Apache
/
druid
mirror of https://github.com/apache/druid.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Fix topN lexicographic sort (#5815) 

* fixes #5814
changes:
* pass `StorageAdapter` to topn algorithms to get things like if column is 'sorted' or if query interval is smaller than segment granularity, instead of using `io.druid.segment.Capabilities`
* remove `io.druid.segment.Capabilities` since it had one purpose, supplying `dimensionValuesSorted` which is now provided directly by `StorageAdapter`.
* added test for topn optimization path checking

* add Capabilities back since StorageAdapter is marked PublicApi

* oops

* add javadoc, fix build i think

* correctly revert api changes

* fix intellij fail

* fix typo :(
This commit is contained in:
Clint Wylie 2018-05-31 09:53:29 -07:00 committed by Charles Allen
parent 50ad7a45ff
commit 2b45a6a42d
17 changed files with 577 additions and 72 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
processing/src/main/java/io/druid/query/topn/AggregateTopNMetricFirstAlgorithm.java
View File

@ -26,8 +26,8 @@ import io.druid.query.ColumnSelectorPlus;
import io.druid.query.aggregation.AggregatorFactory;
import io.druid.query.aggregation.AggregatorUtil;
import io.druid.query.aggregation.PostAggregator;
import io.druid.segment.Capabilities;
import io.druid.segment.Cursor;
import io.druid.segment.StorageAdapter;
import javax.annotation.Nullable;
import java.nio.ByteBuffer;
@ -39,17 +39,17 @@ import java.util.List;
*/
public class AggregateTopNMetricFirstAlgorithm implements TopNAlgorithm<int[], TopNParams>
{
private final Capabilities capabilities;
private final StorageAdapter storageAdapter;
private final TopNQuery query;
private final NonBlockingPool<ByteBuffer> bufferPool;
public AggregateTopNMetricFirstAlgorithm(
Capabilities capabilities,
StorageAdapter storageAdapter,
TopNQuery query,
NonBlockingPool<ByteBuffer> bufferPool
)
{
this.capabilities = capabilities;
this.storageAdapter = storageAdapter;
this.query = query;
this.bufferPool = bufferPool;
}
@ -91,7 +91,7 @@ public class AggregateTopNMetricFirstAlgorithm implements TopNAlgorithm<int[], T
.build();
final TopNResultBuilder singleMetricResultBuilder = BaseTopNAlgorithm.makeResultBuilder(params, singleMetricQuery);
PooledTopNAlgorithm singleMetricAlgo = new PooledTopNAlgorithm(capabilities, singleMetricQuery, bufferPool);
PooledTopNAlgorithm singleMetricAlgo = new PooledTopNAlgorithm(storageAdapter, singleMetricQuery, bufferPool);
PooledTopNAlgorithm.PooledTopNParams singleMetricParam = null;
int[] dimValSelector = null;
try {
@ -110,7 +110,7 @@ public class AggregateTopNMetricFirstAlgorithm implements TopNAlgorithm<int[], T
singleMetricAlgo.cleanup(singleMetricParam);
}
PooledTopNAlgorithm allMetricAlgo = new PooledTopNAlgorithm(capabilities, query, bufferPool);
PooledTopNAlgorithm allMetricAlgo = new PooledTopNAlgorithm(storageAdapter, query, bufferPool);
PooledTopNAlgorithm.PooledTopNParams allMetricsParam = null;
try {
// Run topN for all metrics for top N dimension values

32
processing/src/main/java/io/druid/query/topn/BaseTopNAlgorithm.java
View File

@ -19,6 +19,7 @@
package io.druid.query.topn;
import com.google.common.annotations.VisibleForTesting;
import io.druid.java.util.common.IAE;
import io.druid.java.util.common.Pair;
import io.druid.query.aggregation.Aggregator;
@ -29,6 +30,7 @@ import io.druid.segment.Capabilities;
import io.druid.segment.Cursor;
import io.druid.segment.DimensionSelector;
import io.druid.segment.IdLookup;
import io.druid.segment.StorageAdapter;
import javax.annotation.Nullable;
import java.util.Arrays;
@ -62,11 +64,11 @@ public abstract class BaseTopNAlgorithm<DimValSelector, DimValAggregateStore, Pa
return aggregators;
}
protected final Capabilities capabilities;
protected final StorageAdapter storageAdapter;
protected BaseTopNAlgorithm(Capabilities capabilities)
protected BaseTopNAlgorithm(StorageAdapter storageAdapter)
{
this.capabilities = capabilities;
this.storageAdapter = storageAdapter;
}
@Override
@ -134,6 +136,7 @@ public abstract class BaseTopNAlgorithm<DimValSelector, DimValAggregateStore, Pa
* This function currently handles TopNs on long and float columns, which do not provide cardinality or an ID lookup.
* When cardinality is unknown, process everything in one pass.
* Existing implementations of makeDimValSelector() require cardinality as well, so the DimValSelector is not used.
*
* @param params TopN parameters from run()
* @param resultBuilder Result builder from run()
*/
@ -206,9 +209,14 @@ public abstract class BaseTopNAlgorithm<DimValSelector, DimValAggregateStore, Pa
Aggregator[][] expansionAggs;
int cardinality;
public AggregatorArrayProvider(DimensionSelector dimSelector, TopNQuery query, int cardinality, Capabilities capabilities)
public AggregatorArrayProvider(
DimensionSelector dimSelector,
TopNQuery query,
int cardinality,
StorageAdapter storageAdapter
)
{
super(dimSelector, query, capabilities);
super(dimSelector, query, storageAdapter);
this.expansionAggs = new Aggregator[cardinality][];
this.cardinality = cardinality;
@ -236,17 +244,17 @@ public abstract class BaseTopNAlgorithm<DimValSelector, DimValAggregateStore, Pa
private final IdLookup idLookup;
private final TopNQuery query;
private final Capabilities capabilities;
private final StorageAdapter storageAdapter;
public BaseArrayProvider(
DimensionSelector dimSelector,
TopNQuery query,
Capabilities capabilities
StorageAdapter storageAdapter
)
{
this.idLookup = dimSelector.idLookup();
this.query = query;
this.capabilities = capabilities;
this.storageAdapter = storageAdapter;
previousStop = null;
ignoreAfterThreshold = false;
@ -261,6 +269,7 @@ public abstract class BaseTopNAlgorithm<DimValSelector, DimValAggregateStore, Pa
@Override
public void skipTo(String previousStop)
{
Capabilities capabilities = storageAdapter.getCapabilities();
if (capabilities.dimensionValuesSorted()) {
this.previousStop = previousStop;
}
@ -284,7 +293,8 @@ public abstract class BaseTopNAlgorithm<DimValSelector, DimValAggregateStore, Pa
keepOnlyN = n;
}
protected Pair<Integer, Integer> computeStartEnd(int cardinality)
@VisibleForTesting
public Pair<Integer, Integer> computeStartEnd(int cardinality)
{
int startIndex = ignoreFirstN;
@ -305,7 +315,9 @@ public abstract class BaseTopNAlgorithm<DimValSelector, DimValAggregateStore, Pa
int endIndex = Math.min(ignoreFirstN + keepOnlyN, cardinality);
if (ignoreAfterThreshold && query.getDimensionsFilter() == null) {
if (ignoreAfterThreshold &&
query.getDimensionsFilter() == null &&
query.getIntervals().stream().anyMatch(interval -> interval.contains(storageAdapter.getInterval()))) {
endIndex = Math.min(endIndex, startIndex + query.getThreshold());
}

11
processing/src/main/java/io/druid/query/topn/DimExtractionTopNAlgorithm.java
View File

@ -23,24 +23,25 @@ import com.google.common.base.Function;
import io.druid.query.ColumnSelectorPlus;
import io.druid.query.aggregation.Aggregator;
import io.druid.query.topn.types.TopNColumnSelectorStrategy;
import io.druid.segment.Capabilities;
import io.druid.segment.Cursor;
import io.druid.segment.StorageAdapter;
import java.util.Map;
/**
* This has to be its own strategy because the pooled topn algorithm assumes each index is unique, and cannot handle multiple index numerals referencing the same dimension value.
*/
public class DimExtractionTopNAlgorithm extends BaseTopNAlgorithm<Aggregator[][], Map<Comparable, Aggregator[]>, TopNParams>
public class DimExtractionTopNAlgorithm
extends BaseTopNAlgorithm<Aggregator[][], Map<Comparable, Aggregator[]>, TopNParams>
{
private final TopNQuery query;
public DimExtractionTopNAlgorithm(
Capabilities capabilities,
StorageAdapter storageAdapter,
TopNQuery query
)
{
super(capabilities);
super(storageAdapter);
this.query = query;
}
@ -65,7 +66,7 @@ public class DimExtractionTopNAlgorithm extends BaseTopNAlgorithm<Aggregator[][]
throw new UnsupportedOperationException("Cannot operate on a dimension with unknown cardinality");
}
ColumnSelectorPlus<TopNColumnSelectorStrategy> selectorPlus = params.getSelectorPlus();
return selectorPlus.getColumnSelectorStrategy().getDimExtractionRowSelector(query, params, capabilities);
return selectorPlus.getColumnSelectorStrategy().getDimExtractionRowSelector(query, params, storageAdapter);
}
@Override

13
processing/src/main/java/io/druid/query/topn/PooledTopNAlgorithm.java
View File

@ -33,10 +33,10 @@ import io.druid.query.aggregation.SimpleDoubleBufferAggregator;
import io.druid.query.monomorphicprocessing.SpecializationService;
import io.druid.query.monomorphicprocessing.SpecializationState;
import io.druid.query.monomorphicprocessing.StringRuntimeShape;
import io.druid.segment.Capabilities;
import io.druid.segment.Cursor;
import io.druid.segment.DimensionSelector;
import io.druid.segment.FilteredOffset;
import io.druid.segment.StorageAdapter;
import io.druid.segment.column.ValueType;
import io.druid.segment.data.IndexedInts;
import io.druid.segment.data.Offset;
@ -64,7 +64,9 @@ public class PooledTopNAlgorithm
private static boolean specializeHistoricalSingleValueDimSelector1SimpleDoubleAggPooledTopN =
!Boolean.getBoolean("dontSpecializeHistoricalSingleValueDimSelector1SimpleDoubleAggPooledTopN");
/** See TopNQueryRunnerTest */
/**
* See TopNQueryRunnerTest
*/
@VisibleForTesting
static void setSpecializeGeneric1AggPooledTopN(boolean value)
{
@ -116,6 +118,7 @@ public class PooledTopNAlgorithm
}
private static final List<ScanAndAggregate> specializedScanAndAggregateImplementations = new ArrayList<>();
static {
computeSpecializedScanAndAggregateImplementations();
}
@ -197,12 +200,12 @@ public class PooledTopNAlgorithm
private static final int AGG_UNROLL_COUNT = 8; // Must be able to fit loop below
public PooledTopNAlgorithm(
Capabilities capabilities,
StorageAdapter storageAdapter,
TopNQuery query,
NonBlockingPool<ByteBuffer> bufferPool
)
{
super(capabilities);
super(storageAdapter);
this.query = query;
this.bufferPool = bufferPool;
}
@ -226,7 +229,7 @@ public class PooledTopNAlgorithm
final TopNMetricSpecBuilder<int[]> arrayProvider = new BaseArrayProvider<int[]>(
dimSelector,
query,
capabilities
storageAdapter
)
{
private final int[] positions = new int[cardinality];

8
processing/src/main/java/io/druid/query/topn/TimeExtractionTopNAlgorithm.java
View File

@ -20,11 +20,11 @@
package io.druid.query.topn;
import com.google.common.collect.Maps;
import io.druid.query.aggregation.Aggregator;
import io.druid.query.ColumnSelectorPlus;
import io.druid.segment.Capabilities;
import io.druid.query.aggregation.Aggregator;
import io.druid.segment.Cursor;
import io.druid.segment.DimensionSelector;
import io.druid.segment.StorageAdapter;
import java.util.Map;
@ -33,9 +33,9 @@ public class TimeExtractionTopNAlgorithm extends BaseTopNAlgorithm<int[], Map<St
public static final int[] EMPTY_INTS = new int[]{};
private final TopNQuery query;
public TimeExtractionTopNAlgorithm(Capabilities capabilities, TopNQuery query)
public TimeExtractionTopNAlgorithm(StorageAdapter storageAdapter, TopNQuery query)
{
super(capabilities);
super(storageAdapter);
this.query = query;
}

18
processing/src/main/java/io/druid/query/topn/TopNQueryEngine.java
View File

@ -30,7 +30,6 @@ import io.druid.query.Result;
import io.druid.query.aggregation.AggregatorFactory;
import io.druid.query.extraction.ExtractionFn;
import io.druid.query.filter.Filter;
import io.druid.segment.Capabilities;
import io.druid.segment.Cursor;
import io.druid.segment.SegmentMissingException;
import io.druid.segment.StorageAdapter;
@ -109,7 +108,6 @@ public class TopNQueryEngine
final @Nullable TopNQueryMetrics queryMetrics
)
{
final Capabilities capabilities = adapter.getCapabilities();
final String dimension = query.getDimensionSpec().getDimension();
final int cardinality = adapter.getDimensionCardinality(dimension);
if (queryMetrics != null) {
@ -137,19 +135,19 @@ public class TopNQueryEngine
) {
// A special TimeExtractionTopNAlgorithm is required, since DimExtractionTopNAlgorithm
// currently relies on the dimension cardinality to support lexicographic sorting
topNAlgorithm = new TimeExtractionTopNAlgorithm(capabilities, query);
topNAlgorithm = new TimeExtractionTopNAlgorithm(adapter, query);
} else if (selector.isHasExtractionFn()) {
topNAlgorithm = new DimExtractionTopNAlgorithm(capabilities, query);
topNAlgorithm = new DimExtractionTopNAlgorithm(adapter, query);
} else if (columnCapabilities != null && !(columnCapabilities.getType() == ValueType.STRING
&& columnCapabilities.isDictionaryEncoded())) {
// Use DimExtraction for non-Strings and for non-dictionary-encoded Strings.
topNAlgorithm = new DimExtractionTopNAlgorithm(capabilities, query);
topNAlgorithm = new DimExtractionTopNAlgorithm(adapter, query);
} else if (selector.isAggregateAllMetrics()) {
topNAlgorithm = new PooledTopNAlgorithm(capabilities, query, bufferPool);
topNAlgorithm = new PooledTopNAlgorithm(adapter, query, bufferPool);
} else if (selector.isAggregateTopNMetricFirst() || query.getContextBoolean("doAggregateTopNMetricFirst", false)) {
topNAlgorithm = new AggregateTopNMetricFirstAlgorithm(capabilities, query, bufferPool);
topNAlgorithm = new AggregateTopNMetricFirstAlgorithm(adapter, query, bufferPool);
} else {
topNAlgorithm = new PooledTopNAlgorithm(capabilities, query, bufferPool);
topNAlgorithm = new PooledTopNAlgorithm(adapter, query, bufferPool);
}
if (queryMetrics != null) {
queryMetrics.algorithm(topNAlgorithm);
@ -162,7 +160,9 @@ public class TopNQueryEngine
{
return query.getDimensionSpec() != null
&& query.getDimensionSpec().getExtractionFn() != null
&& ExtractionFn.ExtractionType.ONE_TO_ONE.equals(query.getDimensionSpec().getExtractionFn().getExtractionType())
&& ExtractionFn.ExtractionType.ONE_TO_ONE.equals(query.getDimensionSpec()
.getExtractionFn()
.getExtractionType())
&& query.getTopNMetricSpec().canBeOptimizedUnordered();
}
}

4
processing/src/main/java/io/druid/query/topn/types/NumericTopNColumnSelectorStrategy.java
View File

@ -28,8 +28,8 @@ import io.druid.query.topn.TopNResultBuilder;
import io.druid.segment.BaseDoubleColumnValueSelector;
import io.druid.segment.BaseFloatColumnValueSelector;
import io.druid.segment.BaseLongColumnValueSelector;
import io.druid.segment.Capabilities;
import io.druid.segment.Cursor;
import io.druid.segment.StorageAdapter;
import io.druid.segment.column.ValueType;
import it.unimi.dsi.fastutil.ints.Int2ObjectMap;
import it.unimi.dsi.fastutil.ints.Int2ObjectOpenHashMap;
@ -51,7 +51,7 @@ public abstract class NumericTopNColumnSelectorStrategy<
@Override
public Aggregator[][] getDimExtractionRowSelector(
TopNQuery query, TopNParams params, Capabilities capabilities
TopNQuery query, TopNParams params, StorageAdapter storageAdapter
)
{
return null;

6
processing/src/main/java/io/druid/query/topn/types/StringTopNColumnSelectorStrategy.java
View File

@ -26,9 +26,9 @@ import io.druid.query.topn.BaseTopNAlgorithm;
import io.druid.query.topn.TopNParams;
import io.druid.query.topn.TopNQuery;
import io.druid.query.topn.TopNResultBuilder;
import io.druid.segment.Capabilities;
import io.druid.segment.Cursor;
import io.druid.segment.DimensionSelector;
import io.druid.segment.StorageAdapter;
import io.druid.segment.column.ValueType;
import io.druid.segment.data.IndexedInts;
@ -50,7 +50,7 @@ public class StringTopNColumnSelectorStrategy
}
@Override
public Aggregator[][] getDimExtractionRowSelector(TopNQuery query, TopNParams params, Capabilities capabilities)
public Aggregator[][] getDimExtractionRowSelector(TopNQuery query, TopNParams params, StorageAdapter storageAdapter)
{
if (params.getCardinality() < 0) {
throw new UnsupportedOperationException("Cannot operate on a dimension with unknown cardinality");
@ -64,7 +64,7 @@ public class StringTopNColumnSelectorStrategy
(DimensionSelector) params.getSelectorPlus().getSelector(),
query,
params.getCardinality(),
capabilities
storageAdapter
);
return provider.build();

9
processing/src/main/java/io/druid/query/topn/types/TopNColumnSelectorStrategy.java
View File

@ -25,8 +25,8 @@ import io.druid.query.dimension.ColumnSelectorStrategy;
import io.druid.query.topn.TopNParams;
import io.druid.query.topn.TopNQuery;
import io.druid.query.topn.TopNResultBuilder;
import io.druid.segment.Capabilities;
import io.druid.segment.Cursor;
import io.druid.segment.StorageAdapter;
import io.druid.segment.column.ValueType;
import javax.annotation.Nullable;
@ -53,10 +53,12 @@ public interface TopNColumnSelectorStrategy<ValueSelectorType, DimExtractionAggr
*
* @param query The TopN query being served
* @param params Parameters for the TopN query being served
* @param capabilities Object indicating if dimension values are sorted
* @param storageAdapter Column storage adapter, to provide information about the column that can be used for
* query optimization, e.g. whether dimension values are sorted or not
*
* @return an Aggregator[][] for integer-valued dimensions, null otherwise
*/
Aggregator[][] getDimExtractionRowSelector(TopNQuery query, TopNParams params, Capabilities capabilities);
Aggregator[][] getDimExtractionRowSelector(TopNQuery query, TopNParams params, StorageAdapter storageAdapter);
/**
* Used by DimExtractionTopNAlgorithm.
@ -88,6 +90,7 @@ public interface TopNColumnSelectorStrategy<ValueSelectorType, DimExtractionAggr
* @param cursor Cursor for the segment being queried
* @param rowSelector Integer lookup containing aggregators
* @param aggregatesStore Map containing aggregators
*
* @return the number of processed rows (after postFilters are applied inside the cursor being processed)
*/
long dimExtractionScanAndAggregate(

5
processing/src/main/java/io/druid/segment/Capabilities.java
View File

@ -21,6 +21,7 @@ package io.druid.segment;
/**
*/
public class Capabilities
{
private final boolean dimensionValuesSorted;
@ -37,6 +38,10 @@ public class Capabilities
this.dimensionValuesSorted = dimensionValuesSorted;
}
/**
* Is dimension value dictionary sorted?
* @return
*/
public boolean dimensionValuesSorted()
{
return dimensionValuesSorted;

2
processing/src/main/java/io/druid/segment/DimensionHandlerUtils.java
View File

@ -192,7 +192,7 @@ public final class DimensionHandlerUtils
}
}
// When determining the capabilites of a column during query processing, this function
// When determining the capabilities of a column during query processing, this function
// adjusts the capabilities for columns that cannot be handled as-is to manageable defaults
// (e.g., treating missing columns as empty String columns)
private static ColumnCapabilities getEffectiveCapabilities(

2
processing/src/main/java/io/druid/segment/QueryableIndexColumnSelectorFactory.java
View File

@ -146,6 +146,6 @@ class QueryableIndexColumnSelectorFactory implements ColumnSelectorFactory
return virtualColumns.getColumnCapabilities(columnName);
}
return QueryableIndexStorageAdapter.getColumnCapabilites(index, columnName);
return QueryableIndexStorageAdapter.getColumnCapabilities(index, columnName);
}
}

4
processing/src/main/java/io/druid/segment/QueryableIndexStorageAdapter.java
View File

@ -167,7 +167,7 @@ public class QueryableIndexStorageAdapter implements StorageAdapter
@Nullable
public ColumnCapabilities getColumnCapabilities(String column)
{
return getColumnCapabilites(index, column);
return getColumnCapabilities(index, column);
}
@Override
@ -316,7 +316,7 @@ public class QueryableIndexStorageAdapter implements StorageAdapter
}
@Nullable
static ColumnCapabilities getColumnCapabilites(ColumnSelector index, String columnName)
static ColumnCapabilities getColumnCapabilities(ColumnSelector index, String columnName)
{
Column columnObj = index.getColumn(columnName);
if (columnObj == null) {

4
processing/src/main/java/io/druid/segment/incremental/IncrementalIndex.java
View File

@ -299,7 +299,7 @@ public abstract class IncrementalIndex<AggregatorType> extends AbstractIndex imp
for (DimensionSchema dimSchema : dimensionsSpec.getDimensions()) {
ValueType type = TYPE_MAP.get(dimSchema.getValueType());
String dimName = dimSchema.getName();
ColumnCapabilitiesImpl capabilities = makeCapabilitesFromValueType(type);
ColumnCapabilitiesImpl capabilities = makeCapabilitiesFromValueType(type);
capabilities.setHasBitmapIndexes(dimSchema.hasBitmapIndex());
if (dimSchema.getTypeName().equals(DimensionSchema.SPATIAL_TYPE_NAME)) {
@ -875,7 +875,7 @@ public abstract class IncrementalIndex<AggregatorType> extends AbstractIndex imp
}
}
private ColumnCapabilitiesImpl makeCapabilitesFromValueType(ValueType type)
private ColumnCapabilitiesImpl makeCapabilitiesFromValueType(ValueType type)
{
ColumnCapabilitiesImpl capabilities = new ColumnCapabilitiesImpl();
capabilities.setDictionaryEncoded(type == ValueType.STRING);

1
processing/src/main/java/io/druid/segment/incremental/IncrementalIndexStorageAdapter.java
View File

@ -142,6 +142,7 @@ public class IncrementalIndexStorageAdapter implements StorageAdapter
return indexer.getMaxValue();
}
@Override
public Capabilities getCapabilities()
{

6
processing/src/test/java/io/druid/query/topn/PooledTopNAlgorithmTest.java
View File

@ -20,7 +20,7 @@
package io.druid.query.topn;
import io.druid.collections.ResourceHolder;
import io.druid.segment.Capabilities;
import io.druid.segment.StorageAdapter;
import org.easymock.EasyMock;
import org.junit.Test;
@ -31,14 +31,14 @@ public class PooledTopNAlgorithmTest
@Test
public void testCleanupWithNullParams()
{
PooledTopNAlgorithm pooledTopNAlgorithm = new PooledTopNAlgorithm(Capabilities.builder().build(), null, null);
PooledTopNAlgorithm pooledTopNAlgorithm = new PooledTopNAlgorithm(EasyMock.mock(StorageAdapter.class), null, null);
pooledTopNAlgorithm.cleanup(null);
}
@Test
public void cleanup()
{
PooledTopNAlgorithm pooledTopNAlgorithm = new PooledTopNAlgorithm(Capabilities.builder().build(), null, null);
PooledTopNAlgorithm pooledTopNAlgorithm = new PooledTopNAlgorithm(EasyMock.mock(StorageAdapter.class), null, null);
PooledTopNAlgorithm.PooledTopNParams params = EasyMock.createMock(PooledTopNAlgorithm.PooledTopNParams.class);
ResourceHolder<ByteBuffer> resourceHolder = EasyMock.createMock(ResourceHolder.class);
EasyMock.expect(params.getResultsBufHolder()).andReturn(resourceHolder).times(1);

480
processing/src/test/java/io/druid/query/topn/TopNMetricSpecOptimizationsTest.java Normal file
View File

@ -0,0 +1,480 @@
/*
* Licensed to Metamarkets Group Inc. (Metamarkets) under one
* or more contributor license agreements. See the NOTICE file
* distributed with this work for additional information
* regarding copyright ownership. Metamarkets licenses this file
* to you under the Apache License, Version 2.0 (the
* "License"); you may not use this file except in compliance
* with the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
* KIND, either express or implied. See the License for the
* specific language governing permissions and limitations
* under the License.
*/
package io.druid.query.topn;
import com.google.common.base.Predicate;
import com.google.common.collect.Iterables;
import com.google.common.collect.Lists;
import io.druid.java.util.common.DateTimes;
import io.druid.java.util.common.Intervals;
import io.druid.java.util.common.Pair;
import io.druid.java.util.common.granularity.Granularity;
import io.druid.java.util.common.guava.Sequence;
import io.druid.query.QueryMetrics;
import io.druid.query.aggregation.AggregatorFactory;
import io.druid.query.aggregation.DoubleMaxAggregatorFactory;
import io.druid.query.aggregation.DoubleMinAggregatorFactory;
import io.druid.query.aggregation.PostAggregator;
import io.druid.query.filter.Filter;
import io.druid.query.filter.ValueMatcher;
import io.druid.query.monomorphicprocessing.RuntimeShapeInspector;
import io.druid.segment.Capabilities;
import io.druid.segment.Cursor;
import io.druid.segment.DimensionSelector;
import io.druid.segment.IdLookup;
import io.druid.segment.Metadata;
import io.druid.segment.StorageAdapter;
import io.druid.segment.VirtualColumns;
import io.druid.segment.column.ColumnCapabilities;
import io.druid.segment.data.Indexed;
import io.druid.segment.data.IndexedInts;
import org.joda.time.DateTime;
import org.joda.time.Interval;
import org.junit.Assert;
import org.junit.Test;
import javax.annotation.Nullable;
import java.util.Arrays;
import static io.druid.query.QueryRunnerTestHelper.addRowsIndexConstant;
import static io.druid.query.QueryRunnerTestHelper.allGran;
import static io.druid.query.QueryRunnerTestHelper.commonDoubleAggregators;
import static io.druid.query.QueryRunnerTestHelper.dataSource;
import static io.druid.query.QueryRunnerTestHelper.indexMetric;
import static io.druid.query.QueryRunnerTestHelper.marketDimension;
import static io.druid.query.QueryRunnerTestHelper.qualityDimension;
public class TopNMetricSpecOptimizationsTest
{
@Test
public void testShouldOptimizeLexicographic()
{
// query interval is greater than segment interval, no filters, can ignoreAfterThreshold
int cardinality = 1234;
int threshold = 4;
TopNQuery query = new TopNQueryBuilder()
.dataSource(dataSource)
.granularity(allGran)
.dimension(marketDimension)
.metric(indexMetric)
.threshold(threshold)
.intervals("2018-05-30T00:00:00Z/2018-05-31T00:00:00Z")
.aggregators(
Lists.<AggregatorFactory>newArrayList(
Iterables.concat(
commonDoubleAggregators,
Lists.newArrayList(
new DoubleMaxAggregatorFactory("maxIndex", "index"),
new DoubleMinAggregatorFactory("minIndex", "index")
)
)
)
)
.postAggregators(Arrays.<PostAggregator>asList(addRowsIndexConstant))
.build();
StorageAdapter adapter =
makeFakeStorageAdapter("2018-05-30T00:00:00Z", "2018-05-30T01:00:00Z", cardinality);
DimensionSelector dimSelector = makeFakeDimSelector(cardinality);
BaseTopNAlgorithm.AggregatorArrayProvider arrayProviderToTest = new BaseTopNAlgorithm.AggregatorArrayProvider(
dimSelector,
query,
cardinality,
adapter
);
arrayProviderToTest.ignoreAfterThreshold();
Pair<Integer, Integer> thePair = arrayProviderToTest.computeStartEnd(cardinality);
Assert.assertEquals(new Integer(0), thePair.lhs);
Assert.assertEquals(new Integer(threshold), thePair.rhs);
}
@Test
public void testAlsoShouldOptimizeLexicographic()
{
// query interval is same as segment interval, no filters, can ignoreAfterThreshold
int cardinality = 1234;
int threshold = 4;
TopNQuery query = new TopNQueryBuilder()
.dataSource(dataSource)
.granularity(allGran)
.dimension(marketDimension)
.metric(indexMetric)
.threshold(threshold)
.intervals("2018-05-30T00:00:00Z/2018-05-30T01:00:00Z")
.aggregators(
Lists.<AggregatorFactory>newArrayList(
Iterables.concat(
commonDoubleAggregators,
Lists.newArrayList(
new DoubleMaxAggregatorFactory("maxIndex", "index"),
new DoubleMinAggregatorFactory("minIndex", "index")
)
)
)
)
.postAggregators(Arrays.<PostAggregator>asList(addRowsIndexConstant))
.build();
StorageAdapter adapter =
makeFakeStorageAdapter("2018-05-30T00:00:00Z", "2018-05-30T01:00:00Z", cardinality);
DimensionSelector dimSelector = makeFakeDimSelector(cardinality);
BaseTopNAlgorithm.AggregatorArrayProvider arrayProviderToTest = new BaseTopNAlgorithm.AggregatorArrayProvider(
dimSelector,
query,
cardinality,
adapter
);
arrayProviderToTest.ignoreAfterThreshold();
Pair<Integer, Integer> thePair = arrayProviderToTest.computeStartEnd(cardinality);
Assert.assertEquals(new Integer(0), thePair.lhs);
Assert.assertEquals(new Integer(threshold), thePair.rhs);
}
@Test
public void testShouldNotOptimizeLexicographic()
{
// query interval is smaller than segment interval, no filters, can ignoreAfterThreshold
int cardinality = 1234;
int threshold = 4;
TopNQuery query = new TopNQueryBuilder()
.dataSource(dataSource)
.granularity(allGran)
.dimension(marketDimension)
.metric(indexMetric)
.threshold(threshold)
.intervals("2018-05-30T00:00:00Z/2018-05-30T01:00:00Z")
.aggregators(
Lists.<AggregatorFactory>newArrayList(
Iterables.concat(
commonDoubleAggregators,
Lists.newArrayList(
new DoubleMaxAggregatorFactory("maxIndex", "index"),
new DoubleMinAggregatorFactory("minIndex", "index")
)
)
)
)
.postAggregators(Arrays.<PostAggregator>asList(addRowsIndexConstant))
.build();
StorageAdapter adapter =
makeFakeStorageAdapter("2018-05-30T00:00:00Z", "2018-05-31T00:00:00Z", cardinality);
DimensionSelector dimSelector = makeFakeDimSelector(cardinality);
BaseTopNAlgorithm.AggregatorArrayProvider arrayProviderToTest = new BaseTopNAlgorithm.AggregatorArrayProvider(
dimSelector,
query,
cardinality,
adapter
);
arrayProviderToTest.ignoreAfterThreshold();
Pair<Integer, Integer> thePair = arrayProviderToTest.computeStartEnd(cardinality);
Assert.assertEquals(new Integer(0), thePair.lhs);
Assert.assertEquals(new Integer(cardinality), thePair.rhs);
}
@Test
public void testAlsoShouldNotOptimizeLexicographic()
{
// query interval is larger than segment interval, but has filters, can ignoreAfterThreshold
int cardinality = 1234;
int threshold = 4;
TopNQuery query = new TopNQueryBuilder()
.dataSource(dataSource)
.granularity(allGran)
.dimension(marketDimension)
.filters(qualityDimension, "entertainment")
.metric(indexMetric)
.threshold(threshold)
.intervals("2018-05-30T00:00:00Z/2018-05-31T00:00:00Z")
.aggregators(
Lists.<AggregatorFactory>newArrayList(
Iterables.concat(
commonDoubleAggregators,
Lists.newArrayList(
new DoubleMaxAggregatorFactory("maxIndex", "index"),
new DoubleMinAggregatorFactory("minIndex", "index")
)
)
)
)
.postAggregators(Arrays.<PostAggregator>asList(addRowsIndexConstant))
.build();
StorageAdapter adapter =
makeFakeStorageAdapter("2018-05-30T00:00:00Z", "2018-05-30T01:00:00Z", cardinality);
DimensionSelector dimSelector = makeFakeDimSelector(cardinality);
BaseTopNAlgorithm.AggregatorArrayProvider arrayProviderToTest = new BaseTopNAlgorithm.AggregatorArrayProvider(
dimSelector,
query,
cardinality,
adapter
);
arrayProviderToTest.ignoreAfterThreshold();
Pair<Integer, Integer> thePair = arrayProviderToTest.computeStartEnd(cardinality);
Assert.assertEquals(new Integer(0), thePair.lhs);
Assert.assertEquals(new Integer(cardinality), thePair.rhs);
}
@Test
public void testAgainShouldNotOptimizeLexicographic()
{
// query interval is larger than segment interval, no filters, can NOT ignoreAfterThreshold
int cardinality = 1234;
int threshold = 4;
TopNQuery query = new TopNQueryBuilder()
.dataSource(dataSource)
.granularity(allGran)
.dimension(marketDimension)
.metric(indexMetric)
.threshold(threshold)
.intervals("2018-05-30T00:00:00Z/2018-05-31T00:00:00Z")
.aggregators(
Lists.<AggregatorFactory>newArrayList(
Iterables.concat(
commonDoubleAggregators,
Lists.newArrayList(
new DoubleMaxAggregatorFactory("maxIndex", "index"),
new DoubleMinAggregatorFactory("minIndex", "index")
)
)
)
)
.postAggregators(Arrays.<PostAggregator>asList(addRowsIndexConstant))
.build();
StorageAdapter adapter =
makeFakeStorageAdapter("2018-05-30T00:00:00Z", "2018-05-30T01:00:00Z", cardinality);
DimensionSelector dimSelector = makeFakeDimSelector(cardinality);
BaseTopNAlgorithm.AggregatorArrayProvider arrayProviderToTest = new BaseTopNAlgorithm.AggregatorArrayProvider(
dimSelector,
query,
cardinality,
adapter
);
Pair<Integer, Integer> thePair = arrayProviderToTest.computeStartEnd(cardinality);
Assert.assertEquals(new Integer(0), thePair.lhs);
Assert.assertEquals(new Integer(cardinality), thePair.rhs);
}
private StorageAdapter makeFakeStorageAdapter(String start, String end, int cardinality)
{
StorageAdapter adapter = new StorageAdapter()
{
@Override
public Interval getInterval()
{
return Intervals.of(start + "/" + end);
}
@Override
public int getDimensionCardinality(String column)
{
return cardinality;
}
@Override
public DateTime getMinTime()
{
return DateTimes.of(start);
}
@Override
public DateTime getMaxTime()
{
return DateTimes.of(end);
}
// stubs below this line not important for tests
@Override
public String getSegmentIdentifier()
{
return null;
}
@Override
public Indexed<String> getAvailableDimensions()
{
return null;
}
@Override
public Iterable<String> getAvailableMetrics()
{
return null;
}
@Nullable
@Override
public Comparable getMinValue(String column)
{
return null;
}
@Nullable
@Override
public Comparable getMaxValue(String column)
{
return null;
}
@Override
public Capabilities getCapabilities()
{
return Capabilities.builder().dimensionValuesSorted(true).build();
}
@Nullable
@Override
public ColumnCapabilities getColumnCapabilities(String column)
{
return null;
}
@Nullable
@Override
public String getColumnTypeName(String column)
{
return null;
}
@Override
public int getNumRows()
{
return 0;
}
@Override
public DateTime getMaxIngestedEventTime()
{
return null;
}
@Override
public Metadata getMetadata()
{
return null;
}
@Override
public Sequence<Cursor> makeCursors(
@Nullable Filter filter,
Interval interval,
VirtualColumns virtualColumns,
Granularity gran,
boolean descending,
@Nullable QueryMetrics<?> queryMetrics
)
{
return null;
}
};
return adapter;
}
private DimensionSelector makeFakeDimSelector(int cardinality)
{
DimensionSelector dimSelector = new DimensionSelector()
{
@Override
public int getValueCardinality()
{
return cardinality;
}
// stubs below this line not important for tests
@Override
public IndexedInts getRow()
{
return null;
}
@Override
public ValueMatcher makeValueMatcher(@Nullable String value)
{
return null;
}
@Override
public ValueMatcher makeValueMatcher(Predicate<String> predicate)
{
return null;
}
@Nullable
@Override
public String lookupName(int id)
{
return null;
}
@Override
public boolean nameLookupPossibleInAdvance()
{
return false;
}
@Nullable
@Override
public IdLookup idLookup()
{
return null;
}
@Override
public void inspectRuntimeShape(RuntimeShapeInspector inspector)
{
}
@Nullable
@Override
public Object getObject()
{
return null;
}
@Override
public Class classOfObject()
{
return null;
}
};
return dimSelector;
}
}