001///////////////////////////////////////////////////////////////////////////////////////////////
002// checkstyle: Checks Java source code and other text files for adherence to a set of rules.
003// Copyright (C) 2001-2025 the original author or authors.
004//
005// This library is free software; you can redistribute it and/or
006// modify it under the terms of the GNU Lesser General Public
007// License as published by the Free Software Foundation; either
008// version 2.1 of the License, or (at your option) any later version.
009//
010// This library is distributed in the hope that it will be useful,
011// but WITHOUT ANY WARRANTY; without even the implied warranty of
012// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
013// Lesser General Public License for more details.
014//
015// You should have received a copy of the GNU Lesser General Public
016// License along with this library; if not, write to the Free Software
017// Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
018///////////////////////////////////////////////////////////////////////////////////////////////
019
020package com.puppycrawl.tools.checkstyle.checks;
021
022import java.io.File;
023import java.io.InputStream;
024import java.nio.file.Files;
025import java.nio.file.NoSuchFileException;
026import java.util.Arrays;
027import java.util.Collections;
028import java.util.HashSet;
029import java.util.Locale;
030import java.util.Map;
031import java.util.Map.Entry;
032import java.util.Optional;
033import java.util.Properties;
034import java.util.Set;
035import java.util.SortedSet;
036import java.util.TreeMap;
037import java.util.TreeSet;
038import java.util.regex.Matcher;
039import java.util.regex.Pattern;
040import java.util.stream.Collectors;
041
042import org.apache.commons.logging.Log;
043import org.apache.commons.logging.LogFactory;
044
045import com.puppycrawl.tools.checkstyle.Definitions;
046import com.puppycrawl.tools.checkstyle.GlobalStatefulCheck;
047import com.puppycrawl.tools.checkstyle.LocalizedMessage;
048import com.puppycrawl.tools.checkstyle.api.AbstractFileSetCheck;
049import com.puppycrawl.tools.checkstyle.api.FileText;
050import com.puppycrawl.tools.checkstyle.api.MessageDispatcher;
051import com.puppycrawl.tools.checkstyle.api.Violation;
052import com.puppycrawl.tools.checkstyle.utils.CommonUtil;
053
054/**
055 * <div>
056 * Ensures the correct translation of code by checking property files for consistency
057 * regarding their keys. Two property files describing one and the same context
058 * are consistent if they contain the same keys. TranslationCheck also can check
059 * an existence of required translations which must exist in project, if
060 * {@code requiredTranslations} option is used.
061 * </div>
062 *
063 * <p>
064 * Notes:
065 * Language code for the property {@code requiredTranslations} is composed of
066 * the lowercase, two-letter codes as defined by
067 * <a href="https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes">ISO 639-1</a>.
068 * Default value is empty String Set which means that only the existence of default
069 * translation is checked. Note, if you specify language codes (or just one
070 * language code) of required translations the check will also check for existence
071 * of default translation files in project.
072 * </p>
073 *
074 * <p>
075 * Note: If your project uses preprocessed translation files and the original files do not have the
076 * {@code properties} extension, you can specify additional file extensions
077 * via the {@code fileExtensions} property.
078 * </p>
079 *
080 * <p>
081 * Attention: the check will perform the validation of ISO codes if the option
082 * is used. So, if you specify, for example, "mm" for language code,
083 * TranslationCheck will rise violation that the language code is incorrect.
084 * </p>
085 *
086 * <p>
087 * Attention: this Check could produce false-positives if it is used with
088 * <a href="https://checkstyle.org/config.html#Checker">Checker</a> that use cache
089 * (property "cacheFile") This is known design problem, will be addressed at
090 * <a href="https://github.com/checkstyle/checkstyle/issues/3539">issue</a>.
091 * </p>
092 *
093 * @since 3.0
094 */
095@GlobalStatefulCheck
096public final class TranslationCheck extends AbstractFileSetCheck {
097
098    /**
099     * A key is pointing to the warning message text for missing key
100     * in "messages.properties" file.
101     */
102    public static final String MSG_KEY = "translation.missingKey";
103
104    /**
105     * A key is pointing to the warning message text for missing translation file
106     * in "messages.properties" file.
107     */
108    public static final String MSG_KEY_MISSING_TRANSLATION_FILE =
109        "translation.missingTranslationFile";
110
111    /** Resource bundle which contains messages for TranslationCheck. */
112    private static final String TRANSLATION_BUNDLE =
113        "com.puppycrawl.tools.checkstyle.checks.messages";
114
115    /**
116     * A key is pointing to the warning message text for wrong language code
117     * in "messages.properties" file.
118     */
119    private static final String WRONG_LANGUAGE_CODE_KEY = "translation.wrongLanguageCode";
120
121    /**
122     * Regexp string for default translation files.
123     * For example, messages.properties.
124     */
125    private static final String DEFAULT_TRANSLATION_REGEXP = "^.+\\..+$";
126
127    /**
128     * Regexp pattern for bundles names which end with language code, followed by country code and
129     * variant suffix. For example, messages_es_ES_UNIX.properties.
130     */
131    private static final Pattern LANGUAGE_COUNTRY_VARIANT_PATTERN =
132        CommonUtil.createPattern("^.+\\_[a-z]{2}\\_[A-Z]{2}\\_[A-Za-z]+\\..+$");
133    /**
134     * Regexp pattern for bundles names which end with language code, followed by country code
135     * suffix. For example, messages_es_ES.properties.
136     */
137    private static final Pattern LANGUAGE_COUNTRY_PATTERN =
138        CommonUtil.createPattern("^.+\\_[a-z]{2}\\_[A-Z]{2}\\..+$");
139    /**
140     * Regexp pattern for bundles names which end with language code suffix.
141     * For example, messages_es.properties.
142     */
143    private static final Pattern LANGUAGE_PATTERN =
144        CommonUtil.createPattern("^.+\\_[a-z]{2}\\..+$");
145
146    /** File name format for default translation. */
147    private static final String DEFAULT_TRANSLATION_FILE_NAME_FORMATTER = "%s.%s";
148    /** File name format with language code. */
149    private static final String FILE_NAME_WITH_LANGUAGE_CODE_FORMATTER = "%s_%s.%s";
150
151    /** Formatting string to form regexp to validate required translations file names. */
152    private static final String REGEXP_FORMAT_TO_CHECK_REQUIRED_TRANSLATIONS =
153        "^%1$s\\_%2$s(\\_[A-Z]{2})?\\.%3$s$|^%1$s\\_%2$s\\_[A-Z]{2}\\_[A-Za-z]+\\.%3$s$";
154    /** Formatting string to form regexp to validate default translations file names. */
155    private static final String REGEXP_FORMAT_TO_CHECK_DEFAULT_TRANSLATIONS = "^%s\\.%s$";
156
157    /** Logger for TranslationCheck. */
158    private final Log log;
159
160    /** The files to process. */
161    private final Set<File> filesToProcess = new HashSet<>();
162
163    /**
164     * Specify
165     * <a href="https://docs.oracle.com/en/java/javase/17/docs/api/java.base/java/util/ResourceBundle.html">
166     * Base name</a> of resource bundles which contain message resources.
167     * It helps the check to distinguish config and localization resources.
168     */
169    private Pattern baseName;
170
171    /**
172     * Specify language codes of required translations which must exist in project.
173     */
174    private Set<String> requiredTranslations = new HashSet<>();
175
176    /**
177     * Creates a new {@code TranslationCheck} instance.
178     */
179    public TranslationCheck() {
180        setFileExtensions("properties");
181        baseName = CommonUtil.createPattern("^messages.*$");
182        log = LogFactory.getLog(TranslationCheck.class);
183    }
184
185    /**
186     * Setter to specify
187     * <a href="https://docs.oracle.com/en/java/javase/17/docs/api/java.base/java/util/ResourceBundle.html">
188     * Base name</a> of resource bundles which contain message resources.
189     * It helps the check to distinguish config and localization resources.
190     *
191     * @param baseName base name regexp.
192     * @since 6.17
193     */
194    public void setBaseName(Pattern baseName) {
195        this.baseName = baseName;
196    }
197
198    /**
199     * Setter to specify language codes of required translations which must exist in project.
200     *
201     * @param translationCodes language codes.
202     * @since 6.11
203     */
204    public void setRequiredTranslations(String... translationCodes) {
205        requiredTranslations = Arrays.stream(translationCodes)
206            .collect(Collectors.toUnmodifiableSet());
207        validateUserSpecifiedLanguageCodes(requiredTranslations);
208    }
209
210    /**
211     * Validates the correctness of user specified language codes for the check.
212     *
213     * @param languageCodes user specified language codes for the check.
214     * @throws IllegalArgumentException when any item of languageCodes is not valid language code
215     */
216    private void validateUserSpecifiedLanguageCodes(Set<String> languageCodes) {
217        for (String code : languageCodes) {
218            if (!isValidLanguageCode(code)) {
219                final LocalizedMessage msg = new LocalizedMessage(TRANSLATION_BUNDLE,
220                        getClass(), WRONG_LANGUAGE_CODE_KEY, code);
221                throw new IllegalArgumentException(msg.getMessage());
222            }
223        }
224    }
225
226    /**
227     * Checks whether user specified language code is correct (is contained in available locales).
228     *
229     * @param userSpecifiedLanguageCode user specified language code.
230     * @return true if user specified language code is correct.
231     */
232    private static boolean isValidLanguageCode(final String userSpecifiedLanguageCode) {
233        boolean valid = false;
234        final Locale[] locales = Locale.getAvailableLocales();
235        for (Locale locale : locales) {
236            if (userSpecifiedLanguageCode.equals(locale.toString())) {
237                valid = true;
238                break;
239            }
240        }
241        return valid;
242    }
243
244    @Override
245    public void beginProcessing(String charset) {
246        filesToProcess.clear();
247    }
248
249    @Override
250    protected void processFiltered(File file, FileText fileText) {
251        // We are just collecting files for processing at finishProcessing()
252        filesToProcess.add(file);
253    }
254
255    @Override
256    public void finishProcessing() {
257        final Set<ResourceBundle> bundles = groupFilesIntoBundles(filesToProcess, baseName);
258        for (ResourceBundle currentBundle : bundles) {
259            checkExistenceOfDefaultTranslation(currentBundle);
260            checkExistenceOfRequiredTranslations(currentBundle);
261            checkTranslationKeys(currentBundle);
262        }
263    }
264
265    /**
266     * Checks an existence of default translation file in the resource bundle.
267     *
268     * @param bundle resource bundle.
269     */
270    private void checkExistenceOfDefaultTranslation(ResourceBundle bundle) {
271        getMissingFileName(bundle, null)
272            .ifPresent(fileName -> logMissingTranslation(bundle.getPath(), fileName));
273    }
274
275    /**
276     * Checks an existence of translation files in the resource bundle.
277     * The name of translation file begins with the base name of resource bundle which is followed
278     * by '_' and a language code (country and variant are optional), it ends with the extension
279     * suffix.
280     *
281     * @param bundle resource bundle.
282     */
283    private void checkExistenceOfRequiredTranslations(ResourceBundle bundle) {
284        for (String languageCode : requiredTranslations) {
285            getMissingFileName(bundle, languageCode)
286                .ifPresent(fileName -> logMissingTranslation(bundle.getPath(), fileName));
287        }
288    }
289
290    /**
291     * Returns the name of translation file which is absent in resource bundle or Guava's Optional,
292     * if there is not missing translation.
293     *
294     * @param bundle resource bundle.
295     * @param languageCode language code.
296     * @return the name of translation file which is absent in resource bundle or Guava's Optional,
297     *         if there is not missing translation.
298     */
299    private static Optional<String> getMissingFileName(ResourceBundle bundle, String languageCode) {
300        final String fileNameRegexp;
301        final boolean searchForDefaultTranslation;
302        final String extension = bundle.getExtension();
303        final String baseName = bundle.getBaseName();
304        if (languageCode == null) {
305            searchForDefaultTranslation = true;
306            fileNameRegexp = String.format(Locale.ROOT, REGEXP_FORMAT_TO_CHECK_DEFAULT_TRANSLATIONS,
307                    baseName, extension);
308        }
309        else {
310            searchForDefaultTranslation = false;
311            fileNameRegexp = String.format(Locale.ROOT,
312                REGEXP_FORMAT_TO_CHECK_REQUIRED_TRANSLATIONS, baseName, languageCode, extension);
313        }
314        Optional<String> missingFileName = Optional.empty();
315        if (!bundle.containsFile(fileNameRegexp)) {
316            if (searchForDefaultTranslation) {
317                missingFileName = Optional.of(String.format(Locale.ROOT,
318                        DEFAULT_TRANSLATION_FILE_NAME_FORMATTER, baseName, extension));
319            }
320            else {
321                missingFileName = Optional.of(String.format(Locale.ROOT,
322                        FILE_NAME_WITH_LANGUAGE_CODE_FORMATTER, baseName, languageCode, extension));
323            }
324        }
325        return missingFileName;
326    }
327
328    /**
329     * Logs that translation file is missing.
330     *
331     * @param filePath file path.
332     * @param fileName file name.
333     */
334    private void logMissingTranslation(String filePath, String fileName) {
335        final MessageDispatcher dispatcher = getMessageDispatcher();
336        dispatcher.fireFileStarted(filePath);
337        log(1, MSG_KEY_MISSING_TRANSLATION_FILE, fileName);
338        fireErrors(filePath);
339        dispatcher.fireFileFinished(filePath);
340    }
341
342    /**
343     * Groups a set of files into bundles.
344     * Only files, which names match base name regexp pattern will be grouped.
345     *
346     * @param files set of files.
347     * @param baseNameRegexp base name regexp pattern.
348     * @return set of ResourceBundles.
349     */
350    private static Set<ResourceBundle> groupFilesIntoBundles(Set<File> files,
351                                                             Pattern baseNameRegexp) {
352        final Set<ResourceBundle> resourceBundles = new HashSet<>();
353        for (File currentFile : files) {
354            final String fileName = currentFile.getName();
355            final String baseName = extractBaseName(fileName);
356            final Matcher baseNameMatcher = baseNameRegexp.matcher(baseName);
357            if (baseNameMatcher.matches()) {
358                final String extension = CommonUtil.getFileExtension(fileName);
359                final String path = getPath(currentFile.getAbsolutePath());
360                final ResourceBundle newBundle = new ResourceBundle(baseName, path, extension);
361                final Optional<ResourceBundle> bundle = findBundle(resourceBundles, newBundle);
362                if (bundle.isPresent()) {
363                    bundle.orElseThrow().addFile(currentFile);
364                }
365                else {
366                    newBundle.addFile(currentFile);
367                    resourceBundles.add(newBundle);
368                }
369            }
370        }
371        return resourceBundles;
372    }
373
374    /**
375     * Searches for specific resource bundle in a set of resource bundles.
376     *
377     * @param bundles set of resource bundles.
378     * @param targetBundle target bundle to search for.
379     * @return Guava's Optional of resource bundle (present if target bundle is found).
380     */
381    private static Optional<ResourceBundle> findBundle(Set<ResourceBundle> bundles,
382                                                       ResourceBundle targetBundle) {
383        Optional<ResourceBundle> result = Optional.empty();
384        for (ResourceBundle currentBundle : bundles) {
385            if (targetBundle.getBaseName().equals(currentBundle.getBaseName())
386                    && targetBundle.getExtension().equals(currentBundle.getExtension())
387                    && targetBundle.getPath().equals(currentBundle.getPath())) {
388                result = Optional.of(currentBundle);
389                break;
390            }
391        }
392        return result;
393    }
394
395    /**
396     * Extracts the base name (the unique prefix) of resource bundle from translation file name.
397     * For example "messages" is the base name of "messages.properties",
398     * "messages_de_AT.properties", "messages_en.properties", etc.
399     *
400     * @param fileName the fully qualified name of the translation file.
401     * @return the extracted base name.
402     */
403    private static String extractBaseName(String fileName) {
404        final String regexp;
405        final Matcher languageCountryVariantMatcher =
406            LANGUAGE_COUNTRY_VARIANT_PATTERN.matcher(fileName);
407        final Matcher languageCountryMatcher = LANGUAGE_COUNTRY_PATTERN.matcher(fileName);
408        final Matcher languageMatcher = LANGUAGE_PATTERN.matcher(fileName);
409        if (languageCountryVariantMatcher.matches()) {
410            regexp = LANGUAGE_COUNTRY_VARIANT_PATTERN.pattern();
411        }
412        else if (languageCountryMatcher.matches()) {
413            regexp = LANGUAGE_COUNTRY_PATTERN.pattern();
414        }
415        else if (languageMatcher.matches()) {
416            regexp = LANGUAGE_PATTERN.pattern();
417        }
418        else {
419            regexp = DEFAULT_TRANSLATION_REGEXP;
420        }
421        // We use substring(...) instead of replace(...), so that the regular expression does
422        // not have to be compiled each time it is used inside 'replace' method.
423        final String removePattern = regexp.substring("^.+".length());
424        return fileName.replaceAll(removePattern, "");
425    }
426
427    /**
428     * Extracts path from a file name which contains the path.
429     * For example, if the file name is /xyz/messages.properties,
430     * then the method will return /xyz/.
431     *
432     * @param fileNameWithPath file name which contains the path.
433     * @return file path.
434     */
435    private static String getPath(String fileNameWithPath) {
436        return fileNameWithPath
437            .substring(0, fileNameWithPath.lastIndexOf(File.separator));
438    }
439
440    /**
441     * Checks resource files in bundle for consistency regarding their keys.
442     * All files in bundle must have the same key set. If this is not the case
443     * an audit event message is posted giving information which key misses in which file.
444     *
445     * @param bundle resource bundle.
446     */
447    private void checkTranslationKeys(ResourceBundle bundle) {
448        final Set<File> filesInBundle = bundle.getFiles();
449        // build a map from files to the keys they contain
450        final Set<String> allTranslationKeys = new HashSet<>();
451        final Map<File, Set<String>> filesAssociatedWithKeys = new TreeMap<>();
452        for (File currentFile : filesInBundle) {
453            final Set<String> keysInCurrentFile = getTranslationKeys(currentFile);
454            allTranslationKeys.addAll(keysInCurrentFile);
455            filesAssociatedWithKeys.put(currentFile, keysInCurrentFile);
456        }
457        checkFilesForConsistencyRegardingTheirKeys(filesAssociatedWithKeys, allTranslationKeys);
458    }
459
460    /**
461     * Compares th the specified key set with the key sets of the given translation files (arranged
462     * in a map). All missing keys are reported.
463     *
464     * @param fileKeys a Map from translation files to their key sets.
465     * @param keysThatMustExist the set of keys to compare with.
466     */
467    private void checkFilesForConsistencyRegardingTheirKeys(Map<File, Set<String>> fileKeys,
468                                                            Set<String> keysThatMustExist) {
469        for (Entry<File, Set<String>> fileKey : fileKeys.entrySet()) {
470            final Set<String> currentFileKeys = fileKey.getValue();
471            final Set<String> missingKeys = keysThatMustExist.stream()
472                .filter(key -> !currentFileKeys.contains(key))
473                .collect(Collectors.toUnmodifiableSet());
474            if (!missingKeys.isEmpty()) {
475                final MessageDispatcher dispatcher = getMessageDispatcher();
476                final String path = fileKey.getKey().getAbsolutePath();
477                dispatcher.fireFileStarted(path);
478                for (Object key : missingKeys) {
479                    log(1, MSG_KEY, key);
480                }
481                fireErrors(path);
482                dispatcher.fireFileFinished(path);
483            }
484        }
485    }
486
487    /**
488     * Loads the keys from the specified translation file into a set.
489     *
490     * @param file translation file.
491     * @return a Set object which holds the loaded keys.
492     */
493    private Set<String> getTranslationKeys(File file) {
494        Set<String> keys = new HashSet<>();
495        try (InputStream inStream = Files.newInputStream(file.toPath())) {
496            final Properties translations = new Properties();
497            translations.load(inStream);
498            keys = translations.stringPropertyNames();
499        }
500        // -@cs[IllegalCatch] It is better to catch all exceptions since it can throw
501        // a runtime exception.
502        catch (final Exception exc) {
503            logException(exc, file);
504        }
505        return keys;
506    }
507
508    /**
509     * Helper method to log an exception.
510     *
511     * @param exception the exception that occurred
512     * @param file the file that could not be processed
513     */
514    private void logException(Exception exception, File file) {
515        final String[] args;
516        final String key;
517        if (exception instanceof NoSuchFileException) {
518            args = null;
519            key = "general.fileNotFound";
520        }
521        else {
522            args = new String[] {exception.getMessage()};
523            key = "general.exception";
524        }
525        final Violation message =
526            new Violation(
527                0,
528                Definitions.CHECKSTYLE_BUNDLE,
529                key,
530                args,
531                getId(),
532                getClass(), null);
533        final SortedSet<Violation> messages = new TreeSet<>();
534        messages.add(message);
535        getMessageDispatcher().fireErrors(file.getPath(), messages);
536        log.debug("Exception occurred.", exception);
537    }
538
539    /** Class which represents a resource bundle. */
540    private static final class ResourceBundle {
541
542        /** Bundle base name. */
543        private final String baseName;
544        /** Common extension of files which are included in the resource bundle. */
545        private final String extension;
546        /** Common path of files which are included in the resource bundle. */
547        private final String path;
548        /** Set of files which are included in the resource bundle. */
549        private final Set<File> files;
550
551        /**
552         * Creates a ResourceBundle object with specific base name, common files extension.
553         *
554         * @param baseName bundle base name.
555         * @param path common path of files which are included in the resource bundle.
556         * @param extension common extension of files which are included in the resource bundle.
557         */
558        private ResourceBundle(String baseName, String path, String extension) {
559            this.baseName = baseName;
560            this.path = path;
561            this.extension = extension;
562            files = new HashSet<>();
563        }
564
565        /**
566         * Returns the bundle base name.
567         *
568         * @return the bundle base name
569         */
570        public String getBaseName() {
571            return baseName;
572        }
573
574        /**
575         * Returns the common path of files which are included in the resource bundle.
576         *
577         * @return the common path of files
578         */
579        public String getPath() {
580            return path;
581        }
582
583        /**
584         * Returns the common extension of files which are included in the resource bundle.
585         *
586         * @return the common extension of files
587         */
588        public String getExtension() {
589            return extension;
590        }
591
592        /**
593         * Returns the set of files which are included in the resource bundle.
594         *
595         * @return the set of files
596         */
597        public Set<File> getFiles() {
598            return Collections.unmodifiableSet(files);
599        }
600
601        /**
602         * Adds a file into resource bundle.
603         *
604         * @param file file which should be added into resource bundle.
605         */
606        public void addFile(File file) {
607            files.add(file);
608        }
609
610        /**
611         * Checks whether a resource bundle contains a file which name matches file name regexp.
612         *
613         * @param fileNameRegexp file name regexp.
614         * @return true if a resource bundle contains a file which name matches file name regexp.
615         */
616        public boolean containsFile(String fileNameRegexp) {
617            boolean containsFile = false;
618            for (File currentFile : files) {
619                if (Pattern.matches(fileNameRegexp, currentFile.getName())) {
620                    containsFile = true;
621                    break;
622                }
623            }
624            return containsFile;
625        }
626
627    }
628
629}