#!/usr/bin/env python3
"""
为代码块自动添加语言标注
根据代码块内容智能判断语言类型
"""

import re
import os
from pathlib import Path

def detect_language(content):
 """根据代码块内容检测语言类型"""
 if not content.strip():
 return 'text'
 
 first_line = content.strip().split('\n')[0].strip()
 
 # 检测 Bash/Shell 命令
 bash_patterns = [
 r'^(openclaw|clawhub|npm|pnpm|yarn|bun|git|docker|curl|wget|ssh)\s',
 r'^(cd|ls|mkdir|rm|cp|mv|cat|echo|export|source|chmod|chown)\s',
 r'^\$\s',
 r'^#\s*!/bin/(bash|sh)',
 ]
 for pattern in bash_patterns:
 if re.search(pattern, first_line):
 return 'bash'
 
 # 检测 JSON
 if content.strip().startswith('{') or content.strip().startswith('['):
 if '"' in content and ':' in content:
 return 'json'
 
 # 检测 YAML
 if re.search(r'^[a-zA-Z_][\w-]*:\s*\S', content, re.MULTILINE):
 if '{' not in content[:50]: # 不是 JSON
 return 'yaml'
 
 # 检测 Python
 python_patterns = [
 r'^(def|class|import|from|if __name__|print\(|return)\s',
 r'^\s*(def|class)\s+\w+',
 ]
 for pattern in python_patterns:
 if re.search(pattern, content, re.MULTILINE):
 return 'python'
 
 # 检测 JavaScript/TypeScript
 js_patterns = [
 r'^(const|let|var|function|class|import|export|require\(|module\.exports)\s',
 ]
 ts_patterns = [
 r'^(interface|type|enum|namespace|declare)\s',
 r':\s*(string|number|boolean|any|void|never)\s*[;=]',
 ]
 
 for pattern in ts_patterns:
 if re.search(pattern, content, re.MULTILINE):
 return 'typescript'
 
 for pattern in js_patterns:
 if re.search(pattern, content, re.MULTILINE):
 return 'javascript'
 
 # 检测输出内容
 output_patterns = [
 r'(Output:|Error:|Success:|Warning:|INFO|DEBUG)',
 r'[✅❌⚠️]',
 r'Exit Code:',
 ]
 for pattern in output_patterns:
 if re.search(pattern, content):
 return 'text'
 
 # 默认返回 text
 return 'text'

def process_file(filepath):
 """处理单个 Markdown 文件"""
 with open(filepath, 'r', encoding='utf-8') as f:
 content = f.read()
 
 # 查找所有未标注语言的代码块
 pattern = r'```\s*\n(.*?)\n```'
 
 modified = False
 stats = {'bash': 0, 'json': 0, 'text': 0, 'yaml': 0, 'python': 0, 'javascript': 0, 'typescript': 0}
 
 def replace_code_block(match):
 nonlocal modified, stats
 code_content = match.group(1)
 lang = detect_language(code_content)
 
 if lang:
 modified = True
 stats[lang] = stats.get(lang, 0) + 1
 return f'```{lang}\n{code_content}\n```'
 return match.group(0)
 
 new_content = re.sub(pattern, replace_code_block, content, flags=re.DOTALL)
 
 if modified:
 with open(filepath, 'w', encoding='utf-8') as f:
 f.write(new_content)
 return True, stats
 
 return False, stats

def main():
 docs_dir = Path('docs')
 
 print("🔍 开始为代码块添加语言标注...")
 print()
 
 total_files = 0
 total_stats = {'bash': 0, 'json': 0, 'text': 0, 'yaml': 0, 'python': 0, 'javascript': 0, 'typescript': 0}
 
 # 遍历所有 Markdown 文件
 for md_file in docs_dir.rglob('*.md'):
 modified, stats = process_file(md_file)
 if modified:
 total_files += 1
 print(f"✅ {md_file}")
 for lang, count in stats.items():
 if count > 0:
 total_stats[lang] += count
 
 print()
 print("✅ 处理完成!")
 print()
 print("📊 统计信息:")
 print(f" - 修改文件数: {total_files}")
 print(" - 添加语言标注:")
 for lang, count in sorted(total_stats.items(), key=lambda x: x[1], reverse=True):
 if count > 0:
 print(f" - {lang}: {count}")
 print()

if __name__ == '__main__':
 main()