SweetSpotSimilarity.lengthNorm adds abs(x-min) + abs(x-max) in int, so a field of 1,073,741,848 or more terms (reachable with one custom term frequency) gets a NaN norm: the document is never collected, and default top-k search also skips competitive documents in the same postings block and skip range

Finding LFV-754Area: misc / similarities (SweetSpotSimilarity lengthNorm; NaN effect on impacts-based top-k pruning in core search)Priority: lowImpact: silent wrong resultsNovelty: new

Finding page for LFV-754

AI-assisted finding, confirmed by automated checks, not yet triaged by a human Lucene committer. This site files nothing upstream.

What it shows

You score with SweetSpotSimilarity (default factors) and index one document whose DOCS_AND_FREQS field has a length of at least 1,073,741,848, for example one token with custom term frequency 1.1e9 or about 1.07e9 ordinary tokens, expecting the documented lengthNorm 1/sqrt(x) (about 3.05e-5, the ClassicSimilarity value) and an ordinary, low-ranked hit. Instead lengthNorm's int sum |x-1| + |x-1| wraps negative, so the norm and the score are NaN and the document is never collected (exhaustive collection returns 5,999 of 6,000 matches); because the NaN also makes the postings block's maximum score NaN, default top-k search skips that block and the skip range above it, so search(q, 7) on a 10,000-document index misses three documents that score 1.0 and returns documents that score 0.5 instead. ClassicSimilarity on the same index is correct.

The test shows that SweetSpotSimilarity.lengthNorm with the default factors returns NaN for a field length above 2^30, because abs(x-min) + abs(x-max) overflows int.

Expected vs actual

Expected 1/sqrt(1073741825) = 3.0517578E-5 (the javadoc says the default degrades to 1/sqrt(x)); actual NaN.

Files

• src/org/apache/lucene/misc/TestSweetSpotLengthNormOverflow.java (JUnit test)

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/misc/build/libs/lucene-misc-$V.jar
CP=$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.misc.TestSweetSpotLengthNormOverflow

Gradle 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 2026-11-03. AI-assisted; not yet triaged by a human Lucene committer. Feedback: open an issue in lucene-ai-bug-hunt.