Package org.elasticsearch.search.aggregations

Class AggregatorTestCase

java.lang.Object

org.junit.Assert

org.apache.lucene.util.LuceneTestCase

org.elasticsearch.test.ESTestCase

org.elasticsearch.search.aggregations.AggregatorTestCase

public abstract class AggregatorTestCase
extends ESTestCase

Base class for testing Aggregator implementations. Provides helpers for constructing and searching an Aggregator implementation based on a provided AggregationBuilder instance.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	AggregatorTestCase.InternalAggCardinalityUpperBound

Nested classes/interfaces inherited from class org.elasticsearch.test.ESTestCase

ESTestCase.DeprecationWarning, ESTestCase.GeohashGenerator, ESTestCase.TestAnalysis

Nested classes/interfaces inherited from class org.apache.lucene.util.LuceneTestCase

LuceneTestCase.AwaitsFix, LuceneTestCase.BadApple, LuceneTestCase.Monster, LuceneTestCase.Nightly, LuceneTestCase.Slow, LuceneTestCase.SuppressCodecs, LuceneTestCase.SuppressFileSystems, LuceneTestCase.SuppressFsync, LuceneTestCase.SuppressReproduceLine, LuceneTestCase.SuppressSysoutChecks, LuceneTestCase.SuppressTempFileChecks, LuceneTestCase.ThrowingRunnable, LuceneTestCase.Weekly

Field Summary

Fields

Modifier and Type	Field	Description
protected ValuesSourceRegistry	valuesSourceRegistry

Fields inherited from class org.elasticsearch.test.ESTestCase

checkIndexFailures, DEFAULT_NAMED_WRITABLE_REGISTRY, DEFAULT_TEST_WORKER_ID, failureAndSuccessEvents, FIPS_SYSPROP, JAVA_TIMEZONE_IDS, JAVA_ZONE_IDS, JODA_TIMEZONE_IDS, logger, MIN_PRIVATE_PORT, TEST_WORKER_SYS_PROPERTY, TEST_WORKER_VM_ID

Fields inherited from class org.apache.lucene.util.LuceneTestCase

assertsAreEnabled, classRules, DEFAULT_LINE_DOCS_FILE, INFOSTREAM, JENKINS_LARGE_LINE_DOCS_FILE, LEAVE_TEMPORARY, MAYBE_CACHE_POLICY, RANDOM_MULTIPLIER, ruleChain, suiteFailureMarker, SYSPROP_AWAITSFIX, SYSPROP_BADAPPLES, SYSPROP_FAILFAST, SYSPROP_MAXFAILURES, SYSPROP_MONSTER, SYSPROP_NIGHTLY, SYSPROP_SLOW, SYSPROP_WEEKLY, TEST_ASSERTS_ENABLED, TEST_AWAITSFIX, TEST_BADAPPLES, TEST_CODEC, TEST_DIRECTORY, TEST_DOCVALUESFORMAT, TEST_LINE_DOCS_FILE, TEST_MONSTER, TEST_NIGHTLY, TEST_POSTINGSFORMAT, TEST_SLOW, TEST_THROTTLING, TEST_WEEKLY, VERBOSE

Constructor Summary

Constructors

Constructor	Description
AggregatorTestCase()

Method Summary

Modifier and Type	Method	Description
protected void	afterClose()	Hook for checking things after all Aggregators have been closed.
static AggregationBuilder	aggCardinalityUpperBound(String name)	Request an aggregation that returns the CardinalityUpperBound that was passed to its ctor.
protected FieldMapper	buildMockFieldMapper(MappedFieldType ft)	Build a FieldMapper to create the MappingLookup used for the aggs.
void	cleanupReleasables()
protected AggregationBuilder	createAggBuilderForTypeTest(MappedFieldType fieldType, String fieldName)	This method is invoked each time a field type is tested in testSupportedFieldTypes().
protected AggregationContext	createAggregationContext(org.apache.lucene.search.IndexSearcher indexSearcher, org.apache.lucene.search.Query query, MappedFieldType... fieldTypes)	Create a AggregationContext for testing an Aggregator.
protected AggregationContext	createAggregationContext(org.apache.lucene.search.IndexSearcher indexSearcher, IndexSettings indexSettings, org.apache.lucene.search.Query query, CircuitBreakerService breakerService, long bytesToPreallocate, int maxBucket, MappedFieldType... fieldTypes)	Create a AggregationContext for testing an Aggregator.
protected <A extends Aggregator> A	createAggregator(AggregationBuilder aggregationBuilder, org.apache.lucene.search.IndexSearcher searcher, MappedFieldType... fieldTypes)
protected <A extends Aggregator> A	createAggregator(AggregationBuilder builder, AggregationContext context)
protected AnalysisModule	createAnalysisModule()
protected IndexSettings	createIndexSettings()
protected DateFieldMapper.DateFieldType	dateField(String name, DateFieldMapper.Resolution resolution)	Make a DateFieldMapper.DateFieldType for a date.
protected <R extends InternalAggregation> void	debugTestCase(AggregationBuilder builder, org.apache.lucene.search.Query query, org.apache.lucene.search.IndexSearcher searcher, TriConsumer<R,Class<? extends Aggregator>,Map<String,Map<String,Object>>> verify, MappedFieldType... fieldTypes)	Execute and aggregation and collect its debug information.
protected <R extends InternalAggregation> void	debugTestCase(AggregationBuilder builder, org.apache.lucene.search.Query query, org.elasticsearch.core.CheckedConsumer<RandomIndexWriter,IOException> buildIndex, TriConsumer<R,Class<? extends Aggregator>,Map<String,Map<String,Object>>> verify, MappedFieldType... fieldTypes)	Execute and aggregation and collect its debug information.
protected void	doAssertReducedMultiBucketConsumer(Aggregation agg, MultiBucketConsumerService.MultiBucketConsumer bucketConsumer)
protected NumberFieldMapper.NumberFieldType	doubleField(String name)	Make a NumberFieldMapper.NumberFieldType for a double.
protected GeoPointFieldMapper.GeoPointFieldType	geoPointField(String name)	Make a GeoPointFieldMapper.GeoPointFieldType for a geo_point.
protected ScriptService	getMockScriptService()	Sub-tests that need scripting can override this method to provide a script service and pre-baked scripts
protected List<SearchPlugin>	getSearchPlugins()	Test cases should override this if they have plugins that need to be loaded, e.g.
protected List<ValuesSourceType>	getSupportedValuesSourceTypes()	Implementors should return a list of ValuesSourceType that the aggregator supports.
void	initAnalysisRegistry()
final void	initPlugns()
protected KeywordFieldMapper.KeywordFieldType	keywordField(String name)	Make a DateFieldMapper.DateFieldType for a date.
protected NumberFieldMapper.NumberFieldType	longField(String name)	Make a NumberFieldMapper.NumberFieldType for a long.
protected static IndexReader	maybeWrapReaderEs(DirectoryReader reader)	Added to randomly run with more assertions on the index reader level, like LuceneTestCase.wrapReader(IndexReader), which can't be used because it also wraps in the IndexReader with other implementations that we can't handle.
protected static org.apache.lucene.search.IndexSearcher	newIndexSearcher(IndexReader indexReader)	Added to randomly run with more assertions on the index searcher level, like LuceneTestCase.newSearcher(IndexReader), which can't be used because it also wraps in the IndexSearcher's IndexReader with other implementations that we can't handle.
protected List<ObjectMapper>	objectMappers()	ObjectMappers to add to the lookup.
protected RangeFieldMapper.RangeFieldType	rangeField(String name, RangeType rangeType)	Make a NumberFieldMapper.NumberFieldType for a range.
protected <A extends InternalAggregation, C extends Aggregator> A	searchAndReduce(org.apache.lucene.search.IndexSearcher searcher, org.apache.lucene.search.Query query, AggregationBuilder builder, int maxBucket, MappedFieldType... fieldTypes)
protected <A extends InternalAggregation, C extends Aggregator> A	searchAndReduce(org.apache.lucene.search.IndexSearcher searcher, org.apache.lucene.search.Query query, AggregationBuilder builder, MappedFieldType... fieldTypes)
protected <A extends InternalAggregation, C extends Aggregator> A	searchAndReduce(IndexSettings indexSettings, org.apache.lucene.search.IndexSearcher searcher, org.apache.lucene.search.Query query, AggregationBuilder builder, int maxBucket, boolean splitLeavesIntoSeparateAggregators, MappedFieldType... fieldTypes)	Collects all documents that match the provided query Query and returns the reduced InternalAggregation.
protected <A extends InternalAggregation, C extends Aggregator> A	searchAndReduce(IndexSettings indexSettings, org.apache.lucene.search.IndexSearcher searcher, org.apache.lucene.search.Query query, AggregationBuilder builder, int maxBucket, MappedFieldType... fieldTypes)	Collects all documents that match the provided query Query and returns the reduced InternalAggregation.
protected <A extends InternalAggregation, C extends Aggregator> A	searchAndReduce(IndexSettings indexSettings, org.apache.lucene.search.IndexSearcher searcher, org.apache.lucene.search.Query query, AggregationBuilder builder, MappedFieldType... fieldTypes)
protected <T extends AggregationBuilder, V extends InternalAggregation> void	testCase(T aggregationBuilder, org.apache.lucene.search.Query query, org.elasticsearch.core.CheckedConsumer<RandomIndexWriter,IOException> buildIndex, Consumer<V> verify, MappedFieldType... fieldTypes)
void	testSupportedFieldTypes()	This test will validate that an aggregator succeeds or fails to run against all the field types that are registered in IndicesModule (e.g.
protected List<String>	unsupportedMappedFieldTypes()	A method that allows implementors to specifically blacklist particular field types (based on their content_name).
protected <T extends AggregationBuilder, V extends InternalAggregation> void	verifyOutputFieldNames(T aggregationBuilder, V agg)
protected void	withAggregator(AggregationBuilder aggregationBuilder, org.apache.lucene.search.Query query, org.elasticsearch.core.CheckedConsumer<RandomIndexWriter,IOException> buildIndex, CheckedBiConsumer<org.apache.lucene.search.IndexSearcher,Aggregator,IOException> verify, MappedFieldType... fieldTypes)
protected void	withIndex(org.elasticsearch.core.CheckedConsumer<RandomIndexWriter,IOException> buildIndex, org.elasticsearch.core.CheckedConsumer<org.apache.lucene.search.IndexSearcher,IOException> consume)
protected void	withNonMergingIndex(org.elasticsearch.core.CheckedConsumer<RandomIndexWriter,IOException> buildIndex, org.elasticsearch.core.CheckedConsumer<org.apache.lucene.search.IndexSearcher,IOException> consume)
protected IndexReader	wrapDirectoryReader(DirectoryReader reader)	Override to wrap the DirectoryReader for aggs like NestedAggregationBuilder.
protected static DirectoryReader	wrapInMockESDirectoryReader(DirectoryReader directoryReader)
protected final NamedWriteableRegistry	writableRegistry()	The NamedWriteableRegistry to use for this test.

Methods inherited from class org.elasticsearch.test.ESTestCase

after, afterIfFailed, afterIfSuccessful, allowedWarnings, assertArrayEquals, assertBusy, assertBusy, assertEquals, assertSettingDeprecationsAndWarnings, assertWarnings, assertWarnings, assertWarnings, before, between, buildEnvSettings, buildNewFakeTransportAddress, checkStaticState, clearAdditionalRoles, copyInstance, copyNamedWriteable, copyNamedWriteable, copyWriteable, copyWriteable, createDefaultIndexAnalyzers, createParser, createParser, createParser, createParser, createParser, createParser, createTestAnalysis, createTestAnalysis, createTestAnalysis, enableJodaDeprecationWarningsCheck, enableWarningsCheck, ensureAllSearchContextsReleased, ensureCheckIndexPassed, ensureNoWarnings, ensureSupportedLocale, filteredWarnings, frequently, generateRandomStringArray, generateRandomStringArray, getDataPath, getPortRange, getSuiteFailureMarker, getTestTransportPlugin, getTestTransportType, getWorkerBasePort, inFipsJvm, iterations, maybeSet, mockScript, newEnvironment, newEnvironment, newNodeEnvironment, newNodeEnvironment, randomAlphaOfLength, randomAlphaOfLengthBetween, randomArray, randomArray, randomBigInteger, randomBoolean, randomByte, randomByteArrayOfLength, randomDateFormatterPattern, randomDateTimeZone, randomDouble, randomDoubleBetween, randomFloat, randomFrom, randomFrom, randomFrom, randomFrom, randomFrom, randomFrom, randomGeohash, randomInt, randomInt, randomIntBetween, randomIp, randomList, randomList, randomLong, randomLongBetween, randomMap, randomMillisUpToYear9999, randomNonEmptySubsetOf, randomNonNegativeByte, randomNonNegativeLong, randomPositiveTimeValue, randomRealisticUnicodeOfCodepointLength, randomRealisticUnicodeOfCodepointLengthBetween, randomRealisticUnicodeOfLength, randomRealisticUnicodeOfLengthBetween, randomShort, randomSubsetOf, randomSubsetOf, randomSubsetOf, randomTimeValue, randomTimeValue, randomTimeValue, randomTimeZone, randomUnicodeOfCodepointLength, randomUnicodeOfCodepointLengthBetween, randomUnicodeOfLength, randomUnicodeOfLengthBetween, randomUnique, randomValueOtherThan, randomValueOtherThanMany, randomZone, removeHeaderWarningAppender, resetCheckIndexStatus, resetPortCounter, restoreContentType, restoreFileSystem, scaledRandomIntBetween, setContentType, setFileSystem, setHeaderWarningAppender, settings, shuffleMap, shuffleXContent, shuffleXContent, spinForAtLeastNMilliseconds, spinForAtLeastOneMillisecond, terminate, terminate, tmpPaths, toShuffledXContent, waitUntil, waitUntil, xContentRegistry

Methods inherited from class org.apache.lucene.util.LuceneTestCase

addVirusChecker, assertDeletedDocsEquals, assertDocsAndPositionsEnumEquals, assertDocsEnumEquals, assertDocsSkippingEquals, assertDocValuesEquals, assertDocValuesEquals, assertFieldInfosEquals, assertNormsEquals, assertPointsEquals, assertPositionsSkippingEquals, assertReaderEquals, assertReaderStatisticsEquals, assertStoredFieldEquals, assertStoredFieldsEquals, assertTermsEnumEquals, assertTermsEquals, assertTermsEquals, assertTermsStatisticsEquals, assertTermStatsEquals, assertTermVectorsEquals, asSet, assumeFalse, assumeNoException, assumeTrue, assumeWorkingMMapOnWindows, atLeast, atLeast, callStackContains, callStackContains, callStackContainsAnyOf, closeAfterSuite, closeAfterTest, collate, createTempDir, createTempDir, createTempFile, createTempFile, dumpArray, dumpIterator, ensureSaneIWCOnNightly, expectThrows, expectThrows, expectThrows, expectThrowsAnyOf, expectThrowsAnyOf, getBaseTempDirForTestClass, getDataInputStream, getOnlyLeafReader, getTestClass, getTestName, hasWorkingMMapOnWindows, isTestThread, localeForLanguageTag, maybeChangeLiveIndexWriterConfig, maybeWrapReader, newAlcoholicMergePolicy, newAlcoholicMergePolicy, newBytesRef, newBytesRef, newBytesRef, newBytesRef, newBytesRef, newBytesRef, newDirectory, newDirectory, newDirectory, newDirectory, newDirectory, newField, newField, newFSDirectory, newFSDirectory, newIndexWriterConfig, newIndexWriterConfig, newIndexWriterConfig, newIOContext, newIOContext, newLogMergePolicy, newLogMergePolicy, newLogMergePolicy, newLogMergePolicy, newLogMergePolicy, newMaybeVirusCheckingDirectory, newMaybeVirusCheckingFSDirectory, newMergePolicy, newMergePolicy, newMergePolicy, newMockDirectory, newMockDirectory, newMockDirectory, newMockFSDirectory, newMockFSDirectory, newSearcher, newSearcher, newSearcher, newSearcher, newStringField, newStringField, newStringField, newStringField, newTextField, newTextField, newTieredMergePolicy, newTieredMergePolicy, overrideDefaultQueryCache, overrideTestDefaultQueryCache, random, randomLocale, randomTimeZone, rarely, rarely, replaceMaxFailureRule, resetDefaultQueryCache, restoreCPUCoreCount, restoreIndexWriterMaxDocs, restoreSpins, runWithRestrictedPermissions, setIndexWriterMaxDocs, setUp, setupCPUCoreCount, setupSpins, slowFileExists, tearDown, usually, usually, wrapReader

Methods inherited from class org.junit.Assert

assertArrayEquals, assertArrayEquals, assertArrayEquals, assertArrayEquals, assertArrayEquals, assertArrayEquals, assertArrayEquals, assertArrayEquals, assertArrayEquals, assertArrayEquals, assertArrayEquals, assertArrayEquals, assertArrayEquals, assertArrayEquals, assertArrayEquals, assertArrayEquals, assertArrayEquals, assertArrayEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertEquals, assertFalse, assertFalse, assertNotEquals, assertNotEquals, assertNotEquals, assertNotEquals, assertNotEquals, assertNotEquals, assertNotEquals, assertNotEquals, assertNotNull, assertNotNull, assertNotSame, assertNotSame, assertNull, assertNull, assertSame, assertSame, assertThat, assertThat, assertTrue, assertTrue, fail, fail

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

valuesSourceRegistry

protected ValuesSourceRegistry valuesSourceRegistry

Constructor Details

AggregatorTestCase

public AggregatorTestCase()

Method Details

initPlugns

public final void initPlugns()

initAnalysisRegistry

public void initAnalysisRegistry()
                          throws Exception

Throws:

Exception

createAnalysisModule

protected AnalysisModule createAnalysisModule()
                                       throws Exception

Returns:

a new analysis module. Tests that require a fully constructed analysis module (used to create an analysis registry) should override this method

Throws:

Exception

getSearchPlugins

protected List<SearchPlugin> getSearchPlugins()

Test cases should override this if they have plugins that need to be loaded, e.g. the plugins their aggregators are in.

createAggregator

protected <A extends Aggregator> A createAggregator(AggregationBuilder aggregationBuilder,
 org.apache.lucene.search.IndexSearcher searcher,
 MappedFieldType... fieldTypes)
                                             throws IOException

Throws:

IOException

createAggregator

protected <A extends Aggregator> A createAggregator(AggregationBuilder builder,
 AggregationContext context)
                                             throws IOException

Throws:

IOException

createAggregationContext

protected AggregationContext createAggregationContext(org.apache.lucene.search.IndexSearcher indexSearcher,
 org.apache.lucene.search.Query query,
 MappedFieldType... fieldTypes)
                                               throws IOException

Create a AggregationContext for testing an Aggregator. While AggregationContext is Releasable the caller is not responsible for releasing it. Instead, it is released automatically in in cleanupReleasables().

Throws:

IOException

createAggregationContext

protected AggregationContext createAggregationContext(org.apache.lucene.search.IndexSearcher indexSearcher,
 IndexSettings indexSettings,
 org.apache.lucene.search.Query query,
 CircuitBreakerService breakerService,
 long bytesToPreallocate,
 int maxBucket,
 MappedFieldType... fieldTypes)
                                               throws IOException

Create a AggregationContext for testing an Aggregator. While AggregationContext is Releasable the caller is not responsible for releasing it. Instead, it is released automatically in in cleanupReleasables().

Throws:

IOException

buildMockFieldMapper

protected FieldMapper buildMockFieldMapper(MappedFieldType ft)

Build a FieldMapper to create the MappingLookup used for the aggs. protected so subclasses can have it.

objectMappers

protected List<ObjectMapper> objectMappers()

ObjectMappers to add to the lookup. By default we don't need any ObjectMappers but testing nested objects will require adding some.

createIndexSettings

protected IndexSettings createIndexSettings()

getMockScriptService

protected ScriptService getMockScriptService()

Sub-tests that need scripting can override this method to provide a script service and pre-baked scripts

searchAndReduce

protected <A extends InternalAggregation,
C extends Aggregator>
A searchAndReduce(org.apache.lucene.search.IndexSearcher searcher,
 org.apache.lucene.search.Query query,
 AggregationBuilder builder,
 MappedFieldType... fieldTypes)
           throws IOException

Throws:

IOException

searchAndReduce

protected <A extends InternalAggregation,
C extends Aggregator>
A searchAndReduce(IndexSettings indexSettings,
 org.apache.lucene.search.IndexSearcher searcher,
 org.apache.lucene.search.Query query,
 AggregationBuilder builder,
 MappedFieldType... fieldTypes)
           throws IOException

Throws:

IOException

searchAndReduce

protected <A extends InternalAggregation,
C extends Aggregator>
A searchAndReduce(org.apache.lucene.search.IndexSearcher searcher,
 org.apache.lucene.search.Query query,
 AggregationBuilder builder,
 int maxBucket,
 MappedFieldType... fieldTypes)
           throws IOException

Throws:

IOException

searchAndReduce

protected <A extends InternalAggregation,
C extends Aggregator>
A searchAndReduce(IndexSettings indexSettings,
 org.apache.lucene.search.IndexSearcher searcher,
 org.apache.lucene.search.Query query,
 AggregationBuilder builder,
 int maxBucket,
 MappedFieldType... fieldTypes)
           throws IOException

Collects all documents that match the provided query Query and returns the reduced InternalAggregation.

Half the time it aggregates each leaf individually and reduces all results together. The other half the time it aggregates across the entire index at once and runs a final reduction on the single resulting agg.

Throws:

IOException

searchAndReduce

protected <A extends InternalAggregation,
C extends Aggregator>
A searchAndReduce(IndexSettings indexSettings,
 org.apache.lucene.search.IndexSearcher searcher,
 org.apache.lucene.search.Query query,
 AggregationBuilder builder,
 int maxBucket,
 boolean splitLeavesIntoSeparateAggregators,
 MappedFieldType... fieldTypes)
           throws IOException

Collects all documents that match the provided query Query and returns the reduced InternalAggregation. It runs the aggregation as well using a circuit breaker that randomly throws CircuitBreakingException in order to mak sure the implementation does not leak.

Parameters:

splitLeavesIntoSeparateAggregators - If true this creates a new Aggregator for each leaf as though it were a separate index. If false this aggregates all leaves together, like we do in production.

Throws:

IOException

doAssertReducedMultiBucketConsumer

protected void doAssertReducedMultiBucketConsumer(Aggregation agg,
 MultiBucketConsumerService.MultiBucketConsumer bucketConsumer)

testCase

protected <T extends AggregationBuilder,
V extends InternalAggregation>
void testCase(T aggregationBuilder,
 org.apache.lucene.search.Query query,
 org.elasticsearch.core.CheckedConsumer<RandomIndexWriter,IOException> buildIndex,
 Consumer<V> verify,
 MappedFieldType... fieldTypes)
       throws IOException

Throws:

IOException

withIndex

protected void withIndex(org.elasticsearch.core.CheckedConsumer<RandomIndexWriter,IOException> buildIndex,
 org.elasticsearch.core.CheckedConsumer<org.apache.lucene.search.IndexSearcher,IOException> consume)
                  throws IOException

Throws:

IOException

withNonMergingIndex

protected void withNonMergingIndex(org.elasticsearch.core.CheckedConsumer<RandomIndexWriter,IOException> buildIndex,
 org.elasticsearch.core.CheckedConsumer<org.apache.lucene.search.IndexSearcher,IOException> consume)
                            throws IOException

Throws:

IOException

debugTestCase

protected <R extends InternalAggregation> void debugTestCase(AggregationBuilder builder,
 org.apache.lucene.search.Query query,
 org.elasticsearch.core.CheckedConsumer<RandomIndexWriter,IOException> buildIndex,
 TriConsumer<R,Class<? extends Aggregator>,Map<String,Map<String,Object>>> verify,
 MappedFieldType... fieldTypes)
                                                      throws IOException

Execute and aggregation and collect its debug information. Unlike testCase(T, org.apache.lucene.search.Query, org.elasticsearch.core.CheckedConsumer<org.apache.lucene.index.RandomIndexWriter, java.io.IOException>, java.util.function.Consumer<V>, org.elasticsearch.index.mapper.MappedFieldType...) this doesn't randomly create an Aggregator per leaf and perform partial reductions. It always creates a single Aggregator so we can get consistent debug info.

Throws:

IOException

debugTestCase

protected <R extends InternalAggregation> void debugTestCase(AggregationBuilder builder,
 org.apache.lucene.search.Query query,
 org.apache.lucene.search.IndexSearcher searcher,
 TriConsumer<R,Class<? extends Aggregator>,Map<String,Map<String,Object>>> verify,
 MappedFieldType... fieldTypes)
                                                      throws IOException

Execute and aggregation and collect its debug information. Unlike testCase(T, org.apache.lucene.search.Query, org.elasticsearch.core.CheckedConsumer<org.apache.lucene.index.RandomIndexWriter, java.io.IOException>, java.util.function.Consumer<V>, org.elasticsearch.index.mapper.MappedFieldType...) this doesn't randomly create an Aggregator per leaf and perform partial reductions. It always creates a single Aggregator so we can get consistent debug info.

Throws:

IOException

withAggregator

protected void withAggregator(AggregationBuilder aggregationBuilder,
 org.apache.lucene.search.Query query,
 org.elasticsearch.core.CheckedConsumer<RandomIndexWriter,IOException> buildIndex,
 CheckedBiConsumer<org.apache.lucene.search.IndexSearcher,Aggregator,IOException> verify,
 MappedFieldType... fieldTypes)
                       throws IOException

Throws:

IOException

verifyOutputFieldNames

protected <T extends AggregationBuilder,
V extends InternalAggregation>
void verifyOutputFieldNames(T aggregationBuilder,
 V agg)
                     throws IOException

Throws:

IOException

wrapDirectoryReader

protected IndexReader wrapDirectoryReader(DirectoryReader reader)
                                   throws IOException

Override to wrap the DirectoryReader for aggs like NestedAggregationBuilder.

Throws:

IOException

wrapInMockESDirectoryReader

protected static DirectoryReader wrapInMockESDirectoryReader(DirectoryReader directoryReader)
                                                      throws IOException

Throws:

IOException

newIndexSearcher

protected static org.apache.lucene.search.IndexSearcher newIndexSearcher(IndexReader indexReader)

Added to randomly run with more assertions on the index searcher level, like LuceneTestCase.newSearcher(IndexReader), which can't be used because it also wraps in the IndexSearcher's IndexReader with other implementations that we can't handle. (e.g. ParallelCompositeReader)

maybeWrapReaderEs

protected static IndexReader maybeWrapReaderEs(DirectoryReader reader)
                                        throws IOException

Added to randomly run with more assertions on the index reader level, like LuceneTestCase.wrapReader(IndexReader), which can't be used because it also wraps in the IndexReader with other implementations that we can't handle. (e.g. ParallelCompositeReader)

Throws:

IOException

getSupportedValuesSourceTypes

protected List<ValuesSourceType> getSupportedValuesSourceTypes()

Implementors should return a list of ValuesSourceType that the aggregator supports. This is used to test the matrix of supported/unsupported field types against the aggregator and verify it works (or doesn't) as expected. If this method is implemented, createAggBuilderForTypeTest(MappedFieldType, String) should be implemented as well.

Returns:

list of supported ValuesSourceTypes

createAggBuilderForTypeTest

protected AggregationBuilder createAggBuilderForTypeTest(MappedFieldType fieldType,
 String fieldName)

This method is invoked each time a field type is tested in testSupportedFieldTypes(). The field type and name are provided, and the implementor is expected to return an AggBuilder accordingly. The AggBuilder should be returned even if the aggregation does not support the field type, because the test will check if an exception is thrown in that case. The list of supported types are provided by getSupportedValuesSourceTypes(), which must also be implemented.

Parameters:

fieldType - the type of the field that will be tested

fieldName - the name of the field that will be test

Returns:

an aggregation builder to test against the field

unsupportedMappedFieldTypes

protected List<String> unsupportedMappedFieldTypes()

A method that allows implementors to specifically blacklist particular field types (based on their content_name). This is needed in some areas where the ValuesSourceType is not granular enough, for example integer values vs floating points, or `keyword` bytes vs `binary` bytes (which are not searchable) This is a blacklist instead of a whitelist because there are vastly more field types than ValuesSourceTypes, and it's expected that these unsupported cases are exceptional rather than common

testSupportedFieldTypes

public void testSupportedFieldTypes()
                             throws IOException

This test will validate that an aggregator succeeds or fails to run against all the field types that are registered in IndicesModule (e.g. all the core field types). An aggregator is provided by the implementor class, and it is executed against each field type in turn. If an exception is thrown when the field is supported, that will fail the test. Similarly, if an exception _is not_ thrown when a field is unsupported, that will also fail the test. Exception types/messages are not currently checked, just presence/absence of an exception.

Throws:

IOException

cleanupReleasables

public void cleanupReleasables()

afterClose

protected void afterClose()

Hook for checking things after all Aggregators have been closed.

dateField

protected DateFieldMapper.DateFieldType dateField(String name,
 DateFieldMapper.Resolution resolution)

Make a DateFieldMapper.DateFieldType for a date.

doubleField

protected NumberFieldMapper.NumberFieldType doubleField(String name)

Make a NumberFieldMapper.NumberFieldType for a double.

geoPointField

protected GeoPointFieldMapper.GeoPointFieldType geoPointField(String name)

Make a GeoPointFieldMapper.GeoPointFieldType for a geo_point.

keywordField

protected KeywordFieldMapper.KeywordFieldType keywordField(String name)

Make a DateFieldMapper.DateFieldType for a date.

longField

protected NumberFieldMapper.NumberFieldType longField(String name)

Make a NumberFieldMapper.NumberFieldType for a long.

rangeField

protected RangeFieldMapper.RangeFieldType rangeField(String name,
 RangeType rangeType)

Make a NumberFieldMapper.NumberFieldType for a range.

writableRegistry

protected final NamedWriteableRegistry writableRegistry()

Description copied from class: ESTestCase

The NamedWriteableRegistry to use for this test. Subclasses should override and use liberally.

Overrides:

writableRegistry in class ESTestCase

aggCardinalityUpperBound

public static AggregationBuilder aggCardinalityUpperBound(String name)

Request an aggregation that returns the CardinalityUpperBound that was passed to its ctor.