Python将Markdown文件转换为Word(docx)完整教学

pip install python-docx markdown beautifulsoup4

from docx import Document
from docx.shared import Pt
from bs4 import BeautifulSoup
import markdown
import uuid
import os

FILE_DIR = "static"
os.makedirs(FILE_DIR, exist_ok=True)


def md_to_word(md_text: str) -> str:
    """
    将 Markdown 文本转换为 Word，返回生成的文件名
    """
    html = markdown.markdown(
        md_text,
        extensions=["extra", "tables", "fenced_code"]
    )

    soup = BeautifulSoup(html, "html.parser")
    doc = Document()

    for element in soup.contents:
        handle_element(doc, element)

    filename = f"{uuid.uuid4().hex}.docx"
    file_path = os.path.join(FILE_DIR, filename)
    doc.save(file_path)

    return filename


def handle_element(doc, el):
    if not hasattr(el, "name") or el.name is None:
        return

    # 标题
    if el.name in ["h1", "h2", "h3", "h4", "h5", "h6"]:
        doc.add_heading(el.get_text(), level=int(el.name[1]))
        return

    # 段落
    if el.name == "p":
        p = doc.add_paragraph()
        add_inline_text(p, el)
        return

    # 无序列表
    if el.name == "ul":
        for li in el.find_all("li", recursive=False):
            p = doc.add_paragraph(style="List Bullet")
            add_inline_text(p, li)
        return

    # 有序列表
    if el.name == "ol":
        for li in el.find_all("li", recursive=False):
            p = doc.add_paragraph(style="List Number")
            add_inline_text(p, li)
        return

    # 代码块
    if el.name == "pre":
        code_el = el.find("code")
        code = code_el.get_text() if code_el else el.get_text()
        p = doc.add_paragraph()
        run = p.add_run(code)
        run.font.name = "Courier New"
        run.font.size = Pt(10)
        return

    # 表格
    if el.name == "table":
        rows = el.find_all("tr")
        cols = rows[0].find_all(["th", "td"])
        table = doc.add_table(rows=len(rows), cols=len(cols))
        table.style = "Table Grid"

        for r, row in enumerate(rows):
            for c, cell in enumerate(row.find_all(["th", "td"])):
                paragraph = table.rows[r].cells[c].paragraphs[0]
                run = paragraph.add_run(cell.get_text())
                if cell.name == "th":
                    run.bold = True


def add_inline_text(paragraph, el):
    """
    处理行内样式：加粗、斜体、行内代码
    """
    for node in el.contents:
        if isinstance(node, str):
            paragraph.add_run(node)
        else:
            run = paragraph.add_run(node.get_text())
            if node.name == "strong":
                run.bold = True
            elif node.name == "em":
                run.italic = True
            elif node.name == "code":
                run.font.name = "Courier New"
                run.font.size = Pt(10)


def md_to_word_from_file(md_file_path: str) -> str:
    """
    从 Markdown 文件生成 Word
    """
    with open(md_file_path, "r", encoding="utf-8") as f:
        md_text = f.read()

    return md_to_word(md_text)


if __name__ == "__main__":
    md_path = "jd.md"   # Markdown 文件路径
    filename = md_to_word_from_file(md_path)
    print("生成的 Word 文件：", filename)

python md_to_word.py