Tracking Issue for more_float_constants

[ghost]

Feature gate: #![feature(more_float_constants)]

This is a tracking issue for additional constants in the {f32,f64}::consts modules.

ACP: rust-lang/libs-team#119

Public API

// in core::{f32, f64}::consts

/// The golden ratio (φ)
pub const PHI: f64 = 1.618033988749894848204586834365638118_f64;

/// The Euler-Mascheroni constant (γ)
pub const EGAMMA: f64 = 0.577215664901532860606512090082402431_f64;

/// 1/sqrt(π)
pub const FRAC_1_SQRT_PI: f64 = 0.564189583547756286948079451560772586_f64;

/// sqrt(3)
pub const SQRT_3: f64 = 1.732050807568877293527446341505872367_f64;

/// 1/sqrt(3)
pub const FRAC_1_SQRT_3: f64 = 0.577350269189625764509148780501957456_f64;

Steps / History

• Implementation: Add additional float constants #103836
• Final comment period (FCP)¹
• Stabilization PR

Unresolved Questions

• Naming of the Euler-Mascheroni constant (gamma)
• Any additional constants?
  • ln(3)

Footnotes

1. https://std-dev-guide.rust-lang.org/feature-lifecycle/stabilization.html ↩

[leonardo-m]

On constants defined like:

pub const PHI: f64 = 1.618033988749894848204586834365638118_f64;

Clippy raises a "float has excessive precision" using this lint:

https://rust-lang.github.io/rust-clippy/master/index.html#excessive_precision

[ghost]

The lint also triggers for like every other constant in the {f32,f64}::consts modules. It's not necessarily a bad thing, as rustc just silently truncates them.

[torvalamo]

As for the question of additional constants: SQRT_5, which is an element in the proposed PHI and also used for various geometric and trigonometric calculations. PHI probably covers the most common usage of SQRT_5, but it wouldn't hurt to include it as it is relatively common (more so than EGAMMA I would argue). Just my 2 cents.