dsProject/dsRag/Test/T2_read_word_content.py

import docx


def read_word_content(docx_path):
    res = ""
    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
                res = res + "\n" + "【图片" + str(idx) + "】"
            elif paragraph.text.strip():
                res = res + "\n" + paragraph.text.strip()
        return res
    except Exception as e:
        print(f"处理Word文档时出错: {str(e)}")


def split_into_blocks(text):
    """按行遍历文本，发现'问题X'或'话题X'时开始分割，只移除前缀但保留整行内容"""
    blocks = []
    current_block = []
    in_block = False

    for line in text.splitlines():
        if line.startswith(('问题', '话题')) and any(c.isdigit() for c in line[:5]):
            if in_block:
                blocks.append('\n'.join(current_block))
                current_block = []
            in_block = True
            # 循环移除问题和话题前缀后的数字
            while line and (line.startswith(('问题', '话题')) or (line and line and line[0].isdigit())):
                if line.startswith(('问题', '话题')):
                    line = line[2:] if len(line) > 2 else line
                elif line and line[0].isdigit():
                    line = line[1:] if len(line) > 1 else line
                line = line.strip()
        if in_block and line:  # 只添加非空行
            current_block.append(line)

    if current_block:
        blocks.append('\n'.join(current_block))

    return [(i + 1, block) for i, block in enumerate(blocks)]


if __name__ == "__main__":
    word_document_path = "d:\\dsWork\\dsProject\\dsRag\\static\\Test\\带图的WORD文档_MATH_3.docx"
    res = read_word_content(word_document_path)
    q = split_into_blocks(res)
    for x in q:
        print(x[1])