Java Code Examples for com.jstarcraft.core.utility.StringUtility#SPACE

/**
 * 检测语言
 * 
 * @param text
 * @param options
 * @return
 */
public SortedSet<DetectionLanguage> detectLanguages(String text, Object2BooleanMap<String> options) {
    SortedSet<DetectionLanguage> locales = new TreeSet<>();

    // 最小长度限制
    int size = text.length();
    if (size < minimum) {
        return locales;
    }
    // 最大长度限制
    if (size > maximum) {
        text = text.substring(0, maximum);
        size = maximum;
    }

    // 白名单,黑名单
    Set<String> writes = options.size() == 0 ? Collections.EMPTY_SET : new HashSet<>();
    Set<String> blacks = options.size() == 0 ? Collections.EMPTY_SET : new HashSet<>();
    for (Object2BooleanMap.Entry<String> option : options.object2BooleanEntrySet()) {
        if (option.getBooleanValue()) {
            writes.add(option.getKey());
        } else {
            blacks.add(option.getKey());
        }
    }

    /*
     * Get the script which characters occur the most in `value`.
     */
    int count = -1;
    String script = null;
    for (DetectionPattern regulation : patterns.values()) {
        Pattern pattern = regulation.getPattern();
        Matcher matcher = pattern.matcher(text);
        int match = 0;
        while (matcher.find()) {
            match++;
        }
        if (match > count) {
            count = match;
            script = regulation.getName();
        }
    }
    if (script == null || count <= 0) {
        return locales;
    }

    /* One languages exists for the most-used script. */
    Set<DetectionTrie> dictionaries = tires.get(script);
    if (dictionaries == null) {
        /*
         * If no matches occured, such as a digit only string, or because the language is ignored, exit with `und`.
         */
        if (!checkLanguage(script, writes, blacks)) {
            return locales;
        }
        locales.add(new DetectionLanguage(Locale.forLanguageTag(script), 1D));
        return locales;
    }

    /*
     * Get all distances for a given script, and normalize the distance values.
     */
    // 前后补空格是为了N-Gram处理
    text = StringUtility.SPACE + REPLACE.matcher(text).replaceAll(StringUtility.SPACE).toLowerCase() + StringUtility.SPACE;
    CharacterNgram ngram = new CharacterNgram(3, text);
    Object2IntMap<CharSequence> tuples = new Object2IntOpenHashMap<>();
    for (CharSequence character : ngram) {
        count = tuples.getInt(character);
        tuples.put(character, count + 1);
    }
    for (DetectionTrie dictionary : dictionaries) {
        String language = dictionary.getName();
        if (checkLanguage(language, writes, blacks)) {
            double score = getScore(tuples, dictionary.getTrie());
            DetectionLanguage locale = new DetectionLanguage(Locale.forLanguageTag(language), score);
            locales.add(locale);
        }
    }
    if (!locales.isEmpty()) {
        normalizeScores(text, locales);
    }
    return locales;
}
 

@Override
public String toString() {
    return "(" + left + StringUtility.SPACE + operator.getOperate() + StringUtility.SPACE + right + ")";
}
 

@Override
public String toString() {
    return prefix + index + StringUtility.SPACE + operator.getOperate() + StringUtility.SPACE + (operator == AwkOperator.IN ? property : value);
}