Copilot commented on code in PR #1104: URL: https://github.com/apache/opennlp/pull/1104#discussion_r3449578770
########## opennlp-core/opennlp-runtime/src/main/java/opennlp/tools/tokenize/uax29/ExtendedPictographic.java: ########## @@ -0,0 +1,87 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ +package opennlp.tools.tokenize.uax29; + +import java.io.BufferedReader; +import java.io.IOException; +import java.io.InputStream; +import java.io.InputStreamReader; +import java.io.UncheckedIOException; +import java.nio.charset.StandardCharsets; +import java.util.BitSet; + +/** + * Tests the Unicode {@code Extended_Pictographic} property of a code point. + * + * <p>This is the one extra property the word boundary algorithm needs (rule WB3c), to keep emoji + * zero-width-joiner sequences together. The data is loaded once from the {@code emoji-data.txt} + * derived resource of the Unicode Character Database and stored in a {@link BitSet}, so membership + * is an O(1) bit test.</p> + */ +public final class ExtendedPictographic { + + private static final String RESOURCE = "ExtendedPictographic.txt"; + + private static final BitSet MEMBERS = new BitSet(); + + static { + try (InputStream in = ExtendedPictographic.class.getResourceAsStream(RESOURCE)) { + if (in == null) { + throw new IllegalStateException("Missing Extended_Pictographic data resource: " + RESOURCE); + } + load(in); + } catch (IOException e) { + throw new UncheckedIOException("Unable to read Extended_Pictographic data resource", e); + } Review Comment: The `UncheckedIOException` message omits the resource name, which makes it harder to diagnose failures when the resource is unreadable/corrupted. Include `RESOURCE` in the message (consistent with `WordBreakProperty`). ########## opennlp-core/opennlp-runtime/src/main/java/opennlp/tools/util/normalizer/NormalizationProfiles.java: ########## @@ -0,0 +1,120 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ +package opennlp.tools.util.normalizer; + +import java.util.HashMap; +import java.util.Locale; +import java.util.Map; +import java.util.MissingResourceException; +import java.util.Objects; +import java.util.Optional; +import java.util.Set; + +import opennlp.tools.langdetect.LanguageDetector; +import opennlp.tools.stemmer.snowball.SnowballStemmer; + +/** + * A registry of {@link NormalizationProfile}s by language, with detection-based fallback. This is + * the language dispatch the design note calls for: pick the profile for a requested language, or + * detect the language with a {@link LanguageDetector} when it is unspecified. The covered languages + * are exactly those with a Snowball stemmer. + * + * <p>Profiles are keyed by ISO 639-3 code (what {@link LanguageDetector} produces); + * {@link #forLanguage(String)} also accepts ISO 639-1 two-letter codes.</p> + */ +public final class NormalizationProfiles { + + private static final Map<String, NormalizationProfile> BY_LANGUAGE = build(); + + private NormalizationProfiles() { + } + + private static Map<String, NormalizationProfile> build() { + final Map<String, NormalizationProfile> map = new HashMap<>(); + // The generic accent fold is used for English and the major Romance languages, German uses its + // own ae/oe/ue/ss fold, and folding is disabled elsewhere (Nordic, non-Latin) where diacritics + // mark distinct letters. + final CharSequenceNormalizer latin = AccentFoldCharSequenceNormalizer.getInstance(); + final CharSequenceNormalizer german = GermanUmlautCharSequenceNormalizer.getInstance(); + add(map, "ara", SnowballStemmer.ALGORITHM.ARABIC, null); + add(map, "cat", SnowballStemmer.ALGORITHM.CATALAN, latin); + add(map, "dan", SnowballStemmer.ALGORITHM.DANISH, null); + add(map, "deu", SnowballStemmer.ALGORITHM.GERMAN, german); + add(map, "ell", SnowballStemmer.ALGORITHM.GREEK, null); + add(map, "eng", SnowballStemmer.ALGORITHM.ENGLISH, latin); + add(map, "fin", SnowballStemmer.ALGORITHM.FINNISH, null); + add(map, "fra", SnowballStemmer.ALGORITHM.FRENCH, latin); + add(map, "gle", SnowballStemmer.ALGORITHM.IRISH, null); + add(map, "hun", SnowballStemmer.ALGORITHM.HUNGARIAN, null); + add(map, "ind", SnowballStemmer.ALGORITHM.INDONESIAN, null); + add(map, "ita", SnowballStemmer.ALGORITHM.ITALIAN, latin); + add(map, "nld", SnowballStemmer.ALGORITHM.DUTCH, null); + add(map, "nor", SnowballStemmer.ALGORITHM.NORWEGIAN, null); + add(map, "por", SnowballStemmer.ALGORITHM.PORTUGUESE, latin); + add(map, "ron", SnowballStemmer.ALGORITHM.ROMANIAN, null); + add(map, "rus", SnowballStemmer.ALGORITHM.RUSSIAN, null); + add(map, "spa", SnowballStemmer.ALGORITHM.SPANISH, latin); + add(map, "swe", SnowballStemmer.ALGORITHM.SWEDISH, null); + return Map.copyOf(map); + } + + private static void add(Map<String, NormalizationProfile> map, String language, + SnowballStemmer.ALGORITHM algorithm, CharSequenceNormalizer accentFold) { + map.put(language, new NormalizationProfile(language, algorithm, accentFold)); + } + + /** + * Returns the profile for a language. + * + * @param language An ISO 639-3 or ISO 639-1 language code; case-insensitive. + * @return The profile, or empty if the language has no Snowball stemmer. + */ + public static Optional<NormalizationProfile> forLanguage(String language) { + Objects.requireNonNull(language, "language"); + String code = language.strip().toLowerCase(Locale.ROOT); + if (code.length() == 2) { + try { + final String iso3 = Locale.of(code).getISO3Language(); + if (!iso3.isEmpty()) { + code = iso3; + } + } catch (MissingResourceException ignored) { + // No ISO 639-3 code for this two-letter code; fall through and look up as given. + } + } + return Optional.ofNullable(BY_LANGUAGE.get(code)); + } + + /** + * Detects the language of {@code text} and returns its profile. + * + * @param text The text to detect. + * @param detector The language detector to use. + * @return The profile for the detected language, or empty if it has no Snowball stemmer. + */ + public static Optional<NormalizationProfile> detect(CharSequence text, + LanguageDetector detector) { + return forLanguage(detector.predictLanguage(text).getLang()); + } Review Comment: `detect(...)` is a public entry point but doesn't validate `text` or `detector`. If either is null, the method will throw a `NullPointerException` from inside `predictLanguage(...)`/`getLang()`, which is harder to diagnose than a clear parameter error. Add explicit `Objects.requireNonNull` checks to match `forLanguage(...)`'s fail-fast behavior. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
