最近在作一個小程序項目,須要爬取第三方數據,因而開始重撿起來爬蟲,其實前端爬蟲挺好實現的,但由於如今網頁出現了SPA,因而開始瘋狂踩坑,聊記此文,以慰諸君。html
挺簡單的,一個工具包,幾行代碼就解決了前端
const $ = require('cheerio')
const requestPromise = require('request-promise')
const url = 'https://juejin.im/books';
requestPromise(url)
.then((html) => {
// 利用 cheerio 來分析網頁內容,拿到全部小冊子的描述
const books = $('.info', html)
let totalSold = 0
let totalSale = 0
let totalBooks = books.length
// 遍歷冊子節點,分別統計它的購買人數,和銷售額總和
books.each(function () {
const book = $(this)
const price = $(book.find('.price-text')).text().replace('¥', '')
const count = book.find('.message').last().find('span').text().split('人')[0]
totalSale += Number(price) * Number(count)
totalSold += Number(count)
})
// 最後打印出來
console.log(
`共 ${totalBooks} 本小冊子`,
`共 ${totalSold} 人次購買`,
`約 ${Math.round(totalSale / 10000)} 萬`
)
})
複製代碼
但。。。可是,上面例子爬取掘金是不行的,由於掘金就是經典的SPA,服務器只返回一個空的掛載節點,毫無數據。因而引出無頭瀏覽器puppeteer小程序
const $ = require('cheerio');
const puppeteer = require('puppeteer');
const url = 'https://juejin.im/books';
async function run(params) {
//打開一個瀏覽器
const browser = await puppeteer.launch();
// 打開一個頁面
const page = await browser.newPage();
await page.goto(url, {
waitUntil: 'networkidle0'
});
const html = await page.content();
const books = $('.info', html);
let totalSold = 0
let totalSale = 0
let totalBooks = books.length
// 遍歷冊子節點,分別統計它的購買人數,和銷售額總和
books.each(function () {
const book = $(this)
const price = $(book.find('.price-text')).text().replace('¥', '')
const count = book.find('.message').last().find('span').text().split('人')[0]
totalSale += Number(price) * Number(count)
totalSold += Number(count)
})
// 最後打印出來
console.log(
`共 ${totalBooks} 本小冊子`,
`共 ${totalSold} 人次購買`,
`約 ${Math.round(totalSale / 10000)} 萬`
)
}
run()
複製代碼