LiteyukiBot-TriM/src/nonebot_plugins/trimo_plugin_handle/utils.py

import json
import random
from io import BytesIO
from pathlib import Path
from typing import Dict, List, Tuple, TypedDict, Optional

# from watchdog.observers import Observer
# from watchdog.events import FileSystemEventHandler, FileModifiedEvent

from PIL import ImageFont
from PIL.Image import Image as IMG
from PIL.ImageFont import FreeTypeFont
from pypinyin import Style, pinyin as py_get_pinyin

resource_dir = Path(__file__).parent / "resources"
fonts_dir = resource_dir / "fonts"
data_dir = resource_dir / "data"
handle_common_idiom_path = data_dir / "common_idioms.json"
handle_all_idiom_path = data_dir / "idioms.json"
handle_answer_path = data_dir / "answers.json"


class IdiomEntry(TypedDict):
    """
    成语答案类型
    """

    explanation: str
    pinyin: List[str]


HANDLE_COMMON_PHRASES: List[str] = json.load(
    handle_common_idiom_path.open("r", encoding="utf-8")
)
HANDLE_LEGAL_PHRASES: List[str] = json.load(
    handle_all_idiom_path.open("r", encoding="utf-8")
)
HANDLE_ANSWER_PHRASES: Dict[str, IdiomEntry] = json.load(
    handle_answer_path.open("r", encoding="utf-8")
)

# class LegalPhrasesModifiedHandler(FileSystemEventHandler):
#     """
#     Handler for resource file changes
#     """

#     def on_modified(self, event):
#         print(f"{event.src_path} modified, reloading resource...")
#         if "idioms.txt" in event.src_path:
#             global LEGAL_PHRASES
#             LEGAL_PHRASES = [
#                 idiom.strip()
#                 for idiom in idiom_path.open("r", encoding="utf-8").readlines()
#             ]
#         elif "answers.json" in event.src_path:
#             global ANSWER_PHRASES
#             ANSWER_PHRASES = json.load(
#                 answer_path.open("r", encoding="utf-8")
#             )


# Observer().schedule(
#     LegalPhrasesModifiedHandler(),
#     data_dir,
#     recursive=False,
#     event_filter=FileModifiedEvent,
# )

# 答案转换器
# json.dump({ i["word"]:{"explanation":i["explanation"], "pinyin":(lambda x : [i for j in pinyin(x, style=Style.TONE3, v_to_u=True) for i in j])(i["word"])} for i in json.load(open("answers_hard-old.json",encoding="utf-8"))},open("answers.json", "w",encoding="utf-8"), ensure_ascii=False, indent=4)


def wordbase_updater(
    idiom: str,
    explanation: Optional[str] = None,
    pinyin: Optional[List[str]] = None,
    hard: Optional[bool] = None,
) -> Tuple[bool, str, List[str]]:
    """
    idiom: 需要更新或新增的成语词条
    explanation: 词条的释义，若 `None` 且在词库中不存在，则默认为“未提供该成语的解释说明”
    pinyin: 词条的拼音，若 `None` 则默认求出拼音
    hard: 是否为难词，若 `None` 则不对难易词库修改

    return: 是否为新增成语，成语释义，词条的拼音
    """
    if (not idiom) or (len(idiom) != 4):
        raise ValueError("不可以非四字成语载入词库")

    if (existance := (idiom in HANDLE_LEGAL_PHRASES)) and (explanation is None):
        # 这个判断的顺序必须高于下面的判断语句
        explanation = HANDLE_ANSWER_PHRASES[idiom]["explanation"]

    if explanation is None:
        explanation = "未提供该成语的解释说明"

    if not existance:
        HANDLE_LEGAL_PHRASES.append(idiom)
        json.dump(
            HANDLE_LEGAL_PHRASES,
            handle_all_idiom_path.open("w", encoding="utf-8"),
            ensure_ascii=False,
            indent=4,
            sort_keys=True,
        )
    else:
        pinyin = pinyin or HANDLE_ANSWER_PHRASES[idiom]["pinyin"].copy()
    if hard:
        if idiom in HANDLE_COMMON_PHRASES:
            HANDLE_COMMON_PHRASES.remove(idiom)
            json.dump(
                HANDLE_COMMON_PHRASES,
                handle_common_idiom_path.open("w", encoding="utf-8"),
                ensure_ascii=False,
                indent=4,
                sort_keys=True,
            )
    else:
        if (idiom not in HANDLE_COMMON_PHRASES) and (hard is not None):
            HANDLE_COMMON_PHRASES.append(idiom)
            json.dump(
                HANDLE_COMMON_PHRASES,
                handle_common_idiom_path.open("w", encoding="utf-8"),
                ensure_ascii=False,
                indent=4,
                sort_keys=True,
            )

    HANDLE_ANSWER_PHRASES[idiom] = {
        "explanation": explanation,
        "pinyin": (
            pinyin := (
                pinyin
                or [
                    j
                    for i in py_get_pinyin(idiom, style=Style.TONE3, v_to_u=True)
                    for j in i
                ]
            )
        ),
    }
    json.dump(
        HANDLE_ANSWER_PHRASES,
        handle_answer_path.open("w", encoding="utf-8"),
        ensure_ascii=False,
        indent=4,
        sort_keys=True,
    )
    return not existance, explanation, pinyin


def legal_idiom(word: str) -> bool:
    return word in HANDLE_LEGAL_PHRASES

def random_idiom(is_hard: bool = False) -> Tuple[str, str]:
    answer = random.choice(HANDLE_LEGAL_PHRASES if is_hard else HANDLE_COMMON_PHRASES)
    return answer, HANDLE_ANSWER_PHRASES[answer]["explanation"]


# fmt: off
# 声母
INITIALS = [
    "zh", "z", "y", "x", "w", "t", "sh", "s", "r", "q", "p",
    "n", "m", "l", "k", "j", "h", "g", "f", "d", "ch", "c", "b"
]
# 韵母
FINALS = [
    "ün", "üe", "üan", "ü", "uo", "un", "ui", "ue", "uang",
    "uan", "uai","ua", "ou", "iu", "iong", "ong", "io", "ing",
    "in", "ie", "iao", "iang", "ian", "ia", "er", "eng", "en",
    "ei", "ao", "ang", "an", "ai", "u", "o", "i", "e", "a"
]
# fmt: on


def split_pinyin(py: str) -> Tuple[str, str, str]:
    """
    py: 输入的拼音
    return: 返回一个三元组，分别表示该拼音的声母，韵母，声调
    """

    if py[-1].isdigit():
        tone = py[-1]
        py = py[:-1]
    else:
        tone = ""
    initial = ""
    for i in INITIALS:
        if py.startswith(i):
            initial = i
            break
    final = ""
    for f in FINALS:
        if py.endswith(f):
            final = f
            break
    return initial, final, tone


def get_pinyin(
    idiom: str,
) -> List[Tuple[str, str, str]]:
    """
    idiom: 输入的汉字
    return: 返回一个元组，每个元素是一个三元组，分别表示该汉字的声母，韵母，声调
    """

    return [
        split_pinyin(py)
        for py in (
            wordbase_updater(idiom, explanation=None, pinyin=None, hard=None)[2]
            if HANDLE_ANSWER_PHRASES.get(idiom, None)
            else [
                j
                for i in py_get_pinyin(idiom, style=Style.TONE3, v_to_u=True)
                for j in i
            ]
        )
    ]


def save_jpg(frame: IMG) -> BytesIO:
    output = BytesIO()
    frame = frame.convert("RGB")
    frame.save(output, format="jpeg")
    return output


def load_font(name: str, fontsize: int) -> FreeTypeFont:
    return ImageFont.truetype(str(fonts_dir / name), fontsize, encoding="utf-8")