1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License in the LICENSE-APACHE file or at:
// https://www.apache.org/licenses/LICENSE-2.0
//! Vector types
//!
//! For drawing operations, all dimensions use the `f32` type.
use crate::cast::*;
use crate::dir::Directional;
use crate::geom::{Coord, Offset, Rect, Size};
use std::ops::{Add, AddAssign, Div, Mul, Neg, Sub, SubAssign};
/// Axis-aligned 2D cuboid, specified via two corners `a` and `b`
///
/// Typically it is expected that `a.le(b)`, although this is not required.
#[repr(C)]
#[derive(Clone, Copy, Debug, Default, PartialEq)]
#[cfg_attr(feature = "serde", derive(serde::Serialize, serde::Deserialize))]
pub struct Quad {
pub a: Vec2,
pub b: Vec2,
}
impl Quad {
/// Zero
pub const ZERO: Quad = Quad::from_coords(Vec2::ZERO, Vec2::ZERO);
/// Negative infinity to positive infinity (everything)
pub const INFINITY: Quad = Quad::from_coords(Vec2::NEG_INFINITY, Vec2::INFINITY);
/// Not a Number (NaN)
pub const NAN: Quad = Quad::from_coords(Vec2::NAN, Vec2::NAN);
/// Construct with two coords
#[inline]
pub const fn from_coords(a: Vec2, b: Vec2) -> Self {
Quad { a, b }
}
/// Construct with position and size
#[inline]
pub fn from_pos_and_size(pos: Vec2, size: Vec2) -> Self {
Quad {
a: pos,
b: pos + size,
}
}
/// Get the size
#[inline]
pub fn size(&self) -> Vec2 {
self.b - self.a
}
/// Swizzle coordinates: x from first, y from second point
#[inline]
pub fn ab(&self) -> Vec2 {
Vec2(self.a.0, self.b.1)
}
/// Swizzle coordinates: x from second, y from first point
#[inline]
pub fn ba(&self) -> Vec2 {
Vec2(self.b.0, self.a.1)
}
/// Shrink self in all directions by the given `value`
#[inline]
#[must_use = "method does not modify self but returns a new value"]
pub fn shrink(&self, value: f32) -> Quad {
let a = self.a + value;
let b = self.b - value;
Quad { a, b }
}
/// Grow self in all directions by the given `value`
#[inline]
#[must_use = "method does not modify self but returns a new value"]
pub fn grow(&self, value: f32) -> Quad {
let a = self.a - value;
let b = self.b + value;
Quad { a, b }
}
/// Shrink self in all directions by the given `value`
#[inline]
#[must_use = "method does not modify self but returns a new value"]
pub fn shrink_vec(&self, value: Vec2) -> Quad {
let a = self.a + value;
let b = self.b - value;
Quad { a, b }
}
/// Calculate the intersection of two quads
#[inline]
pub fn intersection(&self, rhs: &Quad) -> Option<Quad> {
let a = Vec2(self.a.0.max(rhs.a.0), self.a.1.max(rhs.a.1));
let x = (self.b.0.min(rhs.b.0) - a.0).max(0.0);
let y = (self.b.1.min(rhs.b.1) - a.1).max(0.0);
if x * y > 0.0 {
Some(Quad::from_pos_and_size(a, Vec2(x, y)))
} else {
None
}
}
}
impl AddAssign<Vec2> for Quad {
#[inline]
fn add_assign(&mut self, rhs: Vec2) {
self.a += rhs;
self.b += rhs;
}
}
impl SubAssign<Vec2> for Quad {
#[inline]
fn sub_assign(&mut self, rhs: Vec2) {
self.a -= rhs;
self.b -= rhs;
}
}
impl Add<Vec2> for Quad {
type Output = Quad;
#[inline]
fn add(mut self, rhs: Vec2) -> Self::Output {
self += rhs;
self
}
}
impl Sub<Vec2> for Quad {
type Output = Quad;
#[inline]
fn sub(mut self, rhs: Vec2) -> Self::Output {
self -= rhs;
self
}
}
impl Conv<Rect> for Quad {
#[inline]
fn try_conv(rect: Rect) -> Result<Self> {
let a = Vec2::try_conv(rect.pos)?;
let b = a + Vec2::try_conv(rect.size)?;
Ok(Quad { a, b })
}
}
/// 2D vector
///
/// Usually used as either a coordinate or a difference of coordinates, but
/// may have some other uses.
///
/// Vectors are partially ordered and support component-wise comparison via
/// methods like `lhs.lt(rhs)`. The `PartialOrd` trait is not implemented since
/// it implements `lhs ≤ rhs` as `lhs < rhs || lhs == rhs` which is wrong for
/// vectors (consider for `lhs = (0, 1), rhs = (1, 0)`).
#[repr(C)]
#[derive(Clone, Copy, Debug, Default, PartialEq)]
#[cfg_attr(feature = "serde", derive(serde::Serialize, serde::Deserialize))]
pub struct Vec2(pub f32, pub f32);
/// 2D vector (double precision)
///
/// Usually used as either a coordinate or a difference of coordinates, but
/// may have some other uses.
///
/// Vectors are partially ordered and support component-wise comparison via
/// methods like `lhs.lt(rhs)`. The `PartialOrd` trait is not implemented since
/// it implements `lhs ≤ rhs` as `lhs < rhs || lhs == rhs` which is wrong for
/// vectors (consider for `lhs = (0, 1), rhs = (1, 0)`).
#[repr(C)]
#[derive(Clone, Copy, Debug, Default, PartialEq)]
#[cfg_attr(feature = "serde", derive(serde::Serialize, serde::Deserialize))]
pub struct DVec2(pub f64, pub f64);
macro_rules! impl_vec2 {
($T:ident, $f:ty) => {
impl $T {
/// Zero
pub const ZERO: $T = $T::splat(0.0);
/// One
pub const ONE: $T = $T::splat(1.0);
/// Negative infinity
pub const NEG_INFINITY: $T = $T::splat(<$f>::NEG_INFINITY);
/// Positive infinity
pub const INFINITY: $T = $T::splat(<$f>::INFINITY);
/// Not a Number (NaN)
pub const NAN: $T = $T::splat(<$f>::NAN);
/// Constructs a new instance with each element initialized to `value`.
#[inline]
pub const fn splat(value: $f) -> Self {
$T(value, value)
}
/// Take the minimum component
#[inline]
pub fn min_comp(self) -> $f {
self.0.min(self.1)
}
/// Take the maximum component
#[inline]
pub fn max_comp(self) -> $f {
self.0.max(self.1)
}
/// Return the minimum, componentwise
#[inline]
#[must_use = "method does not modify self but returns a new value"]
pub fn min(self, other: Self) -> Self {
$T(self.0.min(other.0), self.1.min(other.1))
}
/// Return the maximum, componentwise
#[inline]
#[must_use = "method does not modify self but returns a new value"]
pub fn max(self, other: Self) -> Self {
$T(self.0.max(other.0), self.1.max(other.1))
}
/// Take the absolute value of each component
#[inline]
#[must_use = "method does not modify self but returns a new value"]
pub fn abs(self) -> Self {
$T(self.0.abs(), self.1.abs())
}
/// Take the floor of each component
#[inline]
#[must_use = "method does not modify self but returns a new value"]
pub fn floor(self) -> Self {
$T(self.0.floor(), self.1.floor())
}
/// Take the ceiling of each component
#[inline]
#[must_use = "method does not modify self but returns a new value"]
pub fn ceil(self) -> Self {
$T(self.0.ceil(), self.1.ceil())
}
/// Round each component to the nearest integer
#[inline]
#[must_use = "method does not modify self but returns a new value"]
pub fn round(self) -> Self {
$T(self.0.round(), self.1.round())
}
/// Take the trunc of each component
#[inline]
#[must_use = "method does not modify self but returns a new value"]
pub fn trunc(self) -> Self {
$T(self.0.trunc(), self.1.trunc())
}
/// Take the fract of each component
#[inline]
#[must_use = "method does not modify self but returns a new value"]
pub fn fract(self) -> Self {
$T(self.0.fract(), self.1.fract())
}
/// For each component, return `±1` with the same sign as `self`.
#[inline]
#[must_use = "method does not modify self but returns a new value"]
pub fn sign(self) -> Self {
let one: $f = 1.0;
$T(one.copysign(self.0), one.copysign(self.1))
}
/// True when for all components, `lhs < rhs`
#[inline]
pub fn lt(self, rhs: Self) -> bool {
self.0 < rhs.0 && self.1 < rhs.1
}
/// True when for all components, `lhs ≤ rhs`
#[inline]
pub fn le(self, rhs: Self) -> bool {
self.0 <= rhs.0 && self.1 <= rhs.1
}
/// True when for all components, `lhs ≥ rhs`
#[inline]
pub fn ge(self, rhs: Self) -> bool {
self.0 >= rhs.0 && self.1 >= rhs.1
}
/// True when for all components, `lhs > rhs`
#[inline]
pub fn gt(self, rhs: Self) -> bool {
self.0 > rhs.0 && self.1 > rhs.1
}
/// Multiply two vectors as if they are complex numbers
#[inline]
#[must_use = "method does not modify self but returns a new value"]
pub fn complex_mul(self, rhs: Self) -> Self {
$T(
self.0 * rhs.0 - self.1 * rhs.1,
self.0 * rhs.1 + self.1 * rhs.0,
)
}
/// Divide by a second vector as if they are complex numbers
#[inline]
#[must_use = "method does not modify self but returns a new value"]
pub fn complex_div(self, rhs: Self) -> Self {
self.complex_mul(rhs.complex_inv())
}
/// Take the complex reciprocal
#[inline]
#[must_use = "method does not modify self but returns a new value"]
pub fn complex_inv(self) -> Self {
let ssi = 1.0 / self.sum_square();
$T(self.0 * ssi, -self.1 * ssi)
}
/// Return the sum of the terms
#[inline]
pub fn sum(self) -> $f {
self.0 + self.1
}
/// Return the sum of the square of the terms
#[inline]
pub fn sum_square(self) -> $f {
self.0 * self.0 + self.1 * self.1
}
/// Extract one component, based on a direction
///
/// This merely extracts the horizontal or vertical component.
/// It never negates it, even if the axis is reversed.
#[inline]
pub fn extract<D: Directional>(self, dir: D) -> $f {
match dir.is_vertical() {
false => self.0,
true => self.1,
}
}
}
impl Neg for $T {
type Output = $T;
#[inline]
fn neg(self) -> Self::Output {
$T(-self.0, -self.1)
}
}
impl Add<$T> for $T {
type Output = $T;
#[inline]
fn add(self, rhs: $T) -> Self::Output {
$T(self.0 + rhs.0, self.1 + rhs.1)
}
}
impl Add<$f> for $T {
type Output = $T;
#[inline]
fn add(self, rhs: $f) -> Self::Output {
$T(self.0 + rhs, self.1 + rhs)
}
}
impl AddAssign<$T> for $T {
#[inline]
fn add_assign(&mut self, rhs: $T) {
self.0 += rhs.0;
self.1 += rhs.1;
}
}
impl AddAssign<$f> for $T {
#[inline]
fn add_assign(&mut self, rhs: $f) {
self.0 += rhs;
self.1 += rhs;
}
}
impl Sub<$T> for $T {
type Output = $T;
#[inline]
fn sub(self, rhs: $T) -> Self::Output {
$T(self.0 - rhs.0, self.1 - rhs.1)
}
}
impl Sub<$f> for $T {
type Output = $T;
#[inline]
fn sub(self, rhs: $f) -> Self::Output {
$T(self.0 - rhs, self.1 - rhs)
}
}
impl SubAssign<$T> for $T {
#[inline]
fn sub_assign(&mut self, rhs: $T) {
self.0 -= rhs.0;
self.1 -= rhs.1;
}
}
impl SubAssign<$f> for $T {
#[inline]
fn sub_assign(&mut self, rhs: $f) {
self.0 -= rhs;
self.1 -= rhs;
}
}
impl Mul<$T> for $T {
type Output = $T;
#[inline]
fn mul(self, rhs: $T) -> Self::Output {
$T(self.0 * rhs.0, self.1 * rhs.1)
}
}
impl Mul<$f> for $T {
type Output = $T;
#[inline]
fn mul(self, rhs: $f) -> Self::Output {
$T(self.0 * rhs, self.1 * rhs)
}
}
impl Div<$T> for $T {
type Output = $T;
#[inline]
fn div(self, rhs: $T) -> Self::Output {
$T(self.0 / rhs.0, self.1 / rhs.1)
}
}
impl Div<$f> for $T {
type Output = $T;
#[inline]
fn div(self, rhs: $f) -> Self::Output {
$T(self.0 / rhs, self.1 / rhs)
}
}
impl Div<$T> for $f {
type Output = $T;
#[inline]
fn div(self, rhs: $T) -> Self::Output {
$T(self / rhs.0, self / rhs.1)
}
}
impl From<($f, $f)> for $T {
#[inline]
fn from(arg: ($f, $f)) -> Self {
$T(arg.0, arg.1)
}
}
impl Conv<($f, $f)> for $T {
#[inline]
fn conv(arg: ($f, $f)) -> Self {
$T(arg.0, arg.1)
}
#[inline]
fn try_conv(v: ($f, $f)) -> Result<Self> {
Ok(Self::conv(v))
}
}
impl From<$T> for ($f, $f) {
#[inline]
fn from(v: $T) -> Self {
(v.0, v.1)
}
}
impl Conv<$T> for ($f, $f) {
#[inline]
fn conv(v: $T) -> Self {
(v.0, v.1)
}
#[inline]
fn try_conv(v: $T) -> Result<Self> {
Ok(Self::conv(v))
}
}
};
}
impl From<kas_text::Vec2> for Vec2 {
#[inline]
fn from(size: kas_text::Vec2) -> Self {
Vec2(size.0, size.1)
}
}
impl Conv<kas_text::Vec2> for Vec2 {
#[inline]
fn conv(size: kas_text::Vec2) -> Self {
Vec2(size.0, size.1)
}
#[inline]
fn try_conv(v: kas_text::Vec2) -> Result<Self> {
Ok(Self::conv(v))
}
}
impl From<Vec2> for kas_text::Vec2 {
#[inline]
fn from(size: Vec2) -> kas_text::Vec2 {
kas_text::Vec2(size.0, size.1)
}
}
impl Conv<Vec2> for kas_text::Vec2 {
#[inline]
fn conv(size: Vec2) -> kas_text::Vec2 {
kas_text::Vec2(size.0, size.1)
}
#[inline]
fn try_conv(v: Vec2) -> Result<Self> {
Ok(Self::conv(v))
}
}
impl ConvApprox<DVec2> for Vec2 {
fn try_conv_approx(size: DVec2) -> Result<Vec2> {
Ok(Vec2(size.0.try_cast_approx()?, size.1.try_cast_approx()?))
}
}
impl_vec2!(Vec2, f32);
impl_vec2!(DVec2, f64);
macro_rules! impl_conv_vec2 {
($S:ty, $T:ty) => {
impl Conv<$S> for $T {
#[inline]
fn try_conv(arg: $S) -> Result<Self> {
Ok(Self(arg.0.try_cast()?, arg.1.try_cast()?))
}
}
impl ConvApprox<$T> for $S {
#[inline]
fn try_conv_approx(arg: $T) -> Result<Self> {
Ok(Self(arg.0.try_cast_approx()?, arg.1.try_cast_approx()?))
}
}
impl ConvFloat<$T> for $S {
#[inline]
fn try_conv_trunc(x: $T) -> Result<Self> {
Ok(Self(i32::try_conv_trunc(x.0)?, i32::try_conv_trunc(x.1)?))
}
#[inline]
fn try_conv_nearest(x: $T) -> Result<Self> {
Ok(Self(
i32::try_conv_nearest(x.0)?,
i32::try_conv_nearest(x.1)?,
))
}
#[inline]
fn try_conv_floor(x: $T) -> Result<Self> {
Ok(Self(i32::try_conv_floor(x.0)?, i32::try_conv_floor(x.1)?))
}
#[inline]
fn try_conv_ceil(x: $T) -> Result<Self> {
Ok(Self(i32::try_conv_ceil(x.0)?, i32::try_conv_ceil(x.1)?))
}
}
};
}
impl_conv_vec2!(Coord, Vec2);
impl_conv_vec2!(Size, Vec2);
impl_conv_vec2!(Offset, Vec2);
impl_conv_vec2!(Coord, DVec2);
impl_conv_vec2!(Size, DVec2);
impl_conv_vec2!(Offset, DVec2);
/// 3D vector
///
/// Usually used for a 2D coordinate with a depth value.
#[repr(C)]
#[derive(Clone, Copy, Debug, Default, PartialEq)]
#[cfg_attr(feature = "serde", derive(serde::Serialize, serde::Deserialize))]
pub struct Vec3(pub f32, pub f32, pub f32);
impl Vec3 {
/// Construct from a [`Vec2`] and third value
#[inline]
pub fn from2(v: Vec2, z: f32) -> Self {
Vec3(v.0, v.1, z)
}
}