import re
import subprocess
import os
import uuid


# 获取MathType对应的Latex公式
def get_latex_list(docx_file):
    # 获取当前目录的父级目录
    current_dir = os.path.dirname(os.path.abspath(__file__))
    current_dir = os.path.dirname(current_dir)
    mtef = os.path.join(current_dir, 'mtef-go-3', 'mtef-go.exe')
    res = []
    output = os.path.join(os.environ['TEMP'], uuid.uuid4().hex + '.txt')
    command = mtef + r" -w " + docx_file + " -o " + output
    os.system(command)
    with open(output, 'r', encoding='utf-8') as file:
        for i, line in enumerate(file):
            res.append(line.strip())
    return res


# 结合Pandoc和mtef-go的结果，合并成最终的输出文本
def get_docx_content_by_pandoc(docx_file):
    # 一、获取Latex公式列表
    formula_list = get_latex_list(docx_file)
    # 最后拼接的内容
    content = ""
    # output_file 设置为临时目录下的uuid.md
    temp_markdown = os.path.join(os.environ['TEMP'], uuid.uuid4().hex + '.md')
    # 调用pandoc将docx文件转换成markdown
    subprocess.run(['pandoc', docx_file, '-f', 'docx', '-t', 'markdown', '-o', temp_markdown])
    # 打印 temp_markdown 文件
    # with open(temp_markdown, 'r', encoding='utf-8') as f:
    #     print(f.read())
    # 读取然后修改内容，输出到新的文件
    wmf_idx = 0  # wmf索引
    img_idx = 0  # 图片索引
    with open(temp_markdown, 'r', encoding='utf-8') as f:
        for line in f:
            if line.strip():
                # 改进后的正则表达式，匹配更多格式的MathType公式
                """
                ![](media/image1.wmf)
                问题2 氢气与氧气燃烧的方程式
                .![](media/image2.wmf).
                问题3 我是一个图片
                ![](media/image3.png){width="3.1251607611548557in" height="3.694634733158355in"}
                """
                if line.index("![](media/image") >= 0 and line.index(".wmf") > 0:
                    content = content + formula_list[wmf_idx] + "\n"
                    wmf_idx = wmf_idx + 1
                elif line.index("![](media/image") >= 0 and (
                        line.index(".png") > 0 or line.index(".jpg") > 0 or line.index(".jpeg") > 0):
                    content = content + "【图片" + str(img_idx) + "】\n"
                else:
                    content = content + line.strip() + "\n"

    # 删除临时文件 output_file
    os.remove(temp_markdown)
    return content