At a glance, functions in Elixir are like ones in imperative languages. But that only seems true. They are different in many ways and that influences your code organization inside them a lot.
The easiest way to define a function is to use the construction def/do/end:
def hello do IO.puts("Hello, World!") end
If a function accepts no arguments, then we can omit brackets.
With arguments we write it the same way most of the languages has it:
defmodule Math do # Write name in snake_case def sum_of_values(a, b) do # Basic sum a + b end end # We can call functions without brackets! IO.puts Math.sum_of_values(5, 8) # => 13
In Elixir there's no instruction
return that might stop code execution. The result of function execution is always the last expression result. If the function is empty then it returns nil which is equal to null in other languages.
Another feature is that we can call a function without brackets. But there is a trick: in nested calls, it might cause confusion and unexpected behavior:
IO.puts IO.puts "huh".
For one-liners Elixir supports a special short declaration syntax:
# Note that we put a comma before _do_ and we omit _end_ def sum_of_values(a, b), do: a + b # or def hello, do: IO.puts("Hello, World!")
Functions defined with
def are available outside of its module. If we want to make a private function we should define it using
defmodule Solution do # We can call the same module functions without specifying its module def hello, do: IO.puts(text_for_hello()) defp text_for_hello, do: "Hello, World" end
print_twice(value) that prints provided value twice
Solution.print_twice("WoW") # => WoW # => WoW