Agent-n8n/scripts/fix_all_orders_images.py

#!/usr/bin/env python3
"""
处理所有订单班的图片整理
- 统一图片格式为jpg
- 规范化命名
- 更新Markdown引用
"""

import os
import re
import shutil
from pathlib import Path
from urllib.parse import unquote

def process_all_orders():
    """处理所有订单班"""
    base_path = Path("/Users/xiaoqi/Documents/Dev/Project/2025-09-08_n8nDEMO演示/data/订单班文档资料")

    # 订单班列表（排除已处理的文旅）
    order_classes = [
        "财经商贸", "食品", "智能开发", "智能制造",
        "视觉设计", "交通物流", "土木", "大健康", "能源",
        "化工", "环保"
    ]

    print("=" * 60)
    print("开始处理其他订单班图片")
    print("=" * 60)

    for order_class in order_classes:
        order_dir = base_path / order_class
        if not order_dir.exists():
            print(f"\n⚠ 跳过 {order_class}：目录不存在")
            continue

        notion_dir = order_dir / "notion文稿"
        if not notion_dir.exists():
            print(f"\n⚠ 跳过 {order_class}：notion文稿目录不存在")
            continue

        print(f"\n处理 {order_class}...")
        process_single_order(order_class, notion_dir)

    print("\n" + "=" * 60)
    print("✅ 所有订单班处理完成！")

def process_single_order(order_class: str, notion_dir: Path):
    """处理单个订单班"""
    image_dir = notion_dir / "image"
    image_dir.mkdir(exist_ok=True)

    # 收集所有图片
    all_images = []
    for ext in ['jpg', 'jpeg', 'png', 'gif', 'bmp', 'webp']:
        all_images.extend(notion_dir.glob(f"*.{ext}"))
        all_images.extend(notion_dir.glob(f"*.{ext.upper()}"))
        all_images.extend(image_dir.glob(f"*.{ext}"))
        all_images.extend(image_dir.glob(f"*.{ext.upper()}"))

    # 去重
    all_images = list(set(all_images))

    if not all_images:
        print(f"  没有找到图片")
        return

    print(f"  找到 {len(all_images)} 张图片")

    # 分类计数器
    counters = {
        "设计图": 0,
        "展示图": 0,
        "效果图": 0,
        "流程图": 0,
        "场景图": 0,
        "图片": 0
    }

    # 映射关系
    image_mapping = {}

    # 处理每张图片
    for img_path in sorted(all_images):
        old_name = img_path.name
        new_name = generate_new_name(old_name, counters)
        new_path = image_dir / new_name

        # 移动或重命名
        if img_path != new_path:
            try:
                if img_path.parent != image_dir:
                    # 需要移动到image目录
                    shutil.move(str(img_path), str(new_path))
                else:
                    # 只需重命名
                    img_path.rename(new_path)

                image_mapping[old_name] = new_name
                print(f"    ✓ {old_name} → {new_name}")
            except Exception as e:
                print(f"    ✗ 失败 {old_name}: {e}")

    # 更新Markdown文件
    if image_mapping:
        update_markdown_files(notion_dir, image_mapping)
        create_index(order_class, image_dir, len(all_images))

def generate_new_name(filename: str, counters: dict) -> str:
    """生成新的规范化文件名"""
    name = Path(filename).stem.lower()

    # 根据关键词分类
    if any(kw in name for kw in ['设计', 'design', 'whisk', 'cad', '3d', '三维', '渲染', '效果图', 'render']):
        counters["设计图"] += 1
        return f"设计图_{counters['设计图']:02d}.jpg"
    elif any(kw in name for kw in ['展示', 'display', 'show', '展台', '展位', '展览']):
        counters["展示图"] += 1
        return f"展示图_{counters['展示图']:02d}.jpg"
    elif any(kw in name for kw in ['效果', 'effect', '成品', '最终']):
        counters["效果图"] += 1
        return f"效果图_{counters['效果图']:02d}.jpg"
    elif any(kw in name for kw in ['流程', 'flow', 'process', '步骤', '架构', 'diagram']):
        counters["流程图"] += 1
        return f"流程图_{counters['流程图']:02d}.jpg"
    elif any(kw in name for kw in ['场景', 'scene', '展会', '博览', '会场', '现场', '签到', '试驾', '实景', '照片']):
        counters["场景图"] += 1
        return f"场景图_{counters['场景图']:02d}.jpg"
    elif name[0:1].isdigit():
        # 数字开头的文件
        counters["展示图"] += 1
        return f"展示图_{counters['展示图']:02d}.jpg"
    else:
        # 其他图片
        counters["图片"] += 1
        return f"图片_{counters['图片']:02d}.jpg"

def update_markdown_files(notion_dir: Path, image_mapping: dict):
    """更新Markdown文件中的图片引用"""
    md_files = list(notion_dir.glob("*.md"))

    for md_file in md_files:
        try:
            content = md_file.read_text(encoding='utf-8')
            original_content = content

            # 更新每个图片引用
            for old_name, new_name in image_mapping.items():
                # 处理各种可能的引用格式
                # 1. 普通引用
                content = re.sub(
                    f"\\]\\({re.escape(old_name)}\\)",
                    f"](./image/{new_name})",
                    content
                )
                # 2. 带路径的引用
                content = re.sub(
                    f"\\]\\([^)]*{re.escape(old_name)}\\)",
                    f"](./image/{new_name})",
                    content
                )
                # 3. URL编码的引用
                encoded_name = unquote(old_name)
                if encoded_name != old_name:
                    content = re.sub(
                        f"\\]\\([^)]*{re.escape(encoded_name)}[^)]*\\)",
                        f"](./image/{new_name})",
                        content
                    )

            # 确保所有图片都指向./image/目录
            content = re.sub(
                r'!\[([^\]]*)\]\((?!\.?/image/)([^)]+\.(jpg|jpeg|png|gif))\)',
                r'![\1](./image/\2)',
                content,
                flags=re.IGNORECASE
            )

            if content != original_content:
                # 创建备份
                backup_file = md_file.with_suffix('.md.bak')
                shutil.copy2(md_file, backup_file)

                # 保存更新
                md_file.write_text(content, encoding='utf-8')
                print(f"    ✓ 更新MD: {md_file.name}")
        except Exception as e:
            print(f"    ✗ 更新MD失败 {md_file.name}: {e}")

def create_index(order_class: str, image_dir: Path, total_count: int):
    """创建图片索引文件"""
    images = sorted(image_dir.glob("*.jpg"))

    # 分类
    categories = {
        "设计图": [],
        "展示图": [],
        "效果图": [],
        "流程图": [],
        "场景图": [],
        "其他": []
    }

    for img in images:
        name = img.name
        added = False
        for cat in ["设计图", "展示图", "效果图", "流程图", "场景图"]:
            if name.startswith(cat):
                categories[cat].append(name)
                added = True
                break
        if not added:
            categories["其他"].append(name)

    # 生成索引内容
    content = f"""# {order_class}订单班 - 图片资源索引

**图片总数**: {len(images)} 张
**更新时间**: 2025-09-28
**状态**: ✅ 已完成规范化整理

## 图片分类
"""

    for cat, files in categories.items():
        if files:
            content += f"\n### {cat} ({len(files)}张)\n"
            for file in files:
                content += f"- {file}\n"

    content += """
## 使用说明

### Markdown引用格式
```markdown
![图片描述](./image/图片名.jpg)
```

### 注意事项
- 所有图片已统一为 .jpg 格式
- 图片位于 notion文稿/image/ 目录下
- Markdown文件已创建备份（.bak）
"""

    # 保存索引
    index_file = image_dir.parent / "图片索引.md"
    index_file.write_text(content, encoding='utf-8')
    print(f"    ✓ 创建索引: {index_file.name}")

if __name__ == "__main__":
    process_all_orders()