Examples with Language org.omegat.util.Language used on opensource projects

Example 16 with Language

use of org.omegat.util.Language in project omegat by omegat-org.

the class SpellChecker method initialize.

/**
 * Initialize the library for the given project. Loads the lists of ignored and learned words for the
 * project
 */
public void initialize() {
    Language targetLanguage = Core.getProject().getProjectProperties().getTargetLanguage();
    Stream<String> toCheck = Stream.of(// Full xx_YY
    targetLanguage.getLocaleCode(), // Full xx-YY
    targetLanguage.getLocaleCode().replace('_', '-'), // xx only
    targetLanguage.getLanguageCode());
    checker = toCheck.map(SpellChecker::initializeWithLanguage).filter(Optional::isPresent).findFirst().orElseGet(() -> Optional.of(new SpellCheckerDummy())).get();
    if (checker instanceof SpellCheckerDummy) {
        Log.log("No spell checker found for language " + targetLanguage);
    }
    loadWordLists();
}

Example 17 with Language

use of org.omegat.util.Language in project omegat by omegat-org.

the class FindMatches method search.

public List<NearString> search(final String searchText, final boolean requiresTranslation, final boolean fillSimilarityData, final IStopped stop) throws StoppedException {
    result = new ArrayList<>(OConsts.MAX_NEAR_STRINGS + 1);
    srcText = searchText;
    removedText = "";
    // of the translatable text
    if (removePattern != null) {
        StringBuilder removedBuffer = new StringBuilder();
        Matcher removeMatcher = removePattern.matcher(srcText);
        while (removeMatcher.find()) {
            removedBuffer.append(removeMatcher.group());
        }
        srcText = removeMatcher.replaceAll("");
        removedText = removedBuffer.toString();
    }
    // get tokens for original string
    strTokensStem = tokenizeStem(srcText);
    strTokensNoStem = tokenizeNoStem(srcText);
    strTokensAll = tokenizeAll(srcText);
    // travel by project entries, including orphaned
    if (project.getProjectProperties().isSupportDefaultTranslations()) {
        project.iterateByDefaultTranslations(new DefaultTranslationsIterator() {

            public void iterate(String source, TMXEntry trans) {
                checkStopped(stop);
                if (!searchExactlyTheSame && source.equals(searchText)) {
                    // skip original==original entry comparison
                    return;
                }
                if (requiresTranslation && trans.translation == null) {
                    return;
                }
                String fileName = project.isOrphaned(source) ? ORPHANED_FILE_NAME : null;
                processEntry(null, source, trans.translation, NearString.MATCH_SOURCE.MEMORY, false, 0, fileName, trans.creator, trans.creationDate, trans.changer, trans.changeDate, null);
            }
        });
    }
    project.iterateByMultipleTranslations(new MultipleTranslationsIterator() {

        public void iterate(EntryKey source, TMXEntry trans) {
            checkStopped(stop);
            if (!searchExactlyTheSame && source.sourceText.equals(searchText)) {
                // skip original==original entry comparison
                return;
            }
            if (requiresTranslation && trans.translation == null) {
                return;
            }
            String fileName = project.isOrphaned(source) ? ORPHANED_FILE_NAME : null;
            processEntry(source, source.sourceText, trans.translation, NearString.MATCH_SOURCE.MEMORY, false, 0, fileName, trans.creator, trans.creationDate, trans.changer, trans.changeDate, null);
        }
    });
    // travel by translation memories
    for (Map.Entry<String, ExternalTMX> en : project.getTransMemories().entrySet()) {
        int penalty = 0;
        Matcher matcher = SEARCH_FOR_PENALTY.matcher(en.getKey());
        if (matcher.find()) {
            penalty = Integer.parseInt(matcher.group(1));
        }
        for (PrepareTMXEntry tmen : en.getValue().getEntries()) {
            checkStopped(stop);
            if (tmen.source == null) {
                // Not all TMX entries have a source; in that case there can be no meaningful match, so skip.
                continue;
            }
            if (requiresTranslation && tmen.translation == null) {
                continue;
            }
            processEntry(null, tmen.source, tmen.translation, NearString.MATCH_SOURCE.TM, false, penalty, en.getKey(), tmen.creator, tmen.creationDate, tmen.changer, tmen.changeDate, tmen.otherProperties);
        }
    }
    // travel by all entries for check source file translations
    for (SourceTextEntry ste : project.getAllEntries()) {
        checkStopped(stop);
        if (ste.getSourceTranslation() != null) {
            processEntry(ste.getKey(), ste.getSrcText(), ste.getSourceTranslation(), NearString.MATCH_SOURCE.MEMORY, ste.isSourceTranslationFuzzy(), 0, ste.getKey().file, "", 0, "", 0, null);
        }
    }
    if (separateSegmentMatcher != null) {
        // split paragraph even when segmentation disabled, then find matches for every segment
        List<StringBuilder> spaces = new ArrayList<StringBuilder>();
        List<Rule> brules = new ArrayList<Rule>();
        Language sourceLang = project.getProjectProperties().getSourceLanguage();
        Language targetLang = project.getProjectProperties().getTargetLanguage();
        List<String> segments = Core.getSegmenter().segment(sourceLang, srcText, spaces, brules);
        if (segments.size() > 1) {
            List<String> fsrc = new ArrayList<String>(segments.size());
            List<String> ftrans = new ArrayList<String>(segments.size());
            // multiple segments
            for (short i = 0; i < segments.size(); i++) {
                String onesrc = segments.get(i);
                // find match for separate segment
                List<NearString> segmentMatch = separateSegmentMatcher.search(onesrc, requiresTranslation, false, stop);
                if (!segmentMatch.isEmpty() && segmentMatch.get(0).scores[0].score >= SUBSEGMENT_MATCH_THRESHOLD) {
                    fsrc.add(segmentMatch.get(0).source);
                    ftrans.add(segmentMatch.get(0).translation);
                } else {
                    fsrc.add("");
                    ftrans.add("");
                }
            }
            // glue found sources
            String foundSrc = Core.getSegmenter().glue(sourceLang, sourceLang, fsrc, spaces, brules);
            // glue found translations
            String foundTrans = Core.getSegmenter().glue(sourceLang, targetLang, ftrans, spaces, brules);
            processEntry(null, foundSrc, foundTrans, NearString.MATCH_SOURCE.TM, false, 0, "", "", 0, "", 0, null);
        }
    }
    if (fillSimilarityData) {
        // fill similarity data only for result
        for (NearString near : result) {
            // fix for bug 1586397
            byte[] similarityData = FuzzyMatcher.buildSimilarityData(strTokensAll, tokenizeAll(near.source));
            near.attr = similarityData;
        }
    }
    return result;
}

Example 18 with Language

use of org.omegat.util.Language in project omegat by omegat-org.

the class TeamTool method initTeamProject.

/**
 * Utility function to create a minimal project to serve as a base for a
 * team project. Will add/stage everything if invoked on a path already
 * containing a git working tree or svn checkout.
 *
 * @param dir
 *            Directory in which to create team project
 * @param srcLang
 *            Source language
 * @param trgLang
 *            Target language
 * @param showGui
 *            If true, show the Project Properties dialog
 * @throws Exception
 *             If specified dir is not a directory, is not writeable, etc.
 */
public static void initTeamProject(File dir, String srcLang, String trgLang) throws Exception {
    if (!dir.isDirectory()) {
        throw new IllegalArgumentException("Specified dir is not a directory: " + dir.getPath());
    }
    if (!dir.canWrite()) {
        throw new IOException("Specified dir is not writeable: " + dir.getPath());
    }
    // Create project properties
    ProjectProperties props = new ProjectProperties(dir);
    props.setSourceLanguage(srcLang);
    props.setTargetLanguage(trgLang);
    // Set default tokenizers
    props.setSourceTokenizer(PluginUtils.getTokenizerClassForLanguage(new Language(srcLang)));
    props.setTargetTokenizer(PluginUtils.getTokenizerClassForLanguage(new Language(trgLang)));
    // Create project internal directories
    props.autocreateDirectories();
    // Create version-controlled glossary file
    props.getWritableGlossaryFile().getAsFile().createNewFile();
    ProjectFileStorage.writeProjectFile(props);
    // Create empty project TM
    new ProjectTMX(props.getSourceLanguage(), props.getTargetLanguage(), props.isSentenceSegmentingEnabled(), null, null).save(props, new File(props.getProjectInternal(), OConsts.STATUS_EXTENSION).getPath(), false);
    // and set EOL handling correctly for cross-platform work
    if (new File(dir, ".svn").isDirectory()) {
        SVNClientManager mgr = SVNClientManager.newInstance();
        mgr.getWCClient().doSetProperty(dir, "svn:auto-props", SVNPropertyValue.create("*.txt = svn:eol-style=native\n*.tmx = svn:eol-style=native\n"), false, SVNDepth.EMPTY, null, null);
        mgr.getWCClient().doAdd(dir.listFiles(f -> !f.getName().startsWith(".")), false, false, true, SVNDepth.fromRecurse(true), false, false, false, true);
    } else if (new File(dir, ".git").isDirectory()) {
        try (BufferedWriter writer = Files.newBufferedWriter(new File(dir, ".gitattributes").toPath())) {
            writer.write("* text=auto\n");
            writer.write("*.tmx text\n");
            writer.write("*.txt text\n");
        }
        Git.open(dir).add().addFilepattern(".").call();
    }
    System.out.println(StringUtil.format(OStrings.getString("TEAM_TOOL_INIT_COMPLETE"), srcLang, trgLang));
}

Example 19 with Language

use of org.omegat.util.Language in project omegat by omegat-org.

the class LanguageToolWrapper method setBridgeFromCurrentProject.

/**
 * Set this instance's LanguageTool bridge based on the current project.
 */
public static void setBridgeFromCurrentProject() {
    if (bridge != null) {
        bridge.stop();
    }
    if (Core.getProject().isProjectLoaded()) {
        Language sourceLang = Core.getProject().getProjectProperties().getSourceLanguage();
        Language targetLang = Core.getProject().getProjectProperties().getTargetLanguage();
        bridge = createBridgeFromPrefs(sourceLang, targetLang);
    }
}

Example 20 with Language

use of org.omegat.util.Language in project omegat by omegat-org.

the class HunspellTokenizer method populateInstalledDicts.

private static synchronized void populateInstalledDicts() {
    if (affixFiles != null && dictionaryFiles != null) {
        return;
    }
    affixFiles = new HashMap<>();
    dictionaryFiles = new HashMap<>();
    String dictionaryDirPath = Preferences.getPreference(Preferences.SPELLCHECKER_DICTIONARY_DIRECTORY);
    if (dictionaryDirPath.isEmpty()) {
        return;
    }
    File dictionaryDir = new File(dictionaryDirPath);
    if (!dictionaryDir.isDirectory()) {
        return;
    }
    for (File file : dictionaryDir.listFiles()) {
        String name = file.getName();
        if (name.endsWith(OConsts.SC_AFFIX_EXTENSION)) {
            Language lang = new Language(name.substring(0, name.lastIndexOf(OConsts.SC_AFFIX_EXTENSION)));
            affixFiles.put(lang, file);
            affixFiles.put(new Language(lang.getLanguageCode()), file);
        } else if (name.endsWith(OConsts.SC_DICTIONARY_EXTENSION)) {
            Language lang = new Language(name.substring(0, name.lastIndexOf(OConsts.SC_DICTIONARY_EXTENSION)));
            dictionaryFiles.put(lang, file);
            dictionaryFiles.put(new Language(lang.getLanguageCode()), file);
        }
    }
    Set<Language> commonLangs = new HashSet<>(affixFiles.keySet());
    commonLangs.retainAll(dictionaryFiles.keySet());
    availableDictLangs = langsToStrings(commonLangs);
}