Web::Simple::Application

(source, CPAN)

#NAME

Web::Simple::Application - A base class for your Web-Simple application

#DESCRIPTION

This is a base class for your Web::Simple application. You probably don't need to construct this class yourself, since Web::Simple does the 'heavy lifting' for you in that regards.

#METHODS

This class exposes the following public methods.

#default_config

Merges with the config initializer to provide configuration information for your application. For example:

sub default_config {
  (
    title => 'Bloggery',
    posts_dir => $FindBin::Bin.'/posts',
  );
}

Now, the config attribute of $self will be set to a HashRef containing keys 'title' and 'posts_dir'.

The keys from default_config are merged into any config supplied, so if you construct your application like:

MyWebSimpleApp::Web->new(
  config => { title => 'Spoon', environment => 'dev' }
)

then config will contain:

{
  title => 'Spoon',
  posts_dir => '/path/to/myapp/posts',
  environment => 'dev'
}

#run_if_script

The run_if_script method is designed to be used at the end of the script or .pm file where your application class is defined - for example:

## my_web_simple_app.pl
#!/usr/bin/env perl
use Web::Simple 'HelloWorld';

{
  package HelloWorld;

  sub dispatch_request {
    sub (GET) {
      [ 200, [ 'Content-type', 'text/plain' ], [ 'Hello world!' ] ]
    },
    sub () {
      [ 405, [ 'Content-type', 'text/plain' ], [ 'Method not allowed' ] ]
    }
  }
}

HelloWorld->run_if_script;

This returns a true value, so your file is now valid as a module - so

require 'my_web_simple_app.pl';

my $hw = HelloWorld->new;

will work fine (and you can rename it to lib/HelloWorld.pm later to make it a real use-able module).

However, it detects if it's being run as a script (via testing $0) and if so attempts to do the right thing.

If run under a CGI environment, your application will execute as a CGI.

If run under a FastCGI environment, your application will execute as a FastCGI process (this works both for dynamic shared-hosting-style FastCGI and for apache FastCgiServer style setups).

If run from the commandline with a URL path, it runs a GET request against that path -

$ perl -Ilib examples/hello-world/hello-world.cgi /
200 OK
Content-Type: text/plain

Hello world!

You can also provide a method name -

$ perl -Ilib examples/hello-world/hello-world.cgi POST /
405 Method Not Allowed
Content-Type: text/plain

Method not allowed

For a POST or PUT request, pairs on the command line will be treated as form variables. For any request, pairs on the command line ending in : are treated as headers, and 'Content:' will set the request body -

$ ./myapp POST / Accept: text/html form_field_name form_field_value

$ ./myapp POST / Content-Type: text/json Content: '{ "json": "here" }'

The body of the response is sent to STDOUT and the headers to STDERR, so

$ ./myapp GET / >index.html

will generally do the right thing.

To send basic authentication credentials, use user:pass@ syntax -

$ ./myapp GET bob:secret@/protected/path

Additionally, you can treat the file as though it were a standard PSGI application file (*.psgi). For example you can start up up with plackup

plackup my_web_simple_app.pl

or starman

starman my_web_simple_app.pl

#to_psgi_app

This method is called by "run_if_script" to create the PSGI app coderef for use via Plack and plackup. If you want to globally add middleware, you can override this method:

 use Web::Simple 'HelloWorld';

 {
   package HelloWorld;
   use Plack::Builder;
 
   around 'to_psgi_app', sub {
     my ($orig, $self) = (shift, shift);
     my $app = $self->$orig(@_); 
     builder {
       enable ...; ## whatever middleware you want
       $app;
     };
   };
 }

This method can also be used to mount a Web::Simple application within a separate *.psgi file -

use strictures 1;
use Plack::Builder;
use WSApp;
use AnotherWSApp;

builder {
  mount '/' => WSApp->to_psgi_app;
  mount '/another' => AnotherWSApp->to_psgi_app;
};

This method can be called as a class method, in which case it implicitly calls ->new, or as an object method ... in which case it doesn't.

#run

Used for running your application under stand-alone CGI and FCGI modes.

I should document this more extensively but run_if_script will call it when you need it, so don't worry about it too much.

#run_test_request

my $res = $app->run_test_request(GET => '/' => %headers);

my $res = $app->run_test_request(POST => '/' => %headers_or_form);

my $res = $app->run_test_request($http_request);

Accepts either an HTTP::Request object or ($method, $path) and runs that request against the application, returning an HTTP::Response object.

If the HTTP method is POST or PUT, then a series of pairs can be passed after this to create a form style message body. If you need to test an upload, then create an HTTP::Request object by hand or use the POST subroutine provided by HTTP::Request::Common.

If you prefix the URL with 'user:pass@' this will be converted into an Authorization header for HTTP basic auth:

my $res = $app->run_test_request(
            GET => 'bob:secret@/protected/resource'
          );

If pairs are passed where the key ends in :, it is instead treated as a headers, so:

my $res = $app->run_test_request(
            POST => '/',
           'Accept:' => 'text/html',
            some_form_key => 'value'
          );

will do what you expect. You can also pass a special key of Content: to set the request body:

my $res = $app->run_test_request(
            POST => '/',
            'Content-Type:' => 'text/json',
            'Content:' => '{ "json": "here" }',
          );

#AUTHORS

See Web::Simple for authors.

CONTENTS