Why not adopt me?
NAME
Net::Async::Webservice::UPS - UPS API client, non-blocking
VERSION
version 0.09_2
SYNOPSIS
use IO::Async::Loop;
use Net::Async::Webservice::UPS;
my $loop = IO::Async::Loop->new;
my $ups = Net::Async::Webservice::UPS->new({
config_file => $ENV{HOME}.'/.naws_ups.conf',
loop => $loop,
});
$ups->validate_address($postcode)->then(sub {
my ($response) = @_;
say $_->postal_code for @{$response->addresses};
return Future->wrap();
});
$loop->run;
Alternatively:
use Net::Async::Webservice::UPS;
my $ups = Net::Async::Webservice::UPS->new({
config_file => $ENV{HOME}.'/.naws_ups.conf',
user_agent => LWP::UserAgent->new,
});
my $response = $ups->validate_address($postcode)->get;
say $_->postal_code for @{$response->addresses};
DESCRIPTION
This class implements some of the methods of the UPS API, using Net::Async::HTTP as a user agent by default (you can still pass something like LWP::UserAgent and it will work). All methods that perform API calls return Futures (if using a synchronous user agent, all the Futures will be returned already completed).
NOTE: I've kept many names and codes from the original Net::UPS, so the API of this distribution may look a bit strange. It should make it simpler to migrate from Net::UPS, though.
ATTRIBUTES
live_mode
Boolean, defaults to false. When set to true, the live API endpoint will be used, otherwise the test one will. Flipping this attribute will reset "base_url", so you generally don't want to touch this if you're using some custom API endpoint.
base_url
A URI object, coercible from a string. The base URL to use to send API requests to (actual requests will be POST
ed to an actual URL built from this by appending the appropriate service path). Defaults to the standard UPS endpoints:
https://onlinetools.ups.com/ups.app/xml
for livehttps://wwwcie.ups.com/ups.app/xml
for testing
See also "live_mode".
user_id
password
access_key
Strings, required. Authentication credentials.
account_number
String. Used in some requests as "shipper number".
customer_classification
String, usually one of WHOLESALE
, OCCASIONAL
, RETAIL
. Used when requesting rates.
pickup_type
String, defaults to ONE_TIME
. Used when requesting rates.
cache
Responses are cached if this is set. You can pass your own cache object (that implements the get
and set
methods like CHI does), or use the cache_life
and cache_root
constructor parameters to get a CHI instance based on CHI::Driver::File.
user_agent
A user agent object, looking either like Net::Async::HTTP (has do_request
and POST
) or like LWP::UserAgent (has request
and post
). You can pass the loop
constructor parameter to get a default Net::Async::HTTP instance.
METHODS
does_caching
Returns a true value if caching is enabled.
new
Async:
my $ups = Net::Async::Webservice::UPS->new({
loop => $loop,
config_file => $file_name,
cache_life => 5,
});
Sync:
my $ups = Net::Async::Webservice::UPS->new({
user_agent => LWP::UserAgent->new,
config_file => $file_name,
cache_life => 5,
});
In addition to passing all the various attributes values, you can use a few shortcuts.
loop
-
a IO::Async::Loop; a locally-constructed Net::Async::HTTP will be registered to it and set as "user_agent"
config_file
-
a path name; will be parsed with Config::Any, and the values used as if they had been passed in to the constructor
cache_life
-
lifetime, in minutes, of cache entries; a "cache" will be built automatically if this is set (using CHI with the
File
driver) cache_root
-
where to store the cache files for the default cache object, defaults to
naws_ups
under your system's temporary directory
A few more examples:
no config file, no cache, async:
->new({ user_id=>$user,password=>$pw,access_key=>$ak, loop=>$loop, }),
no config file, no cache, custom user agent (sync or async):
->new({ user_id=>$user,password=>$pw,access_key=>$ak, user_agent=>$ua, }),
it's your job to register the custom user agent to the event loop, if you're using an async agent
config file, async, custom cache:
->new({ loop=>$loop, cache=>CHI->new(...), }),
transaction_reference
Constant data used to fill something in requests. I don't know what it's for, I just copied it from Net::UPS.
access_as_xml
Returns a XML document with the credentials.
request_rate
$ups->request_rate({
from => $address_a,
to => $address_b,
packages => [ $package_1, $package_2 ],
}) ==> (Net::Async::Webservice::UPS::Response::Rate)
from
and to
are instances of Net::Async::Webservice::UPS::Address, or postcode strings that will be coerced to addresses.
packages
is an arrayref of Net::Async::Webservice::UPS::Package (or a single package, will be coerced to a 1-element array ref).
NOTE: the id
field of the packages you pass in will be modified, and set to their position in the array.
Optional parameters:
limit_to
-
only accept some services (see "ServiceLabel" in Net::Async::Webservice::UPS::Types)
exclude
-
exclude some services (see "ServiceLabel" in Net::Async::Webservice::UPS::Types)
mode
-
defaults to
rate
, could beshop
service
-
defaults to
GROUND
, see Net::Async::Webservice::UPS::Service
The Future returned will yield an instance of Net::Async::Webservice::UPS::Response::Rate, or fail with an exception.
Identical requests can be cached.
validate_address
$ups->validate_address($address)
==> (Net::Async::Webservice::UPS::Response::Address)
$ups->validate_address($address,$tolerance)
==> (Net::Async::Webservice::UPS::Response::Address)
$address
is an instance of Net::Async::Webservice::UPS::Address, or a postcode string that will be coerced to an address.
Optional parameter: a tolerance (float, between 0 and 1). Returned addresses with quality below the tolerance will be filtered out.
The Future returned will yield an instance of Net::Async::Webservice::UPS::Response::Address, or fail with an exception.
Identical requests can be cached.
validate_street_address
$ups->validate_street_address($address)
==> (Net::Async::Webservice::UPS::Response::Address)
$address
is an instance of Net::Async::Webservice::UPS::Address, or a postcode string that will be coerced to an address.
The Future returned will yield an instance of Net::Async::Webservice::UPS::Response::Address, or fail with an exception.
Identical requests can be cached.
xml_request
$ups->xml_request({
url_suffix => $string,
data => \%request_data,
XMLout => \%xml_simple_out_options,
XMLin => \%xml_simple_in_options,
}) ==> ($parsed_response);
This method is mostly internal, you shouldn't need to call it.
It builds a request XML document by concatenating the output of "access_as_xml" with whatever XML::Simple produces from the given data
and XMLout
options.
It then posts (possibly asynchronously) this to the URL obtained concatenating "base_url" with url_suffix
(see the "post" method). If the request is successful, it parses the body (with XML::Simple using the XMLin
options) and completes the returned future with the result.
If the parsed response contains a non-zero /Response/ResponseStatusCode
, the returned future will fail with a Net::Async::Webservice::UPS::Exception::UPSError instance.
post
$ups->post($url_suffix,$body) ==> ($decoded_content)
Posts the given $body
to the URL obtained concatenating "base_url" with $url_suffix
. If the request is successful, it completes the returned future with the decoded content of the response, otherwise it fails the future with a Net::Async::Webservice::UPS::Exception::HTTPError instance.
generate_cache_key
Generates a cache key (a string) identifying a request. Two requests with the same cache key should return the same response.
AUTHORS
Gianni Ceccarelli <gianni.ceccarelli@net-a-porter.com>
Sherzod B. Ruzmetov <sherzodr@cpan.org>
COPYRIGHT AND LICENSE
This software is copyright (c) 2014 by Net-a-porter.com.
This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself.