import os
import uuid
from docx import Document

def extract_images_from_word(word_path, output_dir):
    """
    从Word文档中提取图片并保存到指定目录
    :param word_path: Word文档路径
    :param output_dir: 图片输出目录
    """
    doc = Document(word_path)
    
    # 确保输出目录存在
    os.makedirs(output_dir, exist_ok=True)
    
    # 获取文档中的所有图片
    for rel in doc.part.rels.values():
        if "image" in rel.target_ref:
            img_data = rel.target_part.blob
            
            # 使用UUID命名图片
            output_path = os.path.join(output_dir, f"{uuid.uuid4()}.jpg")
            with open(output_path, "wb") as f:
                f.write(img_data)
            print(f"图片已保存到: {output_path}")

if __name__ == "__main__":
    word_path = "d:\\dsWork\\dsProject\\dsRag\\Test\\带图的WORD文档.docx"
    output_dir = os.path.abspath(os.path.join(os.path.dirname(word_path), "..", "static", "Images"))
    
    extract_images_from_word(word_path, output_dir)