Examples with InDimFilter io.druid.query.filter.InDimFilter used on opensource projects

Example 11 with InDimFilter

use of io.druid.query.filter.InDimFilter in project druid by druid-io.

the class FilteredAggregatorBenchmark method setup.

@Setup
public void setup() throws IOException {
    log.info("SETUP CALLED AT " + System.currentTimeMillis());
    if (ComplexMetrics.getSerdeForType("hyperUnique") == null) {
        ComplexMetrics.registerSerde("hyperUnique", new HyperUniquesSerde(HyperLogLogHash.getDefault()));
    }
    schemaInfo = BenchmarkSchemas.SCHEMA_MAP.get(schema);
    BenchmarkDataGenerator gen = new BenchmarkDataGenerator(schemaInfo.getColumnSchemas(), RNG_SEED, schemaInfo.getDataInterval(), rowsPerSegment);
    incIndex = makeIncIndex(schemaInfo.getAggsArray());
    filter = new OrDimFilter(Arrays.asList(new BoundDimFilter("dimSequential", "-1", "-1", true, true, null, null, StringComparators.ALPHANUMERIC), new JavaScriptDimFilter("dimSequential", "function(x) { return false }", null, JavaScriptConfig.getEnabledInstance()), new RegexDimFilter("dimSequential", "X", null), new SearchQueryDimFilter("dimSequential", new ContainsSearchQuerySpec("X", false), null), new InDimFilter("dimSequential", Arrays.asList("X"), null)));
    filteredMetrics = new AggregatorFactory[1];
    filteredMetrics[0] = new FilteredAggregatorFactory(new CountAggregatorFactory("rows"), filter);
    incIndexFilteredAgg = makeIncIndex(filteredMetrics);
    inputRows = new ArrayList<>();
    for (int j = 0; j < rowsPerSegment; j++) {
        InputRow row = gen.nextRow();
        if (j % 10000 == 0) {
            log.info(j + " rows generated.");
        }
        incIndex.add(row);
        inputRows.add(row);
    }
    tmpDir = Files.createTempDir();
    log.info("Using temp dir: " + tmpDir.getAbsolutePath());
    indexFile = INDEX_MERGER_V9.persist(incIndex, tmpDir, new IndexSpec());
    qIndex = INDEX_IO.loadIndex(indexFile);
    factory = new TimeseriesQueryRunnerFactory(new TimeseriesQueryQueryToolChest(QueryBenchmarkUtil.NoopIntervalChunkingQueryRunnerDecorator()), new TimeseriesQueryEngine(), QueryBenchmarkUtil.NOOP_QUERYWATCHER);
    BenchmarkSchemaInfo basicSchema = BenchmarkSchemas.SCHEMA_MAP.get("basic");
    QuerySegmentSpec intervalSpec = new MultipleIntervalSegmentSpec(Arrays.asList(basicSchema.getDataInterval()));
    List<AggregatorFactory> queryAggs = new ArrayList<>();
    queryAggs.add(filteredMetrics[0]);
    query = Druids.newTimeseriesQueryBuilder().dataSource("blah").granularity(Granularities.ALL).intervals(intervalSpec).aggregators(queryAggs).descending(false).build();
}

Example 12 with InDimFilter

use of io.druid.query.filter.InDimFilter in project druid by druid-io.

the class GroupByQueryRunnerTest method testGroupByNumericStringsAsNumericWithDecoration.

@Test
public void testGroupByNumericStringsAsNumericWithDecoration() {
    if (config.getDefaultStrategy().equals(GroupByStrategySelector.STRATEGY_V1)) {
        expectedException.expect(UnsupportedOperationException.class);
        expectedException.expectMessage("GroupBy v1 only supports dimensions with an outputType of STRING.");
    }
    // rows with `technology` have `170000` in the qualityNumericString field
    RegexFilteredDimensionSpec regexSpec = new RegexFilteredDimensionSpec(new DefaultDimensionSpec("qualityNumericString", "ql", ValueType.LONG), "170000");
    ListFilteredDimensionSpec listFilteredSpec = new ListFilteredDimensionSpec(new DefaultDimensionSpec("qualityNumericString", "qf", ValueType.FLOAT), Sets.newHashSet("170000"), true);
    GroupByQuery query = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setQuerySegmentSpec(QueryRunnerTestHelper.firstToThird).setDimensions(Lists.<DimensionSpec>newArrayList(regexSpec, listFilteredSpec)).setDimFilter(new InDimFilter("quality", Arrays.asList("entertainment", "technology"), null)).setAggregatorSpecs(Arrays.<AggregatorFactory>asList(new CountAggregatorFactory("count"))).setGranularity(QueryRunnerTestHelper.allGran).build();
    // "entertainment" rows are excluded by the decorated specs, they become empty rows
    List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "ql", 0L, "qf", 0.0, "count", 2L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "ql", 170000L, "qf", 170000.0, "count", 2L));
    Iterable<Row> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "");
}

Example 13 with InDimFilter

use of io.druid.query.filter.InDimFilter in project druid by druid-io.

the class GroupByQueryRunnerTest method testSubqueryWithOuterTimeFilter.

@Test
public void testSubqueryWithOuterTimeFilter() {
    final GroupByQuery subquery = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setQuerySegmentSpec(QueryRunnerTestHelper.fullOnInterval).setDimensions(Lists.<DimensionSpec>newArrayList(new DefaultDimensionSpec("market", "market"), new DefaultDimensionSpec("quality", "quality"))).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("index", "index"))).setGranularity(QueryRunnerTestHelper.dayGran).build();
    final DimFilter fridayFilter = new SelectorDimFilter(Column.TIME_COLUMN_NAME, "Friday", new TimeFormatExtractionFn("EEEE", null, null, null, false));
    final DimFilter firstDaysFilter = new InDimFilter(Column.TIME_COLUMN_NAME, ImmutableList.of("1", "2", "3"), new TimeFormatExtractionFn("d", null, null, null, false));
    final GroupByQuery query = GroupByQuery.builder().setDataSource(subquery).setQuerySegmentSpec(QueryRunnerTestHelper.fullOnInterval).setDimensions(Lists.<DimensionSpec>newArrayList()).setDimFilter(firstDaysFilter).setAggregatorSpecs(ImmutableList.<AggregatorFactory>of(new FilteredAggregatorFactory(QueryRunnerTestHelper.rowsCount, fridayFilter))).setGranularity(QueryRunnerTestHelper.dayGran).build();
    List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-02-01", "rows", 0L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-02-02", "rows", 0L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-02-03", "rows", 0L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-03-01", "rows", 0L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-03-02", "rows", 0L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-03-03", "rows", 0L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "rows", 13L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "rows", 0L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-03", "rows", 0L));
    Iterable<Row> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "");
}

Example 14 with InDimFilter

use of io.druid.query.filter.InDimFilter in project druid by druid-io.

the class GroupByQueryRunnerTest method testGroupByLongColumnDescending.

@Test
public void testGroupByLongColumnDescending() {
    if (config.getDefaultStrategy().equals(GroupByStrategySelector.STRATEGY_V1)) {
        expectedException.expect(UnsupportedOperationException.class);
        expectedException.expectMessage("GroupBy v1 only supports dimensions with an outputType of STRING.");
    }
    GroupByQuery query = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setQuerySegmentSpec(QueryRunnerTestHelper.firstToThird).setDimensions(Lists.<DimensionSpec>newArrayList(new DefaultDimensionSpec("qualityLong", "ql_alias", ValueType.LONG))).setDimFilter(new InDimFilter("quality", Arrays.asList("entertainment", "technology"), null)).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"))).addOrderByColumn(new OrderByColumnSpec("ql_alias", OrderByColumnSpec.Direction.DESCENDING, StringComparators.NUMERIC)).setGranularity(QueryRunnerTestHelper.allGran).build();
    Assert.assertNotEquals(Functions.<Sequence<Row>>identity(), query.getLimitSpec().build(query.getDimensions(), query.getAggregatorSpecs(), query.getPostAggregatorSpecs()));
    List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "ql_alias", 1700L, "rows", 2L, "idx", 175L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "ql_alias", 1200L, "rows", 2L, "idx", 324L));
    Iterable<Row> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "");
}

Example 15 with InDimFilter

use of io.druid.query.filter.InDimFilter in project druid by druid-io.

the class GroupByQueryRunnerTest method testBySegmentResultsWithAllFiltersWithExtractionFns.

@Test
public void testBySegmentResultsWithAllFiltersWithExtractionFns() {
    int segmentCount = 32;
    Result<BySegmentResultValue> singleSegmentResult = new Result<BySegmentResultValue>(new DateTime("2011-01-12T00:00:00.000Z"), new BySegmentResultValueClass(Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "mezzanine", "rows", 6L, "idx", 4420L)), "testSegment", new Interval("2011-04-02T00:00:00.000Z/2011-04-04T00:00:00.000Z")));
    List<Result> bySegmentResults = Lists.newArrayList();
    for (int i = 0; i < segmentCount; i++) {
        bySegmentResults.add(singleSegmentResult);
    }
    String extractionJsFn = "function(str) { return 'super-' + str; }";
    String jsFn = "function(x) { return(x === 'super-mezzanine') }";
    ExtractionFn extractionFn = new JavaScriptExtractionFn(extractionJsFn, false, JavaScriptConfig.getEnabledInstance());
    List<DimFilter> superFilterList = new ArrayList<>();
    superFilterList.add(new SelectorDimFilter("quality", "super-mezzanine", extractionFn));
    superFilterList.add(new InDimFilter("quality", Arrays.asList("not-super-mezzanine", "FOOBAR", "super-mezzanine"), extractionFn));
    superFilterList.add(new BoundDimFilter("quality", "super-mezzanine", "super-mezzanine", false, false, true, extractionFn, StringComparators.ALPHANUMERIC));
    superFilterList.add(new RegexDimFilter("quality", "super-mezzanine", extractionFn));
    superFilterList.add(new SearchQueryDimFilter("quality", new ContainsSearchQuerySpec("super-mezzanine", true), extractionFn));
    superFilterList.add(new JavaScriptDimFilter("quality", jsFn, extractionFn, JavaScriptConfig.getEnabledInstance()));
    DimFilter superFilter = new AndDimFilter(superFilterList);
    GroupByQuery.Builder builder = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setInterval("2011-04-02/2011-04-04").setDimensions(Lists.<DimensionSpec>newArrayList(new DefaultDimensionSpec("quality", "alias"))).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"))).setGranularity(new PeriodGranularity(new Period("P1M"), null, null)).setDimFilter(superFilter).setContext(ImmutableMap.<String, Object>of("bySegment", true));
    final GroupByQuery fullQuery = builder.build();
    QueryToolChest toolChest = factory.getToolchest();
    List<QueryRunner<Row>> singleSegmentRunners = Lists.newArrayList();
    for (int i = 0; i < segmentCount; i++) {
        singleSegmentRunners.add(toolChest.preMergeQueryDecoration(runner));
    }
    ExecutorService exec = Executors.newCachedThreadPool();
    QueryRunner theRunner = toolChest.postMergeQueryDecoration(new FinalizeResultsQueryRunner<>(toolChest.mergeResults(factory.mergeRunners(Executors.newCachedThreadPool(), singleSegmentRunners)), toolChest));
    TestHelper.assertExpectedObjects(bySegmentResults, theRunner.run(fullQuery, Maps.newHashMap()), "");
    exec.shutdownNow();
}