Dont check WrongEqualsTypeParameter if the function is topLevel

[BraisGabin]

You can have a function named equals as a top level function. It's strange but safe.

[codecov-io]

Codecov Report

Merging #2027 into master will not change coverage.
The diff coverage is 100%.

@@            Coverage Diff            @@
##             master    #2027   +/-   ##
=========================================
  Coverage     80.68%   80.68%           
- Complexity     1985     1986    +1     
=========================================
  Files           329      329           
  Lines          5591     5591           
  Branches       1021     1021           
=========================================
  Hits           4511     4511           
  Misses          539      539           
  Partials        541      541

Impacted Files	Coverage Δ	Complexity Δ
...osch/detekt/rules/bugs/WrongEqualsTypeParameter.kt	100% <100%> (ø)	8 <4> (+1)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d99550d...ed111e8. Read the comment docs.

[3flex]

I think this rule should instead be changed to instead only equals() functions that are overrides (similar to EqualsAlwaysReturnsTrueOrFalse rule) instead of checking if it's a top level function.

I don't think a top-level equals() function would ever have to fulfil this contract.

[BraisGabin]

I think that maybe can be reformuled to: "any non-top-level function called equals that has only one parameter should have the modifier override". The reason is that otherwise we are overloading the equals function and that's exactly what this rule is about. If you agree with this definition I can rewrite this rule.

[3flex]

What I meant was, the rule would check for a function named "equals" that is an override. If those conditions are true, then the rule would check that the return type is as expected.

If the equals function is not an override then this rule shouldn't flag an issue.

[BraisGabin]

If the equals function is not an override then this rule shouldn't flag an issue.

Yes, it should. Think on this code:

fun main(args: Array<String>) {
    A().equals("")  // the output is "wrong"
}

private class A {

    fun equals(other: String): Boolean {
        println("wrong")
        return !super.equals(other)
    }

    override fun equals(other: Any?): Boolean {
        println("correct")
        return super.equals(other)
    }
}

The first equals function doesn't have the override modifier but it's overloading the equals(Any?). So this can lead to very ugly bugs. This is what this rule is cheking.

[3flex]

Thanks for your patience and contribution!