vent_agent_3/utils/dox_builder.py

import os
import re
import uuid

import pypandoc
from docx import Document
from docx.shared import Pt, Cm
from docx.oxml.ns import qn
from docx.enum.text import WD_ALIGN_PARAGRAPH
from docx.enum.table import WD_TABLE_ALIGNMENT
from config.settings import SERVER_HOST, REPORT_OUTPUT_DIR

TEMPLATE_FILENAME = "template.docx"

def convert_markdown_to_docx(md_content: str, save_path: str) -> None:
    # 自动检测并下载 pandoc
    try:
        pypandoc.get_pandoc_path()
    except OSError:
        print("未检测到 pandoc，正在自动下载...")
        pypandoc.download_pandoc()
    report_content = md_content.strip()

    template_file = os.path.join(os.path.dirname(__file__), TEMPLATE_FILENAME)
    if not os.path.exists(template_file):
        raise FileNotFoundError(
            f"样式模板不存在！路径：{template_file}"
        )

    extra_args = [
        "--standalone",
        f"--reference-doc={template_file}"
    ]

    pypandoc.convert_text(
        source=report_content,
        format="md",
        to="docx",
        outputfile=save_path,
        extra_args=extra_args,
        encoding="utf-8"
    )
    print(f"✅ 转换成功！文件输出路径：{save_path}")

def get_docx_download_url(word_filename: str) -> str:
    return f"{SERVER_HOST}/static/report_pdf/{word_filename}"


if __name__ == '__main__':
    # 读取本地 md 文本文件
    try:
        with open("../md.txt", "r", encoding="utf-8") as f:
            md_text = f.read()
        # 输出 word 路径，可自行修改
        word_output_path = "../审查报告.docx"
        convert_markdown_to_docx(md_text, word_output_path)
    except FileNotFoundError:
        print("错误：未找到 ../md.txt 文件，请检查文件路径！")
    except Exception as e:
        print(f"执行异常：{str(e)}")