Puppeteer is one of the greatest tools available on the web in order to perform the automation regarding website or In-Browser Activities. Today, Instead of single huge chunk of code (which does nothing helpful other than only page screenshots ) I will try to explain the several tricks which you can use in your next project or any testing purposes.
Puppeteer is browser automation tool which is used to automated activities on the internet using Headless Chromium Browser.
Puppteer uses the high level API provided by the google in order to interact with headless chrome. It provides high level of flexibility and extensibility in order to modify the default behavior of Puppeteer.
For launching Puppeteer, people normally use basic configuration, but here I will show some advance modification in order to launch Puppeteer in various modes
Here we are using main three options
- Lauching Puppteer using headless flag as false , because it will show the activity of browser on the screen
- Second flag is linux specific. I used it here, because if i excluded it then code will break on ubuntu 20.04
- This option is pretty useful because it removes the tag from the browser window which is "* Google chrome is controlled by the automated software*"
Then all we are doing is launching it using launch() and then redirecting the page to google
Whenever we are doing automation, there is a point when you will be forced to use the incognito mode of the chrome. It is pretty helpful, if a site is preventing you from accessing the useful stuff, because they stored the cookie in you localStorage marking you as a robot and when you are accessing the stuff it is blocking your access
Then solution is here
Add this two line snippet while launching and Puppteer will open automatically in the incognito mode. COOL 8)
Handling navigation is pretty important while you are extending the testing / automation to the multiple pages. If the navigation isn't handled properly then we may get this
Failed to execute. Mostly because context destroyed due to navigation
To avoid this stuff handle navigation as follows
This is the example of performing the search on the google. Type the search term in the search box (with provided delay in order to avoid the bot detection) and then press Enter.
After it, wait until navigation completed and when the DOM content is loaded completely then proceed further.
Further to these, we can use two constraint to check whether the page is loaded completely or not
- 'domcontentloaded' :- as soon as DOM loaded proceed
We have to click on the web in order to get info regarding something. We all Know that but Puppteer doesn't. Hence, forcing it to click on element is one of crucial functionality we all must know.
😂 You will use this a lot, because it allows the smooth scrolling up to the element which is to be clicked. If it is present at the bottom then instead of clicking it directly, it will first scroll up to the point and the click on it.
It is pretty useful, if the scenario occurs where you don't want to interact directly with DOM then we can use the eval() in order to get things from element
For ex. clicking on it, getting text content etc
using $$eval() over the native dom API's like
is preferred by many individual. Because, if we used eval() then Puppteer will efficiently searches through the DOM and intern manages all things internally but by using native API's we are adding additional load on puppteer. Although it is preferential
Puppeteer is pretty useful for automated testing and other stuff. But, you must have to take care of detection. In this article, I touched over the few things (* Less than 1% *) which are important in day to usage. but, if you want to explore more please let me know in comments. I will love to hear your thoughts 💜
🙏 Thanks For Reading....