| use crate::f64::consts; |
| use crate::num::FpCategory as Fp; |
| use crate::num::*; |
| |
| #[test] |
| fn test_num_f64() { |
| test_num(10f64, 2f64); |
| } |
| |
| #[test] |
| fn test_min_nan() { |
| assert_eq!(f64::NAN.min(2.0), 2.0); |
| assert_eq!(2.0f64.min(f64::NAN), 2.0); |
| } |
| |
| #[test] |
| fn test_max_nan() { |
| assert_eq!(f64::NAN.max(2.0), 2.0); |
| assert_eq!(2.0f64.max(f64::NAN), 2.0); |
| } |
| |
| #[test] |
| fn test_nan() { |
| let nan: f64 = f64::NAN; |
| assert!(nan.is_nan()); |
| assert!(!nan.is_infinite()); |
| assert!(!nan.is_finite()); |
| assert!(!nan.is_normal()); |
| assert!(nan.is_sign_positive()); |
| assert!(!nan.is_sign_negative()); |
| assert_eq!(Fp::Nan, nan.classify()); |
| } |
| |
| #[test] |
| fn test_infinity() { |
| let inf: f64 = f64::INFINITY; |
| assert!(inf.is_infinite()); |
| assert!(!inf.is_finite()); |
| assert!(inf.is_sign_positive()); |
| assert!(!inf.is_sign_negative()); |
| assert!(!inf.is_nan()); |
| assert!(!inf.is_normal()); |
| assert_eq!(Fp::Infinite, inf.classify()); |
| } |
| |
| #[test] |
| fn test_neg_infinity() { |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| assert!(neg_inf.is_infinite()); |
| assert!(!neg_inf.is_finite()); |
| assert!(!neg_inf.is_sign_positive()); |
| assert!(neg_inf.is_sign_negative()); |
| assert!(!neg_inf.is_nan()); |
| assert!(!neg_inf.is_normal()); |
| assert_eq!(Fp::Infinite, neg_inf.classify()); |
| } |
| |
| #[test] |
| fn test_zero() { |
| let zero: f64 = 0.0f64; |
| assert_eq!(0.0, zero); |
| assert!(!zero.is_infinite()); |
| assert!(zero.is_finite()); |
| assert!(zero.is_sign_positive()); |
| assert!(!zero.is_sign_negative()); |
| assert!(!zero.is_nan()); |
| assert!(!zero.is_normal()); |
| assert_eq!(Fp::Zero, zero.classify()); |
| } |
| |
| #[test] |
| fn test_neg_zero() { |
| let neg_zero: f64 = -0.0; |
| assert_eq!(0.0, neg_zero); |
| assert!(!neg_zero.is_infinite()); |
| assert!(neg_zero.is_finite()); |
| assert!(!neg_zero.is_sign_positive()); |
| assert!(neg_zero.is_sign_negative()); |
| assert!(!neg_zero.is_nan()); |
| assert!(!neg_zero.is_normal()); |
| assert_eq!(Fp::Zero, neg_zero.classify()); |
| } |
| |
| #[cfg_attr(all(target_arch = "wasm32", target_os = "emscripten"), ignore)] // issue 42630 |
| #[test] |
| fn test_one() { |
| let one: f64 = 1.0f64; |
| assert_eq!(1.0, one); |
| assert!(!one.is_infinite()); |
| assert!(one.is_finite()); |
| assert!(one.is_sign_positive()); |
| assert!(!one.is_sign_negative()); |
| assert!(!one.is_nan()); |
| assert!(one.is_normal()); |
| assert_eq!(Fp::Normal, one.classify()); |
| } |
| |
| #[test] |
| fn test_is_nan() { |
| let nan: f64 = f64::NAN; |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| assert!(nan.is_nan()); |
| assert!(!0.0f64.is_nan()); |
| assert!(!5.3f64.is_nan()); |
| assert!(!(-10.732f64).is_nan()); |
| assert!(!inf.is_nan()); |
| assert!(!neg_inf.is_nan()); |
| } |
| |
| #[test] |
| fn test_is_infinite() { |
| let nan: f64 = f64::NAN; |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| assert!(!nan.is_infinite()); |
| assert!(inf.is_infinite()); |
| assert!(neg_inf.is_infinite()); |
| assert!(!0.0f64.is_infinite()); |
| assert!(!42.8f64.is_infinite()); |
| assert!(!(-109.2f64).is_infinite()); |
| } |
| |
| #[test] |
| fn test_is_finite() { |
| let nan: f64 = f64::NAN; |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| assert!(!nan.is_finite()); |
| assert!(!inf.is_finite()); |
| assert!(!neg_inf.is_finite()); |
| assert!(0.0f64.is_finite()); |
| assert!(42.8f64.is_finite()); |
| assert!((-109.2f64).is_finite()); |
| } |
| |
| #[cfg_attr(all(target_arch = "wasm32", target_os = "emscripten"), ignore)] // issue 42630 |
| #[test] |
| fn test_is_normal() { |
| let nan: f64 = f64::NAN; |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| let zero: f64 = 0.0f64; |
| let neg_zero: f64 = -0.0; |
| assert!(!nan.is_normal()); |
| assert!(!inf.is_normal()); |
| assert!(!neg_inf.is_normal()); |
| assert!(!zero.is_normal()); |
| assert!(!neg_zero.is_normal()); |
| assert!(1f64.is_normal()); |
| assert!(1e-307f64.is_normal()); |
| assert!(!1e-308f64.is_normal()); |
| } |
| |
| #[cfg_attr(all(target_arch = "wasm32", target_os = "emscripten"), ignore)] // issue 42630 |
| #[test] |
| fn test_classify() { |
| let nan: f64 = f64::NAN; |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| let zero: f64 = 0.0f64; |
| let neg_zero: f64 = -0.0; |
| assert_eq!(nan.classify(), Fp::Nan); |
| assert_eq!(inf.classify(), Fp::Infinite); |
| assert_eq!(neg_inf.classify(), Fp::Infinite); |
| assert_eq!(zero.classify(), Fp::Zero); |
| assert_eq!(neg_zero.classify(), Fp::Zero); |
| assert_eq!(1e-307f64.classify(), Fp::Normal); |
| assert_eq!(1e-308f64.classify(), Fp::Subnormal); |
| } |
| |
| #[test] |
| fn test_floor() { |
| assert_approx_eq!(1.0f64.floor(), 1.0f64); |
| assert_approx_eq!(1.3f64.floor(), 1.0f64); |
| assert_approx_eq!(1.5f64.floor(), 1.0f64); |
| assert_approx_eq!(1.7f64.floor(), 1.0f64); |
| assert_approx_eq!(0.0f64.floor(), 0.0f64); |
| assert_approx_eq!((-0.0f64).floor(), -0.0f64); |
| assert_approx_eq!((-1.0f64).floor(), -1.0f64); |
| assert_approx_eq!((-1.3f64).floor(), -2.0f64); |
| assert_approx_eq!((-1.5f64).floor(), -2.0f64); |
| assert_approx_eq!((-1.7f64).floor(), -2.0f64); |
| } |
| |
| #[test] |
| fn test_ceil() { |
| assert_approx_eq!(1.0f64.ceil(), 1.0f64); |
| assert_approx_eq!(1.3f64.ceil(), 2.0f64); |
| assert_approx_eq!(1.5f64.ceil(), 2.0f64); |
| assert_approx_eq!(1.7f64.ceil(), 2.0f64); |
| assert_approx_eq!(0.0f64.ceil(), 0.0f64); |
| assert_approx_eq!((-0.0f64).ceil(), -0.0f64); |
| assert_approx_eq!((-1.0f64).ceil(), -1.0f64); |
| assert_approx_eq!((-1.3f64).ceil(), -1.0f64); |
| assert_approx_eq!((-1.5f64).ceil(), -1.0f64); |
| assert_approx_eq!((-1.7f64).ceil(), -1.0f64); |
| } |
| |
| #[test] |
| fn test_round() { |
| assert_approx_eq!(1.0f64.round(), 1.0f64); |
| assert_approx_eq!(1.3f64.round(), 1.0f64); |
| assert_approx_eq!(1.5f64.round(), 2.0f64); |
| assert_approx_eq!(1.7f64.round(), 2.0f64); |
| assert_approx_eq!(0.0f64.round(), 0.0f64); |
| assert_approx_eq!((-0.0f64).round(), -0.0f64); |
| assert_approx_eq!((-1.0f64).round(), -1.0f64); |
| assert_approx_eq!((-1.3f64).round(), -1.0f64); |
| assert_approx_eq!((-1.5f64).round(), -2.0f64); |
| assert_approx_eq!((-1.7f64).round(), -2.0f64); |
| } |
| |
| #[test] |
| fn test_trunc() { |
| assert_approx_eq!(1.0f64.trunc(), 1.0f64); |
| assert_approx_eq!(1.3f64.trunc(), 1.0f64); |
| assert_approx_eq!(1.5f64.trunc(), 1.0f64); |
| assert_approx_eq!(1.7f64.trunc(), 1.0f64); |
| assert_approx_eq!(0.0f64.trunc(), 0.0f64); |
| assert_approx_eq!((-0.0f64).trunc(), -0.0f64); |
| assert_approx_eq!((-1.0f64).trunc(), -1.0f64); |
| assert_approx_eq!((-1.3f64).trunc(), -1.0f64); |
| assert_approx_eq!((-1.5f64).trunc(), -1.0f64); |
| assert_approx_eq!((-1.7f64).trunc(), -1.0f64); |
| } |
| |
| #[test] |
| fn test_fract() { |
| assert_approx_eq!(1.0f64.fract(), 0.0f64); |
| assert_approx_eq!(1.3f64.fract(), 0.3f64); |
| assert_approx_eq!(1.5f64.fract(), 0.5f64); |
| assert_approx_eq!(1.7f64.fract(), 0.7f64); |
| assert_approx_eq!(0.0f64.fract(), 0.0f64); |
| assert_approx_eq!((-0.0f64).fract(), -0.0f64); |
| assert_approx_eq!((-1.0f64).fract(), -0.0f64); |
| assert_approx_eq!((-1.3f64).fract(), -0.3f64); |
| assert_approx_eq!((-1.5f64).fract(), -0.5f64); |
| assert_approx_eq!((-1.7f64).fract(), -0.7f64); |
| } |
| |
| #[test] |
| fn test_abs() { |
| assert_eq!(f64::INFINITY.abs(), f64::INFINITY); |
| assert_eq!(1f64.abs(), 1f64); |
| assert_eq!(0f64.abs(), 0f64); |
| assert_eq!((-0f64).abs(), 0f64); |
| assert_eq!((-1f64).abs(), 1f64); |
| assert_eq!(f64::NEG_INFINITY.abs(), f64::INFINITY); |
| assert_eq!((1f64 / f64::NEG_INFINITY).abs(), 0f64); |
| assert!(f64::NAN.abs().is_nan()); |
| } |
| |
| #[test] |
| fn test_signum() { |
| assert_eq!(f64::INFINITY.signum(), 1f64); |
| assert_eq!(1f64.signum(), 1f64); |
| assert_eq!(0f64.signum(), 1f64); |
| assert_eq!((-0f64).signum(), -1f64); |
| assert_eq!((-1f64).signum(), -1f64); |
| assert_eq!(f64::NEG_INFINITY.signum(), -1f64); |
| assert_eq!((1f64 / f64::NEG_INFINITY).signum(), -1f64); |
| assert!(f64::NAN.signum().is_nan()); |
| } |
| |
| #[test] |
| fn test_is_sign_positive() { |
| assert!(f64::INFINITY.is_sign_positive()); |
| assert!(1f64.is_sign_positive()); |
| assert!(0f64.is_sign_positive()); |
| assert!(!(-0f64).is_sign_positive()); |
| assert!(!(-1f64).is_sign_positive()); |
| assert!(!f64::NEG_INFINITY.is_sign_positive()); |
| assert!(!(1f64 / f64::NEG_INFINITY).is_sign_positive()); |
| assert!(f64::NAN.is_sign_positive()); |
| assert!(!(-f64::NAN).is_sign_positive()); |
| } |
| |
| #[test] |
| fn test_is_sign_negative() { |
| assert!(!f64::INFINITY.is_sign_negative()); |
| assert!(!1f64.is_sign_negative()); |
| assert!(!0f64.is_sign_negative()); |
| assert!((-0f64).is_sign_negative()); |
| assert!((-1f64).is_sign_negative()); |
| assert!(f64::NEG_INFINITY.is_sign_negative()); |
| assert!((1f64 / f64::NEG_INFINITY).is_sign_negative()); |
| assert!(!f64::NAN.is_sign_negative()); |
| assert!((-f64::NAN).is_sign_negative()); |
| } |
| |
| #[test] |
| fn test_mul_add() { |
| let nan: f64 = f64::NAN; |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| assert_approx_eq!(12.3f64.mul_add(4.5, 6.7), 62.05); |
| assert_approx_eq!((-12.3f64).mul_add(-4.5, -6.7), 48.65); |
| assert_approx_eq!(0.0f64.mul_add(8.9, 1.2), 1.2); |
| assert_approx_eq!(3.4f64.mul_add(-0.0, 5.6), 5.6); |
| assert!(nan.mul_add(7.8, 9.0).is_nan()); |
| assert_eq!(inf.mul_add(7.8, 9.0), inf); |
| assert_eq!(neg_inf.mul_add(7.8, 9.0), neg_inf); |
| assert_eq!(8.9f64.mul_add(inf, 3.2), inf); |
| assert_eq!((-3.2f64).mul_add(2.4, neg_inf), neg_inf); |
| } |
| |
| #[test] |
| fn test_recip() { |
| let nan: f64 = f64::NAN; |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| assert_eq!(1.0f64.recip(), 1.0); |
| assert_eq!(2.0f64.recip(), 0.5); |
| assert_eq!((-0.4f64).recip(), -2.5); |
| assert_eq!(0.0f64.recip(), inf); |
| assert!(nan.recip().is_nan()); |
| assert_eq!(inf.recip(), 0.0); |
| assert_eq!(neg_inf.recip(), 0.0); |
| } |
| |
| #[test] |
| fn test_powi() { |
| let nan: f64 = f64::NAN; |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| assert_eq!(1.0f64.powi(1), 1.0); |
| assert_approx_eq!((-3.1f64).powi(2), 9.61); |
| assert_approx_eq!(5.9f64.powi(-2), 0.028727); |
| assert_eq!(8.3f64.powi(0), 1.0); |
| assert!(nan.powi(2).is_nan()); |
| assert_eq!(inf.powi(3), inf); |
| assert_eq!(neg_inf.powi(2), inf); |
| } |
| |
| #[test] |
| fn test_powf() { |
| let nan: f64 = f64::NAN; |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| assert_eq!(1.0f64.powf(1.0), 1.0); |
| assert_approx_eq!(3.4f64.powf(4.5), 246.408183); |
| assert_approx_eq!(2.7f64.powf(-3.2), 0.041652); |
| assert_approx_eq!((-3.1f64).powf(2.0), 9.61); |
| assert_approx_eq!(5.9f64.powf(-2.0), 0.028727); |
| assert_eq!(8.3f64.powf(0.0), 1.0); |
| assert!(nan.powf(2.0).is_nan()); |
| assert_eq!(inf.powf(2.0), inf); |
| assert_eq!(neg_inf.powf(3.0), neg_inf); |
| } |
| |
| #[test] |
| fn test_sqrt_domain() { |
| assert!(f64::NAN.sqrt().is_nan()); |
| assert!(f64::NEG_INFINITY.sqrt().is_nan()); |
| assert!((-1.0f64).sqrt().is_nan()); |
| assert_eq!((-0.0f64).sqrt(), -0.0); |
| assert_eq!(0.0f64.sqrt(), 0.0); |
| assert_eq!(1.0f64.sqrt(), 1.0); |
| assert_eq!(f64::INFINITY.sqrt(), f64::INFINITY); |
| } |
| |
| #[test] |
| fn test_exp() { |
| assert_eq!(1.0, 0.0f64.exp()); |
| assert_approx_eq!(2.718282, 1.0f64.exp()); |
| assert_approx_eq!(148.413159, 5.0f64.exp()); |
| |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| let nan: f64 = f64::NAN; |
| assert_eq!(inf, inf.exp()); |
| assert_eq!(0.0, neg_inf.exp()); |
| assert!(nan.exp().is_nan()); |
| } |
| |
| #[test] |
| fn test_exp2() { |
| assert_eq!(32.0, 5.0f64.exp2()); |
| assert_eq!(1.0, 0.0f64.exp2()); |
| |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| let nan: f64 = f64::NAN; |
| assert_eq!(inf, inf.exp2()); |
| assert_eq!(0.0, neg_inf.exp2()); |
| assert!(nan.exp2().is_nan()); |
| } |
| |
| #[test] |
| fn test_ln() { |
| let nan: f64 = f64::NAN; |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| assert_approx_eq!(1.0f64.exp().ln(), 1.0); |
| assert!(nan.ln().is_nan()); |
| assert_eq!(inf.ln(), inf); |
| assert!(neg_inf.ln().is_nan()); |
| assert!((-2.3f64).ln().is_nan()); |
| assert_eq!((-0.0f64).ln(), neg_inf); |
| assert_eq!(0.0f64.ln(), neg_inf); |
| assert_approx_eq!(4.0f64.ln(), 1.386294); |
| } |
| |
| #[test] |
| fn test_log() { |
| let nan: f64 = f64::NAN; |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| assert_eq!(10.0f64.log(10.0), 1.0); |
| assert_approx_eq!(2.3f64.log(3.5), 0.664858); |
| assert_eq!(1.0f64.exp().log(1.0f64.exp()), 1.0); |
| assert!(1.0f64.log(1.0).is_nan()); |
| assert!(1.0f64.log(-13.9).is_nan()); |
| assert!(nan.log(2.3).is_nan()); |
| assert_eq!(inf.log(10.0), inf); |
| assert!(neg_inf.log(8.8).is_nan()); |
| assert!((-2.3f64).log(0.1).is_nan()); |
| assert_eq!((-0.0f64).log(2.0), neg_inf); |
| assert_eq!(0.0f64.log(7.0), neg_inf); |
| } |
| |
| #[test] |
| fn test_log2() { |
| let nan: f64 = f64::NAN; |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| assert_approx_eq!(10.0f64.log2(), 3.321928); |
| assert_approx_eq!(2.3f64.log2(), 1.201634); |
| assert_approx_eq!(1.0f64.exp().log2(), 1.442695); |
| assert!(nan.log2().is_nan()); |
| assert_eq!(inf.log2(), inf); |
| assert!(neg_inf.log2().is_nan()); |
| assert!((-2.3f64).log2().is_nan()); |
| assert_eq!((-0.0f64).log2(), neg_inf); |
| assert_eq!(0.0f64.log2(), neg_inf); |
| } |
| |
| #[test] |
| fn test_log10() { |
| let nan: f64 = f64::NAN; |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| assert_eq!(10.0f64.log10(), 1.0); |
| assert_approx_eq!(2.3f64.log10(), 0.361728); |
| assert_approx_eq!(1.0f64.exp().log10(), 0.434294); |
| assert_eq!(1.0f64.log10(), 0.0); |
| assert!(nan.log10().is_nan()); |
| assert_eq!(inf.log10(), inf); |
| assert!(neg_inf.log10().is_nan()); |
| assert!((-2.3f64).log10().is_nan()); |
| assert_eq!((-0.0f64).log10(), neg_inf); |
| assert_eq!(0.0f64.log10(), neg_inf); |
| } |
| |
| #[test] |
| fn test_to_degrees() { |
| let pi: f64 = consts::PI; |
| let nan: f64 = f64::NAN; |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| assert_eq!(0.0f64.to_degrees(), 0.0); |
| assert_approx_eq!((-5.8f64).to_degrees(), -332.315521); |
| assert_eq!(pi.to_degrees(), 180.0); |
| assert!(nan.to_degrees().is_nan()); |
| assert_eq!(inf.to_degrees(), inf); |
| assert_eq!(neg_inf.to_degrees(), neg_inf); |
| } |
| |
| #[test] |
| fn test_to_radians() { |
| let pi: f64 = consts::PI; |
| let nan: f64 = f64::NAN; |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| assert_eq!(0.0f64.to_radians(), 0.0); |
| assert_approx_eq!(154.6f64.to_radians(), 2.698279); |
| assert_approx_eq!((-332.31f64).to_radians(), -5.799903); |
| assert_eq!(180.0f64.to_radians(), pi); |
| assert!(nan.to_radians().is_nan()); |
| assert_eq!(inf.to_radians(), inf); |
| assert_eq!(neg_inf.to_radians(), neg_inf); |
| } |
| |
| #[test] |
| fn test_asinh() { |
| assert_eq!(0.0f64.asinh(), 0.0f64); |
| assert_eq!((-0.0f64).asinh(), -0.0f64); |
| |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| let nan: f64 = f64::NAN; |
| assert_eq!(inf.asinh(), inf); |
| assert_eq!(neg_inf.asinh(), neg_inf); |
| assert!(nan.asinh().is_nan()); |
| assert!((-0.0f64).asinh().is_sign_negative()); |
| // issue 63271 |
| assert_approx_eq!(2.0f64.asinh(), 1.443635475178810342493276740273105f64); |
| assert_approx_eq!((-2.0f64).asinh(), -1.443635475178810342493276740273105f64); |
| // regression test for the catastrophic cancellation fixed in 72486 |
| assert_approx_eq!((-67452098.07139316f64).asinh(), -18.72007542627454439398548429400083); |
| } |
| |
| #[test] |
| fn test_acosh() { |
| assert_eq!(1.0f64.acosh(), 0.0f64); |
| assert!(0.999f64.acosh().is_nan()); |
| |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| let nan: f64 = f64::NAN; |
| assert_eq!(inf.acosh(), inf); |
| assert!(neg_inf.acosh().is_nan()); |
| assert!(nan.acosh().is_nan()); |
| assert_approx_eq!(2.0f64.acosh(), 1.31695789692481670862504634730796844f64); |
| assert_approx_eq!(3.0f64.acosh(), 1.76274717403908605046521864995958461f64); |
| } |
| |
| #[test] |
| fn test_atanh() { |
| assert_eq!(0.0f64.atanh(), 0.0f64); |
| assert_eq!((-0.0f64).atanh(), -0.0f64); |
| |
| let inf: f64 = f64::INFINITY; |
| let neg_inf: f64 = f64::NEG_INFINITY; |
| let nan: f64 = f64::NAN; |
| assert_eq!(1.0f64.atanh(), inf); |
| assert_eq!((-1.0f64).atanh(), neg_inf); |
| assert!(2f64.atanh().atanh().is_nan()); |
| assert!((-2f64).atanh().atanh().is_nan()); |
| assert!(inf.atanh().is_nan()); |
| assert!(neg_inf.atanh().is_nan()); |
| assert!(nan.atanh().is_nan()); |
| assert_approx_eq!(0.5f64.atanh(), 0.54930614433405484569762261846126285f64); |
| assert_approx_eq!((-0.5f64).atanh(), -0.54930614433405484569762261846126285f64); |
| } |
| |
| #[test] |
| fn test_real_consts() { |
| use super::consts; |
| let pi: f64 = consts::PI; |
| let frac_pi_2: f64 = consts::FRAC_PI_2; |
| let frac_pi_3: f64 = consts::FRAC_PI_3; |
| let frac_pi_4: f64 = consts::FRAC_PI_4; |
| let frac_pi_6: f64 = consts::FRAC_PI_6; |
| let frac_pi_8: f64 = consts::FRAC_PI_8; |
| let frac_1_pi: f64 = consts::FRAC_1_PI; |
| let frac_2_pi: f64 = consts::FRAC_2_PI; |
| let frac_2_sqrtpi: f64 = consts::FRAC_2_SQRT_PI; |
| let sqrt2: f64 = consts::SQRT_2; |
| let frac_1_sqrt2: f64 = consts::FRAC_1_SQRT_2; |
| let e: f64 = consts::E; |
| let log2_e: f64 = consts::LOG2_E; |
| let log10_e: f64 = consts::LOG10_E; |
| let ln_2: f64 = consts::LN_2; |
| let ln_10: f64 = consts::LN_10; |
| |
| assert_approx_eq!(frac_pi_2, pi / 2f64); |
| assert_approx_eq!(frac_pi_3, pi / 3f64); |
| assert_approx_eq!(frac_pi_4, pi / 4f64); |
| assert_approx_eq!(frac_pi_6, pi / 6f64); |
| assert_approx_eq!(frac_pi_8, pi / 8f64); |
| assert_approx_eq!(frac_1_pi, 1f64 / pi); |
| assert_approx_eq!(frac_2_pi, 2f64 / pi); |
| assert_approx_eq!(frac_2_sqrtpi, 2f64 / pi.sqrt()); |
| assert_approx_eq!(sqrt2, 2f64.sqrt()); |
| assert_approx_eq!(frac_1_sqrt2, 1f64 / 2f64.sqrt()); |
| assert_approx_eq!(log2_e, e.log2()); |
| assert_approx_eq!(log10_e, e.log10()); |
| assert_approx_eq!(ln_2, 2f64.ln()); |
| assert_approx_eq!(ln_10, 10f64.ln()); |
| } |
| |
| #[test] |
| fn test_float_bits_conv() { |
| assert_eq!((1f64).to_bits(), 0x3ff0000000000000); |
| assert_eq!((12.5f64).to_bits(), 0x4029000000000000); |
| assert_eq!((1337f64).to_bits(), 0x4094e40000000000); |
| assert_eq!((-14.25f64).to_bits(), 0xc02c800000000000); |
| assert_approx_eq!(f64::from_bits(0x3ff0000000000000), 1.0); |
| assert_approx_eq!(f64::from_bits(0x4029000000000000), 12.5); |
| assert_approx_eq!(f64::from_bits(0x4094e40000000000), 1337.0); |
| assert_approx_eq!(f64::from_bits(0xc02c800000000000), -14.25); |
| |
| // Check that NaNs roundtrip their bits regardless of signaling-ness |
| // 0xA is 0b1010; 0x5 is 0b0101 -- so these two together clobbers all the mantissa bits |
| let masked_nan1 = f64::NAN.to_bits() ^ 0x000A_AAAA_AAAA_AAAA; |
| let masked_nan2 = f64::NAN.to_bits() ^ 0x0005_5555_5555_5555; |
| assert!(f64::from_bits(masked_nan1).is_nan()); |
| assert!(f64::from_bits(masked_nan2).is_nan()); |
| |
| assert_eq!(f64::from_bits(masked_nan1).to_bits(), masked_nan1); |
| assert_eq!(f64::from_bits(masked_nan2).to_bits(), masked_nan2); |
| } |
| |
| #[test] |
| #[should_panic] |
| fn test_clamp_min_greater_than_max() { |
| let _ = 1.0f64.clamp(3.0, 1.0); |
| } |
| |
| #[test] |
| #[should_panic] |
| fn test_clamp_min_is_nan() { |
| let _ = 1.0f64.clamp(f64::NAN, 1.0); |
| } |
| |
| #[test] |
| #[should_panic] |
| fn test_clamp_max_is_nan() { |
| let _ = 1.0f64.clamp(3.0, f64::NAN); |
| } |
| |
| #[test] |
| fn test_total_cmp() { |
| use core::cmp::Ordering; |
| |
| fn quiet_bit_mask() -> u64 { |
| 1 << (f64::MANTISSA_DIGITS - 2) |
| } |
| |
| fn min_subnorm() -> f64 { |
| f64::MIN_POSITIVE / f64::powf(2.0, f64::MANTISSA_DIGITS as f64 - 1.0) |
| } |
| |
| fn max_subnorm() -> f64 { |
| f64::MIN_POSITIVE - min_subnorm() |
| } |
| |
| fn q_nan() -> f64 { |
| f64::from_bits(f64::NAN.to_bits() | quiet_bit_mask()) |
| } |
| |
| fn s_nan() -> f64 { |
| f64::from_bits((f64::NAN.to_bits() & !quiet_bit_mask()) + 42) |
| } |
| |
| assert_eq!(Ordering::Equal, (-q_nan()).total_cmp(&-q_nan())); |
| assert_eq!(Ordering::Equal, (-s_nan()).total_cmp(&-s_nan())); |
| assert_eq!(Ordering::Equal, (-f64::INFINITY).total_cmp(&-f64::INFINITY)); |
| assert_eq!(Ordering::Equal, (-f64::MAX).total_cmp(&-f64::MAX)); |
| assert_eq!(Ordering::Equal, (-2.5_f64).total_cmp(&-2.5)); |
| assert_eq!(Ordering::Equal, (-1.0_f64).total_cmp(&-1.0)); |
| assert_eq!(Ordering::Equal, (-1.5_f64).total_cmp(&-1.5)); |
| assert_eq!(Ordering::Equal, (-0.5_f64).total_cmp(&-0.5)); |
| assert_eq!(Ordering::Equal, (-f64::MIN_POSITIVE).total_cmp(&-f64::MIN_POSITIVE)); |
| assert_eq!(Ordering::Equal, (-max_subnorm()).total_cmp(&-max_subnorm())); |
| assert_eq!(Ordering::Equal, (-min_subnorm()).total_cmp(&-min_subnorm())); |
| assert_eq!(Ordering::Equal, (-0.0_f64).total_cmp(&-0.0)); |
| assert_eq!(Ordering::Equal, 0.0_f64.total_cmp(&0.0)); |
| assert_eq!(Ordering::Equal, min_subnorm().total_cmp(&min_subnorm())); |
| assert_eq!(Ordering::Equal, max_subnorm().total_cmp(&max_subnorm())); |
| assert_eq!(Ordering::Equal, f64::MIN_POSITIVE.total_cmp(&f64::MIN_POSITIVE)); |
| assert_eq!(Ordering::Equal, 0.5_f64.total_cmp(&0.5)); |
| assert_eq!(Ordering::Equal, 1.0_f64.total_cmp(&1.0)); |
| assert_eq!(Ordering::Equal, 1.5_f64.total_cmp(&1.5)); |
| assert_eq!(Ordering::Equal, 2.5_f64.total_cmp(&2.5)); |
| assert_eq!(Ordering::Equal, f64::MAX.total_cmp(&f64::MAX)); |
| assert_eq!(Ordering::Equal, f64::INFINITY.total_cmp(&f64::INFINITY)); |
| assert_eq!(Ordering::Equal, s_nan().total_cmp(&s_nan())); |
| assert_eq!(Ordering::Equal, q_nan().total_cmp(&q_nan())); |
| |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&-s_nan())); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&-f64::INFINITY)); |
| assert_eq!(Ordering::Less, (-f64::INFINITY).total_cmp(&-f64::MAX)); |
| assert_eq!(Ordering::Less, (-f64::MAX).total_cmp(&-2.5)); |
| assert_eq!(Ordering::Less, (-2.5_f64).total_cmp(&-1.5)); |
| assert_eq!(Ordering::Less, (-1.5_f64).total_cmp(&-1.0)); |
| assert_eq!(Ordering::Less, (-1.0_f64).total_cmp(&-0.5)); |
| assert_eq!(Ordering::Less, (-0.5_f64).total_cmp(&-f64::MIN_POSITIVE)); |
| assert_eq!(Ordering::Less, (-f64::MIN_POSITIVE).total_cmp(&-max_subnorm())); |
| assert_eq!(Ordering::Less, (-max_subnorm()).total_cmp(&-min_subnorm())); |
| assert_eq!(Ordering::Less, (-min_subnorm()).total_cmp(&-0.0)); |
| assert_eq!(Ordering::Less, (-0.0_f64).total_cmp(&0.0)); |
| assert_eq!(Ordering::Less, 0.0_f64.total_cmp(&min_subnorm())); |
| assert_eq!(Ordering::Less, min_subnorm().total_cmp(&max_subnorm())); |
| assert_eq!(Ordering::Less, max_subnorm().total_cmp(&f64::MIN_POSITIVE)); |
| assert_eq!(Ordering::Less, f64::MIN_POSITIVE.total_cmp(&0.5)); |
| assert_eq!(Ordering::Less, 0.5_f64.total_cmp(&1.0)); |
| assert_eq!(Ordering::Less, 1.0_f64.total_cmp(&1.5)); |
| assert_eq!(Ordering::Less, 1.5_f64.total_cmp(&2.5)); |
| assert_eq!(Ordering::Less, 2.5_f64.total_cmp(&f64::MAX)); |
| assert_eq!(Ordering::Less, f64::MAX.total_cmp(&f64::INFINITY)); |
| assert_eq!(Ordering::Less, f64::INFINITY.total_cmp(&s_nan())); |
| assert_eq!(Ordering::Less, s_nan().total_cmp(&q_nan())); |
| |
| assert_eq!(Ordering::Greater, (-s_nan()).total_cmp(&-q_nan())); |
| assert_eq!(Ordering::Greater, (-f64::INFINITY).total_cmp(&-s_nan())); |
| assert_eq!(Ordering::Greater, (-f64::MAX).total_cmp(&-f64::INFINITY)); |
| assert_eq!(Ordering::Greater, (-2.5_f64).total_cmp(&-f64::MAX)); |
| assert_eq!(Ordering::Greater, (-1.5_f64).total_cmp(&-2.5)); |
| assert_eq!(Ordering::Greater, (-1.0_f64).total_cmp(&-1.5)); |
| assert_eq!(Ordering::Greater, (-0.5_f64).total_cmp(&-1.0)); |
| assert_eq!(Ordering::Greater, (-f64::MIN_POSITIVE).total_cmp(&-0.5)); |
| assert_eq!(Ordering::Greater, (-max_subnorm()).total_cmp(&-f64::MIN_POSITIVE)); |
| assert_eq!(Ordering::Greater, (-min_subnorm()).total_cmp(&-max_subnorm())); |
| assert_eq!(Ordering::Greater, (-0.0_f64).total_cmp(&-min_subnorm())); |
| assert_eq!(Ordering::Greater, 0.0_f64.total_cmp(&-0.0)); |
| assert_eq!(Ordering::Greater, min_subnorm().total_cmp(&0.0)); |
| assert_eq!(Ordering::Greater, max_subnorm().total_cmp(&min_subnorm())); |
| assert_eq!(Ordering::Greater, f64::MIN_POSITIVE.total_cmp(&max_subnorm())); |
| assert_eq!(Ordering::Greater, 0.5_f64.total_cmp(&f64::MIN_POSITIVE)); |
| assert_eq!(Ordering::Greater, 1.0_f64.total_cmp(&0.5)); |
| assert_eq!(Ordering::Greater, 1.5_f64.total_cmp(&1.0)); |
| assert_eq!(Ordering::Greater, 2.5_f64.total_cmp(&1.5)); |
| assert_eq!(Ordering::Greater, f64::MAX.total_cmp(&2.5)); |
| assert_eq!(Ordering::Greater, f64::INFINITY.total_cmp(&f64::MAX)); |
| assert_eq!(Ordering::Greater, s_nan().total_cmp(&f64::INFINITY)); |
| assert_eq!(Ordering::Greater, q_nan().total_cmp(&s_nan())); |
| |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&-s_nan())); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&-f64::INFINITY)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&-f64::MAX)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&-2.5)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&-1.5)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&-1.0)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&-0.5)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&-f64::MIN_POSITIVE)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&-max_subnorm())); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&-min_subnorm())); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&-0.0)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&0.0)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&min_subnorm())); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&max_subnorm())); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&f64::MIN_POSITIVE)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&0.5)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&1.0)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&1.5)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&2.5)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&f64::MAX)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&f64::INFINITY)); |
| assert_eq!(Ordering::Less, (-q_nan()).total_cmp(&s_nan())); |
| |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&-f64::INFINITY)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&-f64::MAX)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&-2.5)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&-1.5)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&-1.0)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&-0.5)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&-f64::MIN_POSITIVE)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&-max_subnorm())); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&-min_subnorm())); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&-0.0)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&0.0)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&min_subnorm())); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&max_subnorm())); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&f64::MIN_POSITIVE)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&0.5)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&1.0)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&1.5)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&2.5)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&f64::MAX)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&f64::INFINITY)); |
| assert_eq!(Ordering::Less, (-s_nan()).total_cmp(&s_nan())); |
| } |
| |
| #[test] |
| fn test_lerp_exact() { |
| // simple values |
| assert_eq!(f64::lerp(0.0, 2.0, 4.0), 2.0); |
| assert_eq!(f64::lerp(1.0, 2.0, 4.0), 4.0); |
| |
| // boundary values |
| assert_eq!(f64::lerp(0.0, f64::MIN, f64::MAX), f64::MIN); |
| assert_eq!(f64::lerp(1.0, f64::MIN, f64::MAX), f64::MAX); |
| } |
| |
| #[test] |
| fn test_lerp_consistent() { |
| assert_eq!(f64::lerp(f64::MAX, f64::MIN, f64::MIN), f64::MIN); |
| assert_eq!(f64::lerp(f64::MIN, f64::MAX, f64::MAX), f64::MAX); |
| |
| // as long as t is finite, a/b can be infinite |
| assert_eq!(f64::lerp(f64::MAX, f64::NEG_INFINITY, f64::NEG_INFINITY), f64::NEG_INFINITY); |
| assert_eq!(f64::lerp(f64::MIN, f64::INFINITY, f64::INFINITY), f64::INFINITY); |
| } |
| |
| #[test] |
| fn test_lerp_nan_infinite() { |
| // non-finite t is not NaN if a/b different |
| assert!(!f64::lerp(f64::INFINITY, f64::MIN, f64::MAX).is_nan()); |
| assert!(!f64::lerp(f64::NEG_INFINITY, f64::MIN, f64::MAX).is_nan()); |
| } |
| |
| #[test] |
| fn test_lerp_values() { |
| // just a few basic values |
| assert_eq!(f64::lerp(0.25, 1.0, 2.0), 1.25); |
| assert_eq!(f64::lerp(0.50, 1.0, 2.0), 1.50); |
| assert_eq!(f64::lerp(0.75, 1.0, 2.0), 1.75); |
| } |
| |
| #[test] |
| fn test_lerp_monotonic() { |
| // near 0 |
| let below_zero = f64::lerp(-f64::EPSILON, f64::MIN, f64::MAX); |
| let zero = f64::lerp(0.0, f64::MIN, f64::MAX); |
| let above_zero = f64::lerp(f64::EPSILON, f64::MIN, f64::MAX); |
| assert!(below_zero <= zero); |
| assert!(zero <= above_zero); |
| assert!(below_zero <= above_zero); |
| |
| // near 1 |
| let below_one = f64::lerp(1.0 - f64::EPSILON, f64::MIN, f64::MAX); |
| let one = f64::lerp(1.0, f64::MIN, f64::MAX); |
| let above_one = f64::lerp(1.0 + f64::EPSILON, f64::MIN, f64::MAX); |
| assert!(below_one <= one); |
| assert!(one <= above_one); |
| assert!(below_one <= above_one); |
| } |