use of io.druid.query.lookup.LookupExtractionFn in project druid by druid-io.
the class GroupByQueryRunnerTest method testGroupByWithSimpleRenameRetainMissingNonInjective.
@Test
public void testGroupByWithSimpleRenameRetainMissingNonInjective() {
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 = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setQuerySegmentSpec(QueryRunnerTestHelper.firstToThird).setDimensions(Lists.<DimensionSpec>newArrayList(new ExtractionDimensionSpec("quality", "alias", new LookupExtractionFn(new MapLookupExtractor(map, false), true, null, false, false)))).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"))).setGranularity(QueryRunnerTestHelper.dayGran).build();
List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "automotive0", "rows", 1L, "idx", 135L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "business0", "rows", 1L, "idx", 118L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "entertainment0", "rows", 1L, "idx", 158L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "health0", "rows", 1L, "idx", 120L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "mezzanine0", "rows", 3L, "idx", 2870L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "news0", "rows", 1L, "idx", 121L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "premium0", "rows", 3L, "idx", 2900L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "technology0", "rows", 1L, "idx", 78L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "travel0", "rows", 1L, "idx", 119L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "automotive0", "rows", 1L, "idx", 147L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "business0", "rows", 1L, "idx", 112L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "entertainment0", "rows", 1L, "idx", 166L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "health0", "rows", 1L, "idx", 113L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "mezzanine0", "rows", 3L, "idx", 2447L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "news0", "rows", 1L, "idx", 114L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "premium0", "rows", 3L, "idx", 2505L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "technology0", "rows", 1L, "idx", 97L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "travel0", "rows", 1L, "idx", 126L));
Iterable<Row> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
TestHelper.assertExpectedObjects(expectedResults, results, "");
}
use of io.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 = new LookupExtractionFn(mapLookupExtractor, false, null, true, true);
GroupByQuery query = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setQuerySegmentSpec(QueryRunnerTestHelper.firstToThird).setDimensions(Lists.<DimensionSpec>newArrayList(new DefaultDimensionSpec("null_column", "alias"))).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"))).setGranularity(QueryRunnerTestHelper.dayGran).setDimFilter(new ExtractionDimFilter("null_column", "EMPTY", lookupExtractionFn, null)).build();
List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", null, "rows", 13L, "idx", 6619L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", null, "rows", 13L, "idx", 5827L));
Iterable<Row> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
TestHelper.assertExpectedObjects(expectedResults, results, "");
}
use of io.druid.query.lookup.LookupExtractionFn in project druid by druid-io.
the class GroupByQueryRunnerTest method testGroupByWithSimpleRenameRetainMissing.
@Test
public void testGroupByWithSimpleRenameRetainMissing() {
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 = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setQuerySegmentSpec(QueryRunnerTestHelper.firstToThird).setDimensions(Lists.<DimensionSpec>newArrayList(new ExtractionDimensionSpec("quality", "alias", new LookupExtractionFn(new MapLookupExtractor(map, false), true, null, true, false)))).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"))).setGranularity(QueryRunnerTestHelper.dayGran).build();
List<Row> expectedResults = Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "automotive0", "rows", 1L, "idx", 135L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "business0", "rows", 1L, "idx", 118L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "entertainment0", "rows", 1L, "idx", 158L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "health0", "rows", 1L, "idx", 120L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "mezzanine0", "rows", 3L, "idx", 2870L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "news0", "rows", 1L, "idx", 121L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "premium0", "rows", 3L, "idx", 2900L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "technology0", "rows", 1L, "idx", 78L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "travel0", "rows", 1L, "idx", 119L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "automotive0", "rows", 1L, "idx", 147L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "business0", "rows", 1L, "idx", 112L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "entertainment0", "rows", 1L, "idx", 166L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "health0", "rows", 1L, "idx", 113L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "mezzanine0", "rows", 3L, "idx", 2447L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "news0", "rows", 1L, "idx", 114L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "premium0", "rows", 3L, "idx", 2505L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "technology0", "rows", 1L, "idx", 97L), GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-02", "alias", "travel0", "rows", 1L, "idx", 126L));
Iterable<Row> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
TestHelper.assertExpectedObjects(expectedResults, results, "");
}
use of io.druid.query.lookup.LookupExtractionFn in project druid by druid-io.
the class GroupByQueryRunnerTest method testBySegmentResultsUnOptimizedDimextraction.
@Test
public void testBySegmentResultsUnOptimizedDimextraction() {
int segmentCount = 32;
Result<BySegmentResultValue> singleSegmentResult = new Result<BySegmentResultValue>(new DateTime("2011-01-12T00:00:00.000Z"), new BySegmentResultValueClass(Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "mezzanine0", "rows", 6L, "idx", 4420L)), "testSegment", new Interval("2011-04-02T00:00:00.000Z/2011-04-04T00:00:00.000Z")));
List<Result> bySegmentResults = Lists.newArrayList();
for (int i = 0; i < segmentCount; i++) {
bySegmentResults.add(singleSegmentResult);
}
GroupByQuery.Builder builder = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setInterval("2011-04-02/2011-04-04").setDimensions(Lists.<DimensionSpec>newArrayList(new ExtractionDimensionSpec("quality", "alias", new LookupExtractionFn(new MapLookupExtractor(ImmutableMap.of("mezzanine", "mezzanine0"), false), false, null, false, false)))).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"))).setGranularity(new PeriodGranularity(new Period("P1M"), null, null)).setDimFilter(new SelectorDimFilter("quality", "mezzanine", null)).setContext(ImmutableMap.<String, Object>of("bySegment", true));
final GroupByQuery fullQuery = builder.build();
QueryToolChest toolChest = factory.getToolchest();
List<QueryRunner<Row>> singleSegmentRunners = Lists.newArrayList();
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(fullQuery, Maps.newHashMap()), "");
exec.shutdownNow();
}
use of io.druid.query.lookup.LookupExtractionFn in project druid by druid-io.
the class GroupByQueryRunnerTest method testBySegmentResultsOptimizedDimextraction.
@Test
public void testBySegmentResultsOptimizedDimextraction() {
int segmentCount = 32;
Result<BySegmentResultValue> singleSegmentResult = new Result<BySegmentResultValue>(new DateTime("2011-01-12T00:00:00.000Z"), new BySegmentResultValueClass(Arrays.asList(GroupByQueryRunnerTestHelper.createExpectedRow("2011-04-01", "alias", "mezzanine0", "rows", 6L, "idx", 4420L)), "testSegment", new Interval("2011-04-02T00:00:00.000Z/2011-04-04T00:00:00.000Z")));
List<Result> bySegmentResults = Lists.newArrayList();
for (int i = 0; i < segmentCount; i++) {
bySegmentResults.add(singleSegmentResult);
}
GroupByQuery.Builder builder = GroupByQuery.builder().setDataSource(QueryRunnerTestHelper.dataSource).setInterval("2011-04-02/2011-04-04").setDimensions(Lists.<DimensionSpec>newArrayList(new ExtractionDimensionSpec("quality", "alias", new LookupExtractionFn(new MapLookupExtractor(ImmutableMap.of("mezzanine", "mezzanine0"), false), false, null, true, false)))).setAggregatorSpecs(Arrays.asList(QueryRunnerTestHelper.rowsCount, new LongSumAggregatorFactory("idx", "index"))).setGranularity(new PeriodGranularity(new Period("P1M"), null, null)).setDimFilter(new SelectorDimFilter("quality", "mezzanine", null)).setContext(ImmutableMap.<String, Object>of("bySegment", true));
final GroupByQuery fullQuery = builder.build();
QueryToolChest toolChest = factory.getToolchest();
List<QueryRunner<Row>> singleSegmentRunners = Lists.newArrayList();
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(fullQuery, Maps.newHashMap()), "");
exec.shutdownNow();
}
Aggregations