bopjesvla

bopjesvla

Elixir and misspelled keyword list options

A lot of functions in the standard library trod along happily if a misspelled keyword list option is passed:

iex(3)> String.split("a1a11a", "1", triim: true)
["a", "a", "", "a"]

Compare to Python, which has keyword arguments baked into the language:

>>> from sklearn.linear_model import LogisticRegression
>>> LogisticRegression(qwerty=5)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: __init__() got an unexpected keyword argument 'qwerty'

The Elixir behavior has bitten me many times. Putting the onus on the developer to check whether options are all valid isn’t working in my opinion, given that these kinds of checks are incredibly rare. One solution is to add a helper function to the standard library:

defaults = [opt: false, another_opt: true, rare_opt: 5]
correct_opts = [opt: true, another_opt: false]

Options.get!(correct_opts, defaults)
# [opt: true, another_opt: false, rare_opt: 5]

misspelled_opts = [blopt: true, another_opt: false]

Options.get!(misspelled_opts, defaults)
# throws, since blopt is not in defaults

This solution is too verbose, I think, since it requires developers to name all options at least once more than they usually would. The reason why this is almost a non-issue in Python is that it would take more effort to allow misspelled keyword arguments. With that in mind, I think the use of macros might be warranted.

def my_fun(positional_arg, opts \\ []) do
  options!(opts, my_opt = false, another_opt = true, rare_opt = 5)
  IO.inspect(rare_opt) # the value passed for opts[:rare_opt]
end

my_fun("pos_arg", my_opt: true)

my_fun("pos_arg", fake_opt: true) # throws

The intended behavior is that of keyword arguments in Python, meaning that the options are accessible as another_opt rather than opts[:another_opt] after the options! macro call.

I don’t necessary like the assignment syntax hijacking, but I do think a solution of this kind is called for.

First 6 of 6 Posts! Switch mode

lpil

lpil

Creator of Gleam

This doesn’t need to be a macro, you could implement a function that takes the desired keywords from the list and throws on any unexpected keywords :slight_smile:

Phillipp

Phillipp

I would find that very annoying during development. Often when playing around, I put a spelling mistake in an optional keyword to temporarily disable it (for fiddling around etc.).

Qqwy

Qqwy

TypeCheck Core Team

I built the library Specify for situations where we want to make explicit what options are supported somewhere (and what defaults are used for them).

As for the behaviour of functions built-in to Elixir: I agree that in many cases it would be preferable for the function to crash rather than to silently ignore an unrecognized keyword. In some cases this currently happens but seemingly not everywhere.

That said, if you were to currently write code like this:

defmodule OptionsExample do
  def foo(normal, arguments) do
    foo(normal, arguments, some_option: true)
  end

  def foo(normal, arguments, some_option: some_option) do
     # ...
     IO.inspect({normal, arguments, some_option})
  end
end

then if we were to call it with OptionsExample.foo(10, 20, unexistent: 42) then we will get a FunctionClauseError that highlights what values were passed as well as which function clauses were attempted:

** (FunctionClauseError) no function clause matching in OptionsExample.foo/3    
    
    The following arguments were given to OptionsExample.foo/3:
    
        # 1
        10
    
        # 2
        20
    
        # 3
        [unexistent: 42]
    
    Attempted function clauses (showing 1 out of 1):
    
        def foo(normal, arguments, [some_option: some_option])

Writing code like this is already a very lightweight way to create the behaviour you are suggesting.

bopjesvla

bopjesvla

That’s only lightweight if a function has 1 option, though, especially since keyword lists are ordered. If a function has 4 options you’d probably need tens of function clauses.

The most radical solution, of course, would be to extend def to deal with parsing options:

  def foo(normal, arguments, !some_option = 5, !other_option = 3) do
     # ...
     IO.inspect({normal, arguments, some_option})
  end

But I think the option! macro I suggested would be sufficient.

Qqwy

Qqwy

TypeCheck Core Team

I stand corrected. :slightly_smiling_face:

slashdotdash

slashdotdash

The NimbleOptions library by Dashbit provides a way to validate Keyword lists by validating the options against a definition. The examples below are from the README.

This library allows you to validate options based on a definition. A definition is a keyword list specifying how the options you want to validate should look like:

definition = [
  connections: [
    type: :non_neg_integer,
    default: 5
  ],
  url: [
    type: :string,
    required: true
  ]
]

Now you can validate options through NimbleOptions.validate/2:

options = [url: "https://example.com"]

NimbleOptions.validate(options, definition)
#=> {:ok, [url: "https://example.com", connections: 5]}

If the options don’t match the definition, an error is returned:

NimbleOptions.validate([connections: 3], schema)
#=> {:error, "required option :url not found, received options: [:connections]"}

Where Next?

Trending in Discussions Top

AstonJ
As the title says, please share what you’ve been up to with Elixir. Whether that’s been learning it, looking into it, making stuff with i...
2976 91332 914
New
f0rest8
Hi everyone :waving_hand: Posting here to showcase and announce that Metamorphic is now officially live on a public-facing domain at htt...
New
byu
@chrismccord : I just saw the Extract AGENTS.md from Phoenix.new into phx.new generator commit to the phoenix project. My initial shotgu...
New
arcanemachine
I was working on an Ecto migration and I needed a timestamp. So, for the nth time, I looked up the different data types for timestamps, a...
New
AstonJ
Just a general thread to post chat/news/info relating to AI/ML stuff that may be relevant for Nx now or in the future. Got anything to sh...
New
type1fool
I just stumbled on a newly redesigned elixir-lang.org. :tada: It looks like @Software_Mansion did the work, and I think it is generally a...
New
juhalehtonen
There has been a thread to discuss the Stack Overflow Developer Survey on this forum every year since 2018, so here’s yet another one for...
New

Other Trending Topics Top

JesseHerrick
Hey, I’m Jesse and I’m the main contributor behind Dexter, a full-featured, lightning-fast Elixir LSP optimized for large codebases. It s...
New
jimsynz
Beam Bots (or just BB for short) is a framework for building fault-tolerant robotics applications in Elixir using familiar OTP patterns. ...
New
Damirados
Hello everyone. After busy few months I am happy to announce v0.1.0 of Emerge &amp; Solve. They are GUI (Emerge) and State management (S...
New
ausimian
Emily is an Elixir library that runs Nx computations on Apple’s MLX. Install it as the default Nx backend and Nx, defn, Axon, Nx.Serving,...
New
akoutmos
@hugobarauna and I (Alex Koutmos) have been hard at work on writing a book on Nerves that takes you from simply blinking LEDs to building...
New
zachdaniel
Introducing AshStorage! Attachment and file management that slots directly into your resources :smiling_face_with_sunglasses: I had hope...
New

We're in Beta

About us Mission Statement