Provides cache and validation for confluent schemas that are pulled from a registry.

It mixes

• ConfluentSchemaRegistry
• ExJsonSchema
• ETS

To provide fast validation for schemas registered remotely on a registry server, for example confluent.cloud.

def deps do
  [
    {:confluent_schema, "~> 0.1.2"}
  ]
end

On application.ex:

  def start(_type, _args) do
    opts = [
      period: :timer.seconds(10),
      debug: true,
      base_url: "https://foobar.region.aws.confluent.cloud",
      username: "key",
      password: "api secret",
    ]

    children = [
      {ConfluentSchema, opts}
    ]

    supervisor_opts = [strategy: :one_for_one, name: MyApp.Supervisor]
    Supervisor.start_link(children, supervisor_opts)
  end

Then you can use confluent schema registry to validate payloads for a subject:

payload = %{foo: "bar"}
ConfluentSchema.validate(payload, "my-subject")

Check out ConfluentSchema.child_spec/1 for documentation about all options.

On test or dev environment, it is common to not have internet access. In this case, we want to load the schemas from local files.

To achieve this, put a subject_name.json file inside priv/confluent_schema/, or run the mix confluent_schema.download mix task defined in Mix.Tasks.ConfluentSchema.Download:

$ mix confluent_schema.download \
  --app-name my_app \
  --username API_KEY \
  --password API_SECRET \
  --base-url https://foobar.region.aws.confluent.cloud 

And configure your application like this:

  # application.ex

  children = [
    {ConfluentSchema, Application.fetch_env!(:my_app, :confluent_schema)}
  ]

  # config.exs
  config :my_app, :confluent_schema, local: true, app_name: :my_app

Now, when ConfluentSchema starts, it will load the schemas from your app's priv/confluent_schema/ directory.