bc518174
 王天杨
提交两个项目文件
|
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
|
import { noop } from '../noop.mjs';
/*
Bezier function generator
This has been modified from Gaëtan Renaudeau's BezierEasing
https://github.com/gre/bezier-easing/blob/master/src/index.js
https://github.com/gre/bezier-easing/blob/master/LICENSE
I've removed the newtonRaphsonIterate algo because in benchmarking it
wasn't noticeably faster than binarySubdivision, indeed removing it
usually improved times, depending on the curve.
I also removed the lookup table, as for the added bundle size and loop we're
only cutting ~4 or so subdivision iterations. I bumped the max iterations up
to 12 to compensate and this still tended to be faster for no perceivable
loss in accuracy.
Usage
const easeOut = cubicBezier(.17,.67,.83,.67);
const x = easeOut(0.5); // returns 0.627...
*/
// Returns x(t) given t, x1, and x2, or y(t) given t, y1, and y2.
const calcBezier = (t, a1, a2) => (((1.0 - 3.0 * a2 + 3.0 * a1) * t + (3.0 * a2 - 6.0 * a1)) * t + 3.0 * a1) *
t;
const subdivisionPrecision = 0.0000001;
const subdivisionMaxIterations = 12;
function binarySubdivide(x, lowerBound, upperBound, mX1, mX2) {
let currentX;
let currentT;
let i = 0;
do {
currentT = lowerBound + (upperBound - lowerBound) / 2.0;
currentX = calcBezier(currentT, mX1, mX2) - x;
if (currentX > 0.0) {
upperBound = currentT;
}
else {
lowerBound = currentT;
}
} while (Math.abs(currentX) > subdivisionPrecision &&
++i < subdivisionMaxIterations);
return currentT;
}
function cubicBezier(mX1, mY1, mX2, mY2) {
// If this is a linear gradient, return linear easing
if (mX1 === mY1 && mX2 === mY2)
return noop;
const getTForX = (aX) => binarySubdivide(aX, 0, 1, mX1, mX2);
// If animation is at start/end, return t without easing
return (t) => t === 0 || t === 1 ? t : calcBezier(getTForX(t), mY1, mY2);
}
export { cubicBezier };
//# sourceMappingURL=cubic-bezier.mjs.map
|
