Examples with SubstringDimExtractionFn org.apache.druid.query.extraction.SubstringDimExtractionFn used on opensource projects

Example 16 with SubstringDimExtractionFn

use of org.apache.druid.query.extraction.SubstringDimExtractionFn in project druid by druid-io.

the class LikeDimFilterTest method testEqualsAndHashCode.

@Test
public void testEqualsAndHashCode() {
    final DimFilter filter = new LikeDimFilter("foo", "bar%", "@", new SubstringDimExtractionFn(1, 2));
    final DimFilter filter2 = new LikeDimFilter("foo", "bar%", "@", new SubstringDimExtractionFn(1, 2));
    final DimFilter filter3 = new LikeDimFilter("foo", "bar%", null, new SubstringDimExtractionFn(1, 2));
    Assert.assertEquals(filter, filter2);
    Assert.assertNotEquals(filter, filter3);
    Assert.assertEquals(filter.hashCode(), filter2.hashCode());
    Assert.assertNotEquals(filter.hashCode(), filter3.hashCode());
}

Example 17 with SubstringDimExtractionFn

use of org.apache.druid.query.extraction.SubstringDimExtractionFn in project druid by druid-io.

the class CalciteQueryTest method testExactCountDistinctOfSemiJoinResult.

@Test
public void testExactCountDistinctOfSemiJoinResult() throws Exception {
    // Cannot vectorize due to extraction dimension spec.
    cannotVectorize();
    testQuery("SELECT COUNT(*)\n" + "FROM (\n" + "  SELECT DISTINCT dim2\n" + "  FROM druid.foo\n" + "  WHERE SUBSTRING(dim2, 1, 1) IN (\n" + "    SELECT SUBSTRING(dim1, 1, 1) FROM druid.foo WHERE dim1 <> ''\n" + "  ) AND __time >= '2000-01-01' AND __time < '2002-01-01'\n" + ")", ImmutableList.of(GroupByQuery.builder().setDataSource(new QueryDataSource(GroupByQuery.builder().setDataSource(join(new TableDataSource(CalciteTests.DATASOURCE1), new QueryDataSource(GroupByQuery.builder().setDataSource(CalciteTests.DATASOURCE1).setInterval(querySegmentSpec(Filtration.eternity())).setGranularity(Granularities.ALL).setDimFilter(not(selector("dim1", "", null))).setDimensions(dimensions(new ExtractionDimensionSpec("dim1", "d0", new SubstringDimExtractionFn(0, 1)))).setContext(QUERY_CONTEXT_DEFAULT).build()), "j0.", equalsCondition(makeExpression("substring(\"dim2\", 0, 1)"), DruidExpression.ofColumn(ColumnType.STRING, "j0.d0")), JoinType.INNER)).setInterval(querySegmentSpec(Intervals.of("2000-01-01/2002-01-01"))).setGranularity(Granularities.ALL).setDimensions(dimensions(new DefaultDimensionSpec("dim2", "d0"))).setContext(QUERY_CONTEXT_DEFAULT).build())).setInterval(querySegmentSpec(Filtration.eternity())).setGranularity(Granularities.ALL).setAggregatorSpecs(aggregators(new CountAggregatorFactory("a0"))).setContext(QUERY_CONTEXT_DEFAULT).build()), ImmutableList.of(new Object[] { 2L }));
}

Example 18 with SubstringDimExtractionFn

use of org.apache.druid.query.extraction.SubstringDimExtractionFn in project druid by druid-io.

the class SubstringOperatorConversion method toDruidExpression.

@Override
public DruidExpression toDruidExpression(final PlannerContext plannerContext, final RowSignature rowSignature, final RexNode rexNode) {
    // Can't simply pass-through operands, since SQL standard args don't match what Druid's expression language wants.
    // SQL is 1-indexed, Druid is 0-indexed.
    final RexCall call = (RexCall) rexNode;
    final DruidExpression input = Expressions.toDruidExpression(plannerContext, rowSignature, call.getOperands().get(0));
    if (input == null) {
        return null;
    }
    final int index = RexLiteral.intValue(call.getOperands().get(1)) - 1;
    final int length;
    if (call.getOperands().size() > 2) {
        length = RexLiteral.intValue(call.getOperands().get(2));
    } else {
        length = -1;
    }
    return input.map(simpleExtraction -> simpleExtraction.cascade(new SubstringDimExtractionFn(index, length < 0 ? null : length)), expression -> StringUtils.format("substring(%s, %s, %s)", expression, DruidExpression.numberLiteral(index), DruidExpression.numberLiteral(length)));
}

Example 19 with SubstringDimExtractionFn

use of org.apache.druid.query.extraction.SubstringDimExtractionFn in project druid by druid-io.

the class CalciteJoinQueryTest method testTopNFilterJoinWithProjection.

@Test
@Parameters(source = QueryContextForJoinProvider.class)
public void testTopNFilterJoinWithProjection(Map<String, Object> queryContext) throws Exception {
    // Cannot vectorize JOIN operator.
    cannotVectorize();
    // Filters on top N values of some dimension by using an inner join. Also projects the outer dimension.
    testQuery("SELECT SUBSTRING(t1.dim1, 1, 10), SUM(t1.cnt)\n" + "FROM druid.foo t1\n" + "  INNER JOIN (\n" + "  SELECT\n" + "    SUM(cnt) AS sum_cnt,\n" + "    dim2\n" + "  FROM druid.foo\n" + "  GROUP BY dim2\n" + "  ORDER BY 1 DESC\n" + "  LIMIT 2\n" + ") t2 ON (t1.dim2 = t2.dim2)\n" + "GROUP BY SUBSTRING(t1.dim1, 1, 10)", queryContext, ImmutableList.of(GroupByQuery.builder().setDataSource(join(new TableDataSource(CalciteTests.DATASOURCE1), new QueryDataSource(new TopNQueryBuilder().dataSource(CalciteTests.DATASOURCE1).intervals(querySegmentSpec(Filtration.eternity())).granularity(Granularities.ALL).dimension(new DefaultDimensionSpec("dim2", "d0")).aggregators(new LongSumAggregatorFactory("a0", "cnt")).metric("a0").threshold(2).context(QUERY_CONTEXT_DEFAULT).build()), "j0.", equalsCondition(makeColumnExpression("dim2"), makeColumnExpression("j0.d0")), JoinType.INNER)).setInterval(querySegmentSpec(Filtration.eternity())).setGranularity(Granularities.ALL).setDimensions(dimensions(new ExtractionDimensionSpec("dim1", "d0", ColumnType.STRING, new SubstringDimExtractionFn(0, 10)))).setAggregatorSpecs(aggregators(new LongSumAggregatorFactory("a0", "cnt"))).setContext(queryContext).build()), ImmutableList.of(new Object[] { NULL_STRING, 1L }, new Object[] { "1", 1L }));
}

Example 20 with SubstringDimExtractionFn

use of org.apache.druid.query.extraction.SubstringDimExtractionFn in project druid by druid-io.

the class CalciteJoinQueryTest method testUsingSubqueryWithExtractionFns.

@Test
@Parameters(source = QueryContextForJoinProvider.class)
public void testUsingSubqueryWithExtractionFns(Map<String, Object> queryContext) throws Exception {
    // Cannot vectorize JOIN operator.
    cannotVectorize();
    testQuery("SELECT dim2, COUNT(*) FROM druid.foo " + "WHERE substring(dim2, 1, 1) IN (SELECT substring(dim1, 1, 1) FROM druid.foo WHERE dim1 <> '')" + "group by dim2", queryContext, ImmutableList.of(GroupByQuery.builder().setDataSource(join(new TableDataSource(CalciteTests.DATASOURCE1), new QueryDataSource(GroupByQuery.builder().setDataSource(CalciteTests.DATASOURCE1).setInterval(querySegmentSpec(Filtration.eternity())).setGranularity(Granularities.ALL).setDimFilter(not(selector("dim1", "", null))).setDimensions(dimensions(new ExtractionDimensionSpec("dim1", "d0", new SubstringDimExtractionFn(0, 1)))).setContext(QUERY_CONTEXT_DEFAULT).build()), "j0.", equalsCondition(makeExpression("substring(\"dim2\", 0, 1)"), makeColumnExpression("j0.d0")), JoinType.INNER)).setInterval(querySegmentSpec(Filtration.eternity())).setGranularity(Granularities.ALL).setDimensions(dimensions(new DefaultDimensionSpec("dim2", "d0"))).setAggregatorSpecs(aggregators(new CountAggregatorFactory("a0"))).setContext(queryContext).build()), ImmutableList.of(new Object[] { "a", 2L }, new Object[] { "abc", 1L }));
}