dsProject/dsRag/Test/T2_read_word_content.py

import docx


def read_word_content(docx_path):
    idx = 0
    """遍历Word文档的每个段落，输出文字或图片标识"""
    try:
        doc = docx.Document(docx_path)

        for paragraph in doc.paragraphs:
            has_image = False
            # 检查段落中是否有图片
            for run in paragraph.runs:
                for element in run._element:
                    if element.tag.endswith('drawing'):
                        # 找到图片元素
                        has_image = True
                        break
                if has_image:
                    break

            if has_image:
                idx = idx + 1
                print("【图片" + str(idx) + "】")
            elif paragraph.text.strip():
                print(paragraph.text.strip())

    except Exception as e:
        print(f"处理Word文档时出错: {str(e)}")


if __name__ == "__main__":
    word_document_path = "d:\\dsWork\\dsProject\\dsRag\\static\\Test\\带图的WORD文档_MATH_3.docx"
    read_word_content(word_document_path)