NAME

MetaCPAN::API - A comprehensive, DWIM-featured API to MetaCPAN

VERSION

version 0.43

SYNOPSIS

# simple usage
my $mcpan  = MetaCPAN::API->new();
my $author = $mcpan->author('XSAWYERX');
my $dist   = $mcpan->release( distribution => 'MetaCPAN-API' );

# advanced usage with cache (contributed by Kent Fredric)
require CHI;
require WWW::Mechanize::Cached;
require HTTP::Tiny::Mech;
require MetaCPAN::API;

my $mcpan = MetaCPAN::API->new(
  ua => HTTP::Tiny::Mech->new(
    mechua => WWW::Mechanize::Cached->new(
      cache => CHI->new(
        driver => 'File',
        root_dir => '/tmp/metacpan-cache',
      ),
    ),
  ),
);

DESCRIPTION

This is a hopefully-complete API-compliant interface to MetaCPAN (https://metacpan.org) with DWIM capabilities, to make your life easier.

This module has three purposes:

• Provide 100% of the beta MetaCPAN API

  This module will be updated regularly on every MetaCPAN API change, and intends to provide the user with as much of the API as possible, no shortcuts. If it's documented in the API, you should be able to do it.

  Because of this design decision, this module has an official MetaCPAN namespace with the blessing of the MetaCPAN developers.

  Notice this module currently only provides the beta API, not the old soon-to-be-deprecated API.

• Be lightweight, to allow flexible usage

  While many modules would help make writing easier, it's important to take into account how they affect your compile-time, run-time and overall memory consumption.

  By providing a slim interface implementation, more users are able to use this module, such as long-running processes (like daemons), CLI or GUI applications, cron jobs, and more.

• DWIM

  While it's possible to access the methods defined by the API spec, there's still a matter of what you're really trying to achieve. For example, when searching for "Dave", you want to find both Dave Cross and Dave Rolsky (and any other Dave), but you also want to search for a PAUSE ID of DAVE, if one exists.

  This is where DWIM comes in. This module provides you with additional generic methods which will try to do what they think you want.

  Of course, this does not prevent you from manually using the API methods. You still have full control over that, if that's what you wish.

  You can (and should) read up on the generic methods, which will explain how their DWIMish nature works, and what searches they run.

ATTRIBUTES

base_url

my $mcpan = MetaCPAN::API->new(
    base_url => 'http://localhost:9999',
);

This attribute is used for REST requests. You should set it to where the MetaCPAN is accessible. By default it's already set correctly, but if you're running a local instance of MetaCPAN, or use a local mirror, or tunnel it through a local port, or any of those stuff, you would want to change this.

Default: http://api.metacpan.org/v0.

This attribute is read-only (immutable), meaning that once it's set on initialize (via new()), you cannot change it. If you need to, create a new instance of MetaCPAN::API. Why is it immutable? Because it's better.

ua

This attribute is used to contain the user agent used for running the REST request to the server. It is specifically set to HTTP::Tiny, so if you want to set it manually, make sure it's of HTTP::Tiny.

HTTP::Tiny is used as part of the philosophy of keeping it tiny.

This attribute is read-only (immutable), meaning that once it's set on initialize (via new()), you cannot change it. If you need to, create a new instance of MetaCPAN::API. Why is it immutable? Because it's better.

ua_args

my $mcpan = MetaCPAN::API->new(
    ua_args => [ agent => 'MyAgent' ],
);

The arguments that will be given to the HTTP::Tiny user agent.

This attribute is read-only (immutable), meaning that once it's set on initialize (via new()), you cannot change it. If you need to, create a new instance of MetaCPAN::API. Why is it immutable? Because it's better.

The default is a user agent string: MetaCPAN::API/$version.

METHODS

fetch

my $result = $mcpan->fetch('/release/distribution/Moose');

# with parameters
my $more = $mcpan->fetch(
    '/release/distribution/Moose',
    param => 'value',
);

This is a helper method for API implementations. It fetches a path from MetaCPAN, decodes the JSON from the content variable and returns it.

You don't really need to use it, but you can in case you want to write your own extension implementation to MetaCPAN::API.

It accepts an additional hash as GET parameters.

post

# /release&content={"query":{"match_all":{}},"filter":{"prefix":{"archive":"Cache-Cache-1.06"}}}
my $result = $mcpan->post(
    'release',
    {
        query  => { match_all => {} },
        filter => { prefix => { archive => 'Cache-Cache-1.06' } },
    },
);

The POST equivalent of the fetch() method. It gets the path and JSON request.

AUTHOR

Sawyer X <xsawyerx@cpan.org>

COPYRIGHT AND LICENSE

This software is copyright (c) 2011 by Sawyer X.

This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself.