defmodule Stripe do
@moduledoc """
A HTTP client for Stripe.
## Configuration
### API Key
You need to set your API key in your application configuration. Typically
this is done in `config/config.exs` or a similar file. For example:
config :stripity_stripe, api_key: "sk_test_abc123456789qwerty"
You can also utilize `System.get_env/1` to retrieve the API key from
an environment variable, but remember that this can cause issues if
you use a release tool like exrm or Distillery.
config :stripity_stripe, api_key: System.get_env("STRIPE_API_KEY")
### Shared Options
Almost all of the requests that can be sent accept the following options:
* `:api_key` - The Stripe API key to use for the request. See
[https://stripe.com/docs/api/authentication](https://stripe.com/docs/api/authentication)
* `:api_version` - The version of the api that is being used, defaults to the
version the library is written for. See [https://stripe.com/docs/api/versioning](https://stripe.com/docs/api/versioning)
* `:connect_account` - The ID of a Stripe Connect account for which the
request should be made, passed through as the "Stripe-Account" header. The
preferred authentication method for Stripe Connect. See
[https://stripe.com/docs/connect/authentication#stripe-account-header](https://stripe.com/docs/connect/authentication#stripe-account-header)
* `:expand` - Takes a list of fields that should be expanded in the response
from Stripe. See [https://stripe.com/docs/api/expanding_objects](https://stripe.com/docs/api/expanding_objects)
* `:idempotency_key` - A string that is passed through as the "Idempotency-Key" header on all POST requests. This is used by Stripe's idempotency layer to manage
duplicate requests to the stripe API. See [https://stripe.com/docs/api/idempotent_requests](https://stripe.com/docs/api/idempotent_requests)
### HTTP Connection Pool
Stripity Stripe is set up to use an HTTP connection pool by default. This
means that it will reuse already opened HTTP connections in order to
minimize the overhead of establishing connections. The pool is directly
supervised by Stripity Stripe. Two configuration options are
available to tune how this pool works: `:timeout` and `:max_connections`.
`:timeout` is the amount of time that a connection will be allowed
to remain open but idle (no data passing over it) before it is closed
and cleaned up. This defaults to 5 seconds.
`:max_connections` is the maximum number of connections that can be
open at any time. This defaults to 10.
Both these settings are located under the `:pool_options` key in
your application configuration:
config :stripity_stripe, :pool_options,
timeout: 5_000,
max_connections: 10
If you prefer, you can also turn pooling off completely using
the `:use_connection_pool` setting:
config :stripity_stripe, use_connection_pool: false
"""
use Application
@type id :: String.t()
@type search_query :: String.t()
@type date_query :: %{
optional(:gt) => timestamp,
optional(:gte) => timestamp,
optional(:lt) => timestamp,
optional(:lte) => timestamp
}
@type integer_query :: %{
optional(:gt) => integer,
optional(:gte) => integer,
optional(:lt) => integer,
optional(:lte) => integer
}
@type options :: Keyword.t()
@type timestamp :: pos_integer
@doc """
Callback for the application
Start the supervision tree including the supervised
HTTP connection pool (if it's being used) when
the VM loads the application pool.
Note that we are taking advantage of the BEAM application
standard in order to start the pool when the application is
started. While we do start a supervisor, the supervisor is only
to comply with the expectations of the BEAM application standard.
It is not given any children to supervise.
"""
@spec start(Application.start_type(), any) :: {:error, any} | {:ok, pid} | {:ok, pid, any}
def start(_start_type, _args) do
import Supervisor.Spec, warn: false
children = Stripe.API.supervisor_children()
opts = [strategy: :one_for_one, name: Stripe.Supervisor]
Supervisor.start_link(children, opts)
end
end