Examples with LookupExtractionFn org.apache.druid.query.lookup.LookupExtractionFn used on opensource projects

Example 16 with LookupExtractionFn

use of org.apache.druid.query.lookup.LookupExtractionFn in project druid by druid-io.

the class GroupByQueryRunnerTest method testGroupByWithRebucketRename.

@Test
public void testGroupByWithRebucketRename() {
    // Cannot vectorize due to extraction dimension spec.
    cannotVectorize();
    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, false, 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, "rebucket-rename");
}

Example 17 with LookupExtractionFn

use of org.apache.druid.query.lookup.LookupExtractionFn in project druid by druid-io.

the class GroupByQueryRunnerTest method testGroupByWithExtractionDimFilterCaseMappingValueIsNullOrEmpty.

@Test
public void testGroupByWithExtractionDimFilterCaseMappingValueIsNullOrEmpty() {
    Map<String, String> extractionMap = new HashMap<>();
    extractionMap.put("automotive", "automotive0");
    extractionMap.put("business", "business0");
    extractionMap.put("entertainment", "entertainment0");
    extractionMap.put("health", "health0");
    extractionMap.put("mezzanine", null);
    extractionMap.put("news", "");
    extractionMap.put("premium", "premium0");
    extractionMap.put("technology", "technology0");
    extractionMap.put("travel", "travel0");
    MapLookupExtractor mapLookupExtractor = new MapLookupExtractor(extractionMap, false);
    LookupExtractionFn lookupExtractionFn = new LookupExtractionFn(mapLookupExtractor, false, null, true, false);
    GroupByQuery query = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FIRST_TO_THIRD).setDimensions(new DefaultDimensionSpec("quality", "alias")).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, new LongSumAggregatorFactory("idx", "index")).setGranularity(QueryRunnerTestHelper.DAY_GRAN).setDimFilter(new ExtractionDimFilter("quality", "", lookupExtractionFn, null)).build();
    List<ResultRow> expectedResults;
    if (NullHandling.replaceWithDefault()) {
        expectedResults = Arrays.asList(makeRow(query, "2011-04-01", "alias", "mezzanine", "rows", 3L, "idx", 2870L), makeRow(query, "2011-04-01", "alias", "news", "rows", 1L, "idx", 121L), makeRow(query, "2011-04-02", "alias", "mezzanine", "rows", 3L, "idx", 2447L), makeRow(query, "2011-04-02", "alias", "news", "rows", 1L, "idx", 114L));
    } else {
        // Only empty string should match, nulls will not match
        expectedResults = Arrays.asList(makeRow(query, "2011-04-01", "alias", "news", "rows", 1L, "idx", 121L), makeRow(query, "2011-04-02", "alias", "news", "rows", 1L, "idx", 114L));
    }
    Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "dim-extraction");
}

Example 18 with LookupExtractionFn

use of org.apache.druid.query.lookup.LookupExtractionFn in project druid by druid-io.

the class GroupByQueryRunnerTest method testBySegmentResultsOptimizedDimextraction.

@Test
public void testBySegmentResultsOptimizedDimextraction() {
    GroupByQuery.Builder builder = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setInterval("2011-04-02/2011-04-04").setDimensions(new ExtractionDimensionSpec("quality", "alias", new LookupExtractionFn(new MapLookupExtractor(ImmutableMap.of("mezzanine", "mezzanine0"), false), false, null, true, false))).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, new LongSumAggregatorFactory("idx", "index")).setGranularity(new PeriodGranularity(new Period("P1M"), null, null)).setDimFilter(new SelectorDimFilter("quality", "mezzanine", null)).overrideContext(ImmutableMap.of(QueryContexts.BY_SEGMENT_KEY, true));
    final GroupByQuery fullQuery = builder.build();
    int segmentCount = 32;
    Result<BySegmentResultValue<ResultRow>> singleSegmentResult = new Result<>(DateTimes.of("2011-01-12T00:00:00.000Z"), new BySegmentResultValueClass<>(Collections.singletonList(makeRow(fullQuery, "2011-04-01", "alias", "mezzanine0", "rows", 6L, "idx", 4420L)), QueryRunnerTestHelper.SEGMENT_ID.toString(), Intervals.of("2011-04-02T00:00:00.000Z/2011-04-04T00:00:00.000Z")));
    List<Result> bySegmentResults = new ArrayList<>();
    for (int i = 0; i < segmentCount; i++) {
        bySegmentResults.add(singleSegmentResult);
    }
    QueryToolChest toolChest = factory.getToolchest();
    List<QueryRunner<ResultRow>> singleSegmentRunners = new ArrayList<>();
    for (int i = 0; i < segmentCount; i++) {
        singleSegmentRunners.add(toolChest.preMergeQueryDecoration(runner));
    }
    ExecutorService exec = Executors.newCachedThreadPool();
    QueryRunner theRunner = toolChest.postMergeQueryDecoration(new FinalizeResultsQueryRunner<>(toolChest.mergeResults(factory.mergeRunners(Executors.newCachedThreadPool(), singleSegmentRunners)), toolChest));
    TestHelper.assertExpectedObjects(bySegmentResults, theRunner.run(QueryPlus.wrap(fullQuery)), "bySegment-dim-extraction");
    exec.shutdownNow();
}

Example 19 with LookupExtractionFn

use of org.apache.druid.query.lookup.LookupExtractionFn in project druid by druid-io.

the class GroupByQueryRunnerTest method testGroupByWithAllFiltersOnNullDimsWithExtractionFns.

@Test
public void testGroupByWithAllFiltersOnNullDimsWithExtractionFns() {
    Map<String, String> extractionMap = new HashMap<>();
    extractionMap.put("", "EMPTY");
    extractionMap.put(null, "EMPTY");
    MapLookupExtractor mapLookupExtractor = new MapLookupExtractor(extractionMap, false);
    LookupExtractionFn extractionFn = new LookupExtractionFn(mapLookupExtractor, false, "EMPTY", true, true);
    String jsFn = "function(x) { return(x === 'EMPTY') }";
    List<DimFilter> superFilterList = new ArrayList<>();
    superFilterList.add(new SelectorDimFilter("null_column", "EMPTY", extractionFn));
    superFilterList.add(new InDimFilter("null_column", Arrays.asList("NOT-EMPTY", "FOOBAR", "EMPTY"), extractionFn));
    superFilterList.add(new BoundDimFilter("null_column", "EMPTY", "EMPTY", false, false, true, extractionFn, StringComparators.ALPHANUMERIC));
    superFilterList.add(new RegexDimFilter("null_column", "EMPTY", extractionFn));
    superFilterList.add(new SearchQueryDimFilter("null_column", new ContainsSearchQuerySpec("EMPTY", true), extractionFn));
    superFilterList.add(new JavaScriptDimFilter("null_column", jsFn, extractionFn, JavaScriptConfig.getEnabledInstance()));
    DimFilter superFilter = new AndDimFilter(superFilterList);
    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(superFilter).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, "extraction");
}

Example 20 with LookupExtractionFn

use of org.apache.druid.query.lookup.LookupExtractionFn in project druid by druid-io.

the class GroupByQueryRunnerTest method testGroupByWithExtractionDimFilterNullDims.

@Test
public void testGroupByWithExtractionDimFilterNullDims() {
    Map<String, String> extractionMap = new HashMap<>();
    extractionMap.put("", "EMPTY");
    MapLookupExtractor mapLookupExtractor = new MapLookupExtractor(extractionMap, false);
    LookupExtractionFn lookupExtractionFn;
    if (NullHandling.replaceWithDefault()) {
        extractionMap.put("", "EMPTY");
        lookupExtractionFn = new LookupExtractionFn(mapLookupExtractor, false, null, true, true);
    } else {
        extractionMap.put("", "SHOULD_NOT_BE_USED");
        lookupExtractionFn = new LookupExtractionFn(mapLookupExtractor, false, "EMPTY", true, true);
    }
    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", "EMPTY", 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, "extraction-dim-filter");
}