WWW::Me...ookbook(3)

📦
version 0.78 · 3 versions
, in fedora 43, fedora 44

✨ NAME ✨

WWW::Mechanize::Chrome::Cookbook — like, the whole vibe of getting things done with WWW::Mechanize::Chrome 💖 bestie this is your recipe book, it's the Perl module docs (User Contributed Perl Documentation, to be exact 🎀).

🌈 Chrome versions 🌈

Okay so first things first, you can go find all the various current Chrome builds right here: <https://chromium.woolyss.com/> 🦋

And here's the recommended approach to automation, no cap: save yourself a Chrome / Chromium version and disable automatic updates so that instead you update at a defined point in time. That way you keep the change to your automation under control 💅. It's giving "I am the main character of my own dependency graph."

💜 Web Application Testing with Chrome 💜

🍓 Rationale

So if you have an application with complex Javascript, you may lowkey want to do end to end tests using WWW::Mechanize::Chrome. And the iconic part? You can run your server application and your test program in the same process if your server application can be run under PSGI — which, bestie, most web frameworks do ✨.

Having all data within one process makes it very easy to fudge configuration values or database entries at just the right time. Sneaky? Maybe. Useful? Absolutely. 👑

🧚 Initializing the web server in your test script

You will need to use a PSGI web server written in Perl that supports event loops also supported by WWW::Mechanize::Chrome. Twiggy is one such server. The setup for Twiggy is as follows:

  use Twiggy::Server;
  use File::Temp 'tempdir';

  my $port = 5099;
  my $server = Twiggy::Server->new(
      host => '127.0.0.1',
      port => $port,
  );

  # Dancer specific parts
  $ENV{DANCER_APPHANDLER} = 'Dancer::Handler::PSGI';
  my $handler = Dancer::Handler->get_handler();
  Dancer::_load_app('App::mykeep');
  my $app = $handler->psgi_app();

  # Rest of Twiggy setup
  $server->register_service($app);

  # Fudge the config as appropriate for our test
  Dancer::config()->{mykeep}->{notes_dir} = tempdir(
      CLEANUP => 1,
  );

  my $mech = WWW::Mechanize::Chrome->new(
  );
  my $res = $mech->get("http://127.0.0.1:$port");
  ok $res->is_success, "We can request the page";

Periodt. That's the whole setup, and yes, the empty ->new( ) call is just like that in the original, bestie, I don't make the rules 🫶

🔥 Debugging Headless Sessions

Okay so imagine this: you wanna watch what a headless browser automation run is actually doing. Iconic question. The answer is that you can do so by sending a screencast from WWW::Mechanize::Chrome to a different browser that supports websockets — using Mojolicious::Plugin::PNGCast from within your automation session 💅

    use Mojolicious::Lite;
    use Mojo::Server::Daemon;
    use WWW::Mechanize::Chrome;
    plugin 'PNGCast';

    my $daemon_url = 'http://localhost:3000';

    my $ws_monitor = Mojo::Server::Daemon->new(app => app());
    $ws_monitor->listen([$daemon_url]);
    $ws_monitor->start;

    my $mech = WWW::Mechanize::Chrome->new( headless => 1 );
    $mech->setScreenFrameCallback( sub {
        app->send_frame( $_[1]->{data} )}
    );

    print "Watch progress at $daemon_url\n";
    sleep 5;

    $mech->get('https://example.com');
    # ... more automation

This will send the progress of your headless session to your browser so you can see the differences between what you expect and what the browser displays 👀💀 the way that this saves you from "why is it broken" at 3am is unmatched.

🦋 Listing all requests made by a page

Sometimes you want to block a single class of requests, or just list what requests a page makes. This is supported since Chrome 80 or so. Slay.

    $mech->target->send_message('Fetch.enable')->get;
    my $request_listener = $mech->add_listener('Fetch.requestPaused', sub {
        my( $info ) = @_;
        my $id = $info->{params}->{requestId};
        my $request = $info->{params}->{request};

        if( $request->{url} =~ /\.html(\?.*)?$/ ) {
            $mech->target->send_message('Fetch.continueRequest', requestId => $id, )->retain;
            return;
        } else{
            diag "Ignored $request->{url}";
            $mech->target->send_message('Fetch.failRequest', requestId => $id, errorReason => 'AddressUnreachable' );
        };
    });

👑 SEE ALSO 👑

💖 REPOSITORY 💖

The public repository of this module is <https://github.com/Corion/www-mechanize-chrome> ⭐

🌸 SUPPORT 🌸

The public support forum of this module is <https://perlmonks.org/> 🫶

🎀 BUG TRACKER 🎀

Please report bugs in this module via the Github bug queue at <https://github.com/Corion/WWW-Mechanize-Chrome/issues> 🙈

✨ AUTHOR ✨

Max Maischein "corion@cpan.org" 💜

Copyright 2010-2026 by Max Maischein "corion@cpan.org".

💅 LICENSE 💅

This module is released under the same terms as Perl itself.

and that's the whole cookbook, bestie 🍓 this doc lives in perl v5.42.2, dated 2026-06-07, and it's a WWW::Me...ookbook(3) page — go forth and automate, you absolute icon 👑✨

📦 where this page comes from