Examples with ExtractionDimensionSpec org.apache.druid.query.dimension.ExtractionDimensionSpec used on opensource projects

Example 26 with ExtractionDimensionSpec

use of org.apache.druid.query.dimension.ExtractionDimensionSpec in project druid by druid-io.

the class SearchQueryRunnerTest method testSearchOnFloatColumnWithExFn.

@Test
public void testSearchOnFloatColumnWithExFn() {
    String jsFn = "function(str) { return 'super-' + str; }";
    ExtractionFn jsExtractionFn = new JavaScriptExtractionFn(jsFn, false, JavaScriptConfig.getEnabledInstance());
    SearchQuery searchQuery = Druids.newSearchQueryBuilder().dimensions(new ExtractionDimensionSpec(QueryRunnerTestHelper.INDEX_METRIC, QueryRunnerTestHelper.INDEX_METRIC, jsExtractionFn)).dataSource(QueryRunnerTestHelper.DATA_SOURCE).granularity(QueryRunnerTestHelper.ALL_GRAN).intervals(QueryRunnerTestHelper.FULL_ON_INTERVAL_SPEC).query("100.7").build();
    List<SearchHit> expectedHits = new ArrayList<>();
    expectedHits.add(new SearchHit(QueryRunnerTestHelper.INDEX_METRIC, "super-100.706057", 1));
    expectedHits.add(new SearchHit(QueryRunnerTestHelper.INDEX_METRIC, "super-100.775597", 1));
    checkSearchQuery(searchQuery, expectedHits);
}

Example 27 with ExtractionDimensionSpec

use of org.apache.druid.query.dimension.ExtractionDimensionSpec in project druid by druid-io.

the class SearchQueryRunnerTest method testSearchWithExtractionFilter1.

@Test
public void testSearchWithExtractionFilter1() {
    final String automotiveSnowman = "automotive☃";
    List<SearchHit> expectedHits = new ArrayList<>();
    expectedHits.add(new SearchHit(QueryRunnerTestHelper.QUALITY_DIMENSION, automotiveSnowman, 93));
    final LookupExtractionFn lookupExtractionFn = new LookupExtractionFn(new MapLookupExtractor(ImmutableMap.of("automotive", automotiveSnowman), false), true, null, true, true);
    SearchQuery query = Druids.newSearchQueryBuilder().dataSource(QueryRunnerTestHelper.DATA_SOURCE).granularity(QueryRunnerTestHelper.ALL_GRAN).filters(new ExtractionDimFilter(QueryRunnerTestHelper.QUALITY_DIMENSION, automotiveSnowman, lookupExtractionFn, null)).intervals(QueryRunnerTestHelper.FULL_ON_INTERVAL_SPEC).dimensions(new ExtractionDimensionSpec(QueryRunnerTestHelper.QUALITY_DIMENSION, null, lookupExtractionFn)).query("☃").build();
    checkSearchQuery(query, expectedHits);
}

Example 28 with ExtractionDimensionSpec

use of org.apache.druid.query.dimension.ExtractionDimensionSpec 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 29 with ExtractionDimensionSpec

use of org.apache.druid.query.dimension.ExtractionDimensionSpec 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 30 with ExtractionDimensionSpec

use of org.apache.druid.query.dimension.ExtractionDimensionSpec in project druid by druid-io.

the class GroupByQueryRunnerTest method testGroupByTimeExtractionWithNulls.

@Test
public void testGroupByTimeExtractionWithNulls() {
    // Cannot vectorize due to extraction dimension specs.
    cannotVectorize();
    final DimExtractionFn nullWednesdays = new DimExtractionFn() {

        @Override
        public String apply(String dimValue) {
            if ("Wednesday".equals(dimValue)) {
                return null;
            } else {
                return dimValue;
            }
        }

        @Override
        public byte[] getCacheKey() {
            throw new UnsupportedOperationException();
        }

        @Override
        public boolean preservesOrdering() {
            return false;
        }

        @Override
        public ExtractionType getExtractionType() {
            return ExtractionType.MANY_TO_ONE;
        }
    };
    GroupByQuery query = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FULL_ON_INTERVAL_SPEC).setDimensions(new DefaultDimensionSpec("market", "market"), new ExtractionDimensionSpec(ColumnHolder.TIME_COLUMN_NAME, "dayOfWeek", new CascadeExtractionFn(new ExtractionFn[] { new TimeFormatExtractionFn("EEEE", null, null, null, false), nullWednesdays }))).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, QueryRunnerTestHelper.INDEX_DOUBLE_SUM).setPostAggregatorSpecs(Collections.singletonList(QueryRunnerTestHelper.ADD_ROWS_INDEX_CONSTANT)).setGranularity(QueryRunnerTestHelper.ALL_GRAN).setDimFilter(new OrDimFilter(Arrays.asList(new SelectorDimFilter("market", "spot", null), new SelectorDimFilter("market", "upfront", null)))).build();
    List<ResultRow> expectedResults = Arrays.asList(makeRow(query, "1970-01-01", "dayOfWeek", null, "market", "spot", "index", 14271.368591308594, "rows", 126L, "addRowsIndexConstant", 14398.368591308594), makeRow(query, "1970-01-01", "dayOfWeek", "Friday", "market", "spot", "index", 13219.574157714844, "rows", 117L, "addRowsIndexConstant", 13337.574157714844), makeRow(query, "1970-01-01", "dayOfWeek", "Monday", "market", "spot", "index", 13557.738830566406, "rows", 117L, "addRowsIndexConstant", 13675.738830566406), makeRow(query, "1970-01-01", "dayOfWeek", "Saturday", "market", "spot", "index", 13493.751281738281, "rows", 117L, "addRowsIndexConstant", 13611.751281738281), makeRow(query, "1970-01-01", "dayOfWeek", "Sunday", "market", "spot", "index", 13585.541015625, "rows", 117L, "addRowsIndexConstant", 13703.541015625), makeRow(query, "1970-01-01", "dayOfWeek", "Thursday", "market", "spot", "index", 14279.127197265625, "rows", 126L, "addRowsIndexConstant", 14406.127197265625), makeRow(query, "1970-01-01", "dayOfWeek", "Tuesday", "market", "spot", "index", 13199.471435546875, "rows", 117L, "addRowsIndexConstant", 13317.471435546875), makeRow(query, "1970-01-01", "dayOfWeek", null, "market", "upfront", "index", 28985.5751953125, "rows", 28L, "addRowsIndexConstant", 29014.5751953125), makeRow(query, "1970-01-01", "dayOfWeek", "Friday", "market", "upfront", "index", 27297.8623046875, "rows", 26L, "addRowsIndexConstant", 27324.8623046875), makeRow(query, "1970-01-01", "dayOfWeek", "Monday", "market", "upfront", "index", 27619.58447265625, "rows", 26L, "addRowsIndexConstant", 27646.58447265625), makeRow(query, "1970-01-01", "dayOfWeek", "Saturday", "market", "upfront", "index", 27820.83154296875, "rows", 26L, "addRowsIndexConstant", 27847.83154296875), makeRow(query, "1970-01-01", "dayOfWeek", "Sunday", "market", "upfront", "index", 24791.223876953125, "rows", 26L, "addRowsIndexConstant", 24818.223876953125), makeRow(query, "1970-01-01", "dayOfWeek", "Thursday", "market", "upfront", "index", 28562.748901367188, "rows", 28L, "addRowsIndexConstant", 28591.748901367188), makeRow(query, "1970-01-01", "dayOfWeek", "Tuesday", "market", "upfront", "index", 26968.280639648438, "rows", 26L, "addRowsIndexConstant", 26995.280639648438));
    Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "time-extraction");
}