ARS-docs/scripts/parse-label.py

from pathlib import Path
from typing import Union, Callable
from pprint import pprint

import mistune

from qtoml.encoder import dumps
from qtoml.decoder import loads
from lib_not_dr import loggers
from lib_not_dr.types.options import Options


ast_markdown = mistune.create_markdown(renderer="ast")
ast_type = list[dict[str, Union[str, dict]]]


def get_all_iter(
    ast: ast_type, type: str, func: Callable[[ast_type], bool] = None
) -> tuple[int, dict]:
    for i, node in enumerate(ast):
        if node["type"] == type:
            if func is not None:
                if not func(node):
                    continue
            yield i, node
    yield -1, {}


def get_text(ast: ast_type) -> str:
    """
    返回第一个找到的字符串
    """
    if ast['type'] == 'text':
        return ast['raw']
    elif ast['type'] in ('link', 'block_text', 'list_item', 'list'):
        return get_text(ast['children'][0])
    print('unkown type', ast['type'], ast)


class TagParser(Options):
    module_root: Path
    tags: dict[str, list[str]] = {}
    tag_map: dict[str, list[str]] = {}
    logger: loggers.logger.Logger = None  # noqa

    def load_module(self, **kwargs):
        for readme in self.module_root.rglob("readme.md"):
            self.logger.debug(readme.absolute(), tag="load file")
            self.get_module_data(readme)

        tag_toml = dumps(self.tags)
        tag_path = self.module_root / ".." / "build" / "tags.toml"
        tag_path.parent.mkdir(parents=True, exist_ok=True)
        tag_path.touch(exist_ok=True)
        with open(tag_path, "w", encoding="utf-8") as file:
            file.write(tag_toml)

    def get_module_data(self, module_path: Path):
        with open(module_path, "r", encoding="utf-8") as f:
            file = f.read()

        ast = ast_markdown(file)
        if len(ast) == 0:
            return

        if ast[0] != {"type": "thematic_break"}:
            # 排除开头不是注释块的
            return

        self.logger.info(f"开始解析 {ast[1]}")
        config_code = ast[1].get("raw", "")
        config_dict = loads(config_code)
        self.logger.trace(config_dict)

        if not (tag_list := config_dict.get("tags")):
            self.logger.warn("未找到 tags", tag=str(module_path))
            return

        for tag in tag_list:
            if tag not in self.tags:
                self.tags[tag] = [module_path.__str__()]
            else:
                self.tags[tag].append(module_path.__str__())

    def init(self, **kwargs) -> bool:
        self.logger = loggers.get_logger()
        self.logger.global_level = 0
        self.load_tags()
        return False

    def load_tags(self):
        tag_list_path = Path("./tags/readme.md")
        if not tag_list_path.exists():
            self.logger.error("未找到 tags/readme.md")
            return False

        with open(tag_list_path, "r", encoding="utf-8") as f:
            file = f.read()

        tag_ast: ast_type = ast_markdown(file)
        start_tag = -1
        # 找到二级标题
        start_tag = get_all_iter(
            tag_ast, "heading", lambda node: node["attrs"]["level"] == 2
        )
        start_tag, node = next(start_tag)
        if start_tag == -1:
            self.logger.error("未找到二级标题")
            return False
        # 获取 tag
        tag_ast: ast_type = tag_ast[start_tag + 1 :]
        start_tag = get_all_iter(tag_ast, "list")
        start_tag, node = next(start_tag)
        if start_tag == -1:
            self.logger.error("未找到 tag 列表")
            return False
        tag_ast: ast_type = [
            item["children"]
            for item in tag_ast[start_tag]["children"]
            if item.get("type") == "list_item"
        ]
        # 顺手过滤一下
        for tag in tag_ast:
            # self.logger.debug([(item['type'], item['children']) for item in tag])
            this_tag = get_text(tag[0])
            if len(tag) == 1:
                # 只有一个 tag, 无别名
                self.tag_map[this_tag] = [this_tag]
                self.logger.debug(f'添加 tag {this_tag}')
            # len > 1
            elif len(tag) >= 1:
                # 寻找 "别名"
                tag_names = tag[1]['children']
                tag_list_start = get_all_iter(tag_names, 'list_item', lambda ast: get_text(ast) == '别名')
                tag_list_start, node = next(tag_list_start)
                if tag_list_start == -1:
                    continue
                sub_names = [this_tag]
                for sub_name in node['children'][1]['children']:
                    sub_names.append(get_text(sub_name))
                self.logger.debug(f"添加 tag {this_tag} 和别名 {sub_names}")
                self.tag_map[this_tag] = sub_names
        # 获取 tag 列表
        self.logger.info(f'可用 tag: {self.tag_map}')
        with open('./tags/tag.toml', 'w', encoding='utf-8') as file:
            file.write(dumps(self.tag_map))


if __name__ == "__main__":
    parser = TagParser(module_root=Path("modules"))
    parser.load_module()
    parser.logger.info(parser.tags)