Nori's DictionaryBuilder checks that rows sharing a left context id have the same part of speech only with a Java assert, so with assertions off a custom system dictionary silently gives those words the part of speech of another row and KoreanPartOfSpeechStopFilter drops or keeps them wrongly

Finding LFV-717Area: analysis / nori system dictionary builder (DictionaryBuilder, TokenInfoDictionaryEntryWriter; custom system dictionaries)Priority: lowImpact: silent wrong resultsNovelty: new

Finding page for LFV-717

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

What it shows

You compile a custom nori system dictionary with DictionaryBuilder from MeCab-format rows in which the noun 가 (NNG) and the particle 는 (JX) share left context id 1, and expect the builder either to reject the source (the compiled format keeps only one part of speech per left id, and kuromoji's builder throws 'Multiple entries found for leftID' for the same case) or each word to keep its own part of speech. Instead, with assertions off, the builder overwrites the single slot for left id 1 with the part of speech of the row it writes last, because the only check is an assert (TokenInfoDictionaryEntryWriter.java:100): 가 then reports left POS JX, and KoreanTokenizer followed by KoreanPartOfSpeechStopFilter with DEFAULT_STOP_TAGS returns no token instead of [가]. With -ea the build fails with a bare AssertionError rather than an IllegalArgumentException that names the problem.

The test shows that nori's DictionaryBuilder checks a left-id/part-of-speech collision (가 NNG and 는 JX on left id 1) only with a Java assert: with -da the build succeeds and 가 silently gets the part of speech of the last row written.

Expected vs actual

Expected: an IllegalArgumentException from DictionaryBuilder.build, or 가 keeps NNG. Actual: with -da expected:<NNG> but was:<JX>; with -ea a bare AssertionError at TokenInfoDictionaryEntryWriter.putEntry:100.

Files

Run it

# 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:$LUCENE/lucene/analysis/nori/build/libs/lucene-analysis-nori-$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.ko.dict.TestNoriBuilderLeftIdPosCollision

The module(s) analysis-nori also need their third-party runtime jars; ./gradlew -p lucene/<module> dependencies lists them.

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).

Formal models and scripts this finding's part of the hunt used: cjk-compiled-dictionary-source-round-trip

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.