拼音分割算法

1. 实现两个词的切割
   如： 开始 kāishǐ ->kāi shǐ

2. 待实现 多字切割

   • simple 1 : 打开门 dakaimen

     1. split to "dakai men"
     2. identify "dakai" can split to " da kai" ,and "men" can't split to "m en"
     3. we have 3 words , so should split first part "da kai"

• simple 2 : 那男人 na'nan'ren
• simple 3 : 小图案 xiao'tu'an

3 . 程序示例

public class Utils {

    private static String[] initialmeta = {"b", "p", "m", "f", "d", "t", "l", "k", "j", "q", "x", "z", "c", "s", "zh", "ch", "sh", "h", "y", "w"};
    private static String[] finalmeta = {"ai", "ao", "an", "ou", "ei", "en", "er", "a", "o", "e"};
    private static String[] finalmetaFull = {"i" + "u", "v", "ia", "ua", "ie", "ve", "uo", "uai", "uei", "iao", "iou",
            "ian", "uan", "van", "in", "uen", "vn", "ang", "iang", "uang", "eng", "ing", "ueng", "ong", "iong"};

    private static String[] voicemark_a = {"ā", "á", "ǎ", "à"};
    private static String[] voicemark_o = {"ō", "ó", "ǒ", "ò"};
    private static String[] voicemark_e = {"ē", "é", "ě", "è"};
    private static String[] voicemark_i = {"ī", "í", "ǐ", "ì"};
    private static String[] voicemark_u = {"ū", "ú", "ǔ", "ù"};
    private static String[] voicemark_v = {"ǖ", "ǘ", "ǚ", "ǜ", "ü"};


    public static String splitPinyin(String str) {
        StringBuilder sb = new StringBuilder(str);
        String after = str.substring(1, str.length());
        int tempVoicemeta;
        if (startWithFinal(str) != 0) {
            tempVoicemeta = startWithFinal(str);
        } else {
            tempVoicemeta = containsInitial(after);
        }
        if (tempVoicemeta != 0) {
            sb.insert(tempVoicemeta, " ");
        } else {
            System.out.println("error : " + str + "\n");
        }
        return sb.toString();
    }

    private static int containsInitial(String str) {
        String metaStr = formatMeta(str);
        for (String s : initialmeta) {
            if (metaStr.substring(1).contains(s))
                return metaStr.substring(1).indexOf(s) + 2;
        }
        if (metaStr.contains("r")) {
            if (metaStr.endsWith("er")) {
                return metaStr.length() - 1;
            }
            if (metaStr.contains("er") && !metaStr.startsWith("er")) {
                // er en 传入r en ，r属于word 1
                //返回r的后一个字母
                return metaStr.indexOf("r") + 2;
            } else {
                // he ren传入e ren，r属于word 2.
                return metaStr.indexOf("r") + 1;
            }
        }
        if (str.contains("gn")) {
            return str.indexOf("gn") + 2;
        }
        if (str.contains("g")) {
            if (str.contains("gg")) {
                return str.indexOf("gg") + 2;
            } else {
                String before = str.substring(0, str.indexOf("g"));
                String after = str.substring(str.indexOf("g") + 1);
                //如果g之后不能独立成拼音 g属于后面
                if (after.length() > 0) {
                    if (equalsFullFinal(after)) {
                        return str.indexOf("g") + 1;
                    }
                }

                //如果 before 不以n结尾 避免了"ng"，可以g分割
                if (!before.endsWith("n")) {
                    return str.indexOf("g") + 1;
                }
                //如果没有 ng  且 before的长度大于1
                if (!str.contains("ng")) {
                    if (!before.startsWith("i") && before.length() > 1) {
                        return str.indexOf("g") + 1;
                    }
                    if (before.startsWith("i") && before.length() > 2) {
                        return str.indexOf("g") + 1;
                    }
                }
            }
        }
        if (str.contains("n")) {
            if (str.contains("nn")) {
                return str.indexOf("nn") + 2;
            } else if (str.indexOf("n") != str.length() - 1) {
                String before = str.substring(0, str.indexOf("n"));
                String after = str.substring(str.indexOf("n") + 1);
                //如果n之后不能独立成拼音 n属于后面
                if (after.length() > 0) {
                    if (equalsFullFinal(after)) {
                        return str.indexOf("n") + 1;
                    }
                }

                if (!before.startsWith("i") && before.length() > 1) {
                    return str.indexOf("n") + 1;
                }
                if (before.startsWith("i") && before.length() > 2) {
                    return str.indexOf("n") + 1;
                }
            }
        }
        String middle = str.substring(0, str.length() - 1);
        if (!middle.contains("n") && !middle.contains("g")) {
            return containsFinal(str);
        }
        return 0;
    }

    public static int startWithFinal(String str) {
        String meta = formatMeta(str);
        for (String s : finalmeta) {
            if (meta.contains(s) && meta.startsWith(s)) {
                return s.length();
            }
        }
        return 0;
    }

    private static int containsFinal(String str) {
        String meta = formatMeta(str);
        for (String s : finalmeta) {
            if (meta.contains(s) && meta.endsWith(s)) {
                return meta.indexOf(s) + 1;
            }
        }
        return 0;
    }

    private static boolean equalsFinal(String str) {
        String meta = formatMeta(str);
        for (String s : finalmeta) {
            if (meta.equals(s)) {
                return true;
            }
        }
        return false;
    }

    private static boolean equalsFullFinal(String str) {
        String meta = formatMeta(str);
        for (String s : finalmetaFull) {
            if (meta.equals(s)) {
                return true;
            }
        }
        return false;
    }

    private static String formatMeta(String str) {
        for (String s : voicemark_a) {
            while (str.contains(s)) {
                str = str.substring(0, str.indexOf(s)) + "a" + str.substring(str.indexOf(s) + 1);
            }
        }
        for (String s : voicemark_o) {
            while (str.contains(s)) {
                str = str.substring(0, str.indexOf(s)) + "o" + str.substring(str.indexOf(s) + 1);
            }
        }
        for (String s : voicemark_e) {
            while (str.contains(s)) {
                str = str.substring(0, str.indexOf(s)) + "e" + str.substring(str.indexOf(s) + 1);
            }
        }
        for (String s : voicemark_i) {
            while (str.contains(s)) {
                str = str.substring(0, str.indexOf(s)) + "i" + str.substring(str.indexOf(s) + 1);
            }
        }
        for (String s : voicemark_u) {
            while (str.contains(s)) {
                str = str.substring(0, str.indexOf(s)) + "u" + str.substring(str.indexOf(s) + 1);
            }
        }
        for (String s : voicemark_v) {
            while (str.contains(s)) {
                str = str.substring(0, str.indexOf(s)) + "v" + str.substring(str.indexOf(s) + 1);
            }
        }
        return str;
    }
}