#!/usr/bin/env python3
"""
阅读理解文章段落拆分脚本 - 简化版
"""

import os
import re
import datetime

def process_file(input_file_path, output_dir):
    """处理单个文件"""
    with open(input_file_path, 'r', encoding='utf-8') as f:
        content = f.read()
    
    # 移除开头的yaml头部
    yaml_end = content.find('---', 3)
    if yaml_end != -1:
        yaml_end = content.find('---', yaml_end + 3)
        if yaml_end != -1:
            content = content[yaml_end + 3:]
    
    # 清理内容
    content = content.strip()
    
    # 移除开头的标题行
    lines = content.split('\n')
    article_lines = []
    
    # 跳过标题和表格行
    skip_title = True
    for line in lines:
        stripped_line = line.strip()
        
        # 跳过标题行
        if stripped_line.startswith('#') or stripped_line.startswith('###'):
            continue
            
        # 跳过表格行
        if stripped_line.startswith('|') and '|' in stripped_line:
            # 检查是否是表头
            if '---' in stripped_line:
                continue
            # 检查是否是数据行
            if len(stripped_line.split('|')) > 2:
                continue
        
        # 跳过题目和答案行
        if re.match(r'^\d+\.', stripped_line) and not re.search(r'[A-D]\.', stripped_line):
            # 这可能是题目行，跳过
            continue
            
        # 跳过答案行
        if re.match(r'^\d+\.\s*[A-Z]', stripped_line):
            continue
            
        # 跳过空行
        if not stripped_line:
            continue
            
        article_lines.append(line)
    
    # 重新组合文章内容
    article_content = '\n'.join(article_lines)
    
    # 按段落分割（以空行分割）
    paragraphs = []
    current_paragraph = ""
    
    # 按行分割内容
    lines = article_content.split('\n')
    
    for line in lines:
        line = line.strip()
        if not line:  # 空行
            if current_paragraph.strip():
                paragraphs.append(current_paragraph.strip())
                current_paragraph = ""
        else:
            if current_paragraph:
                current_paragraph += "\n" + line
            else:
                current_paragraph = line
    
    # 添加最后一个段落
    if current_paragraph.strip():
        paragraphs.append(current_paragraph.strip())
    
    # 创建output目录
    os.makedirs(output_dir, exist_ok=True)
    
    # 为每个段落创建独立的文件
    base_name = os.path.splitext(os.path.basename(input_file_path))[0]
    
    for i, paragraph in enumerate(paragraphs):
        if not paragraph.strip():
            continue
            
        # 生成文件名
        # 提取段落的前几个字作为标题
        title = paragraph[:20].replace('\n', ' ').strip()
        # 移除特殊字符
        title = re.sub(r'[^\w\s\u4e00-\u9fff]', '', title)
        if len(title) > 20:
            title = title[:17] + "..."
        
        # 生成文件名
        filename = f"{i+1}_{title}.md"
        filepath = os.path.join(output_dir, filename)
        
        # 生成tags（从原文中提取）
        tags = ['阅读理解', '文章', '学习']
        
        # 生成日期
        now = datetime.datetime.now()
        date_str = now.strftime("%Y-%m-%d %H:%M:%S")
        
        # 构建文件内容
        file_content = f"""---
date:
  {date_str}: 
tags: [{', '.join(tags)}]
from:
---
{paragraph}
"""
        
        # 写入文件
        with open(filepath, 'w', encoding='utf-8') as f:
            f.write(file_content)
        
        print(f"已创建文件: {filepath}")

def main():
    """主函数"""
    source_dir = '/Users/zhangkun/Documents/myNotes/study/yuedu/source'
    output_dir = '/Users/zhangkun/Documents/myNotes/study/yuedu/output'
    
    # 处理所有文件
    files_to_process = [
        '1-_二阅读理解短文20篇.md',
        '21-_二阅读理解短文20篇.md', 
        '121-_二阅读理解短文20篇.md'
    ]
    
    for filename in files_to_process:
        file_path = f'{source_dir}/{filename}'
        if os.path.exists(file_path):
            print(f"正在处理文件: {filename}")
            process_file(file_path, output_dir)
        else:
            print(f"文件不存在: {file_path}")

if __name__ == "__main__":
    main()