user-fixes.py

import itertools
from collections import defaultdict
from functools import partial

import regex as re
from opencc import OpenCC

generator_base = [
    "-start::!",
    "-start:project:!",
    "-start:template:!",
    "-start:category:!",
]
generator_more = generator_base + ["-start:module:! ", "-start:mediawiki:!"]

base: dict[str] = {
    "regex": True,
    "nocase": True,
    "exceptions": {
        "inside-tags": ["keep", "interwiki"],
    },
}

user_fixes = {}

# region misc
nbsp = "\xa0"

mid_dots_code = [
    721,
    903,
    1468,
    5867,
    8226,
    8231,
    8728,
    8729,
    8901,
    9210,
    9679,
    9702,
    9899,
    10625,
    11824,
    11825,
    11827,
    12539,
    42895,
    65381,
    65793,
]
mid_dots = "[" + "".join(chr(i) for i in mid_dots_code) + "]"
mid_dot = "\xb7"

user_fixes["misc"] = base | {
    "generator": generator_base,
    "replacements": [
        (nbsp, " "),
        (mid_dots, mid_dot),
        ("－－", "——"),
        (r"<!---->|", ""),
        ("“", "「"),
        ("”", "」"),
        ("【", "『"),
        ("】", "』"),
        (r"(?<!==)\s*\n==", r"\n\n=="),
        (r"==\n\s*", r"==\n"),
        (r"\n{3,}", r"\n\n"),
        ("</br>", "<br>"),
        (r"'''(\{\{R\|.*?\}\})'''", r"\1"),
    ],
}

# endregion


# region date
def match_to_yyyymmdd(mouth: int, match: re.Match) -> str:
    return f"{match.group(2)}-{str(mouth).zfill(2)}-{match.group(1).zfill(2)}"


user_fixes["date"] = base | {
    "generator": generator_base,
    "replacements": [
        (
            rf"{month}\s*(\d+)\s*[，,]\s*(\d+)",
            # avoid late binding of i
            partial(match_to_yyyymmdd, i + 1),
        )
        for i, month in enumerate(
            (
                "January",
                "February",
                "March",
                "April",
                "May",
                "June",
                "July",
                "August",
                "September",
                "October",
                "November",
                "December",
            )
        )
    ],
}
# endregion

# region anti-ve
user_fixes["anti-ve"] = {
    "regex": True,
    "nocase": True,
    "exceptions": {
        "inside-tags": ["keep", "interwiki", "template", "table"],
    },
    "generator": generator_base,
    "replacements": [
        ("<br>", r"\n\n"),
    ],
}
# endregion

# region para
user_fixes["para"] = base | {
    "generator": generator_more,
    "replacements": [
        (rf"\|\s*{o}\s*=", f"| {n} =")
        for o, n in [
            ("Name", "name"),
            ("Image", "image"),
            ("Kanji", "name_ja_kanji"),
            ("Romaji", "name_ja_romaji"),
            ("Alias", "alias"),
            ("Nickname", "nickname"),
            ("Race", "race"),
            ("Gender", "gender"),
            ("Birthday", "birthday"),
            ("Age", "age"),
            ("Hair Color", "hair"),
            ("Eye Color", "eyes"),
            ("Height", "height"),
            ("Weight", "weight"),
            ("Affiliation", "affiliation"),
            ("Previous Affiliation", "previous_affiliation"),
            ("Occupation", "occupation"),
            ("Previous Occupation", "previous_occupation"),
            ("Status", "status"),
            ("Relatives", "relatives"),
            ("Magic", "magic"),
            ("Divine Protection", "divine_protection"),
            ("Authority", "authority"),
            ("Weapon", "weapon"),
            ("Equipment", "equipment"),
            ("Anime", "anime"),
            ("Light Novel", "novel"),
            ("Game", "game"),
            ("Manga", "comic"),
            ("Japanese Voice", "voice_ja"),
            ("English Voice", "voice_en"),
            ("Pages", "pages_ja"),
            ("ISBN", "isbn_ja"),
            ("Release Date", "date_ja"),
            ("Painter", "painter"),
            ("Cover", "cover"),
            ("Previous", "该参数已弃用"),
            ("Next", "该参数已弃用"),
            ("Image-Size", "该参数已弃用"),
        ]
    ],
}
# endregion

# region gallery
user_fixes["gallery"] = base | {
    "generator": "-catr:图库",
    "replacements": [
        (o + r"\s*(?==)", n)
        for o, n in [
            ("Anime", "动画"),
            ("Season 1", "第一季"),
            ("Season 2", "第二季"),
            ("Light Novels?", "小说"),
            ("Main Series", "正传"),
            ("Tanpenshuu", "月刊CA短篇"),
            ("Side Content", "特典SS"),
            ("Side Stories", "特典SS"),
            ("Manga", "漫画"),
            ("Daisshou", "第1章"),
            ("Dainishou", "第2章"),
            ("Daisanshou", "第3章"),
            ("Daiyonshou", "第4章"),
            ("Anthology", "官方同人精选集"),
            ("Games?", "游戏"),
            ("-Infinity", "INFINITY"),
            ("The Prophecy of the Throne", "虚假的王选候补"),
            ("Forbidden Book and the Mysterious Spirit", "禁书与谜之精灵"),
            (r"Misc(ellaneous|\.)?", "其他"),
        ]
    ],
}
# endregion

# region heading
user_fixes["heading"] = base | {
    "generator": generator_more,
    "replacements": [
        ("(?<== )" + o + "(?= =)", n)
        for o, n in [
            ("Relationships", "关系"),
            ("Synopsis", "梗概"),
            ("Summary", "梗概"),
            ("Gallery", "图库"),
            ("Image Gallery", "图库"),
            ("Appearance", "外貌"),
            ("Personality", "性格"),
            ("Abilities", "能力"),
            ("Trivia", "你知道吗"),
            ("Lyrics?", "歌词"),
            ("Characters", "登场人物"),
            ("References?", "注释与外部链接"),
        ]
    ],
}
# endregion

# region translation
flatten = itertools.chain.from_iterable
s2t = OpenCC("s2t.json").convert

similar_chars = (
    "珥尔耳鲁露卢勒拉菈利莉丽里吕李",
    "书舒修休杰珠裘鸠吉基其姬奇齐",
    "库克古谷格铬科空柯寇",
    "赫黑海哈",
    "托图多朵特提狄缇德蒂黛",
    "肯卡嘉加伽茄贾",
    "丝斯司兹茨",
    "娅亚雅阿安",
    "菲福飞弗芙伏佛沃",
    "莎沙纱萨",
    "西希席叙",
    "腾滕登坦",
    "娜纳那",
    "艾爱埃",
    "贝培佩",
    "赛瑟札",
    "塞泽佐",
    "琉流硫",
    "乌厄",
    "因茵",
    "奥欧",
    "威维",
    "尤由",
    "昴昂",
    "梅麦",
    "汀丁",
    "碧比",
    "米蜜",
    "蕾雷菜莱",
    "文温",
    "霍荷",
    "谜迷",
    ".·",
)


class SimilarCharsMap(defaultdict):
    """字符到相似字符的映射。"""

    def __missing__(self, key):
        """一个字符总是与它本身相似。"""
        self[key] = key
        return key


sc_map = SimilarCharsMap()  # singleton
sc_map |= {c: sc for sc in similar_chars for c in sc}


def f(chars: str):
    """
    返回匹配相似字符的正则表达式。

    短命名以方便大量使用。

    :param chars: 任意个字符
    :return: "[similar_chars]"
    """
    return (
        "["
        + "".join(sorted(set(flatten(sc_map[c] + s2t(sc_map[c]) for c in chars))))
        + "]"
    )


def p2o(pattern: str):
    """返回传入的正则表达式对应的所有可能译名对应的正则表达式。"""
    return "".join(c if c in "?!(|)=<" else f(c) for c in pattern)


def p2n(pattern: str):
    """返回传入的正则表达式对应的标准译名。"""
    return re.sub(r"\(.*?\)|\?", "", pattern)


def get_repl_func(name: str):
    """返回name对应的替换函数。"""

    def func(match: re.Match) -> str:
        """若为标准译名对应的繁体名则原样返回，否则返回标准译名。"""
        cur = match.group()
        if cur == s2t(name):
            return cur
        return name

    return func


user_fixes["translation"] = base | {
    "generator": generator_more,
    "replacements": [
        (p2o(p), get_repl_func(p2n(p)))
        for p in [  # 普通的
            "丝碧卡",
            "亚拉基亚",
            "亨克尔",
            "伽那库斯",
            "佛拉基亚",
            "佛格",
            "克林德",
            "利布雷",
            "卡佩拉",
            "卡吉雷斯",
            "卡尔兰",
            "卡尔斯腾",
            "卡德蒙",
            "卡拉拉基",
            "卡斯图鲁平原",
            "卡萝",
            "卡蜜拉",
            "古斯提科",
            "史泰德",
            "裘斯",
            "塞西尔斯",
            "夏乌拉",
            "夏库纳尔",
            "多鲁特洛",
            "奇力塔卡",
            "娅艾",
            "安妮罗泽",
            "密涅瓦",
            "尤里乌斯",
            "希尔菲",
            "希斯尼娅",
            "帕克",
            "帕特拉修",
            "库乌德",
            "库奥克",
            "库珥修",
            "库鲁刚",
            "弗琉盖尔",
            "弗莱巴尔",
            "戴因",
            "提姆兹",
            "文森特",
            "斯芬克丝",
            "普拉姆",
            "普莉希拉",
            "普莉斯卡",
            "李凯尔特",
            "查普",
            "格拉姆达特",
            "格拉希丝",
            "格蕾丝",
            "梅卡德",
            "梅娜",
            "梅拉奎拉",
            "欧尔尼娅",
            "欧米伽",
            "比恩",
            "汉娜",
            "波尔多",
            "波尔肯尼卡",
            "泰玛艾",
            "特蕾西亚",
            "玛洛妮",
            "琉兹",
            "琉加",
            "璞可",
            "皮波特",
            "盖因",
            "碧翠丝",
            "福斯特",
            "米尔多",
            "米捷尔",
            "米路德",
            "约书亚",
            "缇丰",
            "缇莉艾娜",
            "罗伊",
            "梅札斯",
            "罗姆爷",
            "艾佐",
            "加德纳",
            "艾力欧尔大森林",
            "艾奇多娜",
            "艾米",
            "艾米莉娅",
            "艾西亚湿地",
            "芙蕾德莉卡",
            "荒地的合辛",
            "莉可莉丝",
            "莉莉安娜",
            "莎克拉",
            "莎缇拉",
            "莱伊",
            "莱普",
            "菜月菜穗子",
            "菜月贤一",
            "菲莉丝",
            "萨尔姆",
            "葛利奇",
            "蒂亚斯",
            "席里乌斯",
            "谢尔盖",
            "贾雷克",
            "赛赫麦特",
            "赫克托尔",
            "赫罗西欧",
            "赫鲁贝尔",
            "达德利",
            "达芙妮",
            "迈尔斯",
            "邱登",
            "里卡多",
            "铁之牙",
            "阿尼茉尼",
            "阿拉姆村",
            "阿汉",
            "阿珍",
            "阿顿",
            "雷古勒斯",
            "露伊",
            "露格尼卡",
            "马可仕",
            "麦克罗托夫",
            "黑塔罗",
            "基尔提拉乌",
            "巴登凯托斯",
            "塞坦塔",
            "特里亚斯",
            "欧德古勒斯",
            "埃尔纱幕",
            "汀泽尔",
            "贝阿托莉丝",
            "奥斯洛",
            "雷金",
            "柯司兹尔",
            "福尔图娜",
            "汤普森",
            "苏文",
            "佳莉华",
            "伊莉雅",
            "塞蕾丝缇雅",
            "禁书与谜之精灵",
            "弗洛普",
            "米迪娅姆",
            "罗安",
            "贾马尔",
            "瓜拉尔",
            "巴多海姆",
            "托斯卡",
            "拉米亚",
            "弗兰德斯",
            "米泽尔妲",
            "塔丽塔",
            "乌塔卡塔",
            "槜克尔",
            "高朗",
            "梅琳达",
            "亚历克",
            "尤尔娜",
            "古斯塔夫",
            "乔拉",
            "贝尔斯特兹",
            "塞丽娜",
            "卡秋娅",
            "法尔塞路",
            "奇夏",
            "秀佐",
            "戴纳斯",
            "拉扎克",
            "巴尔罗伊",
            "莫古洛",
            "葛路比",
            "罗德利格斯",
            "塞尔连",
            "曼弗雷德",
            "莫索里特",
            "马德林",
            "奥尔巴特",
            "卡夫马",
            "塞尔菲苏",
            "瓦尔格伦",
            "奈基",
            "托尔塔",
            "奔加姆",
            "迪德利",
            "拉兹克鲁",
            "利夫坦",
            "希洛洛",
            "伊戈尔",
            "罗格蕾丝",
            "玛卡利斯塔",
            "维库塔",
            "姆斯朗",
            "鲁斯贝尔",
            "基利安",
            "乌比尔克",
            "荷奈特",
            "伊德拉",
            "赫莱茵",
            "玛里乌里",
            "马泽里安",
            "弗拉米尔",
            "雅克托尔",
            "阿尔维里奥",
            "吉奥尼斯",
            "兰德哈尔",
            "杰比聂尔",
            "爱梅拉妲",
            "德莱森",
            "卡欧斯弗雷姆",
            "美佐里亚",
            "托尔塔",
            "莱杰尔",
            "巴罗伊",
        ]
        + [  # 多字少字的
            "菜月·?昴",
            "安娜斯?塔西娅",
            "培提尔其乌?斯",
            "威尔海(鲁)?姆",
            "莱茵哈鲁?特",
            "罗兹瓦尔?",
            "帕尔?米拉",
        ]
        + [  # 需要特判的
            "加菲尔(?!丝|特|艾)",
            "拉菲尔(?!丝|特|艾)",
            "利格鲁(?!卡|姆)",
            "佩特拉(?!其乌斯|姆)",
            "(?<!艾米)莉亚拉",
            "(?<!萨|伽|托)拉姆(?!莉鲁)",
            "(?<!帕)贝尔托",
            "(?<!帕)提修雅",
            "凯缇(?!尔|斯)",
            "(?<!莱茵哈鲁)缇碧(?!翠)",
            "艾尔莎(?!幕)",
            "梅尔蒂(?!典)",
            "弗雷德(?!莉卡)",
            "(?<!丢)芙拉姆",
            "(?<!加)弗利艾",
            "(?<!帕|夏|拉|札|悟)库娜",
            "荷莉(?!蒙)",
            "(?<!法)拉塞尔",
            "威茨(?!利)",
            "(?<!格)莉西亚",
            "(?<!芙蕾|佩)多尔肯(?!罗登|普里恩)",
            "(?<!罗)伊娜",
            "潘多拉(?!顿)",
            "(?<!卡欧斯弗)蕾姆",
        ]
    ]
    + [
        (o, get_repl_func(n))
        for o, n in [  # 手动添加的替换组
            (rf"{f('凛淋萍平苹')}{f('果')}", "{{Ringa}}"),
            (f"{f('流')}{f('丽')}{f('连')}", "琉璃泪"),
            (f"{f('贝')}{f('阿')}{f('托')}{f('莉')}{f('丝')}", "碧翠丝"),
            (f"{f('多')}{f('娜')}{f('狐')}", "领子多娜"),
            (f"{f('欧')}德", "欧德"),
            (f"{f('奥')}托", "奥托"),
            (f"{f('修')}{f('尔')}特", "修尔特"),
            (f"{f('基')}{f('尔')}提", "基尔提"),
            (f"阿{f('尔')}(?!{f('姆伯')}|{f('基')}{f('亚')})", "阿尔"),
            (f"{f('阿')}{f('斯')}{f('特')}{f('雷利')}{f('亚')}", "阿斯特雷亚"),
            (f"欧德(?!{f('古')}{f('勒')}{f('斯')})|{f('魂')}{f('力')}", "{{Od}}"),
            (f"{f('拉')}{f('格')}{f('纳')}|{f('源')}{f('池')}", "{{Laguna}}"),
            (f"{f('空')}{f('斯')}{f('图')}{f('卢')}", "柯司兹尔"),
            (f"其{f('他它她')}", "其他"),
            (r"\{\{Od\}\}\{\{Laguna\}\}", "{{Od}}·{{Laguna}}"),
            (
                "(?<!禁书与谜之|人工|自然|契约|[大邪微准])" f"{f('精')}{f('灵')}" "(?!骑士|[术使])",
                "{{Seirei or Elf}}",
            ),
            (f"{f('妖')}{f('精')}", "{{Yousei or Elf}}"),
            (r"(?<=半)\{\{(Seirei|Yousei) or Elf\}\}", "{{Elf}}"),
            ("风见的加护", "视风的加护"),
            (f"{f('蕾')}泽", "蕾泽"),
            (f"{f('雷')}佐(?!提雅)", "雷佐"),
        ]
    ],
}
_ = [  # 特判太麻烦的，不处理
    "梅丽(?!乌|奎拉|蒂)",
    "(?<!格|芙|·)雷德",
    "(?<!格拉姆)达兹(?!利)",
    "(?<!莉)卢安娜",
    "(?<!文森)狄加",
    "沃尔夫",
    "弗鲁夫",
    "雷诺",
    "(?<!加|拉)菲鲁特(?!娜)",
    "弗尔多(?!娜)",
    "蜜蜜",
    "托托",
    "(?<!赫鲁|卡)贝利",
    "克莱茵",
    "克雷茵",
    "(?<!艾奇|福尔)提娜",
    "亚雷",
    "艾达",
    "坦萨",
    "海伦",
    "欧尔菲",
    "柯林",
    "格林(?!德)",
    "(?<!攻|利|摩)柯蕾特",
    "(?<!多萝西|艾米莉|约书|贝|卡秋)亚齐",
    "(?<!艾尔)萨德(?!拉|兰)",
]
# endregion

fixes: dict
# noinspection PyUnboundLocalVariable
fixes.update(user_fixes)