我试图将util模块对象传递给puppeteer page.evaluate
但没有成功。我理解在How to pass required module object to puppeteer page.evaluate
中提出了这个问题,但提供的解决方案在我的案例中不起作用。 MWE:
const puppeteer = require("puppeteer");
const path = "http://books.toscrape.com/";
// scrape funs
(async () =>{
const browser = await puppeteer.launch({headless: false});
const page = await browser.newPage();
await page.goto(path, {waitUntil: "networkidle2", timeout: 0});
await page.waitFor(1000);
await page.addScriptTag({path: './node_modules/util/util.js'});
// selector with replaceable element
const buttonText = await page.evaluate(() => {
let selectorButton = "#default > div > div > div > div > section > div:nth-child(2) > ol > li:nth-child(%s) > article > div.product_price > form > button";
let buttons = [];
for(let i = 1; i < 21; i ++){
let textOut = document.querySelector(util.format(selectorButton, i)).innerText;
buttons.push(textOut);
};
return buttons;
});
// return
await browse.close();
console.log(buttonText);
})();
显示错误:
UnhandledPromiseRejectionWarning:未处理的承诺拒绝(拒绝ID:1):错误:评估失败:ReferenceError:util未定义
谢谢
在初始行中添加const util = require("util");
并且无法正常工作,如How to pass required module object to puppeteer page.evaluate所示。
即使我使用browserify,我似乎无法将util
模块注入puppeteer页面。步骤进行:
在项目PATH
上,按如下方式创建main.js
:
var util = require('util');
然后在终端PATH
上browserify main.js -o bundle.js
。项目bundle.js
中会显示PATH
个文件。
然后运行以下命令:
const puppeteer = require("puppeteer");
const path = "http://books.toscrape.com/";
// scrape funs
(async () =>{
const browser = await puppeteer.launch({headless: false});
const page = await browser.newPage();
await page.goto(path, {waitUntil: "networkidle2", timeout: 0});
await page.waitFor(1000);
await page.addScriptTag({path: "main.js"});
await page.addScriptTag({path: "bundle.js"});
// selector with replaceable element
const buttonText = await page.evaluate(() => {
let buttons = [];
let selectorButton = "#default > div > div > div > div > section > div:nth-child(2) > ol > li:nth-child(%s) > article > div.product_price > form > button";
for(let i = 1; i < 21; i ++){
let textOut = document.querySelector(util.format(selectorButton, i)).innerText;
buttons.push(textOut);
};
return buttons;
});
// return
await browse.close();
console.log(buttonText);
})();
错误:
UnhandledPromiseRejectionWarning:未处理的承诺拒绝(拒绝ID:1):错误:评估失败:TypeError:无法读取属性&#39;格式&#39;未定义的 at:5:55
答案 0 :(得分:2)
您需要包含与./node_modules/util/util.js
兼容的浏览器。您可以使用browserify
执行此操作或使用其在线服务Browserify Wizard - util下载浏览器版本。
https://try-puppeteer.appspot.com/上的代码
const browser = await puppeteer.launch();
const page = await browser.newPage();
await page.goto("http://books.toscrape.com/", {waitUntil: "networkidle2", timeout: 0});
await page.waitFor(1000);
//Copy of https://wzrd.in/standalone/util@latest
await page.addScriptTag({url: "https://cdn.rawgit.com/brahma-dev/099d0d6d43a5d013603bcd245ee7a862/raw/b0c6bb82905b5b868c287392000dc2487c41994d/util.js"});
// selector with replaceable element
const buttonText = await page.evaluate(() => {
let buttons = [];
let selectorButton = "#default > div > div > div > div > section > div:nth-child(2) > ol > li:nth-child(%s) > article > div.product_price > form > button";
for(let i = 1; i < 21; i ++){
let textOut = document.querySelector(util.format(selectorButton, i)).innerText;
buttons.push(textOut);
};
return buttons;
});
// return
await browser.close();
console.log(buttonText);