Examples with DimFilter org.apache.druid.query.filter.DimFilter used on opensource projects

Example 36 with DimFilter

use of org.apache.druid.query.filter.DimFilter in project druid by druid-io.

the class DataSourceAnalysis method flattenJoin.

/**
 * Flatten a datasource into two parts: the left-hand side datasource (the 'base' datasource), and a list of join
 * clauses, if any.
 *
 * @throws IllegalArgumentException if dataSource cannot be fully flattened.
 */
private static Triple<DataSource, DimFilter, List<PreJoinableClause>> flattenJoin(final JoinDataSource dataSource) {
    DataSource current = dataSource;
    DimFilter currentDimFilter = null;
    final List<PreJoinableClause> preJoinableClauses = new ArrayList<>();
    while (current instanceof JoinDataSource) {
        final JoinDataSource joinDataSource = (JoinDataSource) current;
        current = joinDataSource.getLeft();
        if (currentDimFilter != null) {
            throw new IAE("Left filters are only allowed when left child is direct table access");
        }
        currentDimFilter = joinDataSource.getLeftFilter();
        preJoinableClauses.add(new PreJoinableClause(joinDataSource.getRightPrefix(), joinDataSource.getRight(), joinDataSource.getJoinType(), joinDataSource.getConditionAnalysis()));
    }
    // Join clauses were added in the order we saw them while traversing down, but we need to apply them in the
    // going-up order. So reverse them.
    Collections.reverse(preJoinableClauses);
    return Triple.of(current, currentDimFilter, preJoinableClauses);
}

Example 37 with DimFilter

use of org.apache.druid.query.filter.DimFilter in project druid by druid-io.

the class GroupByQueryRunnerTest method testSubqueryWithOuterTimeFilter.

@Test
public void testSubqueryWithOuterTimeFilter() {
    final GroupByQuery subquery = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FULL_ON_INTERVAL_SPEC).setDimensions(new DefaultDimensionSpec("market", "market"), new DefaultDimensionSpec("quality", "quality")).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, new LongSumAggregatorFactory("index", "index")).setGranularity(QueryRunnerTestHelper.DAY_GRAN).build();
    final DimFilter fridayFilter = new SelectorDimFilter(ColumnHolder.TIME_COLUMN_NAME, "Friday", new TimeFormatExtractionFn("EEEE", null, null, null, false));
    final DimFilter firstDaysFilter = new InDimFilter(ColumnHolder.TIME_COLUMN_NAME, ImmutableList.of("1", "2", "3"), new TimeFormatExtractionFn("d", null, null, null, false));
    final GroupByQuery query = makeQueryBuilder().setDataSource(subquery).setQuerySegmentSpec(QueryRunnerTestHelper.FULL_ON_INTERVAL_SPEC).setDimensions(Collections.emptyList()).setDimFilter(firstDaysFilter).setAggregatorSpecs(new FilteredAggregatorFactory(QueryRunnerTestHelper.ROWS_COUNT, fridayFilter)).setGranularity(QueryRunnerTestHelper.DAY_GRAN).build();
    List<ResultRow> expectedResults = Arrays.asList(makeRow(query, "2011-02-01", "rows", 0L), makeRow(query, "2011-02-02", "rows", 0L), makeRow(query, "2011-02-03", "rows", 0L), makeRow(query, "2011-03-01", "rows", 0L), makeRow(query, "2011-03-02", "rows", 0L), makeRow(query, "2011-03-03", "rows", 0L), makeRow(query, "2011-04-01", "rows", 13L), makeRow(query, "2011-04-02", "rows", 0L), makeRow(query, "2011-04-03", "rows", 0L));
    Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "subquery-time-filter");
}

Example 38 with DimFilter

use of org.apache.druid.query.filter.DimFilter in project druid by druid-io.

the class GroupByQueryRunnerTest method testGroupByWithAggregatorFilterAndExtractionFunction.

@Test
public void testGroupByWithAggregatorFilterAndExtractionFunction() {
    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", "mezzanineANDnews");
    extractionMap.put("news", "mezzanineANDnews");
    extractionMap.put("premium", "premium0");
    extractionMap.put("technology", "technology0");
    extractionMap.put("travel", "travel0");
    MapLookupExtractor mapLookupExtractor = new MapLookupExtractor(extractionMap, false);
    LookupExtractionFn lookupExtractionFn = new LookupExtractionFn(mapLookupExtractor, false, "missing", true, false);
    DimFilter filter = new ExtractionDimFilter("quality", "mezzanineANDnews", lookupExtractionFn, null);
    GroupByQuery query = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FIRST_TO_THIRD).setDimensions(new DefaultDimensionSpec("quality", "alias")).setAggregatorSpecs(new FilteredAggregatorFactory(QueryRunnerTestHelper.ROWS_COUNT, filter), new FilteredAggregatorFactory(new LongSumAggregatorFactory("idx", "index"), filter)).setGranularity(QueryRunnerTestHelper.DAY_GRAN).build();
    List<ResultRow> expectedResults = Arrays.asList(makeRow(query, "2011-04-01", "alias", "automotive", "rows", 0L, "idx", NullHandling.defaultLongValue()), makeRow(query, "2011-04-01", "alias", "business", "rows", 0L, "idx", NullHandling.defaultLongValue()), makeRow(query, "2011-04-01", "alias", "entertainment", "rows", 0L, "idx", NullHandling.defaultLongValue()), makeRow(query, "2011-04-01", "alias", "health", "rows", 0L, "idx", NullHandling.defaultLongValue()), makeRow(query, "2011-04-01", "alias", "mezzanine", "rows", 3L, "idx", 2870L), makeRow(query, "2011-04-01", "alias", "news", "rows", 1L, "idx", 121L), makeRow(query, "2011-04-01", "alias", "premium", "rows", 0L, "idx", NullHandling.defaultLongValue()), makeRow(query, "2011-04-01", "alias", "technology", "rows", 0L, "idx", NullHandling.defaultLongValue()), makeRow(query, "2011-04-01", "alias", "travel", "rows", 0L, "idx", NullHandling.defaultLongValue()), makeRow(query, "2011-04-02", "alias", "automotive", "rows", 0L, "idx", NullHandling.defaultLongValue()), makeRow(query, "2011-04-02", "alias", "business", "rows", 0L, "idx", NullHandling.defaultLongValue()), makeRow(query, "2011-04-02", "alias", "entertainment", "rows", 0L, "idx", NullHandling.defaultLongValue()), makeRow(query, "2011-04-02", "alias", "health", "rows", 0L, "idx", NullHandling.defaultLongValue()), makeRow(query, "2011-04-02", "alias", "mezzanine", "rows", 3L, "idx", 2447L), makeRow(query, "2011-04-02", "alias", "news", "rows", 1L, "idx", 114L), makeRow(query, "2011-04-02", "alias", "premium", "rows", 0L, "idx", NullHandling.defaultLongValue()), makeRow(query, "2011-04-02", "alias", "technology", "rows", 0L, "idx", NullHandling.defaultLongValue()), makeRow(query, "2011-04-02", "alias", "travel", "rows", 0L, "idx", NullHandling.defaultLongValue()));
    Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "agg-filter");
}

Example 39 with DimFilter

use of org.apache.druid.query.filter.DimFilter in project druid by druid-io.

the class GroupByQueryRunnerTest method testSubqueryWithOuterFilterAggregator.

@Test
public void testSubqueryWithOuterFilterAggregator() {
    final GroupByQuery subquery = makeQueryBuilder().setDataSource(QueryRunnerTestHelper.DATA_SOURCE).setQuerySegmentSpec(QueryRunnerTestHelper.FULL_ON_INTERVAL_SPEC).setDimensions(new DefaultDimensionSpec("market", "market"), new DefaultDimensionSpec("quality", "quality")).setAggregatorSpecs(QueryRunnerTestHelper.ROWS_COUNT, new LongSumAggregatorFactory("index", "index")).setGranularity(QueryRunnerTestHelper.DAY_GRAN).build();
    final DimFilter filter = new SelectorDimFilter("market", "spot", null);
    final GroupByQuery query = makeQueryBuilder().setDataSource(subquery).setQuerySegmentSpec(QueryRunnerTestHelper.FULL_ON_INTERVAL_SPEC).setDimensions(Collections.emptyList()).setAggregatorSpecs(new FilteredAggregatorFactory(QueryRunnerTestHelper.ROWS_COUNT, filter)).setGranularity(QueryRunnerTestHelper.ALL_GRAN).build();
    List<ResultRow> expectedResults = Collections.singletonList(makeRow(query, "1970-01-01", "rows", 837L));
    Iterable<ResultRow> results = GroupByQueryRunnerTestHelper.runQuery(factory, runner, query);
    TestHelper.assertExpectedObjects(expectedResults, results, "subquery-filter-agg");
}

Example 40 with DimFilter

use of org.apache.druid.query.filter.DimFilter in project druid by druid-io.

the class DruidSegmentReader method intermediateRowIterator.

@Override
protected CloseableIterator<Map<String, Object>> intermediateRowIterator() throws IOException {
    final CleanableFile segmentFile = source.fetch(temporaryDirectory, null);
    final WindowedStorageAdapter storageAdapter = new WindowedStorageAdapter(new QueryableIndexStorageAdapter(indexIO.loadIndex(segmentFile.file())), source.getIntervalFilter());
    final Sequence<Cursor> cursors = storageAdapter.getAdapter().makeCursors(Filters.toFilter(dimFilter), storageAdapter.getInterval(), VirtualColumns.EMPTY, Granularities.ALL, false, null);
    // Retain order of columns from the original segments. Useful for preserving dimension order if we're in
    // schemaless mode.
    final Set<String> columnsToRead = Sets.newLinkedHashSet(Iterables.filter(storageAdapter.getAdapter().getRowSignature().getColumnNames(), columnsFilter::apply));
    final Sequence<Map<String, Object>> sequence = Sequences.concat(Sequences.map(cursors, cursor -> cursorToSequence(cursor, columnsToRead)));
    return makeCloseableIteratorFromSequenceAndSegmentFile(sequence, segmentFile);
}