Agent-n8n/scripts/fix_image_paths.py

#!/usr/bin/env python3
"""
修复图片路径URL编码问题的独立脚本
可以单独运行，也可以被主脚本调用
"""

import os
import re
from pathlib import Path
from urllib.parse import unquote
from typing import List, Tuple

def fix_markdown_image_paths(md_file: Path) -> int:
    """
    修复单个Markdown文件中的图片路径
    返回修复的引用数量
    """
    try:
        content = md_file.read_text(encoding='utf-8')
        original_content = content
        
        # 查找所有图片引用
        # 支持多种格式: ![alt](path), ![](path), ![[path]]
        patterns = [
            r'!\[(.*?)\]\((.*?)\)',  # 标准Markdown格式
            r'!\[\[(.*?)\]\]',        # Obsidian格式
        ]
        
        fixed_count = 0
        
        for pattern in patterns:
            matches = re.finditer(pattern, content)
            for match in matches:
                if len(match.groups()) == 2:  # 标准Markdown格式
                    alt_text = match.group(1)
                    img_path = match.group(2)
                else:  # Obsidian格式
                    img_path = match.group(1)
                    alt_text = ""
                
                # 跳过外部链接
                if img_path.startswith('http://') or img_path.startswith('https://'):
                    continue
                
                # 检查是否包含URL编码
                if '%' in img_path:
                    # 解码URL
                    decoded_path = unquote(img_path)
                    
                    # 标准化路径
                    if decoded_path.startswith('./image/'):
                        decoded_path = decoded_path[2:]  # 移除 ./
                    elif not decoded_path.startswith('image/'):
                        decoded_path = f"image/{Path(decoded_path).name}"
                    
                    # 构建新的引用
                    new_reference = f"![{alt_text or Path(decoded_path).stem}]({decoded_path})"
                    old_reference = match.group(0)
                    
                    # 替换内容
                    content = content.replace(old_reference, new_reference)
                    fixed_count += 1
                    print(f"      修复: {old_reference} → {new_reference}")
                
                # 检查路径格式是否正确
                elif img_path.startswith('./image/'):
                    # 移除不必要的 ./
                    new_path = img_path[2:]
                    new_reference = f"![{alt_text or Path(new_path).stem}]({new_path})"
                    old_reference = match.group(0)
                    
                    content = content.replace(old_reference, new_reference)
                    fixed_count += 1
                    print(f"      修复: {old_reference} → {new_reference}")
        
        # 如果内容有变化，写回文件
        if content != original_content:
            md_file.write_text(content, encoding='utf-8')
            print(f"    ✅ 已修复 {fixed_count} 个图片引用")
        
        return fixed_count
    
    except Exception as e:
        print(f"    ❌ 处理失败: {e}")
        return 0

def rename_encoded_images(image_dir: Path) -> int:
    """
    重命名包含URL编码的图片文件
    返回重命名的文件数量
    """
    renamed_count = 0
    
    if not image_dir.exists():
        return 0
    
    for img_file in image_dir.iterdir():
        if img_file.is_file() and '%' in img_file.name:
            # 解码文件名
            decoded_name = unquote(img_file.name)
            new_path = img_file.parent / decoded_name
            
            # 如果目标文件不存在，重命名
            if not new_path.exists():
                try:
                    img_file.rename(new_path)
                    print(f"      重命名: {img_file.name} → {decoded_name}")
                    renamed_count += 1
                except Exception as e:
                    print(f"      ❌ 重命名失败 {img_file.name}: {e}")
    
    return renamed_count

def fix_order_class_images(order_dir: Path) -> Tuple[int, int]:
    """
    修复单个订单班的图片路径问题
    返回 (修复的MD引用数, 重命名的图片数)
    """
    print(f"\n处理 {order_dir.name} ...")
    
    if not order_dir.exists():
        print(f"  ⚠️  目录不存在")
        return 0, 0
    
    notion_dir = order_dir / "notion文稿"
    if not notion_dir.exists():
        print(f"  ⚠️  notion文稿目录不存在")
        return 0, 0
    
    total_fixed_refs = 0
    total_renamed_files = 0
    
    # 1. 修复图片文件名
    image_dir = notion_dir / "image"
    if image_dir.exists():
        print("  修复图片文件名...")
        renamed = rename_encoded_images(image_dir)
        total_renamed_files += renamed
        if renamed > 0:
            print(f"    ✅ 重命名了 {renamed} 个图片文件")
    
    # 2. 修复Markdown文件中的引用
    print("  修复Markdown文件中的图片引用...")
    md_files = list(notion_dir.glob("*.md"))
    
    for md_file in md_files:
        print(f"    处理 {md_file.name}...")
        fixed = fix_markdown_image_paths(md_file)
        total_fixed_refs += fixed
    
    if total_fixed_refs == 0 and total_renamed_files == 0:
        print("  ✅ 没有需要修复的问题")
    else:
        print(f"  📊 总计: 修复了 {total_fixed_refs} 个引用, 重命名了 {total_renamed_files} 个文件")
    
    return total_fixed_refs, total_renamed_files

def fix_all_order_classes(order_classes: List[str] = None):
    """
    修复指定的订单班（如果没有指定，修复所有）
    """
    BASE_PATH = Path("/Users/xiaoqi/Documents/Dev/Project/2025-09-08_n8nDEMO演示")
    DATA_PATH = BASE_PATH / "data/订单班文档资料"
    
    ALL_ORDER_CLASSES = [
        "食品", "环保", "财经商贸", "视觉设计", "能源", 
        "交通物流", "智能制造", "文旅", "化工", "大健康", 
        "智能开发", "土木"
    ]
    
    dirs_to_process = order_classes if order_classes else ALL_ORDER_CLASSES
    
    print("=" * 60)
    print("开始修复图片路径URL编码问题")
    print("=" * 60)
    
    total_refs = 0
    total_files = 0
    
    for dir_name in dirs_to_process:
        order_dir = DATA_PATH / dir_name
        if order_dir.exists():
            refs, files = fix_order_class_images(order_dir)
            total_refs += refs
            total_files += files
    
    print("\n" + "=" * 60)
    print(f"✅ 修复完成！")
    print(f"   总计修复 {total_refs} 个Markdown引用")
    print(f"   总计重命名 {total_files} 个图片文件")
    print("=" * 60)

def main():
    """主函数 - 独立运行时使用"""
    import sys
    
    # 可以接受命令行参数指定订单班
    if len(sys.argv) > 1:
        order_classes = sys.argv[1].split(',')
        fix_all_order_classes(order_classes)
    else:
        # 没有参数时处理所有订单班
        fix_all_order_classes()

if __name__ == "__main__":
    main()