#!/usr/bin/env python3
"""律师业务操作指引 - Markdown转PDF"""

import os
import re
import subprocess
from pathlib import Path

BASE_DIR = Path("/root/.openclaw/workspace/知识库/律师业务操作指引")
PDF_DIR = BASE_DIR

def run_cmd(c):
    result = subprocess.run(c, shell=True, capture_output=True, text=True)
    return result.stdout, result.stderr

def md_to_pdf(input_md, output_pdf, title=""):
    """将markdown文件转换为PDF，使用weasyprint"""
    
    # 先转成HTML
    html_file = input_md.with_suffix('.html')
    
    # 读取markdown
    with open(input_md, 'r', encoding='utf-8') as f:
        content = f.read()
    
    # 提取标题（如果没有传入）
    if not title:
        match = re.search(r'^#\s+(.+)$', content, re.MULTILINE)
        if match:
            title = match.group(1)
    
    # 简单的markdown解析器
    def parse_md(md):
        lines = md.split('\n')
        result = []
        in_ul = False
        in_ol = False
        in_table = False
        table_lines = []
        
        i = 0
        while i < len(lines):
            line = lines[i]
            stripped = line.strip()
            
            # 跳过front matter
            if stripped == '---':
                i += 1
                continue
            
            # 标题
            if stripped.startswith('# '):
                if in_ul:
                    result.append('</ul>')
                    in_ul = False
                if in_ol:
                    result.append('</ol>')
                    in_ol = False
                if in_table:
                    result.append('</table>')
                    in_table = False
                level = len(stripped) - len(stripped.lstrip('#'))
                tag = f'h{level}'
                text = stripped[level+1:]
                result.append(f'<{tag}>{text}</{tag}>')
            
            # h2-h6
            elif stripped.startswith('## '):
                if in_ul: result.append('</ul>'); in_ul = False
                if in_ol: result.append('</ol>'); in_ol = False
                if in_table: result.append('</table>'); in_table = False
                result.append(f'<h2>{stripped[3:]}</h2>')
            elif stripped.startswith('### '):
                if in_ul: result.append('</ul>'); in_ul = False
                if in_ol: result.append('</ol>'); in_ol = False
                if in_table: result.append('</table>'); in_table = False
                result.append(f'<h3>{stripped[4:]}</h3>')
            elif stripped.startswith('#### '):
                if in_ul: result.append('</ul>'); in_ul = False
                if in_ol: result.append('</ol>'); in_ol = False
                if in_table: result.append('</table>'); in_table = False
                result.append(f'<h4>{stripped[5:]}</h4>')
            
            # 分隔线
            elif stripped in ('---', '***', '___'):
                if in_ul: result.append('</ul>'); in_ul = False
                if in_ol: result.append('</ol>'); in_ol = False
                if in_table: result.append('</table>'); in_table = False
                result.append('<hr/>')
            
            # 表格
            elif '|' in stripped:
                if not in_table:
                    in_table = True
                    table_lines = []
                if not re.match(r'^[\|\-\s:]+$', stripped):
                    cells = [c.strip() for c in stripped.split('|')[1:-1]]
                    table_lines.append(cells)
                i += 1
                continue
            else:
                if in_table:
                    in_table = False
                    result.append('<table>')
                    for j, row in enumerate(table_lines):
                        tag = 'th' if j == 0 else 'td'
                        cells_html = ''.join(f'<{tag}>{c}</{tag}>' for c in row)
                        result.append(f'<tr>{cells_html}</tr>')
                    result.append('</table>')
                    table_lines = []
                
                # 无序列表
                if stripped.startswith('- ') or stripped.startswith('* '):
                    if not in_ul:
                        result.append('<ul>')
                        in_ul = True
                    result.append(f'<li>{stripped[2:]}</li>')
                # 有序列表
                elif re.match(r'^\d+\.\s+', stripped):
                    if not in_ol:
                        result.append('<ol>')
                        in_ol = True
                    num_match = re.match(r'^(\d+)\.\s+(.*)', stripped)
                    result.append(f'<li>{num_match.group(2)}</li>')
                # 空行
                elif stripped == '':
                    if in_ul:
                        result.append('</ul>')
                        in_ul = False
                    if in_ol:
                        result.append('</ol>')
                        in_ol = False
                    result.append('<p>&nbsp;</p>')
                # 段落
                else:
                    if in_ul:
                        result.append('</ul>')
                        in_ul = False
                    if in_ol:
                        result.append('</ol>')
                        in_ol = False
                    # 处理行内样式
                    text = stripped
                    # 粗体 **text**
                    text = re.sub(r'\*\*(.+?)\*\*', r'<strong>\1</strong>', text)
                    # 链接 [text](url) -> text
                    text = re.sub(r'\[(.+?)\]\(.+?\)', r'\1', text)
                    # 行内代码 `code`
                    text = re.sub(r'`(.+?)`', r'<code>\1</code>', text)
                    result.append(f'<p>{text}</p>')
            
            i += 1
        
        # 关闭打开的标签
        if in_ul: result.append('</ul>')
        if in_ol: result.append('</ol>')
        if in_table:
            result.append('<table>')
            for j, row in enumerate(table_lines):
                tag = 'th' if j == 0 else 'td'
                cells_html = ''.join(f'<{tag}>{c}</{tag}>' for c in row)
                result.append(f'<tr>{cells_html}</tr>')
            result.append('</table>')
        
        return '\n'.join(result)
    
    body_html = parse_md(content)
    
    html = f'''<!DOCTYPE html>
<html lang="zh-CN">
<head>
<meta charset="utf-8">
<title>{title}</title>
<style>
@import url('https://fonts.googleapis.com/css2?family=Noto+Sans+SC:wght@400;700&display=swap');

body {{
  font-family: 'AR PL UMing CN', 'Noto Sans CJK SC', 'WenQuanYi Micro Hei', 
               'Source Han Sans CN', 'Droid Sans Fallback', sans-serif;
  font-size: 10.5pt;
  line-height: 1.8;
  text-align: justify;
  text-indent: 2em;
  margin: 2cm 2.5cm;
  color: #222;
}}

h1 {{
  font-size: 18pt;
  text-align: center;
  margin: 0.8em 0 0.4em;
  color: #1a1a1a;
  font-weight: bold;
}}

h2 {{
  font-size: 14pt;
  margin: 1em 0 0.4em;
  color: #1e4d8c;
  border-bottom: 1.5px solid #1e4d8c;
  padding-bottom: 0.2em;
}}

h3 {{
  font-size: 12pt;
  margin: 0.8em 0 0.3em;
  color: #2c5aa0;
}}

h4 {{
  font-size: 10.5pt;
  margin: 0.6em 0 0.3em;
  color: #444;
}}

p {{
  margin: 0.25em 0;
  text-indent: 2em;
  line-height: 1.8;
}}

ul, ol {{
  margin: 0.3em 0 0.3em 2em;
  padding-left: 1em;
}}

li {{
  margin: 0.15em 0;
  line-height: 1.8;
}}

table {{
  border-collapse: collapse;
  width: 100%;
  margin: 0.6em 0;
  font-size: 9.5pt;
}}

th {{
  background-color: #1e4d8c;
  color: white;
  padding: 6px 8px;
  text-align: center;
  font-weight: bold;
}}

td {{
  padding: 4px 8px;
  border: 1px solid #aaa;
  vertical-align: top;
}}

tr:nth-child(even) {{
  background-color: #f4f7fb;
}}

hr {{
  border: none;
  border-top: 1px solid #ccc;
  margin: 0.8em 0;
}}

strong {{
  font-weight: bold;
  color: #1a1a1a;
}}

code {{
  font-family: monospace;
  background-color: #f5f5f5;
  padding: 0 3px;
}}

.source-info {{
  text-align: center;
  color: #666;
  font-size: 9pt;
  margin: 0.3em 0;
}}

.title-block {{
  text-align: center;
  margin: 1em 0;
}}

.toc-box {{
  background: #f0f4ff;
  border-left: 4px solid #1e4d8c;
  padding: 1em 1.2em;
  margin: 1em 0;
}}

.toc-box h2 {{
  border: none;
  margin-top: 0;
  color: #1e4d8c;
}}
</style>
</head>
<body>
{body_html}
</body>
</html>'''
    
    with open(html_file, 'w', encoding='utf-8') as f:
        f.write(html)
    
    # 转换PDF
    cmd_str = f"weasyprint {html_file} {output_pdf} 2>&1"
    stdout, stderr = run_cmd(cmd_str)
    
    # 清理临时HTML
    try:
        html_file.unlink()
    except:
        pass
    
    if os.path.exists(output_pdf):
        size = os.path.getsize(output_pdf)
        return True, f"成功 ({size/1024:.1f}KB)"
    else:
        return False, f"失败: {stderr[:300]}"

# 已收集的指引列表
guides = [
    ("上海律协_律师从事关税法律业务操作指引_2025", "关税"),
    ("上海律协_律师代理医疗科技成果转化业务操作指引_2024", "医药健康"),
    ("上海律协_律师办理公司对外担保业务操作指引_2024", "公司与商事"),
    ("上海律协_律师签发律师函业务操作指引_2021", "民事"),
]

print("="*60)
print("律师业务操作指引 PDF 生成工具")
print("="*60)
print(f"注意: 需要先有markdown文件才能生成PDF")
print(f"Markdown目录: {BASE_DIR}")
print()
print("当前收集的指引:")
for name, cat in guides:
    md_file = BASE_DIR / f"{name}.md"
    pdf_file = PDF_DIR / f"{name}.pdf"
    exists = "✓" if md_file.exists() else "✗"
    print(f"  [{exists}] {cat}: {name}")