Examples with ExtractionDimensionSpec org.apache.druid.query.dimension.ExtractionDimensionSpec used on opensource projects

Example 51 with ExtractionDimensionSpec

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

the class GroupByQueryRunnerTest method testGroupByOnMissingColumn.

@Test
public void testGroupByOnMissingColumn() {
    // Cannot vectorize due to extraction dimension spec.
    cannotVectorize();
    GroupByQuery query = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FIRST_TO_THIRD).setDimensions(new DefaultDimensionSpec("nonexistent0", "alias0"), new ExtractionDimensionSpec("nonexistent1", "alias1", new StringFormatExtractionFn("foo"))).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT).setGranularity(QueryRunnerTestHelper.ALL_GRAN).build();
    List<ResultRow> expectedResults = Collections.singletonList(makeRow(query, "2011-04-01", "alias0", null, "alias1", "foo", "rows", 26L));
    Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "missing-column");
}

Example 52 with ExtractionDimensionSpec

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

the class GroupByQueryRunnerTest method testGroupByWithSimpleRename.

@Test
public void testGroupByWithSimpleRename() {
    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 = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FIRST_TO_THIRD).setDimensions(new ExtractionDimensionSpec("quality", "alias", new LookupExtractionFn(new MapLookupExtractor(map, false), false, null, true, false))).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, new LongSumAggregatorFactory("idx", "index")).setGranularity(QueryRunnerTestHelper.DAY_GRAN).build();
    List<ResultRow> expectedResults = Arrays.asList(makeRow(query, "2011-04-01", "alias", "automotive0", "rows", 1L, "idx", 135L), makeRow(query, "2011-04-01", "alias", "business0", "rows", 1L, "idx", 118L), makeRow(query, "2011-04-01", "alias", "entertainment0", "rows", 1L, "idx", 158L), makeRow(query, "2011-04-01", "alias", "health0", "rows", 1L, "idx", 120L), makeRow(query, "2011-04-01", "alias", "mezzanine0", "rows", 3L, "idx", 2870L), makeRow(query, "2011-04-01", "alias", "news0", "rows", 1L, "idx", 121L), makeRow(query, "2011-04-01", "alias", "premium0", "rows", 3L, "idx", 2900L), makeRow(query, "2011-04-01", "alias", "technology0", "rows", 1L, "idx", 78L), makeRow(query, "2011-04-01", "alias", "travel0", "rows", 1L, "idx", 119L), makeRow(query, "2011-04-02", "alias", "automotive0", "rows", 1L, "idx", 147L), makeRow(query, "2011-04-02", "alias", "business0", "rows", 1L, "idx", 112L), makeRow(query, "2011-04-02", "alias", "entertainment0", "rows", 1L, "idx", 166L), makeRow(query, "2011-04-02", "alias", "health0", "rows", 1L, "idx", 113L), makeRow(query, "2011-04-02", "alias", "mezzanine0", "rows", 3L, "idx", 2447L), makeRow(query, "2011-04-02", "alias", "news0", "rows", 1L, "idx", 114L), makeRow(query, "2011-04-02", "alias", "premium0", "rows", 3L, "idx", 2505L), makeRow(query, "2011-04-02", "alias", "technology0", "rows", 1L, "idx", 97L), makeRow(query, "2011-04-02", "alias", "travel0", "rows", 1L, "idx", 126L));
    Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "simple-rename");
}

Example 53 with ExtractionDimensionSpec

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

the class GroupByQueryRunnerTest method testGroupByWithNullProducingDimExtractionFn.

@Test
public void testGroupByWithNullProducingDimExtractionFn() {
    // Cannot vectorize due to extraction dimension spec.
    cannotVectorize();
    final ExtractionFn nullExtractionFn = new RegexDimExtractionFn("(\\w{1})", false, null) {

        @Override
        public byte[] getCacheKey() {
            return new byte[] { (byte) 0xFF };
        }

        @Override
        public String apply(String dimValue) {
            return "mezzanine".equals(dimValue) ? null : super.apply(dimValue);
        }
    };
    GroupByQuery query = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FIRST_TO_THIRD).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, new LongSumAggregatorFactory("idx", "index")).setGranularity(QueryRunnerTestHelper.DAY_GRAN).setDimensions(new ExtractionDimensionSpec("quality", "alias", nullExtractionFn)).build();
    List<ResultRow> expectedResults = Arrays.asList(makeRow(query, "2011-04-01", "alias", null, "rows", 3L, "idx", 2870L), makeRow(query, "2011-04-01", "alias", "a", "rows", 1L, "idx", 135L), makeRow(query, "2011-04-01", "alias", "b", "rows", 1L, "idx", 118L), makeRow(query, "2011-04-01", "alias", "e", "rows", 1L, "idx", 158L), makeRow(query, "2011-04-01", "alias", "h", "rows", 1L, "idx", 120L), makeRow(query, "2011-04-01", "alias", "n", "rows", 1L, "idx", 121L), makeRow(query, "2011-04-01", "alias", "p", "rows", 3L, "idx", 2900L), makeRow(query, "2011-04-01", "alias", "t", "rows", 2L, "idx", 197L), makeRow(query, "2011-04-02", "alias", null, "rows", 3L, "idx", 2447L), makeRow(query, "2011-04-02", "alias", "a", "rows", 1L, "idx", 147L), makeRow(query, "2011-04-02", "alias", "b", "rows", 1L, "idx", 112L), makeRow(query, "2011-04-02", "alias", "e", "rows", 1L, "idx", 166L), makeRow(query, "2011-04-02", "alias", "h", "rows", 1L, "idx", 113L), makeRow(query, "2011-04-02", "alias", "n", "rows", 1L, "idx", 114L), makeRow(query, "2011-04-02", "alias", "p", "rows", 3L, "idx", 2505L), makeRow(query, "2011-04-02", "alias", "t", "rows", 2L, "idx", 223L));
    TestHelper.assertExpectedObjects(expectedResults, GroupByQueryRunnerTestHelper.runQuery(factory, runner, query), "null-dimextraction");
}

Example 54 with ExtractionDimensionSpec

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

the class GroupByQueryRunnerTest method testTypeConversionWithMergingChainedExecutionRunner.

@Test
public void testTypeConversionWithMergingChainedExecutionRunner() {
    // Cannot vectorize due to extraction dimension spec.
    cannotVectorize();
    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 = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FIRST_TO_THIRD).setDimensions(new DefaultDimensionSpec("quality", "alias"), new ExtractionDimensionSpec("quality", "qualityLen", ColumnType.LONG, StrlenExtractionFn.instance())).setDimFilter(new SelectorDimFilter("quality", "technology", null)).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, new LongSumAggregatorFactory("idx", "index")).setGranularity(QueryRunnerTestHelper.DAY_GRAN).build();
    List<ResultRow> expectedResults = Arrays.asList(makeRow(query, "2011-04-01", "alias", "technology", "qualityLen", 10L, "rows", 2L, "idx", 156L), makeRow(query, "2011-04-02", "alias", "technology", "qualityLen", 10L, "rows", 2L, "idx", 194L));
    ChainedExecutionQueryRunner ceqr = new ChainedExecutionQueryRunner(DirectQueryProcessingPool.INSTANCE, (query1, future) -> {
        return;
    }, ImmutableList.of(runner, runner));
    QueryRunner<ResultRow> mergingRunner = factory.mergeRunners(Execs.directExecutor(), ImmutableList.of(ceqr));
    Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, mergingRunner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "type-conversion");
}

Example 55 with ExtractionDimensionSpec

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

the class GroupByQueryRunnerTest method testGroupByWithEmptyStringProducingDimExtractionFn.

@Test
@Ignore
public /**
 * This test exists only to show what the current behavior is and not necessarily to define that this is
 * correct behavior.  In fact, the behavior when returning the empty string from a DimExtractionFn is, by
 * contract, undefined, so this can do anything.
 */
void testGroupByWithEmptyStringProducingDimExtractionFn() {
    final ExtractionFn emptyStringExtractionFn = new RegexDimExtractionFn("(\\w{1})", false, null) {

        @Override
        public byte[] getCacheKey() {
            return new byte[] { (byte) 0xFF };
        }

        @Override
        public String apply(String dimValue) {
            return "mezzanine".equals(dimValue) ? "" : super.apply(dimValue);
        }
    };
    GroupByQuery query = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FIRST_TO_THIRD).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, new LongSumAggregatorFactory("idx", "index")).setGranularity(QueryRunnerTestHelper.DAY_GRAN).setDimensions(new ExtractionDimensionSpec("quality", "alias", emptyStringExtractionFn)).build();
    List<ResultRow> expectedResults = Arrays.asList(makeRow(query, "2011-04-01", "alias", "", "rows", 3L, "idx", 2870L), makeRow(query, "2011-04-01", "alias", "a", "rows", 1L, "idx", 135L), makeRow(query, "2011-04-01", "alias", "b", "rows", 1L, "idx", 118L), makeRow(query, "2011-04-01", "alias", "e", "rows", 1L, "idx", 158L), makeRow(query, "2011-04-01", "alias", "h", "rows", 1L, "idx", 120L), makeRow(query, "2011-04-01", "alias", "n", "rows", 1L, "idx", 121L), makeRow(query, "2011-04-01", "alias", "p", "rows", 3L, "idx", 2900L), makeRow(query, "2011-04-01", "alias", "t", "rows", 2L, "idx", 197L), makeRow(query, "2011-04-02", "alias", "", "rows", 3L, "idx", 2447L), makeRow(query, "2011-04-02", "alias", "a", "rows", 1L, "idx", 147L), makeRow(query, "2011-04-02", "alias", "b", "rows", 1L, "idx", 112L), makeRow(query, "2011-04-02", "alias", "e", "rows", 1L, "idx", 166L), makeRow(query, "2011-04-02", "alias", "h", "rows", 1L, "idx", 113L), makeRow(query, "2011-04-02", "alias", "n", "rows", 1L, "idx", 114L), makeRow(query, "2011-04-02", "alias", "p", "rows", 3L, "idx", 2505L), makeRow(query, "2011-04-02", "alias", "t", "rows", 2L, "idx", 223L));
    TestHelper.assertExpectedObjects(expectedResults, GroupByQueryRunnerTestHelper.runQuery(factory, runner, query), "empty-string-dimextraction");
}