📎 Implement noMagicNumbers - eslint/no-magic-numbers, typescript-eslint/no-magic-numbers

[Conaclos]

Description

Implement eslint/no-magic-numbers and typescript-eslint/no-magic-numbers.

Want to contribute? Lets you know you are interested! We will assign you to the issue to prevent several people to work on the same issue. Don't worry, we can unassign you later if you are no longer interested in the issue! Read our contributing guide and analyzer contributing guide.

See the related discussion for context.

The ESLint rule has many options.
I suggest implementing a single option ignoreIntegers (See the Options section) for the time being.

Here are the exceptions we should bake in:

• ignore assignment to constant-like variables, i.e. const, var, and static class properties

  const A = 0.25;
  const B = 0.25;
  class Foo {
    static C = 0.5;
  }

• Ignore common numbers
  We should always ignore 0/0n/0., and maybe 2/2n, 1/1n, and -1/-1n.

  -1; 0; 1; -1n; 0n; 1n;

  Should we ignore more integers?

• Ignore integer indexing (corresponds setting ESLints' ignoreArrayIndexes and ignoreTypeIndexes to true)

  arr[10];
  type Foo = Bar[10];
  type Baz = Parameters<Foo>[10];

• Ignore integer Enum values (corresponds setting ESLint's ignoreEnums to true)

  enum E {
    A = 3;
    B = 4;
  }

• ignore numeric types (corresponds setting ESLint's ignoreNumericLiteralTypes to true)

  type SmallPrimes = 2 | 3 | 5 | 7 | 11;
  function f(x: 100): 100 { return x }

• Ignore integer binary operations

  const x = 1 << 4;
  const x = a & 0xff;
  const x = a | 0x0f;

• Others?

Options

We could introduce an ignoreIntegers option (turned off by default).
When the option is set to true, all integers values are ignored:

1795;
1795n;

[Conaclos]

@fellipeutaka Could you add a comment on this task. I am not able to assign you to the task if you don't participate here.

[fellipeutaka]

@fellipeutaka Could you add a comment on this task. I am not able to assign you to the task if you don't participate here.

Oh, sorry. I'm ready to contribute to this task :)

[Conaclos]

Oh, sorry. I'm ready to contribute to this task :)

No problem. I was not aware of this limitation on GitHub ^^
I guess this is to avoid assigning a wrong person.

[kerolloz]

Hey folks,

Any updates on this one?

CC: @Conaclos, @fellipeutaka, @ematipico

[ematipico]

@kerolloz do you want to help and implement it?

[kerolloz]

Yup sure, I'd love to.
@ematipico

[fellipeutaka]

Hey folks,

Any updates on this one?

CC: @Conaclos, @fellipeutaka, @ematipico

Hey, I wanted to apologize for the delay in updating this task. After some time using the no-magic-numbers rule, I realized it wasn't the best fit for my workflow, especially when working with the motion library, as it generated too many false positives.

In my experience, the best way to prevent magic numbers in a codebase is through consistent and thorough code reviews. They allow for context-specific discussions and ensure that meaningful constants or descriptive variable names are used where necessary.

To @kerolloz: best of luck with this task! It’s an exciting challenge, and I’m sure your implementation will bring great value to the community. Feel free to reach out if you need any insights or suggestions from my experience with the rule. Looking forward to seeing your work!