014102201\", \"Y\":\"y.example.com\"}", "{\"ts\":\"2014102202\", \"M\":\"m.example.com\"}", "{\"ts\":\"2014102203\", \"Q\":\"q.example.com\"}", "{\"ts\":\"2014102204\", \"B\":\"b.example.com\"}", "{\"ts\":\"2014102205\", \"F\":\"f.example.com\"}"), null, new StringInputRowParser(new JSONParseSpec(new TimestampSpec("ts", "yyyyMMddHH", null), new DimensionsSpec(null, null, null), null, null), null), // force 1 row max per index for easier testing
    1, aggs2, "inherit_dims" }, { // Tests that pre-specified dim order is maintained across indexes.
    false, "hashed", "2014-10-22T00:00:00Z/P1D", new Integer[][][] { { { 0, 1 } } }, ImmutableList.of("{\"ts\":\"2014102200\", \"X\":\"x.example.com\"}", "{\"ts\":\"2014102201\", \"Y\":\"y.example.com\"}", "{\"ts\":\"2014102202\", \"M\":\"m.example.com\"}", "{\"ts\":\"2014102203\", \"Q\":\"q.example.com\"}", "{\"ts\":\"2014102204\", \"B\":\"b.example.com\"}", "{\"ts\":\"2014102205\", \"F\":\"f.example.com\"}"), null, new StringInputRowParser(new JSONParseSpec(new TimestampSpec("ts", "yyyyMMddHH", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(ImmutableList.of("B", "F", "M", "Q", "X", "Y")), null, null), null, null), null), // force 1 row max per index for easier testing
    1, aggs2, "inherit_dims2" } });
    // Run each baseConstructor with/without buildV9Directly and forceExtendableShardSpecs.
    final List<Object[]> constructors = Lists.newArrayList();
    for (Object[] baseConstructor : baseConstructors) {
        for (int buildV9Directly = 0; buildV9Directly < 2; buildV9Directly++) {
            for (int forceExtendableShardSpecs = 0; forceExtendableShardSpecs < 2; forceExtendableShardSpecs++) {
                final Object[] fullConstructor = new Object[baseConstructor.length + 2];
                System.arraycopy(baseConstructor, 0, fullConstructor, 0, baseConstructor.length);
                fullConstructor[baseConstructor.length] = forceExtendableShardSpecs == 0;
                fullConstructor[baseConstructor.length + 1] = buildV9Directly == 0;
                constructors.add(fullConstructor);
            }
        }
    }
    return constructors;
}

Example 2 with CSVParseSpec

use of io.druid.data.input.impl.CSVParseSpec in project druid by druid-io.

the class BatchDeltaIngestionTest method makeHadoopDruidIndexerConfig.

private HadoopDruidIndexerConfig makeHadoopDruidIndexerConfig(Map<String, Object> inputSpec, File tmpDir) throws Exception {
    HadoopDruidIndexerConfig config = new HadoopDruidIndexerConfig(new HadoopIngestionSpec(new DataSchema("website", MAPPER.convertValue(new StringInputRowParser(new CSVParseSpec(new TimestampSpec("timestamp", "yyyyMMddHH", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(ImmutableList.of("host")), null, null), null, ImmutableList.of("timestamp", "host", "host2", "visited_num")), null), Map.class), new AggregatorFactory[] { new LongSumAggregatorFactory("visited_sum", "visited_num"), new HyperUniquesAggregatorFactory("unique_hosts", "host2") }, new UniformGranularitySpec(Granularities.DAY, Granularities.NONE, ImmutableList.of(INTERVAL_FULL)), MAPPER), new HadoopIOConfig(inputSpec, null, tmpDir.getCanonicalPath()), new HadoopTuningConfig(tmpDir.getCanonicalPath(), null, null, null, null, null, false, false, false, false, null, false, false, null, null, null, false, false)));
    config.setShardSpecs(ImmutableMap.<Long, List<HadoopyShardSpec>>of(INTERVAL_FULL.getStartMillis(), ImmutableList.of(new HadoopyShardSpec(new HashBasedNumberedShardSpec(0, 1, null, HadoopDruidIndexerConfig.JSON_MAPPER), 0))));
    config = HadoopDruidIndexerConfig.fromSpec(config.getSchema());
    return config;
}

Example 3 with CSVParseSpec

use of io.druid.data.input.impl.CSVParseSpec in project druid by druid-io.

the class JobHelperTest method setup.

@Before
public void setup() throws Exception {
    tmpDir = temporaryFolder.newFile();
    dataFile = temporaryFolder.newFile();
    config = new HadoopDruidIndexerConfig(new HadoopIngestionSpec(new DataSchema("website", HadoopDruidIndexerConfig.JSON_MAPPER.convertValue(new StringInputRowParser(new CSVParseSpec(new TimestampSpec("timestamp", "yyyyMMddHH", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(ImmutableList.of("host")), null, null), null, ImmutableList.of("timestamp", "host", "visited_num")), null), Map.class), new AggregatorFactory[] { new LongSumAggregatorFactory("visited_num", "visited_num") }, new UniformGranularitySpec(Granularities.DAY, Granularities.NONE, ImmutableList.of(this.interval)), HadoopDruidIndexerConfig.JSON_MAPPER), new HadoopIOConfig(ImmutableMap.<String, Object>of("paths", dataFile.getCanonicalPath(), "type", "static"), null, tmpDir.getCanonicalPath()), new HadoopTuningConfig(tmpDir.getCanonicalPath(), null, null, null, null, null, false, false, false, false, //Map of job properties
    ImmutableMap.of("fs.s3.impl", "org.apache.hadoop.fs.s3native.NativeS3FileSystem", "fs.s3.awsAccessKeyId", "THISISMYACCESSKEY"), false, false, null, null, null, false, false)));
}

Example 4 with CSVParseSpec

use of io.druid.data.input.impl.CSVParseSpec in project druid by druid-io.

the class GroupByQueryRunnerFactoryTest method createSegment.

private Segment createSegment() throws Exception {
    IncrementalIndex incrementalIndex = new OnheapIncrementalIndex(0, Granularities.NONE, new AggregatorFactory[] { new CountAggregatorFactory("count") }, true, true, true, 5000);
    StringInputRowParser parser = new StringInputRowParser(new CSVParseSpec(new TimestampSpec("timestamp", "iso", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(ImmutableList.of("product", "tags")), null, null), "\t", ImmutableList.of("timestamp", "product", "tags")), "UTF-8");
    String[] rows = new String[] { "2011-01-12T00:00:00.000Z,product_1,t1", "2011-01-13T00:00:00.000Z,product_2,t2", "2011-01-14T00:00:00.000Z,product_3,t2" };
    for (String row : rows) {
        incrementalIndex.add(parser.parse(row));
    }
    closerRule.closeLater(incrementalIndex);
    return new IncrementalIndexSegment(incrementalIndex, "test");
}

Example 5 with CSVParseSpec

use of io.druid.data.input.impl.CSVParseSpec in project druid by druid-io.

the class MultiValuedDimensionTest method setupClass.

@BeforeClass
public static void setupClass() throws Exception {
    incrementalIndex = new OnheapIncrementalIndex(0, Granularities.NONE, new AggregatorFactory[] { new CountAggregatorFactory("count") }, true, true, true, 5000);
    StringInputRowParser parser = new StringInputRowParser(new CSVParseSpec(new TimestampSpec("timestamp", "iso", null), new DimensionsSpec(DimensionsSpec.getDefaultSchemas(ImmutableList.of("product", "tags")), null, null), "\t", ImmutableList.of("timestamp", "product", "tags")), "UTF-8");
    String[] rows = new String[] { "2011-01-12T00:00:00.000Z,product_1,t1\tt2\tt3", "2011-01-13T00:00:00.000Z,product_2,t3\tt4\tt5", "2011-01-14T00:00:00.000Z,product_3,t5\tt6\tt7", "2011-01-14T00:00:00.000Z,product_4" };
    for (String row : rows) {
        incrementalIndex.add(parser.parse(row));
    }
    persistedSegmentDir = Files.createTempDir();
    TestHelper.getTestIndexMerger().persist(incrementalIndex, persistedSegmentDir, new IndexSpec());
    queryableIndex = TestHelper.getTestIndexIO().loadIndex(persistedSegmentDir);
}