Examples with ExtractionDimensionSpec io.druid.query.dimension.ExtractionDimensionSpec used on opensource projects

Example 41 with ExtractionDimensionSpec

use of io.druid.query.dimension.ExtractionDimensionSpec in project druid by druid-io.

the class GroupByQueryRunnerTest method testGroupByWithSimpleRenameRetainMissing.

@Test
public void testGroupByWithSimpleRenameRetainMissing() {
    Map<String, String> map = new HashMap<>();
    map.put("automotive", "automotive0");
    map.put("business", "business0");
    map.put("entertainment", "entertainment0");
    map.put("health", "health0");
    map.put("mezzanine", "mezzanine0");
    map.put("news", "news0");
    map.put("premium", "premium0");
    map.put("technology", "technology0");
    map.put("travel", "travel0");
    GroupByQuery query = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setQuerySegmentSpec(QueryRunnerTestHelper.firstToThird).setDimensions(Lists.<DimensionSpec>newArrayList(new ExtractionDimensionSpec("quality", "alias", new LookupExtractionFn(new MapLookupExtractor(map, false), true, null, true, false)))).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"))).setGranularity(QueryRunnerTestHelper.dayGran).build();
    List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "automotive0", "rows", 1L, "idx", 135L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "business0", "rows", 1L, "idx", 118L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "entertainment0", "rows", 1L, "idx", 158L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "health0", "rows", 1L, "idx", 120L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "mezzanine0", "rows", 3L, "idx", 2870L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "news0", "rows", 1L, "idx", 121L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "premium0", "rows", 3L, "idx", 2900L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "technology0", "rows", 1L, "idx", 78L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "travel0", "rows", 1L, "idx", 119L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "automotive0", "rows", 1L, "idx", 147L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "business0", "rows", 1L, "idx", 112L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "entertainment0", "rows", 1L, "idx", 166L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "health0", "rows", 1L, "idx", 113L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "mezzanine0", "rows", 3L, "idx", 2447L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "news0", "rows", 1L, "idx", 114L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "premium0", "rows", 3L, "idx", 2505L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "technology0", "rows", 1L, "idx", 97L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "travel0", "rows", 1L, "idx", 126L));
    Iterable<Row> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "");
}

Example 42 with ExtractionDimensionSpec

use of io.druid.query.dimension.ExtractionDimensionSpec in project druid by druid-io.

the class GroupByQueryRunnerTest method testBySegmentResultsUnOptimizedDimextraction.

@Test
public void testBySegmentResultsUnOptimizedDimextraction() {
    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", "mezzanine0", "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);
    }
    GroupByQuery.Builder builder = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setInterval("2011-04-02/2011-04-04").setDimensions(Lists.<DimensionSpec>newArrayList(new ExtractionDimensionSpec("quality", "alias", new LookupExtractionFn(new MapLookupExtractor(ImmutableMap.of("mezzanine", "mezzanine0"), false), false, null, false, false)))).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"))).setGranularity(new PeriodGranularity(new Period("P1M"), null, null)).setDimFilter(new SelectorDimFilter("quality", "mezzanine", null)).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();
}

Example 43 with ExtractionDimensionSpec

use of io.druid.query.dimension.ExtractionDimensionSpec in project druid by druid-io.

the class GroupByQueryRunnerTest method testBySegmentResultsOptimizedDimextraction.

@Test
public void testBySegmentResultsOptimizedDimextraction() {
    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", "mezzanine0", "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);
    }
    GroupByQuery.Builder builder = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setInterval("2011-04-02/2011-04-04").setDimensions(Lists.<DimensionSpec>newArrayList(new ExtractionDimensionSpec("quality", "alias", new LookupExtractionFn(new MapLookupExtractor(ImmutableMap.of("mezzanine", "mezzanine0"), false), false, null, true, false)))).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"))).setGranularity(new PeriodGranularity(new Period("P1M"), null, null)).setDimFilter(new SelectorDimFilter("quality", "mezzanine", null)).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();
}

Example 44 with ExtractionDimensionSpec

use of io.druid.query.dimension.ExtractionDimensionSpec in project druid by druid-io.

the class GroupByQueryRunnerTest method testGroupByStringOutputAsLong.

@Test
public void testGroupByStringOutputAsLong() {
    if (config.getDefaultStrategy().equals(GroupByStrategySelector.STRATEGY_V1)) {
        expectedException.expect(UnsupportedOperationException.class);
        expectedException.expectMessage("GroupBy v1 only supports dimensions with an outputType of STRING.");
    }
    ExtractionFn strlenFn = StrlenExtractionFn.instance();
    GroupByQuery query = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setQuerySegmentSpec(QueryRunnerTestHelper.firstToThird).setDimensions(Lists.<DimensionSpec>newArrayList(new ExtractionDimensionSpec(QueryRunnerTestHelper.qualityDimension, "alias", ValueType.LONG, strlenFn))).setDimFilter(new SelectorDimFilter(QueryRunnerTestHelper.qualityDimension, "entertainment", null)).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"))).setGranularity(QueryRunnerTestHelper.dayGran).build();
    List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", 13L, "rows", 1L, "idx", 158L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", 13L, "rows", 1L, "idx", 166L));
    Iterable<Row> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "");
}

Example 45 with ExtractionDimensionSpec

use of io.druid.query.dimension.ExtractionDimensionSpec in project druid by druid-io.

the class GroupByQueryRunnerTest method testGroupByLongColumnWithExFn.

@Test
public void testGroupByLongColumnWithExFn() {
    if (config.getDefaultStrategy().equals(GroupByStrategySelector.STRATEGY_V1)) {
        expectedException.expect(UnsupportedOperationException.class);
        expectedException.expectMessage("GroupBy v1 does not support dimension selectors with unknown cardinality.");
    }
    String jsFn = "function(str) { return 'super-' + str; }";
    ExtractionFn jsExtractionFn = new JavaScriptExtractionFn(jsFn, false, JavaScriptConfig.getEnabledInstance());
    GroupByQuery query = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setQuerySegmentSpec(QueryRunnerTestHelper.firstToThird).setDimensions(Lists.<DimensionSpec>newArrayList(new ExtractionDimensionSpec("qualityLong", "ql_alias", jsExtractionFn))).setDimFilter(new SelectorDimFilter("quality", "entertainment", null)).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"))).setGranularity(QueryRunnerTestHelper.dayGran).build();
    List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "ql_alias", "super-1200", "rows", 1L, "idx", 158L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "ql_alias", "super-1200", "rows", 1L, "idx", 166L));
    Iterable<Row> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "");
}