Hunspell getAllWordForms, generateAllSimpleWords and compress return words that spell() rejects: the sanity check canStemToOriginal accepts any affix path that strips a form back to the stem text, so with 'NEEDAFFIX X', 'SFX A 0 s/XB', an unrelated 'SFX C 0 s' and foo/A, getAllWordForms("foo") returns foos

Finding LFV-638Area: analysis / hunspell (WordFormGenerator: canStemToOriginal, getAllWordForms, generateAllSimpleWords, compress)Priority: lowImpact: silent wrong resultsNovelty: new

Finding page for LFV-638

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

What it shows

You load a Hunspell dictionary with 'NEEDAFFIX X', 'SFX A 0 s/XB' (so foos needs one more suffix), 'SFX B 0 en' and an unrelated class 'SFX C 0 s', with the entries foo/A and bar/C, and call Hunspell.getAllWordForms("foo"). You expect [foo, foosen]: spell() rejects foos (C Hunspell 1.7.2 does too), and the generator's check canStemToOriginal is documented to keep only forms that are 'accepted by the spell-checker and analyzed to be the form of the original dictionary entry'. Instead you get [foo, foos, foosen], because canStemToOriginal accepts as soon as any affix-stripping path gives back the stem text, here class C, which foo does not even have, and ignores which affixes were stripped (WordFormGenerator.java:211); generateAllSimpleWords leaks the same words, and compress([kat, kater, katern]) suggests kat/PQ, which does not accept kater.

The test shows that with NEEDAFFIX X, SFX A 0 s/XB, SFX B 0 en, an unused SFX C 0 s and the entry foo/A, Hunspell.getAllWordForms("foo") returns foos, which the same dictionary's spell() rejects.

Expected vs actual

Expected: every generated form passes spell() (empty list of rejected forms); actual (-ea and -da): [foos], because canStemToOriginal accepts class C's path that strips foos back to the text foo.

Files

• src/org/apache/lucene/analysis/hunspell/TestWordFormsNeedAffixLeak.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/analysis/common/build/libs/lucene-analysis-common-$V.jar
CP=$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.analysis.hunspell.TestWordFormsNeedAffixLeak

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 2026-11-04. AI-assisted; not yet triaged by a human Lucene committer. Feedback: open an issue in lucene-ai-bug-hunt.