Skip to content

Instantly share code, notes, and snippets.

@chrisabrams

chrisabrams/get-text.js

Last active April 19, 2018 17:47
Show Gist options
  • Save chrisabrams/a2a098d1b2d697f349f4e1ad1be98602 to your computer and use it in GitHub Desktop.
Save chrisabrams/a2a098d1b2d697f349f4e1ad1be98602 to your computer and use it in GitHub Desktop.
Download ZIP

Revisions

  1. chrisabrams revised this gist Apr 19, 2018. No changes.
  2. chrisabrams revised this gist Apr 19, 2018. 1 changed file with 1 addition and 1 deletion.
    2 changes: 1 addition & 1 deletion get-text.js
    View file
    Original file line number Diff line number Diff line change
    @@ -35,7 +35,7 @@ const puppeteer = require('puppeteer');
    const browser = await puppeteer.launch();
    const page = await browser.newPage();
    await page.goto('http://www.greenmatters.com')
    const nodes = await page.evaluate(matchNodes, 'news)
    const nodes = await page.evaluate(matchNodes, 'news')
    console.log('nodes', nodes) // This will be undefined
    await browser.close();
    })();
  3. chrisabrams created this gist Apr 19, 2018.
    41 changes: 41 additions & 0 deletions get-text.js
    View file
    Original file line number Diff line number Diff line change
    @@ -0,0 +1,41 @@
    function matchNodes(text) {

    try {

    const filter = {
    acceptNode: function(node){
    if(node.nodeValue.toLowerCase().includes(text)){
    return NodeFilter.FILTER_ACCEPT
    }

    return NodeFilter.FILTER_REJECT
    }
    }

    const nodes = []
    const rootElement = document.body
    const walker = document.createTreeWalker(rootElement, NodeFilter.SHOW_TEXT, filter, false)

    while(walker.nextNode()) {
    nodes.push(walker.currentNode.parentNode)
    }
    console.log(nodes) // This will contain nodes
    return nodes
    }
    catch(e) {
    console.error(e)
    return e
    }

    }

    const puppeteer = require('puppeteer');

    (async () => {
    const browser = await puppeteer.launch();
    const page = await browser.newPage();
    await page.goto('http://www.greenmatters.com')
    const nodes = await page.evaluate(matchNodes, 'news)
    console.log('nodes', nodes) // This will be undefined
    await browser.close();
    })();