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

LUCENE-9103: WANDScorer can miss some hits in some rare conditions.

This commit is contained in:
Adrien Grand 2019-12-19 16:51:02 +01:00
parent 2f7276f935
commit 907d1142fa
2 changed files with 32 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
lucene/CHANGES.txt
View File

@ -174,6 +174,8 @@ Bug Fixes
* LUCENE-9055: Fix the detection of lines crossing triangles through edge points. * LUCENE-9055: Fix the detection of lines crossing triangles through edge points.
(Ignacio Vera) (Ignacio Vera)
* LUCENE-9103: Disjunctions can miss some hits in some rare conditions. (Adrien Grand)
Other Other
* LUCENE-8979: Code Cleanup: Use entryset for map iteration wherever possible. - Part 2 (Koen De Groote) * LUCENE-8979: Code Cleanup: Use entryset for map iteration wherever possible. - Part 2 (Koen De Groote)

44
lucene/core/src/java/org/apache/lucene/search/WANDScorer.java
View File

@ -185,6 +185,7 @@ final class WANDScorer extends Scorer {
} }
assert minCompetitiveScore == 0 || tailMaxScore < minCompetitiveScore; assert minCompetitiveScore == 0 || tailMaxScore < minCompetitiveScore;
assert doc <= upTo;
return true; return true;
} }
@ -374,17 +375,34 @@ final class WANDScorer extends Scorer {
} }
} }
/**
* Update {@code upTo} and maximum scores of sub scorers so that {@code upTo}
* is greater than or equal to the next candidate after {@code target}, i.e.
* the top of `head`.
*/
private void updateMaxScoresIfNecessary(int target) throws IOException { private void updateMaxScoresIfNecessary(int target) throws IOException {
assert lead == null; assert lead == null;
if (head.size() == 0) { // no matches in the current block while (upTo < DocIdSetIterator.NO_MORE_DOCS) {
if (upTo != DocIdSetIterator.NO_MORE_DOCS) { if (head.size() == 0) {
updateMaxScores(Math.max(target, upTo + 1)); // All clauses could fit in the tail, which means that the sum of the
// maximum scores of sub clauses is less than the minimum competitive score.
// Move to the next block until this condition becomes false.
target = Math.max(target, upTo + 1);
updateMaxScores(target);
} else if (head.top().doc > upTo) {
// We have a next candidate but it's not in the current block. We need to
// move to the next block in order to not miss any potential hits between
// `target` and `head.top().doc`.
assert head.top().doc >= target;
updateMaxScores(target);
break;
} else {
break;
} }
} else if (head.top().doc > upTo) { // the next candidate is in a different block
assert head.top().doc >= target;
updateMaxScores(target);
} }
assert upTo == DocIdSetIterator.NO_MORE_DOCS || (head.size() > 0 && head.top().doc <= upTo);
} }
/** Set 'doc' to the next potential match, and move all disis of 'head' that /** Set 'doc' to the next potential match, and move all disis of 'head' that
@ -394,14 +412,12 @@ final class WANDScorer extends Scorer {
updateMaxScoresIfNecessary(target); updateMaxScoresIfNecessary(target);
assert upTo >= target; assert upTo >= target;
// If the head is empty, it means that the sum of all max scores is not // updateMaxScores tries to move forward until a block with matches is found
// enough to produce a competitive score. So we jump to the next block. // so if the head is empty it means there are no matches at all anymore
while (head.size() == 0) { if (head.size() == 0) {
if (upTo == DocIdSetIterator.NO_MORE_DOCS) { assert upTo == DocIdSetIterator.NO_MORE_DOCS;
doc = DocIdSetIterator.NO_MORE_DOCS; doc = DocIdSetIterator.NO_MORE_DOCS;
return; return;
}
updateMaxScores(upTo + 1);
} }
// The top of `head` defines the next potential match // The top of `head` defines the next potential match