Obsidian-Knowledge-Base/X-Knowledge-Base/raw/Joplin/_resources/zhconv_txt.py

#!/usr/bin/env python3
import sys
import os
import argparse
from pathlib import Path

try:
    from zhconv import convert
except ImportError:
    sys.exit("错误：请先安装依赖库，执行命令：pip install zhconv")

def safe_filename(base_path, prefix='converted_'):
    """生成唯一的新文件名"""
    path = Path(base_path)
    new_stem = f"{prefix}{path.stem}"
    new_path = path.with_stem(new_stem)
    
    counter = 0
    while new_path.exists():
        counter += 1
        new_stem = f"{prefix}{counter}_{path.stem}"
        new_path = path.with_stem(new_stem)
    
    return new_path

def detect_and_read(filepath):
    """智能检测编码并读取内容"""
    encodings = ['utf-8', 'gb18030', 'big5', 'latin-1']
    with open(filepath, 'rb') as f:
        raw_data = f.read()
        
    for enc in encodings:
        try:
            return raw_data.decode(enc)
        except UnicodeDecodeError:
            continue
    raise UnicodeDecodeError("无法识别的文件编码")

def convert_file(input_path, prefix='', to_traditional=False, region='tw'):
    """支持多地区转换的核心函数"""
    try:
        content = detect_and_read(input_path)
        
        # 确定转换方向
        if to_traditional:
            target = 'zh-hk' if region == 'hk' else 'zh-tw'
            direction = f"s2t_{region}_"
        else:
            target = 'zh-cn'
            direction = "t2s_"
        
        converted = convert(content, target)
        
        # 生成智能前缀
        final_prefix = f"{prefix}{direction}" if prefix else f"{direction}"
        
        output_path = safe_filename(input_path, final_prefix)
        
        # 原子写入
        temp_path = f"{output_path}.tmp"
        with open(temp_path, 'w', encoding='utf-8', errors='strict') as f:
            f.write(converted)
        
        os.replace(temp_path, output_path)
        return output_path
        
    except Exception as e:
        if 'temp_path' in locals() and os.path.exists(temp_path):
            os.remove(temp_path)
        raise

def main():
    parser = argparse.ArgumentParser(
        description='简繁转换工具（支持台湾/香港地区变体）',
        formatter_class=argparse.ArgumentDefaultsHelpFormatter
    )
    parser.add_argument('filepath', help="输入文件路径")
    parser.add_argument('--prefix', default='', 
                      help="自定义文件名前缀")
    parser.add_argument('--to-traditional', action='store_true',
                      help="启用简体转繁体模式")
    parser.add_argument('--region', choices=['tw', 'hk'], default='tw',
                      help="繁体地区变体：tw(台湾，默认) 或 hk(香港)")

    args = parser.parse_args()
    
    input_path = Path(args.filepath)
    if not input_path.exists():
        sys.exit(f"错误：文件不存在 '{input_path}'")
    
    if args.region == 'hk' and not args.to_traditional:
        print("警告：--region 参数仅在启用 --to-traditional 时生效")
    
    try:
        result_path = convert_file(
            input_path,
            prefix=args.prefix,
            to_traditional=args.to_traditional,
            region=args.region
        )
        print(f"生成文件：{result_path}")
    except Exception as e:
        sys.exit(f"转换失败：{str(e)}")

if __name__ == "__main__":
    main()