Easily build composable queries for Elixir.

JsonApiAssert is built and maintained by DockYard, contact us for expert Elixir and Phoenix consulting.

JsonApiAssert is a collection of composable test helpers to ease the pain of testing JSON API payloads.

You can use the functions individually but they are optimally used in a composable fashion with the pipe operator:

import JsonApiAssert

payload
|> assert_data(user1)
|> assert_data(user2)
|> refute_data(user3)
|> assert_relationship(pet1, as: "pets", for: user1)
|> assert_relationship(pet2, as: "pets", for: user2)
|> assert_included(pet1)
|> assert_included(pet2)

The records passed must already be serialized. Read more about serializers below to see how to easily manage this with structs or Ecto models.

If you've tested JSON API payloads before the benefits of this pattern should be obvious. Hundreds of lines of codes can be reduced to just a handful. Brittle tests are now flexible and don't care about inseration / render order.

The assert/refute functions expect json-api serialized maps. You can write these yourself but that can be verbose. Instead it is easier to manage your data as structs or better from Ecto models. Just use our built-in serializers.

import JsonApiAssert.Serializer, only: [serialize: 1]

user = %User{email: "[email protected]"}

user_json = serialize(user)

payload
|> assert_data(user_json)

As a convenience you can import the short-hand s function

import JsonApiAssert.Serializer, only: [s: 1]

payload
|> assert_data(s(user))

If you want to serialize JSON-API payloads from maps instead of structs, the only extra work you must do is provide a type argument

import JsonApiAssert.Serializer, only: [serialize: 2]

user_params = %{email: "[email protected]"}

user_json = serialize(user_params, type: "user")

payload
|> assert_data(user_json)

The built-in serializers should be good enough for most cases. However, they are not a requirement. Feel free to use any serializer you'd like. The final schema just needs to match the json-api resource schema:

%{
  "id" => 1,
  "type" => "author",
  "attributes" => %{
    "first-name" => "Brian",
    "last-name" => "Cardarella"
  }
}

Use a Regex as the value if that value is non-deterministic.

record = %{
  "id" => ~r/\d+/,
  "type" => "author",
  "attributes" => %{
    "first-name" => "Brian",
    "last-name" => "Cardarella"
  }
}

assert_data(payload, record)

A common use-case for this is when you are creating new records. The resulting value for the id is likely unknown and assigned at the run-time of the test suite. Using a Regex will allow you to assert that the id value is present and conforms to a value range and pattern that you expect.

Be careful however, you may get false-positives as the Regex matching can be done upon data as well as datum sets.

• Brian Cardarella

We are very thankful for the many contributors

This library follows Semantic Versioning

Please do! We are always looking to improve this library. Please see our Contribution Guidelines on how to properly submit issues and pull requests.

DockYard, Inc. © 2016

@dockyard

Licensed under the MIT license