NAME

Mojolicious::Lite - Micro Web Framework

SYNOPSIS

# Using Mojolicious::Lite will enable "strict" and "warnings"
use Mojolicious::Lite;

# Route with placeholder
get '/:foo' => sub {
    my $self = shift;
    my $foo  = $self->param('foo');
    $self->render(text => "Hello from $foo!");
};

# Start the Mojolicious command system
app->start;

DESCRIPTION

Mojolicous::Lite is a micro web framework built around Mojolicious.

A minimal Hello World application looks like this, strict and warnings are automatically enabled and a few functions imported when you use Mojolicious::Lite, turning your script into a full featured web application.

#!/usr/bin/env perl

use Mojolicious::Lite;

get '/' => sub { shift->render(text => 'Hello World!') };

app->start;

There is also a helper command to generate a small example application.

% mojolicious generate lite_app

All the normal Mojolicious command options are available from the command line. Note that CGI, FastCGI and PSGI environments can usually be auto detected and will just work without commands.

% ./myapp.pl daemon
Server available at http://127.0.0.1:3000.

% ./myapp.pl daemon --listen http://*:8080
Server available at http://127.0.0.1:8080.

% ./myapp.pl daemon_prefork
Server available at http://127.0.0.1:3000.

% ./myapp.pl cgi
...CGI output...

% ./myapp.pl fastcgi
...Blocking FastCGI main loop...

% ./myapp.pl
...List of available commands (or automatically detected environment)...

The app->start call that starts the Mojolicious command system can be customized to override normal @ARGV use.

app->start('cgi');

Your application will automatically reload itself if you set the --reload option, so you don't have to restart the server after every change.

% ./myapp.pl daemon --reload
Server available at http://127.0.0.1:3000.

Routes are basically just fancy paths that can contain different kinds of placeholders.

# /foo
get '/foo' => sub {
    my $self = shift;
    $self->render(text => 'Hello World!');
};

All routes can have a name associated with them, this allows automatic template detection and back referencing with url_for, link_to and form_for. Names are always the last argument, the value * means that the name is simply equal to the route without non-word characters.

# /
get '/' => 'index';

# /foo
get '/foo' => '*';

# /bar
get '/bar' => sub {
    my $self = shift;
    $self->render(text => 'Hi!')
} => 'bar';

__DATA__

@@ index.html.ep
<%= link_to foo => {%>
    Foo
<%}%>.
<%= link_to bar => {%>
    Bar
<%}%>.

@@ foo.html.ep
<a href="<%= url_for 'index' %>">Home</a>.

Templates can have layouts.

# GET /with_layout
get '/with_layout' => sub {
    my $self = shift;
    $self->render('with_layout', layout => 'green');
};

__DATA__

@@ with_layout.html.ep
We've got content!

@@ layouts/green.html.ep
<!doctype html><html>
    <head><title>Green!</title></head>
    <body><%= content %></body>
</html>

Template blocks can be reused like functions in Perl scripts.

# GET /with_block
get '/with_block' => 'block';

__DATA__

@@ block.html.ep
<% my $link = {%>
    <% my ($url, $name) = @_; %>
    Try <%= link_to $url => {%><%= $name %><%}%>!
<%}%>
<!doctype html><html>
    <head><title>Sebastians Frameworks!</title></head>
    <body>
        <%== $link->('http://mojolicious.org', 'Mojolicious') %>
        <%== $link->('http://catalystframework.org', 'Catalyst') %>
    </body>
</html>

Templates can also pass around blocks of captured content and extend each other.

# GET /
get '/' => 'first';

# GET /second
get '/second' => 'second';

__DATA__

@@ first.html.ep
<!doctype html><html>
    <head><%= content header => {%><title>Hi!</title><%}%></head>
    <body><%= content body => {%>First page!<%}%></body>
</html>

@@ second.html.ep
% extends 'first';
<% content header => {%>
    <title>Howdy!</title>
<%}%>
<% content body => {%>
    Second page!
<%}%>

Route placeholders allow capturing parts of a request path until a / or . separator occurs, results will be stored by name in the stash and param.

# /foo/* (everything except "/" and ".")
# /foo/test
# /foo/test123
get '/foo/:bar' => sub {
    my $self = shift;
    my $bar  = $self->stash('bar');
    $self->render(text => "Our :bar placeholder matched $bar");
};

# /*something/foo (everything except "/" and ".")
# /test/foo
# /test123/foo
get '/(:bar)something/foo' => sub {
    my $self = shift;
    my $bar  = $self->param('bar');
    $self->render(text => "Our :bar placeholder matched $bar");
};

Relaxed placeholders allow matching of everything until a / occurs.

# /*/hello (everything except "/")
# /test/hello
# /test123/hello
# /test.123/hello
get '/(.you)/hello' => sub {
    shift->render('groovy');
};

__DATA__

@@ groovy.html.ep
Your name is <%= $you %>.

Wildcard placeholders allow matching absolutely everything, including / and ..

# /hello/* (everything)
# /hello/test
# /hello/test123
# /hello/test.123/test/123
get '/hello/(*you)' => sub {
    shift->render('groovy');
};

__DATA__

@@ groovy.html.ep
Your name is <%= $you %>.

Routes can be restricted to specific request methods.

# GET /bye
get '/bye' => sub { shift->render(text => 'Bye!') };

# POST /bye
post '/bye' => sub { shift->render(text => 'Bye!') };

# GET|POST|DELETE /bye
any [qw/get post delete/] => '/bye' => sub {
    shift->render(text => 'Bye!');
};

# /baz
any '/baz' => sub {
    my $self   = shift;
    my $method = $self->req->method;
    $self->render(text => "You called /baz with $method");
};

All placeholders get compiled to a regex internally, with regex constraints this process can be easily customized.

# /*
any '/:bar' => [bar => qr/\d+/] => sub {
    my $self = shift;
    my $bar  = $self->param('bar');
    $self->render(text => "Our :bar placeholder matched $bar");
};

Routes allow default values to make placeholders optional.

# /hello/*
get '/hello/:name' => {name => 'Sebastian'} => sub {
    my $self = shift;
    $self->render('groovy', format => 'txt');
};

__DATA__

@@ groovy.txt.ep
My name is <%= $name %>.

All those features can be easily used together.

# /everything/*?name=*
get '/everything/:stuff' => [stuff => qr/\d+/] => {stuff => 23} => sub {
    shift->render('welcome');
};

__DATA__

@@ welcome.html.ep
Stuff is <%= $stuff %>.
Query param name is <%= param 'name' %>.

Here's a fully functional example for a html form handling application using multiple features at once.

#!/usr/bin/env perl

use Mojolicious::Lite;

get '/' => 'index';

post '/test' => sub {
    my $self = shift;

    my $groovy = $self->param('groovy') || 'Austin Powers';
    $groovy =~ s/[^\w\s]+//g;

    $self->render(
        template => 'welcome',
        layout   => 'funky',
        groovy   => $groovy
    );
} => 'test';

app->start;
__DATA__

@@ index.html.ep
% layout 'funky';
Who is groovy?
<%= form_for test => (method => 'post') => {%>
    <%= input 'groovy', type => 'text' %>
    <input type="submit" value="Woosh!" />
<%}%>

@@ welcome.html.ep
<%= $groovy %> is groovy!
<%= include 'menu' %>

@@ menu.html.ep
<%= link_to index => {%>
    Try again
<%}%>

@@ layouts/funky.html.ep
<!doctype html><html>
    <head><title>Funky!</title></head>
    <body><%= content %>
    </body>
</html>

Authentication and code shared between multiple routes can be realized easily with the under statement. All following routes are only evaluated if the under callback returned a true value.

use Mojolicious::Lite;

# Authenticate based on name parameter
under sub {
    my $self = shift;

    # Authenticated
    my $name = $self->param('name') || '';
    return 1 if $name eq 'Bender';

    # Not authenticated
    $self->render('denied');
    return;
};

# GET / (with authentication)
get '/' => 'index';

app->start;
__DATA__;

@@ denied.html.ep
You are not Bender, permission denied!

@@ index.html.ep
Hi Bender!

Conditions such as agent allow even more powerful route constructs.

# /foo
get '/foo' => (agent => qr/Firefox/) => sub {
    shift->render(
        text => 'Congratulations, you are using a cool browser!');
}

# /foo
get '/foo' => (agent => qr/Internet Explorer/) => sub {
    shift->render(
        text => 'Dude, you really need to upgrade to Firefox!');
}

Formats can be automatically detected by looking at file extensions.

# /detection.html
# /detection.txt
get '/detection' => sub {
    my $self = shift;
    $self->render('detected');
};

__DATA__

@@ detected.html.ep
<!doctype html><html>
    <head><title>Detected!</title></head>
    <body>HTML was detected.</body>
</html>

@@ detected.txt.ep
TXT was detected.

Signed cookie based sessions just work out of the box as soon as you start using them. The flash can be used to store values that will only be available for one request, this is very useful in combination with redirect_to.

use Mojolicious::Lite;

get '/login' => sub {
    my $self = shift;
    my $name = $self->param('name') || '';
    my $pass = $self->param('pass') || '';
    return $self->render unless $name eq 'sebastian' && $pass eq '1234';
    $self->session(name => $name);
    $self->flash(message => 'Thanks for logging in!');
    $self->redirect_to('index');
} => 'login';

get '/' => sub {
    my $self = shift;
    return $self->redirect_to('login') unless $self->session('name');
    $self->render;
} => 'index';

get '/logout' => sub {
    my $self = shift;
    $self->session(expires => 1);
    $self->redirect_to('index');
} => 'logout';

app->start;
__DATA__

@@ layouts/default.html.ep
<!doctype html><html>
    <head><title>Mojolicious rocks!</title></head>
    <body><%= content %></body>
</html>

@@ login.html.ep
% layout 'default';
<%= form_for login => {%>
    <% if (param 'name') { %>
        <b>Wrong name or password, please try again.</b><br />
    <% } %>
    Name:<br />
    <%= input name => (type => 'text') %><br />
    Password:<br />
    <%= input pass => (type => 'text') %><br />
    <input type="submit" value="Login" />
<%}%>

@@ index.html.ep
% layout 'default';
<% if (my $message = flash 'message' ) { %>
    <b><%= $message %></b><br />
<% } %>
Welcome <%= session 'name' %>!<br />
<%= link_to logout => {%>
    Logout
<%}%>

Note that you should use a custom secret to make signed cookies really secure.

app->secret('My secret passphrase here!');

A full featured HTTP 1.1 and WebSocket client is built right in. Especially in combination with Mojo::JSON and Mojo::DOM this can be a very powerful tool.

get '/test' => sub {
    my $self = shift;
    $self->render(
        data => $self->client->get('http://mojolicious.org')->res->body);
};

WebSocket applications have never been this easy before.

websocket '/echo' => sub {
    my $self = shift;
    $self->receive_message(sub {
        my ($self, $message) = @_;
        $self->send_message("echo: $message");
    });
};

External templates will be searched by the renderer in a templates directory.

# /external
any '/external' => sub {
    my $self = shift;

    # templates/foo/bar.html.ep
    $self->render('foo/bar');
};

Static files will be automatically served from the DATA section (even Base 64 encoded) or a public directory if it exists.

@@ something.js
alert('hello!');

@@ test.txt (base64)
dGVzdCAxMjMKbGFsYWxh

% mkdir public
% mv something.js public/something.js

Testing your application is as easy as creating a t directory and filling it with normal Perl unit tests like t/funky.t.

use Test::More tests => 3;
use Test::Mojo;

use FindBin;
$ENV{MOJO_HOME} = "$FindBin::Bin/../";
require "$ENV{MOJO_HOME}/myapp.pl";

my $t = Test::Mojo->new;
$t->get_ok('/')->status_is(200)->content_like(qr/Funky!/);

Run all unit tests with the test command.

% ./myapp.pl test

To make your tests less noisy you can also change the application log level directly in your test files.

$t->app->log->level('error');

To disable debug messages later in a production setup you can change the Mojolicious mode, default will be development.

% MOJO_MODE=production ./myapp.pl

Log messages will be automatically written to a log/$mode.log file if a log directory exists.

% mkdir log

For more control the Mojolicious instance can be accessed directly.

app->log->level('error');
app->routes->route('/foo/:bar')->via('get')->to(cb => sub {
    my $self = shift;
    $self->render(text => 'Hello Mojo!');
});

In case a lite app needs to grow, lite and real Mojolicous applications can be easily mixed to make the transition process very smooth.

package MyApp::Foo;
use base 'Mojolicious::Controller';

sub index { shift->render(text => 'It works!') }

package main;
use Mojolicious::Lite;

get '/bar' => sub { shift->render(text => 'This too!') };

app->routes->namespace('MyApp');
app->routes->route('/foo/:action')->via('get')->to('foo#index');

app->start;

There is also a helper command to generate a full Mojolicious example that will let you explore the astonishing similarities between Mojolicious::Lite and Mojolicious applications. Both share about 99% of the same code, so almost everything you learned in this tutorial applies there too. :)

% mojolicious generate app

Have fun!

ATTRIBUTES

Mojolicious::Lite inherits all attributes from Mojolicious.

METHODS

Mojolicious::Lite inherits all methods from Mojolicious.

SEE ALSO

Mojolicious, Mojolicious::Guides, http://mojolicious.org.