Examples with FilteredAggregatorFactory org.apache.druid.query.aggregation.FilteredAggregatorFactory used on opensource projects

Example 26 with FilteredAggregatorFactory

use of org.apache.druid.query.aggregation.FilteredAggregatorFactory in project druid by druid-io.

the class DoublesSketchSqlAggregatorTest method testQuantileOnCastedString.

@Test
public void testQuantileOnCastedString() throws Exception {
    final List<Object[]> expectedResults;
    if (NullHandling.replaceWithDefault()) {
        expectedResults = ImmutableList.of(new Object[] { 0.0, 1.0, 10.1, 10.1, 20.2, 0.0, 10.1, 0.0 });
    } else {
        expectedResults = ImmutableList.of(new Object[] { 1.0, 2.0, 10.1, 10.1, 20.2, Double.NaN, 10.1, Double.NaN });
    }
    testQuery("SELECT\n" + "APPROX_QUANTILE_DS(CAST(dim1 as DOUBLE), 0.01),\n" + "APPROX_QUANTILE_DS(CAST(dim1 as DOUBLE), 0.5, 64),\n" + "APPROX_QUANTILE_DS(CAST(dim1 as DOUBLE), 0.98, 256),\n" + "APPROX_QUANTILE_DS(CAST(dim1 as DOUBLE), 0.99),\n" + "APPROX_QUANTILE_DS(CAST(dim1 as DOUBLE) * 2, 0.97),\n" + "APPROX_QUANTILE_DS(CAST(dim1 as DOUBLE), 0.99) FILTER(WHERE dim2 = 'abc'),\n" + "APPROX_QUANTILE_DS(CAST(dim1 as DOUBLE), 0.999) FILTER(WHERE dim2 <> 'abc'),\n" + "APPROX_QUANTILE_DS(CAST(dim1 as DOUBLE), 0.999) FILTER(WHERE dim2 = 'abc')\n" + "FROM foo", ImmutableList.of(Druids.newTimeseriesQueryBuilder().dataSource(CalciteTests.DATASOURCE1).intervals(new MultipleIntervalSegmentSpec(ImmutableList.of(Filtration.eternity()))).granularity(Granularities.ALL).virtualColumns(new ExpressionVirtualColumn("v0", "CAST(\"dim1\", 'DOUBLE')", ColumnType.FLOAT, TestExprMacroTable.INSTANCE), new ExpressionVirtualColumn("v1", "(CAST(\"dim1\", 'DOUBLE') * 2)", ColumnType.FLOAT, TestExprMacroTable.INSTANCE)).aggregators(ImmutableList.of(new DoublesSketchAggregatorFactory("a0:agg", "v0", 128), new DoublesSketchAggregatorFactory("a1:agg", "v0", 64), new DoublesSketchAggregatorFactory("a2:agg", "v0", 256), new DoublesSketchAggregatorFactory("a4:agg", "v1", 128), new FilteredAggregatorFactory(new DoublesSketchAggregatorFactory("a5:agg", "v0", 128), new SelectorDimFilter("dim2", "abc", null)), new FilteredAggregatorFactory(new DoublesSketchAggregatorFactory("a6:agg", "v0", 128), new NotDimFilter(new SelectorDimFilter("dim2", "abc", null))))).postAggregators(new DoublesSketchToQuantilePostAggregator("a0", makeFieldAccessPostAgg("a0:agg"), 0.01f), new DoublesSketchToQuantilePostAggregator("a1", makeFieldAccessPostAgg("a1:agg"), 0.50f), new DoublesSketchToQuantilePostAggregator("a2", makeFieldAccessPostAgg("a2:agg"), 0.98f), new DoublesSketchToQuantilePostAggregator("a3", makeFieldAccessPostAgg("a0:agg"), 0.99f), new DoublesSketchToQuantilePostAggregator("a4", makeFieldAccessPostAgg("a4:agg"), 0.97f), new DoublesSketchToQuantilePostAggregator("a5", makeFieldAccessPostAgg("a5:agg"), 0.99f), new DoublesSketchToQuantilePostAggregator("a6", makeFieldAccessPostAgg("a6:agg"), 0.999f), new DoublesSketchToQuantilePostAggregator("a7", makeFieldAccessPostAgg("a5:agg"), 0.999f)).context(QUERY_CONTEXT_DEFAULT).build()), expectedResults);
}

Example 27 with FilteredAggregatorFactory

use of org.apache.druid.query.aggregation.FilteredAggregatorFactory in project druid by druid-io.

the class ThetaSketchSqlAggregatorTest method testAvgDailyCountDistinctThetaSketch.

@Test
public void testAvgDailyCountDistinctThetaSketch() throws Exception {
    // Can't vectorize due to outer query (it operates on an inlined data source, which cannot be vectorized).
    cannotVectorize();
    final List<Object[]> expectedResults = ImmutableList.of(new Object[] { 1L });
    testQuery("SELECT\n" + "  AVG(u)\n" + "FROM (SELECT FLOOR(__time TO DAY), APPROX_COUNT_DISTINCT_DS_THETA(cnt) AS u FROM druid.foo GROUP BY 1)", ImmutableList.of(GroupByQuery.builder().setDataSource(new QueryDataSource(Druids.newTimeseriesQueryBuilder().dataSource(CalciteTests.DATASOURCE1).intervals(new MultipleIntervalSegmentSpec(ImmutableList.of(Filtration.eternity()))).granularity(new PeriodGranularity(Period.days(1), null, DateTimeZone.UTC)).aggregators(Collections.singletonList(new SketchMergeAggregatorFactory("a0:a", "cnt", null, null, null, null))).postAggregators(ImmutableList.of(new FinalizingFieldAccessPostAggregator("a0", "a0:a"))).context(TIMESERIES_CONTEXT_BY_GRAN).build().withOverriddenContext(BaseCalciteQueryTest.getTimeseriesContextWithFloorTime(TIMESERIES_CONTEXT_BY_GRAN, "d0")))).setInterval(new MultipleIntervalSegmentSpec(ImmutableList.of(Filtration.eternity()))).setGranularity(Granularities.ALL).setAggregatorSpecs(NullHandling.replaceWithDefault() ? Arrays.asList(new LongSumAggregatorFactory("_a0:sum", "a0"), new CountAggregatorFactory("_a0:count")) : Arrays.asList(new LongSumAggregatorFactory("_a0:sum", "a0"), new FilteredAggregatorFactory(new CountAggregatorFactory("_a0:count"), BaseCalciteQueryTest.not(BaseCalciteQueryTest.selector("a0", null, null))))).setPostAggregatorSpecs(ImmutableList.of(new ArithmeticPostAggregator("_a0", "quotient", ImmutableList.of(new FieldAccessPostAggregator(null, "_a0:sum"), new FieldAccessPostAggregator(null, "_a0:count"))))).setContext(QUERY_CONTEXT_DEFAULT).build()), expectedResults);
}

Example 28 with FilteredAggregatorFactory

use of org.apache.druid.query.aggregation.FilteredAggregatorFactory in project druid by druid-io.

the class ThetaSketchSqlAggregatorTest method testApproxCountDistinctThetaSketch.

@Test
public void testApproxCountDistinctThetaSketch() throws Exception {
    // Cannot vectorize due to SUBSTRING.
    cannotVectorize();
    final String sql = "SELECT\n" + "  SUM(cnt),\n" + "  APPROX_COUNT_DISTINCT_DS_THETA(dim2),\n" + // uppercase
    "  APPROX_COUNT_DISTINCT_DS_THETA(dim2) FILTER(WHERE dim2 <> ''),\n" + // lowercase; also, filtered
    "  APPROX_COUNT_DISTINCT(SUBSTRING(dim2, 1, 1)),\n" + // on extractionFn, using A.C.D.
    "  COUNT(DISTINCT SUBSTRING(dim2, 1, 1) || 'x'),\n" + // on expression, using COUNT DISTINCT
    "  APPROX_COUNT_DISTINCT_DS_THETA(thetasketch_dim1, 32768),\n" + // on native theta sketch column
    "  APPROX_COUNT_DISTINCT_DS_THETA(thetasketch_dim1)\n" + // on native theta sketch column
    "FROM druid.foo";
    final List<Object[]> expectedResults;
    if (NullHandling.replaceWithDefault()) {
        expectedResults = ImmutableList.of(new Object[] { 6L, 2L, 2L, 1L, 2L, 5L, 5L });
    } else {
        expectedResults = ImmutableList.of(new Object[] { 6L, 2L, 2L, 1L, 1L, 5L, 5L });
    }
    testQuery(sql, ImmutableList.of(Druids.newTimeseriesQueryBuilder().dataSource(CalciteTests.DATASOURCE1).intervals(new MultipleIntervalSegmentSpec(ImmutableList.of(Filtration.eternity()))).granularity(Granularities.ALL).virtualColumns(new ExpressionVirtualColumn("v0", "substring(\"dim2\", 0, 1)", ColumnType.STRING, TestExprMacroTable.INSTANCE), new ExpressionVirtualColumn("v1", "concat(substring(\"dim2\", 0, 1),'x')", ColumnType.STRING, TestExprMacroTable.INSTANCE)).aggregators(ImmutableList.of(new LongSumAggregatorFactory("a0", "cnt"), new SketchMergeAggregatorFactory("a1", "dim2", null, null, null, null), new FilteredAggregatorFactory(new SketchMergeAggregatorFactory("a2", "dim2", null, null, null, null), BaseCalciteQueryTest.not(BaseCalciteQueryTest.selector("dim2", "", null))), new SketchMergeAggregatorFactory("a3", "v0", null, null, null, null), new SketchMergeAggregatorFactory("a4", "v1", null, null, null, null), new SketchMergeAggregatorFactory("a5", "thetasketch_dim1", 32768, null, null, null), new SketchMergeAggregatorFactory("a6", "thetasketch_dim1", null, null, null, null))).context(QUERY_CONTEXT_DEFAULT).build()), expectedResults);
}

Example 29 with FilteredAggregatorFactory

use of org.apache.druid.query.aggregation.FilteredAggregatorFactory in project druid by druid-io.

the class FixedBucketsHistogramQuantileSqlAggregatorTest method testQuantileOnFloatAndLongs.

@Test
public void testQuantileOnFloatAndLongs() throws Exception {
    final List<Object[]> expectedResults = ImmutableList.of(new Object[] { 1.0299999713897705, 3.5, 6.440000057220459, 6.470000267028809, 12.40999984741211, 6.494999885559082, 5.497499942779541, 6.499499797821045, 1.25 });
    testQuery("SELECT\n" + "APPROX_QUANTILE_FIXED_BUCKETS(m1, 0.01, 20, 0.0, 10.0),\n" + "APPROX_QUANTILE_FIXED_BUCKETS(m1, 0.5, 20, 0.0, 10.0),\n" + "APPROX_QUANTILE_FIXED_BUCKETS(m1, 0.98, 20, 0.0, 10.0),\n" + "APPROX_QUANTILE_FIXED_BUCKETS(m1, 0.99, 20, 0.0, 10.0),\n" + "APPROX_QUANTILE_FIXED_BUCKETS(m1 * 2, 0.97, 40, 0.0, 20.0),\n" + "APPROX_QUANTILE_FIXED_BUCKETS(m1, 0.99, 20, 0.0, 10.0) FILTER(WHERE dim1 = 'abc'),\n" + "APPROX_QUANTILE_FIXED_BUCKETS(m1, 0.999, 20, 0.0, 10.0) FILTER(WHERE dim1 <> 'abc'),\n" + "APPROX_QUANTILE_FIXED_BUCKETS(m1, 0.999, 20, 0.0, 10.0) FILTER(WHERE dim1 = 'abc'),\n" + "APPROX_QUANTILE_FIXED_BUCKETS(cnt, 0.5, 20, 0.0, 10.0)\n" + "FROM foo", ImmutableList.of(Druids.newTimeseriesQueryBuilder().dataSource(CalciteTests.DATASOURCE1).intervals(new MultipleIntervalSegmentSpec(ImmutableList.of(Filtration.eternity()))).granularity(Granularities.ALL).virtualColumns(new ExpressionVirtualColumn("v0", "(\"m1\" * 2)", ColumnType.FLOAT, TestExprMacroTable.INSTANCE)).aggregators(ImmutableList.of(new FixedBucketsHistogramAggregatorFactory("a0:agg", "m1", 20, 0.0d, 10.0d, FixedBucketsHistogram.OutlierHandlingMode.IGNORE, false), new FixedBucketsHistogramAggregatorFactory("a4:agg", "v0", 40, 0.0d, 20.0d, FixedBucketsHistogram.OutlierHandlingMode.IGNORE, false), new FilteredAggregatorFactory(new FixedBucketsHistogramAggregatorFactory("a5:agg", "m1", 20, 0.0d, 10.0d, FixedBucketsHistogram.OutlierHandlingMode.IGNORE, false), new SelectorDimFilter("dim1", "abc", null)), new FilteredAggregatorFactory(new FixedBucketsHistogramAggregatorFactory("a6:agg", "m1", 20, 0.0d, 10.0d, FixedBucketsHistogram.OutlierHandlingMode.IGNORE, false), new NotDimFilter(new SelectorDimFilter("dim1", "abc", null))), new FixedBucketsHistogramAggregatorFactory("a8:agg", "cnt", 20, 0.0d, 10.0d, FixedBucketsHistogram.OutlierHandlingMode.IGNORE, false))).postAggregators(new QuantilePostAggregator("a0", "a0:agg", 0.01f), new QuantilePostAggregator("a1", "a0:agg", 0.50f), new QuantilePostAggregator("a2", "a0:agg", 0.98f), new QuantilePostAggregator("a3", "a0:agg", 0.99f), new QuantilePostAggregator("a4", "a4:agg", 0.97f), new QuantilePostAggregator("a5", "a5:agg", 0.99f), new QuantilePostAggregator("a6", "a6:agg", 0.999f), new QuantilePostAggregator("a7", "a5:agg", 0.999f), new QuantilePostAggregator("a8", "a8:agg", 0.50f)).context(QUERY_CONTEXT_DEFAULT).build()), expectedResults);
}

Example 30 with FilteredAggregatorFactory

use of org.apache.druid.query.aggregation.FilteredAggregatorFactory in project druid by druid-io.

the class FixedBucketsHistogramQuantileSqlAggregatorTest method testQuantileOnComplexColumn.

@Test
public void testQuantileOnComplexColumn() throws Exception {
    cannotVectorize();
    final List<Object[]> expectedResults = ImmutableList.of(new Object[] { 1.0299999713897705, 3.5, 6.293333530426025, 6.470000267028809, 6.494999885559082, 5.497499942779541, 6.499499797821045 });
    testQuery("SELECT\n" + "APPROX_QUANTILE_FIXED_BUCKETS(fbhist_m1, 0.01, 20, 0.0, 10.0),\n" + "APPROX_QUANTILE_FIXED_BUCKETS(fbhist_m1, 0.5, 20, 0.0, 10.0),\n" + "APPROX_QUANTILE_FIXED_BUCKETS(fbhist_m1, 0.98, 30, 0.0, 10.0),\n" + "APPROX_QUANTILE_FIXED_BUCKETS(fbhist_m1, 0.99, 20, 0.0, 10.0),\n" + "APPROX_QUANTILE_FIXED_BUCKETS(fbhist_m1, 0.99, 20, 0.0, 10.0) FILTER(WHERE dim1 = 'abc'),\n" + "APPROX_QUANTILE_FIXED_BUCKETS(fbhist_m1, 0.999, 20, 0.0, 10.0) FILTER(WHERE dim1 <> 'abc'),\n" + "APPROX_QUANTILE_FIXED_BUCKETS(fbhist_m1, 0.999, 20, 0.0, 10.0) FILTER(WHERE dim1 = 'abc')\n" + "FROM foo", ImmutableList.of(Druids.newTimeseriesQueryBuilder().dataSource(CalciteTests.DATASOURCE1).intervals(new MultipleIntervalSegmentSpec(ImmutableList.of(Filtration.eternity()))).granularity(Granularities.ALL).aggregators(ImmutableList.of(new FixedBucketsHistogramAggregatorFactory("a0:agg", "fbhist_m1", 20, 0.0, 10.0, FixedBucketsHistogram.OutlierHandlingMode.IGNORE, false), new FixedBucketsHistogramAggregatorFactory("a2:agg", "fbhist_m1", 30, 0.0, 10.0, FixedBucketsHistogram.OutlierHandlingMode.IGNORE, false), new FilteredAggregatorFactory(new FixedBucketsHistogramAggregatorFactory("a4:agg", "fbhist_m1", 20, 0.0, 10.0, FixedBucketsHistogram.OutlierHandlingMode.IGNORE, false), new SelectorDimFilter("dim1", "abc", null)), new FilteredAggregatorFactory(new FixedBucketsHistogramAggregatorFactory("a5:agg", "fbhist_m1", 20, 0.0, 10.0, FixedBucketsHistogram.OutlierHandlingMode.IGNORE, false), new NotDimFilter(new SelectorDimFilter("dim1", "abc", null))))).postAggregators(new QuantilePostAggregator("a0", "a0:agg", 0.01f), new QuantilePostAggregator("a1", "a0:agg", 0.50f), new QuantilePostAggregator("a2", "a2:agg", 0.98f), new QuantilePostAggregator("a3", "a0:agg", 0.99f), new QuantilePostAggregator("a4", "a4:agg", 0.99f), new QuantilePostAggregator("a5", "a5:agg", 0.999f), new QuantilePostAggregator("a6", "a4:agg", 0.999f)).context(QUERY_CONTEXT_DEFAULT).build()), expectedResults);
}