FreeTextSuggester.decodeWeight casts n-gram counts to int, so a word or context seen more than Integer.MAX_VALUE times gets a negative or wrapped count: lookup returns scores outside the documented 0.0 to 1.0 range and can rank the most frequent word last

Finding LFV-661Area: suggest / FreeTextSuggester (n-gram counts stored as long FST outputs; lookup-side decode)Priority: lowImpact: silent wrong resultsNovelty: known upstream

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

What it shows

You build a FreeTextSuggester from text in which one word occurs more than Integer.MAX_VALUE (2,147,483,647) times and expect lookup to rank that word first with a score near 1.0, because the class javadoc says each result value is Long.MAX_VALUE times a score from 0.0 to 1.0 and build() counts n-grams as long. Instead decodeWeight returns (int) (Long.MAX_VALUE - output), so a count of 2,148,532,224 comes back as -2,146,435,072 and lookup('a', 2) returns [ab score 0.0, a score -0.999]: a negative score, with the most frequent word ranked below a word seen once. The same cast decodes context counts, so every prediction from a context seen that often is scored wrongly, and counts above 2^32 wrap to small positive numbers that stay inside the 0..1 range and are silently wrong.

Expected vs actual

Expected: The test passes: it asserts the documented behaviour.
Actual: decodeWeight returns (int) (Long.MAX_VALUE - output), so a count of 2,148,532,224 comes back as -2,146,435,072 and lookup('a', 2) returns [ab score 0.0, a score -0.999]: a negative score, with the most frequent word ranked below a word seen once.

Files

src/org/apache/lucene/search/suggest/analyzing/TestFreeTextSuggesterCountAboveIntMax.java (JUnit test)
src/FreeTextCountOverflow.java (main() class)

Run it

Tested on Lucene main e1faeba08bde (2026-11-02): still reproduces.
Tested on Lucene main a9779ef3c193 (2026-10-26): reproduced.
How to read the outcome: a failing test is the finding reproducing (the failure message states expected against actual); a main() class that exits non-zero, with an exception or AssertionError, is the finding reproducing.
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 off (production-like), seed DEADBEEF
java -da -dsa -Dtests.asserts=false --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.suggest.analyzing.TestFreeTextSuggesterCountAboveIntMax

# run 2: assertions on, seed DEADBEEF, default flags
java -ea -esa --add-modules jdk.incubator.vector --enable-native-access=ALL-UNNAMED -Dtests.seed=DEADBEEF \
     -cp "classes:$CP" FreeTextCountOverflow

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.