如何使用cheerio或puppeteer抓取所有href?

时间:2020-02-14 07:24:30

标签: javascript node.js web-scraping puppeteer cheerio

我有一个场景,需要从网站获取所有社交媒体链接。如果我认为所有社交链接都在主页上,那么每个社交链接都将获取每个社交媒体链接。下面是代码示例:

使用cheeriojs

const cheerio = require('cheerio')
const axios = require('axios')
const https = require('https');

const agent = new https.Agent({
    rejectUnauthorized: false
});
// process.env.NODE_TLS_REJECT_UNAUTHORIZED = '0';

const getLinks = async (url) => {

    try {
        let body = await axios.get(url, { httpsAgent: agent })
        let hrefs = []
        let $ = cheerio.load(body.data)
        let links = $('a')

        links.each((i, link) => {
            hrefs.push($(link).attr('href'))
        })
        return hrefs
    } catch (error) {
        return error
    }

}

const getSocialLinks = async (socialLinks, url) => {
    try {
        let hrefs = await getLinks(url)

        let handles = []
        hrefs.filter(href => {
            if (href) {
                for (const link of socialLinks) {
                    if (href.includes(link)) {
                        handles.push({ platform: link, handle: href })
                        break
                    }
                }
            }
        })

        console.log(handles);
    } catch (error) {
        console.log(error)
    }
}

getSocialLinks(['facebook', 'twitter', 'instagram', 'youtube', 'linkedin'], 'https://synavos.com')

如果所有社交媒体链接都在主页上,则效果很好,但是 我无法弄清楚社交媒体链接是否打开 给定网站的其他页面。

以下是使用 puppeteer 的相同代码示例:

const puppeteer = require('puppeteer')

const getHrefsAttributes = async (website) => {
    try {
        const browser = await puppeteer.launch({
            headless: true,
            ignoreHTTPSErrors: true
        });
        const [page] = await browser.pages();

        await page.goto(website, { waitUntil: 'networkidle2', timeout: 3000000 });

        const hrefs = await page.evaluate(() => Array.from(document.querySelectorAll('a[href]'), a => a.getAttribute('href')));

        await browser.close();
        return hrefs
    } catch (err) {
        console.error(err);
    }
}

const getSocialLinks = async (url, socialLinks) => {
    let hrefs = await getHrefsAttributes(url)
    // add array for social links which you want to fetch
    let handles = []
    hrefs.filter(href => {
        for (const link of socialLinks) {
            if (href.includes(link)) {
                handles.push({ platform: link, handle: href })
                break
            }
        }
    })

    console.log(handles);
}

getSocialLinks('https://synavos.com/', ['facebook', 'twitter', 'instagram', 'youtube', 'linkedin'])

例如,此URL https://netsoltech.com/在主页上没有其社交媒体链接。

1 个答案:

答案 0 :(得分:1)

您可以使用堆栈/递归 (depth-first) 或队列 (breadth-first) 并运行搜索到一定深度,保留一组访问过的 URL 以避免循环。

一个好的设计可能会使用 generator,因此您可以无限期地继续搜索,直到您达到一定深度或找到一定数量的结果等。这为调用者提供了更大的灵活性,但会增加冗长.

您可能希望调整 href 检查或 URL 以确保您始终在基本站点上并且不会从其他域中提取链接,但基本思想是相同的。

还要注意,这纯粹是顺序的,因此速度很慢。您可能希望使用 asynchronous task queue 并行化请求。

const axios = require("axios");
const cheerio = require("cheerio");

const headers = {
  "User-Agent": "Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:52.0) Gecko/20100101 Firefox/52.0",
};

async function *findLinksRecursively(baseUrl, maxDepth) {
  const visited = new Set(baseUrl);

  for (const queue = [{url: baseUrl, depth: 0}]; queue.length;) {
    const {url, depth} = queue.shift();

    if (depth < maxDepth && !visited.has(url)) {
      visited.add(url);
      const {data} = await axios.get(url, {headers})
        .catch(() => ({data: {}}))
      ;
      const $ = cheerio.load(data);
      const links = [...$('a[href]:not(a[href^="#"])')
        .map((i, e) => e.attribs["href"])
      ].map(e => e.startsWith("http") ? e : baseUrl + e);
      queue.push(...links.map(e => ({url: e, depth: depth + 1})));
      yield links;
    }
  }
}

(async () => {
  const baseUrl = "https://netsoltech.com";
  const socialKeywords = [
    "facebook", "twitter", "instagram", "youtube", "linkedin"
  ];
  const links = new Set();

  for (const gen = findLinksRecursively(baseUrl, 2);;) {
    const {value, done} = await gen.next();

    if (done) break;

    value.forEach(e => links.add(e));
  }
  
  const socialLinks = [...links]
    .filter(e => socialKeywords.find(x => e.includes(x)))
  ;
  console.log(socialLinks);
})();

Puppeteer 中的概念几乎相同,只是您使用 page.goto 代替 Axios 和 page.$eval 代替 Cheerio。