Source code

Revision control

Copy as Markdown

Other Tools

// Copyright 2014 The Servo Project Developers. See the COPYRIGHT
// file at the top-level directory of this distribution.
//
// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
// option. This file may not be copied, modified, or distributed
// except according to those terms.
//! A one-dimensional length, tagged with its units.
use crate::approxeq::ApproxEq;
use crate::approxord::{max, min};
use crate::num::Zero;
use crate::scale::Scale;
use crate::num::One;
#[cfg(feature = "bytemuck")]
use bytemuck::{Pod, Zeroable};
use core::cmp::Ordering;
use core::fmt;
use core::hash::{Hash, Hasher};
use core::iter::Sum;
use core::marker::PhantomData;
use core::ops::{Add, Div, Mul, Neg, Sub};
use core::ops::{AddAssign, DivAssign, MulAssign, SubAssign};
use num_traits::{NumCast, Saturating};
#[cfg(feature = "serde")]
use serde::{Deserialize, Deserializer, Serialize, Serializer};
/// A one-dimensional distance, with value represented by `T` and unit of measurement `Unit`.
///
/// `T` can be any numeric type, for example a primitive type like `u64` or `f32`.
///
/// `Unit` is not used in the representation of a `Length` value. It is used only at compile time
/// to ensure that a `Length` stored with one unit is converted explicitly before being used in an
/// expression that requires a different unit. It may be a type without values, such as an empty
/// enum.
///
/// You can multiply a `Length` by a `scale::Scale` to convert it from one unit to
/// another. See the [`Scale`] docs for an example.
///
/// [`Scale`]: struct.Scale.html
#[repr(C)]
pub struct Length<T, Unit>(pub T, #[doc(hidden)] pub PhantomData<Unit>);
impl<T: Clone, U> Clone for Length<T, U> {
fn clone(&self) -> Self {
Length(self.0.clone(), PhantomData)
}
}
impl<T: Copy, U> Copy for Length<T, U> {}
#[cfg(feature = "serde")]
impl<'de, T, U> Deserialize<'de> for Length<T, U>
where
T: Deserialize<'de>,
{
fn deserialize<D>(deserializer: D) -> Result<Self, D::Error>
where
D: Deserializer<'de>,
{
Ok(Length(Deserialize::deserialize(deserializer)?, PhantomData))
}
}
#[cfg(feature = "serde")]
impl<T, U> Serialize for Length<T, U>
where
T: Serialize,
{
fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
where
S: Serializer,
{
self.0.serialize(serializer)
}
}
#[cfg(feature = "bytemuck")]
unsafe impl<T: Zeroable, U> Zeroable for Length<T, U> {}
#[cfg(feature = "bytemuck")]
unsafe impl<T: Pod, U: 'static> Pod for Length<T, U> {}
impl<T, U> Length<T, U> {
/// Associate a value with a unit of measure.
#[inline]
pub const fn new(x: T) -> Self {
Length(x, PhantomData)
}
}
impl<T: Clone, U> Length<T, U> {
/// Unpack the underlying value from the wrapper.
pub fn get(self) -> T {
self.0
}
/// Cast the unit
#[inline]
pub fn cast_unit<V>(self) -> Length<T, V> {
Length::new(self.0)
}
/// Linearly interpolate between this length and another length.
///
/// # Example
///
/// ```rust
/// use euclid::default::Length;
///
/// let from = Length::new(0.0);
/// let to = Length::new(8.0);
///
/// assert_eq!(from.lerp(to, -1.0), Length::new(-8.0));
/// assert_eq!(from.lerp(to, 0.0), Length::new( 0.0));
/// assert_eq!(from.lerp(to, 0.5), Length::new( 4.0));
/// assert_eq!(from.lerp(to, 1.0), Length::new( 8.0));
/// assert_eq!(from.lerp(to, 2.0), Length::new(16.0));
/// ```
#[inline]
pub fn lerp(self, other: Self, t: T) -> Self
where
T: One + Sub<Output = T> + Mul<Output = T> + Add<Output = T>,
{
let one_t = T::one() - t.clone();
Length::new(one_t * self.0.clone() + t * other.0)
}
}
impl<T: PartialOrd, U> Length<T, U> {
/// Returns minimum between this length and another length.
#[inline]
pub fn min(self, other: Self) -> Self {
min(self, other)
}
/// Returns maximum between this length and another length.
#[inline]
pub fn max(self, other: Self) -> Self {
max(self, other)
}
}
impl<T: NumCast + Clone, U> Length<T, U> {
/// Cast from one numeric representation to another, preserving the units.
#[inline]
pub fn cast<NewT: NumCast>(self) -> Length<NewT, U> {
self.try_cast().unwrap()
}
/// Fallible cast from one numeric representation to another, preserving the units.
pub fn try_cast<NewT: NumCast>(self) -> Option<Length<NewT, U>> {
NumCast::from(self.0).map(Length::new)
}
}
impl<T: fmt::Debug, U> fmt::Debug for Length<T, U> {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
self.0.fmt(f)
}
}
impl<T: Default, U> Default for Length<T, U> {
#[inline]
fn default() -> Self {
Length::new(Default::default())
}
}
impl<T: Hash, U> Hash for Length<T, U> {
fn hash<H: Hasher>(&self, h: &mut H) {
self.0.hash(h);
}
}
// length + length
impl<T: Add, U> Add for Length<T, U> {
type Output = Length<T::Output, U>;
fn add(self, other: Self) -> Self::Output {
Length::new(self.0 + other.0)
}
}
// length + &length
impl<T: Add + Copy, U> Add<&Self> for Length<T, U> {
type Output = Length<T::Output, U>;
fn add(self, other: &Self) -> Self::Output {
Length::new(self.0 + other.0)
}
}
// length_iter.copied().sum()
impl<T: Add<Output = T> + Zero, U> Sum for Length<T, U> {
fn sum<I: Iterator<Item = Self>>(iter: I) -> Self {
iter.fold(Self::zero(), Add::add)
}
}
// length_iter.sum()
impl<'a, T: 'a + Add<Output = T> + Copy + Zero, U: 'a> Sum<&'a Self> for Length<T, U> {
fn sum<I: Iterator<Item = &'a Self>>(iter: I) -> Self {
iter.fold(Self::zero(), Add::add)
}
}
// length += length
impl<T: AddAssign, U> AddAssign for Length<T, U> {
fn add_assign(&mut self, other: Self) {
self.0 += other.0;
}
}
// length - length
impl<T: Sub, U> Sub for Length<T, U> {
type Output = Length<T::Output, U>;
fn sub(self, other: Length<T, U>) -> Self::Output {
Length::new(self.0 - other.0)
}
}
// length -= length
impl<T: SubAssign, U> SubAssign for Length<T, U> {
fn sub_assign(&mut self, other: Self) {
self.0 -= other.0;
}
}
// Saturating length + length and length - length.
impl<T: Saturating, U> Saturating for Length<T, U> {
fn saturating_add(self, other: Self) -> Self {
Length::new(self.0.saturating_add(other.0))
}
fn saturating_sub(self, other: Self) -> Self {
Length::new(self.0.saturating_sub(other.0))
}
}
// length / length
impl<Src, Dst, T: Div> Div<Length<T, Src>> for Length<T, Dst> {
type Output = Scale<T::Output, Src, Dst>;
#[inline]
fn div(self, other: Length<T, Src>) -> Self::Output {
Scale::new(self.0 / other.0)
}
}
// length * scalar
impl<T: Mul, U> Mul<T> for Length<T, U> {
type Output = Length<T::Output, U>;
#[inline]
fn mul(self, scale: T) -> Self::Output {
Length::new(self.0 * scale)
}
}
// length *= scalar
impl<T: Copy + Mul<T, Output = T>, U> MulAssign<T> for Length<T, U> {
#[inline]
fn mul_assign(&mut self, scale: T) {
*self = *self * scale
}
}
// length / scalar
impl<T: Div, U> Div<T> for Length<T, U> {
type Output = Length<T::Output, U>;
#[inline]
fn div(self, scale: T) -> Self::Output {
Length::new(self.0 / scale)
}
}
// length /= scalar
impl<T: Copy + Div<T, Output = T>, U> DivAssign<T> for Length<T, U> {
#[inline]
fn div_assign(&mut self, scale: T) {
*self = *self / scale
}
}
// length * scaleFactor
impl<Src, Dst, T: Mul> Mul<Scale<T, Src, Dst>> for Length<T, Src> {
type Output = Length<T::Output, Dst>;
#[inline]
fn mul(self, scale: Scale<T, Src, Dst>) -> Self::Output {
Length::new(self.0 * scale.0)
}
}
// length / scaleFactor
impl<Src, Dst, T: Div> Div<Scale<T, Src, Dst>> for Length<T, Dst> {
type Output = Length<T::Output, Src>;
#[inline]
fn div(self, scale: Scale<T, Src, Dst>) -> Self::Output {
Length::new(self.0 / scale.0)
}
}
// -length
impl<U, T: Neg> Neg for Length<T, U> {
type Output = Length<T::Output, U>;
#[inline]
fn neg(self) -> Self::Output {
Length::new(-self.0)
}
}
impl<T: PartialEq, U> PartialEq for Length<T, U> {
fn eq(&self, other: &Self) -> bool {
self.0.eq(&other.0)
}
}
impl<T: PartialOrd, U> PartialOrd for Length<T, U> {
fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
self.0.partial_cmp(&other.0)
}
}
impl<T: Eq, U> Eq for Length<T, U> {}
impl<T: Ord, U> Ord for Length<T, U> {
fn cmp(&self, other: &Self) -> Ordering {
self.0.cmp(&other.0)
}
}
impl<T: Zero, U> Zero for Length<T, U> {
#[inline]
fn zero() -> Self {
Length::new(Zero::zero())
}
}
impl<U, T: ApproxEq<T>> ApproxEq<T> for Length<T, U> {
#[inline]
fn approx_epsilon() -> T {
T::approx_epsilon()
}
#[inline]
fn approx_eq_eps(&self, other: &Length<T, U>, approx_epsilon: &T) -> bool {
self.0.approx_eq_eps(&other.0, approx_epsilon)
}
}
#[cfg(test)]
mod tests {
use super::Length;
use crate::num::Zero;
use crate::scale::Scale;
use core::f32::INFINITY;
use num_traits::Saturating;
enum Inch {}
enum Mm {}
enum Cm {}
enum Second {}
#[cfg(feature = "serde")]
mod serde {
use super::*;
extern crate serde_test;
use self::serde_test::assert_tokens;
use self::serde_test::Token;
#[test]
fn test_length_serde() {
let one_cm: Length<f32, Mm> = Length::new(10.0);
assert_tokens(&one_cm, &[Token::F32(10.0)]);
}
}
#[test]
fn test_clone() {
// A cloned Length is a separate length with the state matching the
// original Length at the point it was cloned.
let mut variable_length: Length<f32, Inch> = Length::new(12.0);
let one_foot = variable_length.clone();
variable_length.0 = 24.0;
assert_eq!(one_foot.get(), 12.0);
assert_eq!(variable_length.get(), 24.0);
}
#[test]
fn test_add() {
let length1: Length<u8, Mm> = Length::new(250);
let length2: Length<u8, Mm> = Length::new(5);
assert_eq!((length1 + length2).get(), 255);
assert_eq!((length1 + &length2).get(), 255);
}
#[test]
fn test_sum() {
type L = Length<f32, Mm>;
let lengths = [L::new(1.0), L::new(2.0), L::new(3.0)];
assert_eq!(lengths.iter().sum::<L>(), L::new(6.0));
}
#[test]
fn test_addassign() {
let one_cm: Length<f32, Mm> = Length::new(10.0);
let mut measurement: Length<f32, Mm> = Length::new(5.0);
measurement += one_cm;
assert_eq!(measurement.get(), 15.0);
}
#[test]
fn test_sub() {
let length1: Length<u8, Mm> = Length::new(250);
let length2: Length<u8, Mm> = Length::new(5);
let result = length1 - length2;
assert_eq!(result.get(), 245);
}
#[test]
fn test_subassign() {
let one_cm: Length<f32, Mm> = Length::new(10.0);
let mut measurement: Length<f32, Mm> = Length::new(5.0);
measurement -= one_cm;
assert_eq!(measurement.get(), -5.0);
}
#[test]
fn test_saturating_add() {
let length1: Length<u8, Mm> = Length::new(250);
let length2: Length<u8, Mm> = Length::new(6);
let result = length1.saturating_add(length2);
assert_eq!(result.get(), 255);
}
#[test]
fn test_saturating_sub() {
let length1: Length<u8, Mm> = Length::new(5);
let length2: Length<u8, Mm> = Length::new(10);
let result = length1.saturating_sub(length2);
assert_eq!(result.get(), 0);
}
#[test]
fn test_division_by_length() {
// Division results in a Scale from denominator units
// to numerator units.
let length: Length<f32, Cm> = Length::new(5.0);
let duration: Length<f32, Second> = Length::new(10.0);
let result = length / duration;
let expected: Scale<f32, Second, Cm> = Scale::new(0.5);
assert_eq!(result, expected);
}
#[test]
fn test_multiplication() {
let length_mm: Length<f32, Mm> = Length::new(10.0);
let cm_per_mm: Scale<f32, Mm, Cm> = Scale::new(0.1);
let result = length_mm * cm_per_mm;
let expected: Length<f32, Cm> = Length::new(1.0);
assert_eq!(result, expected);
}
#[test]
fn test_multiplication_with_scalar() {
let length_mm: Length<f32, Mm> = Length::new(10.0);
let result = length_mm * 2.0;
let expected: Length<f32, Mm> = Length::new(20.0);
assert_eq!(result, expected);
}
#[test]
fn test_multiplication_assignment() {
let mut length: Length<f32, Mm> = Length::new(10.0);
length *= 2.0;
let expected: Length<f32, Mm> = Length::new(20.0);
assert_eq!(length, expected);
}
#[test]
fn test_division_by_scalefactor() {
let length: Length<f32, Cm> = Length::new(5.0);
let cm_per_second: Scale<f32, Second, Cm> = Scale::new(10.0);
let result = length / cm_per_second;
let expected: Length<f32, Second> = Length::new(0.5);
assert_eq!(result, expected);
}
#[test]
fn test_division_by_scalar() {
let length: Length<f32, Cm> = Length::new(5.0);
let result = length / 2.0;
let expected: Length<f32, Cm> = Length::new(2.5);
assert_eq!(result, expected);
}
#[test]
fn test_division_assignment() {
let mut length: Length<f32, Mm> = Length::new(10.0);
length /= 2.0;
let expected: Length<f32, Mm> = Length::new(5.0);
assert_eq!(length, expected);
}
#[test]
fn test_negation() {
let length: Length<f32, Cm> = Length::new(5.0);
let result = -length;
let expected: Length<f32, Cm> = Length::new(-5.0);
assert_eq!(result, expected);
}
#[test]
fn test_cast() {
let length_as_i32: Length<i32, Cm> = Length::new(5);
let result: Length<f32, Cm> = length_as_i32.cast();
let length_as_f32: Length<f32, Cm> = Length::new(5.0);
assert_eq!(result, length_as_f32);
}
#[test]
fn test_equality() {
let length_5_point_0: Length<f32, Cm> = Length::new(5.0);
let length_5_point_1: Length<f32, Cm> = Length::new(5.1);
let length_0_point_1: Length<f32, Cm> = Length::new(0.1);
assert!(length_5_point_0 == length_5_point_1 - length_0_point_1);
assert!(length_5_point_0 != length_5_point_1);
}
#[test]
fn test_order() {
let length_5_point_0: Length<f32, Cm> = Length::new(5.0);
let length_5_point_1: Length<f32, Cm> = Length::new(5.1);
let length_0_point_1: Length<f32, Cm> = Length::new(0.1);
assert!(length_5_point_0 < length_5_point_1);
assert!(length_5_point_0 <= length_5_point_1);
assert!(length_5_point_0 <= length_5_point_1 - length_0_point_1);
assert!(length_5_point_1 > length_5_point_0);
assert!(length_5_point_1 >= length_5_point_0);
assert!(length_5_point_0 >= length_5_point_1 - length_0_point_1);
}
#[test]
fn test_zero_add() {
type LengthCm = Length<f32, Cm>;
let length: LengthCm = Length::new(5.0);
let result = length - LengthCm::zero();
assert_eq!(result, length);
}
#[test]
fn test_zero_division() {
type LengthCm = Length<f32, Cm>;
let length: LengthCm = Length::new(5.0);
let length_zero: LengthCm = Length::zero();
let result = length / length_zero;
let expected: Scale<f32, Cm, Cm> = Scale::new(INFINITY);
assert_eq!(result, expected);
}
}