Stop Using Selenium or Playwright for Web Scraping

Поділитися
Вставка
  • Опубліковано 25 лис 2024

КОМЕНТАРІ • 52

  • @hydrosis-v3k
    @hydrosis-v3k 29 днів тому +5

    holy shit, literally the first proper guide to a decent alternative to web automation.

  • @andydataguy
    @andydataguy 28 днів тому +4

    Your timing could not have been more perfect on this video brother. Thank you!! 🙌🏾💜
    Idk if you have a community or program, but as soon as i get hired im buying it to help support you.
    As a selftaught dev (formerly digital marketer) your videos were vital in being able to finally wrap my head around data extraction

  • @SigKappel
    @SigKappel Місяць тому +2

    Thanks John for this update. I have some mission critical scrapes for inventory and this will come in handy when I need to be more stealth.

  • @EnglishRain
    @EnglishRain 24 дні тому +1

    Thank you i didn't know Playwright isn't really recommended for scraping

  • @zvnman
    @zvnman 28 днів тому +1

    Thanks for the kind advice! In gratitude to all my clients, I give your referral link)) Keep going bro!!!

  • @frankcasanova2132
    @frankcasanova2132 Місяць тому +3

    JUST WHAT I NEEDED WHEN I NEEDED IT

  • @realitywords-17398
    @realitywords-17398 23 дні тому

    Great Man.......... You are struggling a lot...... Keep your morale high!

  • @davidl3383
    @davidl3383 20 годин тому

    excellente and easier than playwrigth etc !! thank you so much

  • @kinuthiamatata6040
    @kinuthiamatata6040 Місяць тому +9

    working with a website that uses WebSocket connections instead of traditional XHR/fetch requests. What's the best way to intercept traffic for scraping?

    • @JohnWatsonRooney
      @JohnWatsonRooney  Місяць тому +10

      honestly I haven't got a lot of experience with websockets, but i know you can connect to them via requests/httpx but i dont have any practical exp sorry

    • @munchcup
      @munchcup Місяць тому +3

      Search on UA-cam scraping Ohio state student emails but It's in js which same works for python.

    • @mehdikaraouet7980
      @mehdikaraouet7980 Місяць тому

      ​@JohnWatsonRooney all the respect for honesty and humbleness ❤

    • @eyayawb
      @eyayawb Місяць тому

      Reverse engineer the data transfer process. I successfully reverse-engineered an app that pulls data from a Firebase database using a WebSocket connection. I used websockets package to replicate the communication pattern.

  • @mvace
    @mvace 29 днів тому +1

    Thanks John, I was struggling to login to one website for a few days now. It was detecting my automation script. I used selenium-driverless and it works great now. So perfect timing with this video! Do you have any tips on how to handle uploads with selenium-driverless? After I log in to the website I need to upload a file from my local machine but the upload seems to be working differently for selenium-driverless compared to selenium. There is not much in the documentation regarding uploads in selenium-driverless.

  • @gamehubler
    @gamehubler 23 дні тому

    I learn a lot from your video explanations and like your style of sharing it with us, and comes with the fresh air because it is up to date.
    Can you maybe bring some more complex web scraping examples? Mostly the pages in your showcase were viewable directly or with a login.
    Would it be possible to bring some examples in the future with GQL sites, it is more difficult to solve these types.

  • @abdelkoddouslaarif1295
    @abdelkoddouslaarif1295 28 днів тому

    okay! this is all good and well, impressive even, but I wanna ask a slightly different question, how do you stay updated with the new alternatives that come out? you always seem to know when something better is out or if something old isn't as reliable anymore or not being maintained... what tech news forums/ community chats do you follow?

  • @dragon3602010
    @dragon3602010 29 днів тому +3

    so when choosing "nodriver" instead of "SeleniumBase" ?
    thanks

    • @JohnWatsonRooney
      @JohnWatsonRooney  29 днів тому

      Try both and see which works best for what you are trying to do

  • @utsavgoswami5263
    @utsavgoswami5263 19 днів тому

    thank you so much for this. if it is okay, are there any better alternatives (i.e. cheaper) to proxyscrape?

  • @nickwoodward819
    @nickwoodward819 Місяць тому +2

    i'm a bit confused by "using the chrome already installed on your machine" - do you mean the server? or is this a python thing? (I'm using js and would be scraping either in a serverless function or in node)

    • @kinuthiamatata6040
      @kinuthiamatata6040 Місяць тому

      i agree this might not be easily "automateable" , "You may consider Playwright's connectOverCDP() with a containerized Chrome. Just make sure to expose Chrome's debug port via socat (socat TCP-LISTEN:9222,fork TCP:localhost:9222) and connect to that. This works great with Node/serverless and gives you full browser auto-capabilities."

    • @JohnWatsonRooney
      @JohnWatsonRooney  Місяць тому +2

      Sorry to clarify - when you install selenium or playwright,m they install their own version of chrome. these two use the existing install on your pc already (if you have it) so you install chrome as you would normally, no extra install steps etc

  • @return_1101
    @return_1101 28 днів тому

    The best one! Thank you very much, mr. Rooney!

  • @cstanleyhns
    @cstanleyhns 13 днів тому

    Hi, is it possible to run this kind of thing in docker and then ultimately in a lambda?

  • @Optimusjf
    @Optimusjf 11 днів тому

    How can I send a PFX client certificate in Selenium-driverless?

  • @Analyse_US
    @Analyse_US Місяць тому +1

    This is gold! Thanks.

  • @rewazilol
    @rewazilol 29 днів тому +1

    Do you have experience using Playwright and there being issues? On one hand I know its a test lib, but on the other hand its extremely well funded and well maintained. The nodriver lib looks like it was built by one guy. I'm still kind of not sure which way to go

    • @JohnWatsonRooney
      @JohnWatsonRooney  29 днів тому +3

      Playwright it an amazing library, it’s easy to use and works extremely well. But for web scraping you have to monkey around with it as even the most basic WAF can detect it and throw a captcha. Nodriver is build by one guy but it’s open source and removes a lot of the obvious flags that the browser is being controller automatically which is why I’m talking about it here

  • @saadkhan883
    @saadkhan883 25 днів тому

    Sir I am using playwright library to scrap Google maps so please tell is it a good choice to scrap data from GMpaps , because I am still thinking about blocking or recaptcha ...

  • @graczew
    @graczew Місяць тому +1

    Thanks, mate. This is really helpful.

  • @yafethtb
    @yafethtb 28 днів тому

    At last, a web scraping library that does not need another Chromium and uses only my Chrome! I'm waiting for something like nodriver.

  • @EmanueleCannizzaro
    @EmanueleCannizzaro Місяць тому +1

    As usual a great content explained in plain English.

  • @serkhetreo2489
    @serkhetreo2489 28 днів тому

    Hi, what if i want to build an unofficial api for a site . Is there a better way

  • @amrogendiah198
    @amrogendiah198 27 днів тому

    Could you make a video about scraping using AI with python ?

  • @green-forest-23
    @green-forest-23 Місяць тому +1

    There's something like this for typescript?

    • @JohnWatsonRooney
      @JohnWatsonRooney  Місяць тому +1

      Yes but I forget what it’s called - hopefully
      Someone else will know

    • @boniazdaniel
      @boniazdaniel 29 днів тому

      Puppeteer extra plugin? The stealth one

  • @phantazzor
    @phantazzor 29 днів тому

    how to scrape an app if there are no web version

  • @hamzahalli3500
    @hamzahalli3500 29 днів тому +1

    Thank you

  • @wisjnujudho3152
    @wisjnujudho3152 20 днів тому

    wow. i think i need to modify all of my scrapers.

  • @hw5622
    @hw5622 25 днів тому

    great video! thx!!!!!!!!

  • @DavidChavez-z2v
    @DavidChavez-z2v 23 дні тому

    Como sabes estas cosas excelente

  • @MohanadSaid-u8x
    @MohanadSaid-u8x Місяць тому +1

    Liiiiiit 🔥🔥🔥🔥

  • @luisechevarria186
    @luisechevarria186 27 днів тому +1

    Where does one begin for web-scraping? I am trying to scrape, JS heavy websites, all I see are recommendations to use Selenium/Playwright, this video is not too clear about what to use instead.

    • @JohnWatsonRooney
      @JohnWatsonRooney  27 днів тому +2

      Try using selenium-driverless. Similar to selenium but less detectable causing you less issues scraping

  • @Canda-fh4xc
    @Canda-fh4xc 27 днів тому

    Thank you