# -*- coding = utf-8 -*-
# @Project : Logseq标题优化
# @File : Logseq格式优化.py
# @time : 2024/05/11
# @Author : Vandee
# @Description ：适合yaml为Logseq的::样式



import re
import os
# Frontmatter consts for start and end of Frontmatter YAML heading in Markdown
FRONTMATTER_START_STR = "---\n"
FRONTMATTER_END_STR = "---\n\n"
FRONTMATTER_PARAM_NAME_REGEXP: str = r"[A-Za-z0-9-_.]+::\s"
LOGSEQ_LIST_REGEXP: str = r"^[\s\t]*- "


def load_logseq_sanitized(file_path: str, encoding: str = "utf-8") -> list[str]:

    with open(file_path, "r", encoding=encoding) as f:
        lines: list[str] = f.readlines()

    return_lines: list[str] = []
    for line in lines:
        # we skip empty lines
        if line in ("\n", "- \n", "-\n", "- \n"):
            continue

        # we remove "- " or "  " from the beginning of line as
        # it's Logseq specific "everyghing is a list" approach
        if line.startswith(("- ", "  ")):
            line = line[2:]

        # we remove first occurance of tab character ("\\t") from a line as
        # it's indicating list item
        if line.startswith("\t"):
            line = line.replace("\t", "", 1)

        return_lines.append(line)

    return return_lines


def logseq2markdown(logseq_lines: list[str]) -> str:
    """Goes through list of Logseq sanitized lines (provided by `load_logseq_sanitized()`) and
    translates them to proper Markdown and Frontmatter.

    Args:
        logseq_lines (list[str]): List of sanitized lines from Logseq file loader

    Returns:
        str: String containing Frontmatter header in YML format followed by proper Markdown.
    """
    mk_content: list[str] = []

    # Using dict here as we don't want to have duplicate parameter names in Frontmatter
    # (each should have unique indentifier).

    mk_frontmatter: dict[str, str] = {}

    param_regex = re.compile(FRONTMATTER_PARAM_NAME_REGEXP)
    logseq_list_regexp = re.compile(LOGSEQ_LIST_REGEXP)

    for line in logseq_lines:
        params_result = param_regex.findall(line)
        logseq_list_result = logseq_list_regexp.findall(line)

        # if line containts "logseq.order-list-type:: number" it should be
        # numbered list and this line should be ommited
        if -1 != line.find("logseq.order-list-type:: number"):
            line = mk_content.pop()
            line = line.lstrip("\n")
            line = line.replace("- ", "1. ", 1)
            line = line.replace("\t", "    ")
            mk_content.append(line)

        # if line containts "logseq.order-list-type:: bulllet" it should be
        # bullet point list and this line should be ommited
        elif -1 != line.find("logseq.order-list-type:: bullet"):
            line = mk_content.pop()
            line = line.lstrip("\n")
            line = line.replace("- ", "* ", 1)
            line = line.replace("\t", "    ")
            mk_content.append(line)

        # if line is proper unordered list we parse it as such
        elif logseq_list_result:
            line = line.lstrip("\n")
            line = line.replace("\t", "    ")
            mk_content.append(line)

        # if line starts with "# " (meaning h1 in html) we parse it as Frontmatter "title:" param
        elif line.startswith("# "):
            mk_frontmatter["title"] = '"' + line[2:].strip() + '"'

        # if line doesn't have any Logseq-specific parameters like numbered or bullet list in it
        # and has Frontmatter param format we add it to Frontmatter header
        elif params_result:
            mk_frontmatter[params_result[0][0:-3]] = line[len(params_result[0]) :]

        # otherwise we add it to content as any other Markdown element
        else:
            mk_content.append("\n" + line)

    return_string: str = ""
    if mk_frontmatter:
        return_string = (
            FRONTMATTER_START_STR
            + "\n".join(
                [
                    f"{fm_item[0]}: {fm_item[1].strip()}"
                    for fm_item in list(mk_frontmatter.items())
                ]
            )
            + "\n"
            + FRONTMATTER_END_STR
        )

    return return_string + "".join(mk_content)

def process_folder(input_folder: str, output_folder: str):
    """处理输入文件夹中的文件，并将结果导出到输出文件夹。

    Args:
        input_folder (str): 输入文件夹路径。
        output_folder (str): 输出文件夹路径。
    """
    # 确保输出文件夹存在，如果不存在则创建
    if not os.path.exists(output_folder):
        os.makedirs(output_folder)

    # 遍历输入文件夹中的每个文件
    for file_name in os.listdir(input_folder):
        if file_name.endswith(".md"):  # 仅处理以 .md 结尾的文件
            input_file_path = os.path.join(input_folder, file_name)
            output_file_path = os.path.join(output_folder, file_name)

            # 加载、处理和导出文件
            logseq_lines = load_logseq_sanitized(file_path=input_file_path)
            markdown_content = logseq2markdown(logseq_lines)

            # 将转换后的内容写入到输出文件中
            with open(output_file_path, "w", encoding="utf-8") as f:
                f.write(markdown_content)

            print(f"文件 '{file_name}' 已处理并导出到 '{output_folder}'")

# 主程序中使用命令行提示用户输入文件夹路径
if __name__ == "__main__":
    # 提示用户输入输入文件夹路径
    input_folder = input("请输入输入文件夹路径：")

    # 提示用户输入输出文件夹路径
    output_folder = input("请输入输出文件夹路径：")

    process_folder(input_folder, output_folder)