Package org.apache.lucene.search

Class CheckHits

java.lang.Object

org.apache.lucene.search.CheckHits

public class CheckHits
extends Object

Utility class for asserting expected hits in tests.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	CheckHits.ExplanationAsserter	Asserts that the score explanation for every document matching a query corresponds with the true score.
static class	CheckHits.ExplanationAssertingSearcher	an IndexSearcher that implicitly checks hte explanation of every match whenever it executes a search.
static class	CheckHits.MatchesAsserter	Asserts that the Matches from a query is non-null whenever the document its created for is a hit.
static class	CheckHits.SetCollector	Just collects document ids into a set.

Field Summary

Fields

Modifier and Type	Field	Description
static float	EXPLAIN_SCORE_TOLERANCE_DELTA	Some explains methods calculate their values though a slightly different order of operations from the actual scoring method ...
static float	EXPLAIN_SCORE_TOLERANCE_MINIMUM	In general we use a relative epsilon, but some tests do crazy things like boost documents with 0, creating tiny tiny scores where the relative difference is large but the absolute difference is tiny.

Constructor Summary

Constructors

Constructor	Description
CheckHits()

Method Summary

Modifier and Type	Method	Description
static void	checkDocIds(String mes, int[] results, ScoreDoc[] hits)	Tests that a Hits has an expected order of documents
static void	checkEqual(Query query, ScoreDoc[] hits1, ScoreDoc[] hits2)
static void	checkExplanations(Query query, String defaultFieldName, IndexSearcher searcher)	Asserts that the explanation value for every document matching a query corresponds with the true score.
static void	checkExplanations(Query query, String defaultFieldName, IndexSearcher searcher, boolean deep)	Asserts that the explanation value for every document matching a query corresponds with the true score.
static void	checkHitCollector(Random random, Query query, String defaultFieldName, IndexSearcher searcher, int[] results)	Tests that a query matches the an expected set of documents using a HitCollector.
static void	checkHits(Random random, Query query, String defaultFieldName, IndexSearcher searcher, int[] results)	Tests that a query matches the an expected set of documents using Hits.
static void	checkHitsQuery(Query query, ScoreDoc[] hits1, ScoreDoc[] hits2, int[] results)	Tests that two queries have an expected order of documents, and that the two queries have the same score values.
static void	checkMatches(Query query, IndexSearcher searcher)	Asserts that the result of calling Weight.matches(LeafReaderContext, int) for every document matching a query returns a non-null Matches
static void	checkNoMatchExplanations(Query q, String defaultFieldName, IndexSearcher searcher, int[] results)	Tests that all documents up to maxDoc which are *not* in the expected result set, have an explanation which indicates that the document does not match
static float	explainToleranceDelta(float f1, float f2)	returns a reasonable epsilon for comparing two floats, where minor differences are acceptable such as score vs.
static String	hits2str(ScoreDoc[] hits1, ScoreDoc[] hits2, int start, int end)
static String	topdocsString(TopDocs docs, int start, int end)
static void	verifyExplanation(String q, int doc, float score, boolean deep, Explanation expl)	Assert that an explanation has the expected score, and optionally that its sub-details max/sum/factor match to that score.

Methods inherited from class java.lang.Object

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

Field Detail

EXPLAIN_SCORE_TOLERANCE_DELTA

public static float EXPLAIN_SCORE_TOLERANCE_DELTA

Some explains methods calculate their values though a slightly different order of operations from the actual scoring method ... this allows for a small amount of relative variation

EXPLAIN_SCORE_TOLERANCE_MINIMUM

public static float EXPLAIN_SCORE_TOLERANCE_MINIMUM

In general we use a relative epsilon, but some tests do crazy things like boost documents with 0, creating tiny tiny scores where the relative difference is large but the absolute difference is tiny. we ensure the the epsilon is always at least this big.

Constructor Detail

CheckHits

public CheckHits()

Method Detail

checkNoMatchExplanations

public static void checkNoMatchExplanations(Query q,
                                            String defaultFieldName,
                                            IndexSearcher searcher,
                                            int[] results)
                                     throws IOException

Tests that all documents up to maxDoc which are *not* in the expected result set, have an explanation which indicates that the document does not match

Throws:

IOException

checkHitCollector

public static void checkHitCollector(Random random,
                                     Query query,
                                     String defaultFieldName,
                                     IndexSearcher searcher,
                                     int[] results)
                              throws IOException

Tests that a query matches the an expected set of documents using a HitCollector.

Note that when using the HitCollector API, documents will be collected if they "match" regardless of what their score is.

Parameters:

query - the query to test

searcher - the searcher to test the query against

defaultFieldName - used for displaying the query in assertion messages

results - a list of documentIds that must match the query

Throws:

IOException

See Also:

checkHits(java.util.Random, org.apache.lucene.search.Query, java.lang.String, org.apache.lucene.search.IndexSearcher, int[])

checkHits

public static void checkHits(Random random,
                             Query query,
                             String defaultFieldName,
                             IndexSearcher searcher,
                             int[] results)
                      throws IOException

Tests that a query matches the an expected set of documents using Hits.

Note that when using the Hits API, documents will only be returned if they have a positive normalized score.

Parameters:

query - the query to test

searcher - the searcher to test the query against

defaultFieldName - used for displaing the query in assertion messages

results - a list of documentIds that must match the query

Throws:

IOException

See Also:

checkHitCollector(java.util.Random, org.apache.lucene.search.Query, java.lang.String, org.apache.lucene.search.IndexSearcher, int[])

checkDocIds

public static void checkDocIds(String mes,
                               int[] results,
                               ScoreDoc[] hits)

Tests that a Hits has an expected order of documents

checkHitsQuery

public static void checkHitsQuery(Query query,
                                  ScoreDoc[] hits1,
                                  ScoreDoc[] hits2,
                                  int[] results)

Tests that two queries have an expected order of documents, and that the two queries have the same score values.

checkEqual

public static void checkEqual(Query query,
                              ScoreDoc[] hits1,
                              ScoreDoc[] hits2)

hits2str

public static String hits2str(ScoreDoc[] hits1,
                              ScoreDoc[] hits2,
                              int start,
                              int end)

topdocsString

public static String topdocsString(TopDocs docs,
                                   int start,
                                   int end)

checkExplanations

public static void checkExplanations(Query query,
                                     String defaultFieldName,
                                     IndexSearcher searcher)
                              throws IOException

Asserts that the explanation value for every document matching a query corresponds with the true score.

Parameters:

query - the query to test

searcher - the searcher to test the query against

defaultFieldName - used for displaing the query in assertion messages

Throws:

IOException

See Also:

CheckHits.ExplanationAsserter, for a "deep" testing of the explanation details.

checkExplanations

public static void checkExplanations(Query query,
                                     String defaultFieldName,
                                     IndexSearcher searcher,
                                     boolean deep)
                              throws IOException

Asserts that the explanation value for every document matching a query corresponds with the true score. Optionally does "deep" testing of the explanation details.

Parameters:

query - the query to test

searcher - the searcher to test the query against

defaultFieldName - used for displaing the query in assertion messages

deep - indicates whether a deep comparison of sub-Explanation details should be executed

Throws:

IOException

See Also:

CheckHits.ExplanationAsserter

explainToleranceDelta

public static float explainToleranceDelta(float f1,
                                          float f2)

returns a reasonable epsilon for comparing two floats, where minor differences are acceptable such as score vs. explain

checkMatches

public static void checkMatches(Query query,
                                IndexSearcher searcher)
                         throws IOException

Asserts that the result of calling Weight.matches(LeafReaderContext, int) for every document matching a query returns a non-null Matches

Parameters:

query - the query to test

searcher - the search to test against

Throws:

IOException

verifyExplanation

public static void verifyExplanation(String q,
                                     int doc,
                                     float score,
                                     boolean deep,
                                     Explanation expl)

Assert that an explanation has the expected score, and optionally that its sub-details max/sum/factor match to that score.

Parameters:

q - String representation of the query for assertion messages

doc - Document ID for assertion messages

score - Real score value of doc with query q

deep - indicates whether a deep comparison of sub-Explanation details should be executed

expl - The Explanation to match against score