The point of this project is to make web scraping easier for developers in any language. This allows you to send a URL as a parameter to a Heroku application via a GET request and receive the scraped HTML as a result. The most helpful part of this project is that it returns the web page after it has been dynamically populated by JavaScript, so you can scrape nearly any page.
Simply send a GET request to https://dynamic-page-retrieval.herokuapp.com/scrape with a URL
parameter, which should be formatted like so: ?URL=https://www.google.com.
So, the entire URL for your GET request, if you were going to use the pre-hosted Heroku
application, would be https://dynamic-page-retrieval.herokuapp.com/scrape?URL=https://www.google.com
if you wanted to scrape https://www.google.com.
An example of how to format this GET request in JavaScript:
const Http = new XMLHttpRequest();
const url = "https://dynamic-page-retrieval.herokuapp.com/scrape?URL=https://www.google.com";
Http.open("GET", url);
Http.send();
Http.onreadystatechange=(e)=>{
// Replace console.log() with what you need the HTML for,
// or assign it to a global variable for use elsewhere
console.log(Http.responseText)
}First, create a free Heroku account. If you already have one, there is no need to make a new one.
Next, make sure you have Node.js and npm installed locally. In the creation of this project, I used Node v9.3.0 and npm v6.4.1, but it shouldn't matter that much since you are just going to be deploying to Heroku. If you are going to run this locally, then version will likely be more of a factor.
Clone this project to your machine and open a terminal in the folder. Enter the following sequence of commands:
heroku create
heroku buildpacks:add https://github.com/jontewks/puppeteer-heroku-buildpack
git push heroku master
heroku ps:scale web=1
heroku open
And you should have a working copy of the project!
I am using kaffeine to keep my dyno alive to reduce loading times. It is currently set to sleep at 12:00 AM to conserve hours, so it will not be awake from 12:00-6:00 unless someone sends a request during that time interval. An alternative is to add something like this to the app to help it keep itself awake:
var http = require("http");
setInterval(function() {
http.get("http://<your app name>.herokuapp.com");
}, 300000); // every 5 minutes (300000)Feel free to open a PR for README additions of GET requests in other languages, making a pretty homepage and displaying the information on the scraped page in a nicer format, better tests, better error handling, etc.
- Make npm package where you just put in the URL and get back the scraped content
- Make similar projects in other languages (even though that was a bust before)