NAME

Role::REST::Client - REST Client Role

VERSION

version 0.23

SYNOPSIS

        {
                package RESTExample;

                use Moose;
                with 'Role::REST::Client';

                sub bar {
                        my ($self) = @_;
                        my $res = $self->post('/foo/bar/baz', {foo => 'bar'});
                        my $code = $res->code;
                        my $data = $res->data;
                        return $data if $code == 200;
           }

        }

        my $foo = RESTExample->new(
                server =>      'http://localhost:3000',
                type   =>      'application/json',
                clientattrs => {timeout => 5},
        );

        $foo->bar;

        # controller
        sub foo : Local {
                my ($self, $c) = @_;
                my $res = $c->model('MyData')->post('/foo/bar/baz', {foo => 'bar'});
                my $code = $res->code;
                my $data = $res->data;
                ...
        }

DESCRIPTION

This REST Client role makes REST connectivity easy.

Role::REST::Client will handle encoding and decoding when using the HTTP verbs.

        GET
        HEAD
        PUT
        POST
        DELETE
        OPTIONS

Currently Role::REST::Client supports these encodings

        application/json
        application/x-www-form-urlencoded
        application/xml
        application/yaml

x-www-form-urlencoded only works for GET and POST, and only for encoding, not decoding.

Responses which claim to not be serialised data (eg text/plain, application/octet-stream) will by default not be serialised. When the response is none of these, and it is impossible to determine what encoding is used, the content will be treated as JSON by default.

NAME

Role::REST::Client - REST Client Role

METHODS

methods

Role::REST::Client implements the standard HTTP 1.1 verbs as methods

These methods can NOT have a request body

        get
        head

These methods can take a request body.

        post
        put
        delete
        options

All methods take these parameters

        url - The REST service
        data - The data structure (hashref, arrayref) to send. The data will be encoded
                according to the value of the I<type> attribute.
        args - hashref with arguments to augment the way the call is handled.

args - the optional argument parameter can have these entries

        deserializer - if you KNOW that the content-type of the response is incorrect,
        you can supply the correct content type, like

        my $res = $self->post('/foo/bar/baz', {foo => 'bar'}, {deserializer => 'application/yaml'});

        Alternatively, if you KNOW that the response is not serial data, you can
        disable deserialization by setting this to undef.

        preserve_headers - set this to true if you want to keep the headers between calls

All methods return a response object dictated by _rest_response_class. Set to Role::REST::Client::Response by default.

ATTRIBUTES

user_agent

  sub _build_user_agent { HTTP::Thin->new }

A User Agent object which has a ->request method suitably compatible with HTTP::Tiny. It should accept arguments like this: $ua->request($method, $uri, $opts), and needs to return a hashref as HTTP::Tiny does, or an HTTP::Response object. To set your own default, use a _build_user_agent method.

server

URL of the REST server.

e.g. 'http://localhost:3000'

type

MIME Content-Type header,

e.g. application/json

persistent_headers

  $self->set_persistent_header('Header' => 'foo', ... );
  $self->get_persistent_header('Header-Name');
  $self->has_no_persistent_headers;
  $self->clear_persistent_headers;

A hashref containing headers you want to use for all requests. Use the methods described above to manipulate it.

To set your own defaults, override the default or call set_persistent_header() in your BUILD method.

  has '+persistent_headers' => (
    default => sub { ... },
  );

httpheaders

  $self->set_header('Header' => 'foo', ... );
  $self->get_header('Header-Name');
  $self->has_no_headers;
  $self->clear_headers;

You can set any http header you like with set_header, e.g. $self->set_header($key, $value) but the content-type header will be overridden.

http_headers will be reset after each request, unless there's a reserve_headers argument, but it's a hack. The recommended way to keep headers across requests is to store them in the persistent_headers.

$self->httpheaders will return the combined hashref of persistent_headers and what's been added with set_header.

For historical reasons, the two methods clear_headers and reset_headers are equal. Both will clear the headers for the current request, but NOT the persistent headers.

To clear ALL headers, use

  $self->clear_all_headers;

clientattrs

Attributes to feed the user agent object (which defaults to HTTP::Thin)

e.g. {timeout => 10}

serializer_class

You can override the serializer class and use your own. Default is 'Role::REST::Client::Serializer'

serializer_options

Options for the serializer instantiation.

CONTRIBUTORS

Breno G. de Oliveira, <garu@cpan.org>

Mark Stosberg, <mark@stosberg.com>

Matt Phillips, (cpan:MATTP) <mattp@cpan.org>

Wallace Reis, <wallace@reis.me>

BUGS

Please report any bugs or feature requests to bug-role-rest-client at rt.cpan.org, or through the web interface at http://rt.cpan.org/NoAuth/ReportBug.html?Queue=Role-REST-Client.

AUTHOR

Kaare Rasmussen <kaare at cpan dot org>

COPYRIGHT AND LICENSE

This software is copyright (c) 2018 by Kaare Rasmussen.

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