爬取掘金文档并转化为 md

const puppeteer = require("puppeteer");
const turndown = require("turndown");
const fs = require("fs");
(async () => {
  const browser = await puppeteer.launch({
    headless: "new",
  });
  const page = await browser.newPage();
  await page.goto("https://juejin.cn/post/7240730470975111227");

  await page.evaluate(() => {
    const headers = document.querySelectorAll(".code-block-extension-header");
    headers.forEach((header) => header.remove());
  });

  await page.evaluate(() => {
    const codeBlocks = document.querySelectorAll("code.hljs");
    codeBlocks.forEach((codeBlock) => {
      // const code = codeBlock.innerHTML;
      const code = codeBlock.innerHTML.trim().replace(/^\s+/gm, "");
      codeBlock.innerHTML = "```js\n" + code + "\n```";
    });
  });

  const articleContent = await page.evaluate(() => {
    return document.querySelector(".article-content").innerHTML;
  });

  const turndownService = new turndown();
  const markdownContent = turndownService
    .turndown(articleContent)
    .split("\n")
    .map((line) => line.trim())
    .join("\n");

  fs.writeFile("article.md", markdownContent, (err) => {
    if (err) throw err;
    console.log("Markdown file saved!");
  });

  await browser.close();
})();