use of org.apache.druid.query.dimension.DefaultDimensionSpec in project druid by druid-io.
the class GroupByQueryRunnerTest method testGroupByNumericStringsAsNumericWithDecoration.
@Test
public void testGroupByNumericStringsAsNumericWithDecoration() {
// Cannot vectorize due to regex-filtered 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.");
}
// rows with `technology` have `170000` in the qualityNumericString field
RegexFilteredDimensionSpec regexSpec = new RegexFilteredDimensionSpec(new DefaultDimensionSpec("qualityNumericString", "ql", ColumnType.LONG), "170000");
ListFilteredDimensionSpec listFilteredSpec = new ListFilteredDimensionSpec(new DefaultDimensionSpec("qualityNumericString", "qf", ColumnType.FLOAT), Sets.newHashSet("170000"), true);
GroupByQuery query = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FIRST_TO_THIRD).setDimensions(regexSpec, listFilteredSpec).setDimFilter(new InDimFilter("quality", Arrays.asList("entertainment", "technology"), null)).setAggregatorSpecs(new CountAggregatorFactory("count")).setGranularity(QueryRunnerTestHelper.ALL_GRAN).addOrderByColumn("ql").build();
List<ResultRow> expectedResults;
// "entertainment" rows are excluded by the decorated specs, they become empty rows
expectedResults = Arrays.asList(makeRow(query, "2011-04-01", "ql", NullHandling.defaultLongValue(), "qf", NullHandling.defaultDoubleValue(), "count", 2L), makeRow(query, "2011-04-01", "ql", 170000L, "qf", 170000.0, "count", 2L));
Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
TestHelper.assertExpectedObjects(expectedResults, results, "numeric-string");
}
use of org.apache.druid.query.dimension.DefaultDimensionSpec in project druid by druid-io.
the class GroupByQueryRunnerTest method testIdenticalSubquery.
// A subquery identical to the query should yield identical results
@Test
public void testIdenticalSubquery() {
GroupByQuery subquery = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FIRST_TO_THIRD).setDimensions(new DefaultDimensionSpec("quality", "alias")).setDimFilter(new JavaScriptDimFilter("quality", "function(dim){ return true; }", null, JavaScriptConfig.getEnabledInstance())).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, new LongSumAggregatorFactory("idx", "index"), new LongSumAggregatorFactory("indexMaxPlusTen", "indexMaxPlusTen")).setGranularity(QueryRunnerTestHelper.DAY_GRAN).build();
GroupByQuery query = makeQueryBuilder().setDataSource(subquery).setQuerySegmentSpec(QueryRunnerTestHelper.FIRST_TO_THIRD).setDimensions(new DefaultDimensionSpec("alias", "alias")).setAggregatorSpecs(new LongSumAggregatorFactory("rows", "rows"), new LongSumAggregatorFactory("idx", "idx")).setGranularity(QueryRunnerTestHelper.DAY_GRAN).build();
List<ResultRow> expectedResults = Arrays.asList(makeRow(query, "2011-04-01", "alias", "automotive", "rows", 1L, "idx", 135L), makeRow(query, "2011-04-01", "alias", "business", "rows", 1L, "idx", 118L), makeRow(query, "2011-04-01", "alias", "entertainment", "rows", 1L, "idx", 158L), makeRow(query, "2011-04-01", "alias", "health", "rows", 1L, "idx", 120L), makeRow(query, "2011-04-01", "alias", "mezzanine", "rows", 3L, "idx", 2870L), makeRow(query, "2011-04-01", "alias", "news", "rows", 1L, "idx", 121L), makeRow(query, "2011-04-01", "alias", "premium", "rows", 3L, "idx", 2900L), makeRow(query, "2011-04-01", "alias", "technology", "rows", 1L, "idx", 78L), makeRow(query, "2011-04-01", "alias", "travel", "rows", 1L, "idx", 119L), makeRow(query, "2011-04-02", "alias", "automotive", "rows", 1L, "idx", 147L), makeRow(query, "2011-04-02", "alias", "business", "rows", 1L, "idx", 112L), makeRow(query, "2011-04-02", "alias", "entertainment", "rows", 1L, "idx", 166L), makeRow(query, "2011-04-02", "alias", "health", "rows", 1L, "idx", 113L), makeRow(query, "2011-04-02", "alias", "mezzanine", "rows", 3L, "idx", 2447L), makeRow(query, "2011-04-02", "alias", "news", "rows", 1L, "idx", 114L), makeRow(query, "2011-04-02", "alias", "premium", "rows", 3L, "idx", 2505L), makeRow(query, "2011-04-02", "alias", "technology", "rows", 1L, "idx", 97L), makeRow(query, "2011-04-02", "alias", "travel", "rows", 1L, "idx", 126L));
// Subqueries are handled by the ToolChest
Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
TestHelper.assertExpectedObjects(expectedResults, results, "sub-query");
}
use of org.apache.druid.query.dimension.DefaultDimensionSpec in project druid by druid-io.
the class GroupByQueryRunnerTest method testGroupByWithExtractionDimFilterKeyisNull.
@Test
public void testGroupByWithExtractionDimFilterKeyisNull() {
Map<String, String> extractionMap = new HashMap<>();
MapLookupExtractor mapLookupExtractor = new MapLookupExtractor(extractionMap, false);
LookupExtractionFn lookupExtractionFn;
if (NullHandling.replaceWithDefault()) {
lookupExtractionFn = new LookupExtractionFn(mapLookupExtractor, false, null, true, false);
extractionMap.put("", "REPLACED_VALUE");
} else {
lookupExtractionFn = new LookupExtractionFn(mapLookupExtractor, false, "REPLACED_VALUE", true, false);
extractionMap.put("", "NOT_USED");
}
GroupByQuery query = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FIRST_TO_THIRD).setDimensions(new DefaultDimensionSpec("null_column", "alias")).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, new LongSumAggregatorFactory("idx", "index")).setGranularity(QueryRunnerTestHelper.DAY_GRAN).setDimFilter(new ExtractionDimFilter("null_column", "REPLACED_VALUE", lookupExtractionFn, null)).build();
List<ResultRow> expectedResults = Arrays.asList(makeRow(query, "2011-04-01", "alias", null, "rows", 13L, "idx", 6619L), makeRow(query, "2011-04-02", "alias", null, "rows", 13L, "idx", 5827L));
Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
TestHelper.assertExpectedObjects(expectedResults, results, "dim-extraction");
}
use of org.apache.druid.query.dimension.DefaultDimensionSpec in project druid by druid-io.
the class GroupByQueryRunnerTest method testGroupByLongColumn.
@Test
public void testGroupByLongColumn() {
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("qualityLong", "ql_alias", ColumnType.LONG)).setDimFilter(new SelectorDimFilter("quality", "entertainment", null)).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, new LongSumAggregatorFactory("idx", "index")).addOrderByColumn(new OrderByColumnSpec("ql_alias", OrderByColumnSpec.Direction.ASCENDING, StringComparators.NUMERIC)).setGranularity(QueryRunnerTestHelper.DAY_GRAN).build();
Assert.assertEquals(Functions.<Sequence<ResultRow>>identity(), query.getLimitSpec().build(query));
List<ResultRow> expectedResults = Arrays.asList(makeRow(query, "2011-04-01", "ql_alias", 1200L, "rows", 1L, "idx", 158L), makeRow(query, "2011-04-02", "ql_alias", 1200L, "rows", 1L, "idx", 166L));
Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
TestHelper.assertExpectedObjects(expectedResults, results, "long");
}
use of org.apache.druid.query.dimension.DefaultDimensionSpec in project druid by druid-io.
the class GroupByQueryRunnerTest method testGroupByWithRegEx.
@Test
public void testGroupByWithRegEx() {
GroupByQuery.Builder builder = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setInterval("2011-04-02/2011-04-04").setDimFilter(new RegexDimFilter("quality", "auto.*", null)).setDimensions(new DefaultDimensionSpec("quality", "quality")).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT).setGranularity(new PeriodGranularity(new Period("P1M"), null, null));
final GroupByQuery query = builder.build();
List<ResultRow> expectedResults = Collections.singletonList(makeRow(query, "2011-04-01", "quality", "automotive", "rows", 2L));
QueryRunner<ResultRow> mergeRunner = factory.getToolchest().mergeResults(runner);
TestHelper.assertExpectedObjects(expectedResults, mergeRunner.run(QueryPlus.wrap(query)), "no-limit");
}
Aggregations