Search in sources :

Example 11 with LookupExtractionFn

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

the class TopNQueryRunnerTest method testTopNWithExtractionFilter.

@Test
public void testTopNWithExtractionFilter() {
    Map<String, String> extractionMap = new HashMap<>();
    extractionMap.put("spot", "spot0");
    MapLookupExtractor mapLookupExtractor = new MapLookupExtractor(extractionMap, false);
    LookupExtractionFn lookupExtractionFn = new LookupExtractionFn(mapLookupExtractor, false, null, true, false);
    TopNQuery query = new TopNQueryBuilder().dataSource(QueryRunnerTestHelper.dataSource).granularity(QueryRunnerTestHelper.allGran).dimension(QueryRunnerTestHelper.marketDimension).metric("rows").threshold(3).intervals(QueryRunnerTestHelper.firstToThird).aggregators(QueryRunnerTestHelper.commonAggregators).postAggregators(Arrays.<PostAggregator>asList(QueryRunnerTestHelper.addRowsIndexConstant)).filters(new ExtractionDimFilter(QueryRunnerTestHelper.marketDimension, "spot0", lookupExtractionFn, null)).build();
    List<Result<TopNResultValue>> expectedResults = Arrays.asList(new Result<>(new DateTime("2011-04-01T00:00:00.000Z"), new TopNResultValue(Arrays.<Map<String, Object>>asList(ImmutableMap.<String, Object>of(QueryRunnerTestHelper.marketDimension, "spot", "rows", 18L, "index", 2231.8768157958984D, "addRowsIndexConstant", 2250.8768157958984D, "uniques", QueryRunnerTestHelper.UNIQUES_9)))));
    assertExpectedResults(expectedResults, query);
    // Assert the optimization path as well
    final Sequence<Result<TopNResultValue>> retval = runWithPreMergeAndMerge(query);
    TestHelper.assertExpectedResults(expectedResults, retval);
}
Also used : HashMap(java.util.HashMap) LinkedHashMap(java.util.LinkedHashMap) ExtractionDimFilter(io.druid.query.filter.ExtractionDimFilter) DateTime(org.joda.time.DateTime) Result(io.druid.query.Result) LookupExtractionFn(io.druid.query.lookup.LookupExtractionFn) MapLookupExtractor(io.druid.query.extraction.MapLookupExtractor) Test(org.junit.Test)

Example 12 with LookupExtractionFn

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

the class TopNQueryRunnerTest method testTopNDimExtractionFastTopNUnOptimal.

@Test
public // Test query path that must rebucket the data
void testTopNDimExtractionFastTopNUnOptimal() {
    TopNQuery query = new TopNQueryBuilder().dataSource(QueryRunnerTestHelper.dataSource).granularity(QueryRunnerTestHelper.allGran).dimension(new ExtractionDimensionSpec(QueryRunnerTestHelper.marketDimension, QueryRunnerTestHelper.marketDimension, new LookupExtractionFn(new MapLookupExtractor(ImmutableMap.of("spot", "spot0", "total_market", "total_market0", "upfront", "upfront0"), false), true, null, false, false))).metric("rows").threshold(4).intervals(QueryRunnerTestHelper.firstToThird).aggregators(QueryRunnerTestHelper.commonAggregators).postAggregators(Arrays.<PostAggregator>asList(QueryRunnerTestHelper.addRowsIndexConstant)).build();
    List<Result<TopNResultValue>> expectedResults = Arrays.asList(new Result<>(new DateTime("2011-04-01T00:00:00.000Z"), new TopNResultValue(Arrays.<Map<String, Object>>asList(ImmutableMap.<String, Object>of(QueryRunnerTestHelper.marketDimension, "spot0", "rows", 18L, "index", 2231.8768157958984D, "addRowsIndexConstant", 2250.8768157958984D, "uniques", QueryRunnerTestHelper.UNIQUES_9), ImmutableMap.<String, Object>of(QueryRunnerTestHelper.marketDimension, "total_market0", "rows", 4L, "index", 5351.814697265625D, "addRowsIndexConstant", 5356.814697265625D, "uniques", QueryRunnerTestHelper.UNIQUES_2), ImmutableMap.<String, Object>of(QueryRunnerTestHelper.marketDimension, "upfront0", "rows", 4L, "index", 4875.669677734375D, "addRowsIndexConstant", 4880.669677734375D, "uniques", QueryRunnerTestHelper.UNIQUES_2)))));
    assertExpectedResults(expectedResults, query);
}
Also used : HyperUniqueFinalizingPostAggregator(io.druid.query.aggregation.hyperloglog.HyperUniqueFinalizingPostAggregator) PostAggregator(io.druid.query.aggregation.PostAggregator) DateTime(org.joda.time.DateTime) Result(io.druid.query.Result) LookupExtractionFn(io.druid.query.lookup.LookupExtractionFn) MapLookupExtractor(io.druid.query.extraction.MapLookupExtractor) ExtractionDimensionSpec(io.druid.query.dimension.ExtractionDimensionSpec) Test(org.junit.Test)

Example 13 with LookupExtractionFn

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

the class TopNQueryRunnerTest method testTopNDimExtractionFastTopNUnOptimalWithReplaceMissing.

@Test
public void testTopNDimExtractionFastTopNUnOptimalWithReplaceMissing() {
    TopNQuery query = new TopNQueryBuilder().dataSource(QueryRunnerTestHelper.dataSource).granularity(QueryRunnerTestHelper.allGran).dimension(new ExtractionDimensionSpec(QueryRunnerTestHelper.marketDimension, QueryRunnerTestHelper.marketDimension, new LookupExtractionFn(new MapLookupExtractor(ImmutableMap.of("spot", "2spot0", "total_market", "1total_market0", "upfront", "3upfront0"), false), false, "MISSING", false, false))).metric("rows").threshold(4).intervals(QueryRunnerTestHelper.firstToThird).aggregators(QueryRunnerTestHelper.commonAggregators).postAggregators(Arrays.<PostAggregator>asList(QueryRunnerTestHelper.addRowsIndexConstant)).build();
    List<Result<TopNResultValue>> expectedResults = Arrays.asList(new Result<>(new DateTime("2011-04-01T00:00:00.000Z"), new TopNResultValue(Arrays.<Map<String, Object>>asList(ImmutableMap.<String, Object>of(QueryRunnerTestHelper.marketDimension, "2spot0", "rows", 18L, "index", 2231.8768157958984D, "addRowsIndexConstant", 2250.8768157958984D, "uniques", QueryRunnerTestHelper.UNIQUES_9), ImmutableMap.<String, Object>of(QueryRunnerTestHelper.marketDimension, "1total_market0", "rows", 4L, "index", 5351.814697265625D, "addRowsIndexConstant", 5356.814697265625D, "uniques", QueryRunnerTestHelper.UNIQUES_2), ImmutableMap.<String, Object>of(QueryRunnerTestHelper.marketDimension, "3upfront0", "rows", 4L, "index", 4875.669677734375D, "addRowsIndexConstant", 4880.669677734375D, "uniques", QueryRunnerTestHelper.UNIQUES_2)))));
    assertExpectedResults(expectedResults, query);
}
Also used : HyperUniqueFinalizingPostAggregator(io.druid.query.aggregation.hyperloglog.HyperUniqueFinalizingPostAggregator) PostAggregator(io.druid.query.aggregation.PostAggregator) DateTime(org.joda.time.DateTime) Result(io.druid.query.Result) LookupExtractionFn(io.druid.query.lookup.LookupExtractionFn) MapLookupExtractor(io.druid.query.extraction.MapLookupExtractor) ExtractionDimensionSpec(io.druid.query.dimension.ExtractionDimensionSpec) Test(org.junit.Test)

Example 14 with LookupExtractionFn

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

the class TopNQueryRunnerTest method testTopNWithExtractionFilterNoExistingValue.

@Test
public void testTopNWithExtractionFilterNoExistingValue() {
    Map<String, String> extractionMap = new HashMap<>();
    extractionMap.put("", "NULL");
    MapLookupExtractor mapLookupExtractor = new MapLookupExtractor(extractionMap, false);
    LookupExtractionFn lookupExtractionFn = new LookupExtractionFn(mapLookupExtractor, false, null, true, true);
    DimFilter extractionFilter = new ExtractionDimFilter("null_column", "NULL", lookupExtractionFn, null);
    TopNQueryBuilder topNQueryBuilder = new TopNQueryBuilder().dataSource(QueryRunnerTestHelper.dataSource).granularity(QueryRunnerTestHelper.allGran).dimension("null_column").metric(QueryRunnerTestHelper.indexMetric).threshold(4).intervals(QueryRunnerTestHelper.fullOnInterval).aggregators(Lists.newArrayList(Iterables.concat(QueryRunnerTestHelper.commonAggregators, Lists.newArrayList(new FilteredAggregatorFactory(new DoubleMaxAggregatorFactory("maxIndex", "index"), extractionFilter), //new DoubleMaxAggregatorFactory("maxIndex", "index"),
    new DoubleMinAggregatorFactory("minIndex", "index"))))).postAggregators(Arrays.<PostAggregator>asList(QueryRunnerTestHelper.addRowsIndexConstant));
    TopNQuery topNQueryWithNULLValueExtraction = topNQueryBuilder.filters(extractionFilter).build();
    Map<String, Object> map = Maps.newHashMap();
    map.put("null_column", null);
    map.put("rows", 1209L);
    map.put("index", 503332.5071372986D);
    map.put("addRowsIndexConstant", 504542.5071372986D);
    map.put("uniques", QueryRunnerTestHelper.UNIQUES_9);
    map.put("maxIndex", 1870.06103515625D);
    map.put("minIndex", 59.02102279663086D);
    List<Result<TopNResultValue>> expectedResults = Arrays.asList(new Result<>(new DateTime("2011-01-12T00:00:00.000Z"), new TopNResultValue(Arrays.asList(map))));
    assertExpectedResults(expectedResults, topNQueryWithNULLValueExtraction);
    // Assert the optimization path as well
    final Sequence<Result<TopNResultValue>> retval = runWithPreMergeAndMerge(topNQueryWithNULLValueExtraction);
    TestHelper.assertExpectedResults(expectedResults, retval);
}
Also used : FilteredAggregatorFactory(io.druid.query.aggregation.FilteredAggregatorFactory) DoubleMaxAggregatorFactory(io.druid.query.aggregation.DoubleMaxAggregatorFactory) HashMap(java.util.HashMap) LinkedHashMap(java.util.LinkedHashMap) ExtractionDimFilter(io.druid.query.filter.ExtractionDimFilter) DoubleMinAggregatorFactory(io.druid.query.aggregation.DoubleMinAggregatorFactory) DateTime(org.joda.time.DateTime) Result(io.druid.query.Result) LookupExtractionFn(io.druid.query.lookup.LookupExtractionFn) MapLookupExtractor(io.druid.query.extraction.MapLookupExtractor) SelectorDimFilter(io.druid.query.filter.SelectorDimFilter) DimFilter(io.druid.query.filter.DimFilter) ExtractionDimFilter(io.druid.query.filter.ExtractionDimFilter) AndDimFilter(io.druid.query.filter.AndDimFilter) Test(org.junit.Test)

Example 15 with LookupExtractionFn

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

the class FloatFilteringTest method testFloatFilterWithExtractionFn.

@Test
public void testFloatFilterWithExtractionFn() {
    final Map<String, String> stringMap = new HashMap<>();
    stringMap.put("1.0", "Monday");
    stringMap.put("2.0", "Tuesday");
    stringMap.put("3.0", "Wednesday");
    stringMap.put("4.0", "Thursday");
    stringMap.put("5.0", "Friday");
    stringMap.put("6.0", "Saturday");
    LookupExtractor mapExtractor = new MapLookupExtractor(stringMap, false);
    LookupExtractionFn exfn = new LookupExtractionFn(mapExtractor, false, "UNKNOWN", false, true);
    assertFilterMatches(new SelectorDimFilter(FLOAT_COLUMN, "Monday", exfn), ImmutableList.<String>of("1"));
    assertFilterMatches(new SelectorDimFilter(FLOAT_COLUMN, "Notaday", exfn), ImmutableList.<String>of());
    assertFilterMatches(new BoundDimFilter(FLOAT_COLUMN, "Fridax", "Fridaz", false, false, null, exfn, StringComparators.ALPHANUMERIC), ImmutableList.<String>of("5"));
    assertFilterMatches(new BoundDimFilter(FLOAT_COLUMN, "Friday", "Friday", true, true, null, exfn, StringComparators.ALPHANUMERIC), ImmutableList.<String>of());
    assertFilterMatches(new InDimFilter(FLOAT_COLUMN, Arrays.asList("Caturday", "Saturday", "Tuesday"), exfn), ImmutableList.<String>of("2", "6"));
    // test InFilter HashSet implementation
    List<String> bigList = Arrays.asList("Saturday", "Tuesday", "Caturday", "Xanaday", "Vojuday", "Gribaday", "Kipoday", "Dheferday", "Fakeday", "Qeearaday", "Hello", "World", "1", "2", "3", "4", "5", "6", "7");
    assertFilterMatches(new InDimFilter(FLOAT_COLUMN, bigList, exfn), ImmutableList.<String>of("2", "6"));
    String jsFn = "function(x) { return(x === 'Wednesday' || x === 'Thursday') }";
    assertFilterMatches(new JavaScriptDimFilter(FLOAT_COLUMN, jsFn, exfn, JavaScriptConfig.getEnabledInstance()), ImmutableList.<String>of("3", "4"));
    assertFilterMatches(new RegexDimFilter(FLOAT_COLUMN, ".*day", exfn), ImmutableList.<String>of("1", "2", "3", "4", "5", "6"));
    assertFilterMatches(new SearchQueryDimFilter(FLOAT_COLUMN, new ContainsSearchQuerySpec("s", true), exfn), ImmutableList.<String>of("2", "3", "4"));
}
Also used : RegexDimFilter(io.druid.query.filter.RegexDimFilter) BoundDimFilter(io.druid.query.filter.BoundDimFilter) HashMap(java.util.HashMap) ContainsSearchQuerySpec(io.druid.query.search.search.ContainsSearchQuerySpec) MapLookupExtractor(io.druid.query.extraction.MapLookupExtractor) LookupExtractor(io.druid.query.lookup.LookupExtractor) LookupExtractionFn(io.druid.query.lookup.LookupExtractionFn) SelectorDimFilter(io.druid.query.filter.SelectorDimFilter) InDimFilter(io.druid.query.filter.InDimFilter) SearchQueryDimFilter(io.druid.query.filter.SearchQueryDimFilter) MapLookupExtractor(io.druid.query.extraction.MapLookupExtractor) JavaScriptDimFilter(io.druid.query.filter.JavaScriptDimFilter) Test(org.junit.Test)

Aggregations

LookupExtractionFn (io.druid.query.lookup.LookupExtractionFn)39 MapLookupExtractor (io.druid.query.extraction.MapLookupExtractor)38 Test (org.junit.Test)38 HashMap (java.util.HashMap)24 ExtractionDimensionSpec (io.druid.query.dimension.ExtractionDimensionSpec)19 LongSumAggregatorFactory (io.druid.query.aggregation.LongSumAggregatorFactory)16 Result (io.druid.query.Result)15 Row (io.druid.data.input.Row)14 DefaultDimensionSpec (io.druid.query.dimension.DefaultDimensionSpec)14 SelectorDimFilter (io.druid.query.filter.SelectorDimFilter)14 DateTime (org.joda.time.DateTime)14 ExtractionDimFilter (io.druid.query.filter.ExtractionDimFilter)13 PostAggregator (io.druid.query.aggregation.PostAggregator)8 HyperUniqueFinalizingPostAggregator (io.druid.query.aggregation.hyperloglog.HyperUniqueFinalizingPostAggregator)7 DimensionSpec (io.druid.query.dimension.DimensionSpec)7 InDimFilter (io.druid.query.filter.InDimFilter)7 LookupExtractor (io.druid.query.lookup.LookupExtractor)7 ListFilteredDimensionSpec (io.druid.query.dimension.ListFilteredDimensionSpec)6 RegexFilteredDimensionSpec (io.druid.query.dimension.RegexFilteredDimensionSpec)6 BoundDimFilter (io.druid.query.filter.BoundDimFilter)6