Is there a way to add script to add new functions in evaluate() context of chrome+puppeeter?

僤鯓⒐⒋嵵緔 提交于 2019-12-02 07:10:18

You can register helper functions in separate page.evaluate() function. page.exposeFunction() looks temptingly, but it don't have access to browser context (and you need document object).

Here is an example of registering helper function with $x():

const puppeteer = require('puppeteer');

const helperFunctions = () => {
    window.$x = xPath => document
        .evaluate(
            xPath,
            document,
            null,
            XPathResult.FIRST_ORDERED_NODE_TYPE,
            null
        )
        .singleNodeValue;
};

(async () => {
    const browser = await puppeteer.launch();
    const page = await browser.newPage();
    await page.goto('https://en.wikipedia.org', { waitUntil: 'networkidle2' });

    await page.evaluate(helperFunctions);

    const text = await page.evaluate(() => {
        // $x() is now available
        const featureArticle = $x('//*[@id="mp-tfa"]');

        return featureArticle.textContent;
    });
    console.log(text);
    await browser.close();
})();

(edit - add helpers from a file)

You can also keep helpers in a separate file and inject it into browser context by page.addScriptTag(). Here is an example of it:

helperFunctions.js

window.$x = xPath => document
    .evaluate(
        xPath,
        document,
        null,
        XPathResult.FIRST_ORDERED_NODE_TYPE,
        null
    )
    .singleNodeValue;

And use it:

const puppeteer = require('puppeteer');

(async () => {
    const browser = await puppeteer.launch();
    const page = await browser.newPage();
    await page.goto('https://en.wikipedia.org', { waitUntil: 'networkidle2' });

    await page.addScriptTag({ path: './helperFunctions.js' });

    const text = await page.evaluate(() => {
        // $x() is now available
        const featureArticle = $x('//*[@id="mp-tfa"]');

        return featureArticle.textContent;
    });
    console.log(text);
    await browser.close();
})();

Another solution based on getElementByXPath() and getElementsByXPath(). The advantage is that we can use an xpath expression against a particular node (second argument).

window.$x = xPath => document
    .evaluate(
        xPath,
        document,
        null,
        XPathResult.FIRST_ORDERED_NODE_TYPE,
        null
    )
    .singleNodeValue;

window.getElementByXPath = function getElementByXPath(expression, scope) {
    scope = scope || document;
    var a = document.evaluate(expression, scope, null, XPathResult.ORDERED_NODE_SNAPSHOT_TYPE, null);
    if (a.snapshotLength > 0) {
        return a.snapshotItem(0);
    }
};

window.getElementsByXPath = function getElementsByXPath(expression, scope) {
    scope = scope || document;
    var nodes = [];
    var a = document.evaluate(expression, scope, null, XPathResult.ORDERED_NODE_SNAPSHOT_TYPE, null);
    for (var i = 0; i < a.snapshotLength; i++) {
        nodes.push(a.snapshotItem(i));
    }
    return nodes;
};

Real life code sample :

const puppeteer = require('puppeteer');

(async () => {

    const browser = await puppeteer.launch();
    const page = await browser.newPage();

    await page.goto('https://99bitcoins.com/bitcoin-rich-list-top100/#addresses', { waitUntil: 'networkidle2' });
    await page.addScriptTag({ path: './helperFunctions.js' });

    const result = await page.evaluate(() => {
        var obj = {};
        var data = getElementsByXPath('//table[@class="t99btc-rich-list"]//tr');
        for (var i = 1; i<=100; i++) {
           obj[i] = {
               "hash": getElementByXPath('./td/a', data[i]).innerText,
               "balance": getElementByXPath('./td[3]', data[i]).innerText
           }
        }

        return obj;

    });
    console.log(JSON.stringify(result, null, 4));
    await browser.close();

})();
易学教程内所有资源均来自网络或用户发布的内容,如有违反法律规定的内容欢迎反馈
该文章没有解决你所遇到的问题?点击提问,说说你的问题,让更多的人一起探讨吧!