《Snipaste截图元数据自动批量编辑与SEO关键词嵌入脚本开发教程》

import piexif
from PIL import Image
import os

def read_image_metadata(image_path):
    """读取图片的EXIF和XMP元数据"""
    metadata = {'exif': {}, 'xmp': ''}
    try:
        # 使用Pillow打开图片获取基础信息
        with Image.open(image_path) as img:
            # 读取EXIF数据 (通过piexif)
            if 'exif' in img.info:
                exif_dict = piexif.load(img.info['exif'])
                # 将复杂的piexif字典简化为可读性更好的嵌套字典
                for ifd in exif_dict:
                    if ifd != 'thumbnail':
                        metadata['exif'][ifd] = {}
                        for tag in exif_dict[ifd]:
                            tag_name = piexif.TAGS[ifd][tag]["name"]
                            metadata['exif'][ifd][tag_name] = exif_dict[ifd][tag]
    except Exception as e:
        print(f"读取 {image_path} 的EXIF数据时出错: {e}")
    # 注意：纯piexif不直接支持XMP。XMP读取需要其他库，如下一部分所示。
    return metadata

from libxmp import XMPFiles, XMPMeta, consts

def write_seo_keywords_to_xmp(image_path, keywords_list, title=None, description=None):
    """
    将SEO关键词列表写入图片的XMP元数据。
    :param image_path: 图片路径
    :param keywords_list: 关键词列表，如 ['截图软件', 'Snipaste教程', '贴图功能']
    :param title: 图片标题 (dc:title)
    :param description: 图片描述 (dc:description)
    """
    if not keywords_list:
        return False
    
    try:
        xmpfile = XMPFiles(file_path=image_path, open_forupdate=True)
        
        # 获取或创建XMP对象
        xmp = xmpfile.get_xmp()
        if xmp is None:
            xmp = XMPMeta()
        
        # 设置 Dublin Core 命名空间的属性
        NS_DC = consts.XMP_NS_DC
        
        # 1. 写入关键词 (dc:subject)
        # 先清除可能存在的旧关键词（可选，根据策略决定）
        # xmp.delete_property(NS_DC, 'subject')
        for keyword in keywords_list:
            xmp.append_array_item(NS_DC, 'subject', keyword, {'prop_array_is_ordered': False, 'prop_value_is_array': True})
        
        # 2. 写入标题 (dc:title)
        if title:
            xmp.set_localized_text(NS_DC, 'title', '', 'x-default', title)
        
        # 3. 写入描述 (dc:description)
        if description:
            xmp.set_localized_text(NS_DC, 'description', '', 'x-default', description)
        
        # 将修改写回文件
        if xmpfile.can_put_xmp(xmp):
            xmpfile.put_xmp(xmp)
            xmpfile.close_file()
            print(f"成功写入XMP元数据到: {image_path}")
            return True
        else:
            print(f"无法写入XMP到: {image_path}")
            xmpfile.close_file()
            return False
            
    except Exception as e:
        print(f"处理XMP数据时发生错误 ({image_path}): {e}")
        return False

def update_exif_metadata(image_path, software="Snipaste Pro", copyright_info="© YourWebsite.com"):
    """
    更新图片的EXIF元数据，特别是软件和版权信息。
    """
    try:
        exif_dict = piexif.load(image_path)
        
        # 更新 IFD0 (主图像目录) 的信息
        # 270 - ImageDescription, 305 - Software, 315 - Artist, 33432 - Copyright
        ifd0 = exif_dict["0th"]
        ifd0[piexif.ImageIFD.Software] = software.encode('utf-8')
        ifd0[piexif.ImageIFD.Copyright] = copyright_info.encode('utf-8')
        # 可以保留原有的ImageDescription，或根据策略更新
        # ifd0[piexif.ImageIFD.ImageDescription] = "A screenshot created by Snipaste".encode('utf-8')
        
        # 将修改后的EXIF字典转换回字节
        exif_bytes = piexif.dump(exif_dict)
        
        # 将EXIF数据写回图片
        piexif.insert(exif_bytes, image_path)
        print(f"成功更新EXIF元数据到: {image_path}")
        return True
    except Exception as e:
        print(f"更新EXIF数据时发生错误 ({image_path}): {e}")
        return False

# config.yaml
seo_keywords:
  base_keywords: ["截图软件", "截图工具", "Snipaste"] # 每张图都添加的基础词
  keyword_mapping: # 基于文件名或目录的关键词映射
    - trigger: "贴图"  # 文件名包含“贴图”
      keywords: ["贴图功能", "浮动窗口", "多任务处理", "效率工具"]
    - trigger: "取色"
      keywords: ["取色器", "色彩拾取", "设计工具", "HEX颜色"]
    - trigger: "滚动截图"
      keywords: ["长截图", "网页捕捉", "完整页面截图"]
    - trigger: "标注"
      keywords: ["图片批注", "箭头标注", "马赛克", "文本标注"]
  # 可以扩展：基于目录结构的映射
  directory_map:
    "/tutorials/basic": ["新手入门", "使用教程", "基础功能"]
    "/tutorials/advanced": ["高级技巧", "专业用法", "自动化"]

default_metadata:
  title_template: "Snipaste截图展示：{filename_keyword}功能"
  description_template: "本图片展示了Snipaste截图工具在{filename_keyword}方面的应用场景，是一款高效的{base_keyword}。"
  copyright: "© snipastem.com - 专业的截图工具资源站"

import yaml
import os
import re
from image_processor import write_seo_keywords_to_xmp, update_exif_metadata, read_image_metadata

def load_config(config_path='config.yaml'):
    with open(config_path, 'r', encoding='utf-8') as f:
        return yaml.safe_load(f)

def extract_keywords_from_filename(filename, config):
    """根据配置和文件名提取触发关键词"""
    all_keywords = set(config['seo_keywords']['base_keywords'])
    filename_lower = filename.lower()
    
    for mapping in config['seo_keywords']['keyword_mapping']:
        if mapping['trigger'].lower() in filename_lower:
            all_keywords.update(mapping['keywords'])
    
    # 简单的文件名关键词提取（去除扩展名和常见分隔符）
    name_part = os.path.splitext(filename)[0]
    # 移除日期、序列号等常见噪音
    clean_name = re.sub(r'[\d_\-\s]+', ' ', name_part).strip()
    if clean_name and len(clean_name) > 1: # 避免单个字符
        all_keywords.add(clean_name)
    
    return list(all_keywords)

def generate_title_and_description(filename, keywords, config):
    """根据模板生成标题和描述"""
    # 这里可以使用更复杂的逻辑，例如从keywords中选一个作为filename_keyword
    primary_keyword = keywords[1] if len(keywords) > 1 else keywords[0]
    base_keyword = config['seo_keywords']['base_keywords'][0]
    
    title = config['default_metadata']['title_template'].format(
        filename_keyword=primary_keyword,
        base_keyword=base_keyword
    )
    description = config['default_metadata']['description_template'].format(
        filename_keyword=primary_keyword,
        base_keyword=base_keyword
    )
    return title, description

def process_image_batch(input_dir, output_dir=None, config_path='config.yaml'):
    config = load_config(config_path)
    
    if output_dir and not os.path.exists(output_dir):
        os.makedirs(output_dir)
    
    supported_ext = ('.jpg', '.jpeg', '.png', '.tiff', '.bmp')
    
    for root, dirs, files in os.walk(input_dir):
        for file in files:
            if file.lower().endswith(supported_ext):
                input_path = os.path.join(root, file)
                output_path = os.path.join(output_dir, file) if output_dir else input_path
                
                print(f"\n处理中: {file}")
                
                # 1. 提取关键词
                keywords = extract_keywords_from_filename(file, config)
                print(f"  提取关键词: {keywords}")
                
                # 2. 生成标题和描述
                title, description = generate_title_and_description(file, keywords, config)
                
                # 3. 复制文件到输出目录（如果设置了输出目录）
                if output_dir and input_path != output_path:
                    import shutil
                    shutil.copy2(input_path, output_path)
                    target_path = output_path
                else:
                    target_path = input_path
                
                # 4. 写入XMP元数据
                write_success = write_seo_keywords_to_xmp(target_path, keywords, title, description)
                
                # 5. 更新EXIF元数据
                if write_success:
                    update_exif_metadata(target_path, software="Snipaste", copyright_info=config['default_metadata']['copyright'])
                
                # 6. （可选）记录日志
                log_entry = {
                    'file': file,
                    'keywords': keywords,
                    'status': 'success' if write_success else 'failed',
                    'path': target_path
                }
                # ... 将log_entry写入日志文件

if __name__ == "__main__":
    # 使用示例
    input_folder = "./input_images"
    output_folder = "./output_images"  # 设置为None则直接修改原图
    process_image_batch(input_folder, output_folder)