Typespecs

Mix.install([
  {:jason, "~> 1.4"},
  {:kino, "~> 0.9", override: true},
  {:youtube, github: "brooklinjazz/youtube"},
  {:hidden_cell, github: "brooklinjazz/hidden_cell"}
])

Navigation

Typespecs

Elixir comes with a notation for declaring types and specifications.

We can use the @spec module attribute to define the signature of a function.

For example, we could add a @spec for the math module’s add/2 function.

@spec defines the function name, then the types for each argument, then the return value type separated by the :: symbol.

defmodule IntegerTypespecExample do
  @spec double(integer()) :: integer()
  def double(my_int) do
    my_int * 2
  end
end

Basic Types

number() is a built-in type. Here are a few other common types you may find useful.

any()
atom()
map()
tuple()
list()
list(type) # a list where the elements are particular type
float()
integer()
number() # both integers and floats
String.t() # the string type

For a full list of built in types you can see the Basic Types section of the Elixir Typespecs documentation.

Your Turn

In the Adder module below, create an @spec for integers with the add/2 function.

defmodule Adder do
  def add(integer1, integer2) do
    integer1 + integer2
  end
end

Combing Types

The | symbol allows us to combine multiple types. For example we can combine integer() and float() in our Number module

defmodule NumberTypespecExample do
  @spec double(integer() | float()) :: integer() | float()
  def double(number) do
    number * 2
  end
end

Custom Types

We can use @type to define custom types. We give our custom type a name then define it’s value after the :: symbol. Once we’ve defined a custom type, we can use it anywhere in our application. This way we don’t have to repeatedly define the same types over and over.

Below, we’ve defined a custom my_number() type for the sake of example. However, in the real-world we should use the built-in number() type instead.

defmodule CustomTypeExample do
  @type my_number() :: integer() | float()

  @spec double(my_number()) :: my_number()
  def double(number) do
    number * 2
  end
end

Your Turn

In the Math module below, define a custom @type input() that can be a list, integer, or string.

defmodule Math do
end

Dialyzer

Dialyzer is a static analysis tool used to provide warnings about your code, such as mismatched types, unreachable code, and other common issues.

To use Dialyzer, we install Dialyxir, which provides conveniences for working with Dialyzer in an Elixir project.

We can add :dialyxir as a dependency in the mix.exs file in any mix project.

 defp deps do
    [
      {:dialyxir, "~> 1.0", only: :dev, runtime: false}
    ]
  end

Make sure to install dependencies.

$ mix deps.get

We can then run Dialyzer by running the following in the command line.

$ mix dialyzer
...
Total errors: 0, Skipped: 0, Unnecessary Skips: 0
done in 0m0.82s

Your Turn

Previously you converted a Math module into a mix project in the ExUnit with Mix section.

Add dialyxir as a dependency to your mix.exs file in the math project.

defp deps do
  [
    {:dialyxir, "~> 1.0", only: :dev, runtime: false}
  ]
end

Add typespecs for Math.add/2 and Math.subtract/2 for lists, strings, and integers. Once complete, if you include a failing example in your Math module like so, you should see a warning in the Visual Studio Code editor.

Math.add(%{}, %{})

Run dialyzer and you should see some errors because the typespec catches that %{} is an invalid input.

$ mix dialyzer

Commit Your Progress

DockYard Academy now recommends you use the latest Release rather than forking or cloning our repository.

Run git status to ensure there are no undesirable changes. Then run the following in your command line from the curriculum folder to commit your progress.

$ git add .
$ git commit -m "finish Typespecs reading"
$ git push

We’re proud to offer our open-source curriculum free of charge for anyone to learn from at their own pace.

We also offer a paid course where you can learn from an instructor alongside a cohort of your peers. We will accept applications for the June-August 2023 cohort soon.

Navigation