Tracking Issue for float conversion methods (float_conversions)

[valentynkit]

Feature gate: #![feature(float_conversions)]

This is a tracking issue for the float conversion methods from ACP rust-lang/libs-team#810: cast (float to float), to_int_saturating, to_int_checked, and to_int_strict on f16, f32, f64, and f128.

Public API

// core::convert
pub trait FloatToFloat<Flt> { /* sealed */ }

impl f32 { // and f16, f64, f128
    pub fn cast<Flt>(self) -> Flt where Self: FloatToFloat<Flt>;
    pub fn to_int_saturating<Int>(self) -> Int where Self: FloatToInt<Int>;
    pub fn to_int_checked<Int>(self) -> Option<Int> where Self: FloatToInt<Int>;
    pub fn to_int_strict<Int>(self) -> Int where Self: FloatToInt<Int>;
}

cast is self as Flt. to_int_saturating is self as Int (saturating, NaN to 0, per the libs-api decision). to_int_checked truncates toward zero and returns None on NaN, infinity, or out of range. to_int_strict is to_int_checked().unwrap().

Steps / History

(Remember to update the S-tracking-* label when checking boxes.)

• ACP: ACP: Missing floating point conversions libs-team#810
• Implementation: core: implement float conversion methods #159954
• Final comment period (FCP)
• Stabilization PR

Unresolved Questions

• The ACP also proposes cast from integers to floats. It shares the cast method name on integer types with the integer-to-integer cast from ACP: convenient numeric conversions dependent on overflow-checks libs-team#811, which is being unified with the rest of the integer cast methods in the meta-ACP Meta-ACP: Integer cast methods libs-team#833, so it is left out here until that naming settles.
• to_int_wrapping (float to int) is an open question in the ACP and is not included.