Any advantage of using this with Kotlin?

[LifeIsStrange]

I wonder how does this compare to kotlin tailrec keyword implementation.

[Sipkab]

I haven't used Kotlin, but here's what I presume based on the docs for tailrec. (https://kotlinlang.org/docs/reference/functions.html#tail-recursive-functions)

There's no verification. If you apply the optimization provided by this library, you won't get a negative feedback if a method cannot be optimized. Accidental breakage may happen if you're not careful.

---

Kotlin performs weaker analysis. The following is not optimized:

val eps = 1E-10 // "good enough", could be 10^-15
tailrec fun findFixPoint(x: Double = 1.0): Double
{
        if (Math.abs(x - Math.cos(x)) < eps) 
    		return x 
    	else {
    		val r = findFixPoint(Math.cos(x));
    		return r
        }
}

You can see that the variable r is directly returned, however, a warning is issued about the function not being optimized. See this playground link.

This library supports this.

---

For Kotlin:

You cannot use tail recursion when there is more code after the recursive call, and you cannot use it within try/catch/finally blocks.

This is the same for this library. You can use the optimization before and after try-catch blocks, but not inside them.

---

This what I've found based on my understanding of Kotlins tailrec, but there might be more to it.

[LifeIsStrange]

Kotlin performs weaker analysis.

VERY interesting, I wonder if Kotlin compiler developpers could take inspiration / reuse your work or even do a collab with you! Anyway you show a suboptimal case that should be fixed, I will open a Kotlin issue when I find the time