fake_braintree, a Braintree fake

This library is a way to test Braintree code without hitting Braintree's servers. It uses Capybara::Server to intercept all of the calls from Braintree's Ruby library and returns XML that the Braintree library can parse. The whole point is not to hit the Braintree API.

It supports a lot of Braintree methods, but it does not support every single one of them (yet).

Supported API methods

Address

Braintree::Address.create

CreditCard

Braintree::CreditCard.create
Braintree::CreditCard.find
Braintree::CreditCard.sale
Braintree::CreditCard.update

Customer

Braintree::Customer.create (including adding add-ons and discounts)
Braintree::Customer.delete
Braintree::Customer.find
Braintree::Customer.update

Subscription

Braintree::Subscription.cancel
Braintree::Subscription.create
Braintree::Subscription.find
Braintree::Subscription.update
Braintree::Subscription.retry_charge

Transaction

Braintree::Transaction.find
Braintree::Transaction.refund
Braintree::Transaction.sale
Braintree::Transaction.void
Braintree::Transaction.submit_for_settlement

TransparentRedirect

Braintree::TransparentRedirect.confirm (only for creating customers)
Braintree::TransparentRedirect.url

Quick start

Require the library and activate it to start the API server:

require 'fake_braintree'
FakeBraintree.activate!

To run the server on a specific port, pass in the :gateway_port option:

FakeBraintree.activate!(gateway_port: 1234)

FakeBraintree.clear! will clear all data, which you almost certainly want to do before each test.

Full example:

# spec/spec_helper.rb
require 'fake_braintree'
FakeBraintree.activate!

RSpec.configure do |c|
  c.before do
    FakeBraintree.clear!
  end
end

If you're using Cucumber, add this too:

# features/support/env.rb
require 'fake_braintree'
FakeBraintree.activate!

Before do
  FakeBraintree.clear!
end

It is advised to run your tests with js: true (RSpec) or @javascript (Cucumber), so that the requests correctly go through FakeBraintree. You might want to take a look at capybara-webkit.

Don't set the Braintree environment

fake_braintree sets Braintree::Configuration.environment = :development. If your code sets it to anything else (like :sandbox), then fake_braintree won't work.

Credit Cards

credit_card.card_type will always be "FakeBraintree".

Verifying credit cards

To verify every credit card you try to use, call:

FakeBraintree.verify_all_cards!

This will stay "on" until you set

FakeBraintree.verify_all_cards = false

Calling FakeBraintree.clear! will not change this setting. It does very basic verification: it only matches the credit card number against these: http://www.braintreepayments.com/docs/ruby/reference/sandbox and rejects them if they aren't one of the listed numbers.

Declining credit cards

To decline every card you try, call:

FakeBraintree.decline_all_cards!

This will decline all cards until you call

FakeBraintree.clear!

This behavior is different from FakeBraintree.verify_all_cards, which will stay on even when clear! is called.

Note that after decline_all_cards! is set, Braintree will still create customers, but will not be able to charge them (so charging for e.g. a subscription will fail). Setting verify_all_cards!, on the other hand, will prevent creation of customers with bad credit cards - Braintree won't even get to trying to charge them.

Generating transactions

You can generate a transaction using FakeBraintree.generate_transaction. This is for use in testing, e.g. before { user.transaction = FakeBraintree.generate_transaction }.

It takes the following options:

:subscription_id: the ID of the subscription associated with the transaction.
:created_at: when the transaction was created (defaults to Time.now)
:amount: the amount of the transaction
:status: the status of the transaction, e.g. Braintree::Transaction::Status::Failed

Any or all of these can be nil, and in fact are nil by default. You can also call it with no arguments.

Full example:

transaction = FakeBraintree.generate_transaction(
  amount: '20.00',
  status: Braintree::Transaction::Status::Settled,
  subscription_id: 'foobar',
  created_at: Time.now + 60
)

p transaction
# {
#   "status_history" =>
#     [{
#       "timestamp"  => 2011-11-20 12:57:25 -0500,
#       "amount"     => "20.00",
#       "status"     => "settled",
#       "created_at" => 2011-11-20 12:58:25 -0500
#     }],
#   "subscription_id" => "foobar"
# }

Note that the generated transaction is not saved in fake_braintree - the method just gives you a hash.

Adding your own transactions

If you want fake_braintree to be aware of a transaction, you can add it to the FakeBraintree.registry.transactions hash like this:

transaction_id = "something"
example_response = { "id" => transaction_id, "amount" => "10.0", "type" => "credit", "status" => "authorized" }
FakeBraintree.registry.transactions[transaction_id] = example_response

Now you can do Braintree::Transaction.find("something") and it will find that transaction.

Not all of the keys in example_response are necessary, but you'll probably want at least id and amount, depending on the type of response.

FakeBraintree.registry.transactions will be cleared when you call FakeBraintree.clear!.

Running the tests

During tests, debug-level logs will be sent to tmp/braintree_log. This is useful for seeing which URLs Braintree is actually hitting.

Credits

Fake Braintree is maintained and funded by thoughtbot, inc

Thank you to all the contributors!

The names and logos for thoughtbot are trademarks of thoughtbot, inc.

mockra/fake_braintree