use of io.druid.query.dimension.DimensionSpec in project druid by druid-io.
the class DataSourceTest method testQueryDataSource.
@Test
public void testQueryDataSource() throws IOException {
GroupByQuery query = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setQuerySegmentSpec(QueryRunnerTestHelper.firstToThird).setDimensions(Lists.<DimensionSpec>newArrayList(new DefaultDimensionSpec("quality", "alias"))).setAggregatorSpecs(Arrays.<AggregatorFactory>asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"))).setGranularity(QueryRunnerTestHelper.dayGran).build();
String dataSourceJSON = "{\"type\":\"query\", \"query\":" + jsonMapper.writeValueAsString(query) + "}";
DataSource dataSource = jsonMapper.readValue(dataSourceJSON, DataSource.class);
Assert.assertEquals(new QueryDataSource(query), dataSource);
}
use of io.druid.query.dimension.DimensionSpec in project druid by druid-io.
the class MultiValuedDimensionTest method testGroupByWithDimFilterAndWithFilteredDimSpec.
@Test
public void testGroupByWithDimFilterAndWithFilteredDimSpec() throws Exception {
GroupByQuery query = GroupByQuery.builder().setDataSource("xx").setQuerySegmentSpec(new LegacySegmentSpec("1970/3000")).setGranularity(Granularities.ALL).setDimensions(Lists.<DimensionSpec>newArrayList(new RegexFilteredDimensionSpec(new DefaultDimensionSpec("tags", "tags"), "t3"))).setAggregatorSpecs(Arrays.asList(new AggregatorFactory[] { new CountAggregatorFactory("count") })).setDimFilter(new SelectorDimFilter("tags", "t3", null)).build();
Sequence<Row> result = helper.runQueryOnSegmentsObjs(ImmutableList.<Segment>of(new QueryableIndexSegment("sid1", queryableIndex), new IncrementalIndexSegment(incrementalIndex, "sid2")), query);
List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("1970-01-01T00:00:00.000Z", "tags", "t3", "count", 4L));
TestHelper.assertExpectedObjects(expectedResults, Sequences.toList(result, new ArrayList<Row>()), "");
}
use of io.druid.query.dimension.DimensionSpec in project druid by druid-io.
the class IncrementalIndex method makeColumnSelectorFactory.
/**
* Column selector used at ingestion time for inputs to aggregators.
*
* @param agg the aggregator
* @param in ingestion-time input row supplier
* @param deserializeComplexMetrics whether complex objects should be deserialized by a {@link ComplexMetricExtractor}
*
* @return column selector factory
*/
public static ColumnSelectorFactory makeColumnSelectorFactory(final VirtualColumns virtualColumns, final AggregatorFactory agg, final Supplier<InputRow> in, final boolean deserializeComplexMetrics) {
final RowBasedColumnSelectorFactory baseSelectorFactory = RowBasedColumnSelectorFactory.create(in, null);
class IncrementalIndexInputRowColumnSelectorFactory implements ColumnSelectorFactory {
@Override
public LongColumnSelector makeLongColumnSelector(final String columnName) {
return baseSelectorFactory.makeLongColumnSelector(columnName);
}
@Override
public FloatColumnSelector makeFloatColumnSelector(final String columnName) {
return baseSelectorFactory.makeFloatColumnSelector(columnName);
}
@Override
public ObjectColumnSelector makeObjectColumnSelector(final String column) {
final String typeName = agg.getTypeName();
final ObjectColumnSelector rawColumnSelector = baseSelectorFactory.makeObjectColumnSelector(column);
if ((Enums.getIfPresent(ValueType.class, typeName.toUpperCase()).isPresent() && !typeName.equalsIgnoreCase(ValueType.COMPLEX.name())) || !deserializeComplexMetrics) {
return rawColumnSelector;
} else {
final ComplexMetricSerde serde = ComplexMetrics.getSerdeForType(typeName);
if (serde == null) {
throw new ISE("Don't know how to handle type[%s]", typeName);
}
final ComplexMetricExtractor extractor = serde.getExtractor();
return new ObjectColumnSelector() {
@Override
public Class classOfObject() {
return extractor.extractedClass();
}
@Override
public Object get() {
return extractor.extractValue(in.get(), column);
}
};
}
}
@Override
public DimensionSelector makeDimensionSelector(DimensionSpec dimensionSpec) {
return baseSelectorFactory.makeDimensionSelector(dimensionSpec);
}
@Nullable
@Override
public ColumnCapabilities getColumnCapabilities(String columnName) {
return baseSelectorFactory.getColumnCapabilities(columnName);
}
}
return virtualColumns.wrap(new IncrementalIndexInputRowColumnSelectorFactory());
}
use of io.druid.query.dimension.DimensionSpec in project druid by druid-io.
the class GroupByQueryRunnerTest method testSubqueryWithPostAggregators.
@Test
public void testSubqueryWithPostAggregators() {
final GroupByQuery subquery = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setQuerySegmentSpec(QueryRunnerTestHelper.firstToThird).setDimensions(Lists.<DimensionSpec>newArrayList(new DefaultDimensionSpec("quality", "alias"))).setDimFilter(new JavaScriptDimFilter("quality", "function(dim){ return true; }", null, JavaScriptConfig.getEnabledInstance())).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx_subagg", "index"))).setPostAggregatorSpecs(Arrays.<PostAggregator>asList(new ArithmeticPostAggregator("idx_subpostagg", "+", Arrays.asList(new FieldAccessPostAggregator("the_idx_subagg", "idx_subagg"), new ConstantPostAggregator("thousand", 1000))))).setGranularity(QueryRunnerTestHelper.dayGran).build();
final GroupByQuery query = GroupByQuery.builder().setDataSource(subquery).setQuerySegmentSpec(QueryRunnerTestHelper.firstToThird).setDimensions(Lists.<DimensionSpec>newArrayList(new DefaultDimensionSpec("alias", "alias"))).setAggregatorSpecs(Arrays.<AggregatorFactory>asList(new LongSumAggregatorFactory("rows", "rows"), new LongSumAggregatorFactory("idx", "idx_subpostagg"))).setPostAggregatorSpecs(Arrays.<PostAggregator>asList(new ArithmeticPostAggregator("idx_post", "+", Arrays.asList(new FieldAccessPostAggregator("the_idx_agg", "idx"), new ConstantPostAggregator("ten_thousand", 10000))))).setGranularity(QueryRunnerTestHelper.dayGran).build();
List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "automotive", "rows", 1L, "idx_post", 11135.0, "idx", 1135L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "business", "rows", 1L, "idx_post", 11118.0, "idx", 1118L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "entertainment", "rows", 1L, "idx_post", 11158.0, "idx", 1158L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "health", "rows", 1L, "idx_post", 11120.0, "idx", 1120L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "mezzanine", "rows", 3L, "idx_post", 13870.0, "idx", 3870L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "news", "rows", 1L, "idx_post", 11121.0, "idx", 1121L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "premium", "rows", 3L, "idx_post", 13900.0, "idx", 3900L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "technology", "rows", 1L, "idx_post", 11078.0, "idx", 1078L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "travel", "rows", 1L, "idx_post", 11119.0, "idx", 1119L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "automotive", "rows", 1L, "idx_post", 11147.0, "idx", 1147L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "business", "rows", 1L, "idx_post", 11112.0, "idx", 1112L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "entertainment", "rows", 1L, "idx_post", 11166.0, "idx", 1166L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "health", "rows", 1L, "idx_post", 11113.0, "idx", 1113L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "mezzanine", "rows", 3L, "idx_post", 13447.0, "idx", 3447L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "news", "rows", 1L, "idx_post", 11114.0, "idx", 1114L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "premium", "rows", 3L, "idx_post", 13505.0, "idx", 3505L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "technology", "rows", 1L, "idx_post", 11097.0, "idx", 1097L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "travel", "rows", 1L, "idx_post", 11126.0, "idx", 1126L));
// Subqueries are handled by the ToolChest
Iterable<Row> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
TestHelper.assertExpectedObjects(expectedResults, results, "");
}
use of io.druid.query.dimension.DimensionSpec 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 dimValue.equals("mezzanine") ? "" : super.apply(dimValue);
}
};
GroupByQuery query = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setQuerySegmentSpec(QueryRunnerTestHelper.firstToThird).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"))).setGranularity(QueryRunnerTestHelper.dayGran).setDimensions(Lists.<DimensionSpec>newArrayList(new ExtractionDimensionSpec("quality", "alias", emptyStringExtractionFn))).build();
List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "", "rows", 3L, "idx", 2870L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "a", "rows", 1L, "idx", 135L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "b", "rows", 1L, "idx", 118L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "e", "rows", 1L, "idx", 158L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "h", "rows", 1L, "idx", 120L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "n", "rows", 1L, "idx", 121L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "p", "rows", 3L, "idx", 2900L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "t", "rows", 2L, "idx", 197L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "", "rows", 3L, "idx", 2447L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "a", "rows", 1L, "idx", 147L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "b", "rows", 1L, "idx", 112L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "e", "rows", 1L, "idx", 166L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "h", "rows", 1L, "idx", 113L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "n", "rows", 1L, "idx", 114L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "p", "rows", 3L, "idx", 2505L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "t", "rows", 2L, "idx", 223L));
TestHelper.assertExpectedObjects(expectedResults, GroupByQueryRunnerTestHelper.runQuery(factory, runner, query), "");
}
Aggregations