TypeError: Cannot read property 'replace' of undefined

JavaScript
1

So I tried to learn how to do this on my own by taking two video courses about node and cheerio on udemy but I haven’t gotten any better at it. The code produces an array of image URLs when I do this:

const request = require("request-promise");
const cheerio = require("cheerio");

const url = "https://www.example.com";

const scrapeResults = [];
async function scrapeJobHeader() {
  try {
    const htmlResult = await request.get(url);
    const $ = await cheerio.load(htmlResult);
	$("td.productListing-data > a ").each((index, element) => {
      const resultTitle = $(element).children("img");
	  
	  const img_url = resultTitle.attr("src");
	  
      const scrapeResult = { img_url };
    
      scrapeResults.push(scrapeResult);
    });
    return scrapeResults;
  } catch (err) {
    console.error(err);
  }
}

async function scrapeWebsite() {
  const jobsWithHeaders = await scrapeJobHeader();
 console.log(jobsWithHeaders);

}

scrapeWebsite();

but I get the error “TypeError: Cannot read property ‘replace’ of undefined” when I do this:

const img_url = resultTitle.attr("src").replace("images\\/more_color.png","");

The images that are returned include “images/more_color.png” but I just want to return the actual product images and not the pngs.

The results look like this:

{ img_url: 'images/more_color.png' },
  { img_url: undefined },
  { img_url: undefined },
  {
    img_url: 'images/20191206/thumb/AK1501-@RH-CRY-LOVE@22X06-825_3L@467400@200@01@200.jpg'
  },
  { img_url: 'images/more_color.png' },
  { img_url: undefined },
  { img_url: undefined },
  {
    img_url: 'images/20191206/thumb/AK1501-@GD-CRY-LOVE@22X06-825_3L@467399@200@01@200.jpg'
  },
  { img_url: 'images/more_color.png' },
  { img_url: undefined },
  { img_url: undefined },
  {
    img_url: 'images/20191206/thumb/AK1500-@GD-CRY-QUEEN@3X06-825_3L@467397@200@01@200.jpg'
  },
  { img_url: 'images/more_color.png' },
  { img_url: undefined },
  { img_url: undefined },

The next step would be to get rid of the undefined image URLs but I haven’t gotten that far yet.

2

I got rid of the undefined ones by using:

 if (img_url != undefined){
      scrapeResults.push(scrapeResult);
	 }
3

so if attr(“src”) is undefined, you can’t do a .replace on it - it’s not a string, it’s undefined.

4

The solution to this problem was published at https://www.freecodecamp.org/forum/t/web-scraper-outputs-just-one-item-instead-of-many/360427/46

5

This topic was automatically closed 91 days after the last reply. New replies are no longer allowed.