DirectSpellChecker with a non-default StringDistance prunes candidates by comparing the metric's score with FuzzyTermsEnum boosts, so better candidates are never inspected (regression since 7.0)
Finding LFV-240Area: suggest / spell checking (DirectSpellChecker.suggestSimilar with a non-default StringDistance; FuzzyTermsEnum competitive-boost pruning)Priority: lowImpact: silent wrong resultsNovelty: new
AI-assisted finding, confirmed by automated checks, not yet triaged by a human Lucene committer. This site files nothing upstream.
What it shows
You give DirectSpellChecker a metric such as LevenshteinDistance or JaroWinklerDistance and expect what setDistance documents: candidates drawn from the term dictionary by Damerau-Levenshtein (the internal similarity), then ranked by your metric. Since 7.0 the code stores your metric's score in ScoreTerm.boost, a field documented as the internal similarity, and then skips later candidates and tells FuzzyTermsEnum to lower its edit distance by comparing their internal similarity with that number, which is on a different scale, so better terms are never looked at. With default settings, the index {lucen, lucena, lucenb, lucenc, lucend, lucenes} and the query 'lucene', LevenshteinDistance returns 'lucen' (0.833) instead of 'lucenes' (0.857), and with maxInspections=1 and the index {lucen, lucenes} all three shipped metrics return 'lucen' although 'lucenes' scores higher under the internal similarity and under each metric.
The test shows that DirectSpellChecker with LevenshteinDistance and setMaxInspections(1) never inspects the better candidate, because it compares the metric's score with FuzzyTermsEnum's internal-similarity boosts to prune candidates.
Expected vs actual
Expected: suggestSimilar('abcd', 1, SUGGEST_ALWAYS) on the index {abc, abcde} returns abcde (closer under both similarities). Actual: abc (expected:<abc[de]> but was:<abc[]>).
Files
Run it
- Tested on Lucene main e1faeba08bde (2026-11-02): still reproduces.
- Tested on Lucene main 645d4fe14a20 (2026-10-28): reproduced.
- Recorded outcome of run 1 on Lucene main 645d4fe14a20: 1 test(s) run, 1 failed.
- How to read the outcome: a failing test is the finding reproducing (the failure message states expected against actual).
- Seeds: a rerun with the same seed replays the same random choices. Assertions off (-da -dsa -Dtests.asserts=false) is the production-like setting.
# Lucene checkout at the tested commit, built with: ./gradlew --no-daemon assemble (JDK 25)
LUCENE=/path/to/lucene; V=12.0.0-SNAPSHOT
CP=$LUCENE/lucene/core/build/libs/lucene-core-$V.jar
CP=$CP:$LUCENE/lucene/codecs/build/libs/lucene-codecs-$V.jar
CP=$CP:$LUCENE/lucene/test-framework/build/libs/lucene-test-framework-$V.jar
CP=$CP:$LUCENE/lucene/analysis/common/build/libs/lucene-analysis-common-$V.jar
CP=$CP:$LUCENE/lucene/suggest/build/libs/lucene-suggest-$V.jar
CP=$CP:randomizedtesting-runner-2.9.2.jar:junit-4.13.2.jar:hamcrest-3.0.jar # test-framework dependencies (Maven Central)
CP=$CP:junit-platform-console-standalone-6.1.3.jar # JUnit Platform console launcher (Maven Central)
javac -d classes -cp "$CP" $(find src -name '*.java')
# run 1: assertions on, seed DEADBEEF
java -ea -esa --add-modules jdk.incubator.vector --enable-native-access=ALL-UNNAMED -Dtests.seed=DEADBEEF \
-cp "classes:$CP" org.junit.platform.console.ConsoleLauncher \
execute --select-class org.apache.lucene.search.spell.TestDirectSpellCheckerLevenshteinDistanceGradle alternative: copy the test into the matching module's src/test/ tree of the checkout and run ./gradlew -p lucene/<module> test --tests <class> -Ptests.seed=<seed> (add -Ptests.asserts=false for a run without assertions).
Feedback
Open an issue in lucene-ai-bug-hunt. Please do not open issues or pull requests in apache/lucene about this finding unless a Lucene committer agrees it should be filed.
Snapshot of 2026-11-04. AI-assisted; not yet triaged by a human Lucene committer. Feedback: open an issue in lucene-ai-bug-hunt.