"""
conda activate rag
pip install pypandoc
"""

import pypandoc


def docx_to_latex(docx_path):
    latex_content = pypandoc.convert_file(docx_path, 'latex')
    import re
    # 替换公式格式
    latex_content = re.sub(r'\\\[(.*?)\\\]', r'$$\1$$', latex_content)
    latex_content = re.sub(r'\\\((.*?)\\\)', r'$\1$', latex_content)
    # 替换图片路径为【图片X】格式
    img_count = 1

    def replacer(match):
        nonlocal img_count
        result = f'【图片{img_count}】'
        img_count += 1
        return result

    latex_content = re.sub(r'\\includegraphics\[.*?\]\{.*?\}', replacer, latex_content)
    return latex_content


latex_content = docx_to_latex('带公式的WORD文档.docx')

# 遍历字符串的每一行
for line in latex_content.split('\n'):
    if len(line.strip()) > 0:
        print(line.strip())