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

use rust_gpu_bridge::prelude::{Vec2, Vec2Swizzles, Vec3};
use type_fields::Field;

use crate::{
    default,
    prelude::{Distance, Normal, Operator, SignedDistanceField, SignedDistanceOperator},
};

#[derive(Debug, Default, Copy, Clone, PartialEq, Eq, PartialOrd, Ord, Hash)]
pub struct NormalizeOp;

impl SignedDistanceOperator<f32, Normal<f32>> for NormalizeOp {
    fn operator<Sdf>(&self, sdf: &Sdf, p: f32) -> Normal<f32>
    where
        Sdf: SignedDistanceField<f32, Normal<f32>>,
    {
        Normal((*sdf.evaluate(p)).signum())
    }
}

impl SignedDistanceOperator<Vec2, Normal<Vec2>> for NormalizeOp {
    fn operator<Sdf>(&self, sdf: &Sdf, p: Vec2) -> Normal<Vec2>
    where
        Sdf: SignedDistanceField<Vec2, Normal<Vec2>>,
    {
        Normal((*sdf.evaluate(p)).normalize())
    }
}

impl SignedDistanceOperator<Vec3, Normal<Vec3>> for NormalizeOp {
    fn operator<Sdf>(&self, sdf: &Sdf, p: Vec3) -> Normal<Vec3>
    where
        Sdf: SignedDistanceField<Vec3, Normal<Vec3>>,
    {
        Normal((*sdf.evaluate(p)).normalize())
    }
}

pub type Normalize<Sdf> = Operator<NormalizeOp, Sdf>;

#[derive(Debug, Default, Copy, Clone, PartialEq, PartialOrd, Field)]
pub struct TetrahedronDerivative<Sdf> {
    pub sdf: Sdf,
    pub epsilon: f32,
}

impl<Sdf> SignedDistanceField<Vec2, Normal<Vec2>> for TetrahedronDerivative<Sdf>
where
    Sdf: SignedDistanceField<Vec2, Distance>,
{
    fn evaluate(&self, p: Vec2) -> Normal<Vec2> {
        let k = Vec2::new(1.0, -1.0);
        (k.xy() * *self.sdf.evaluate(p + k.xy() * self.epsilon)
            + k.yy() * *self.sdf.evaluate(p + k.yy() * self.epsilon)
            + k.yx() * *self.sdf.evaluate(p + k.yx() * self.epsilon)
            + k.xx() * *self.sdf.evaluate(p + k.xx() * self.epsilon))
        .into()
    }
}

impl<Sdf> SignedDistanceField<Vec3, Normal<Vec3>> for TetrahedronDerivative<Sdf>
where
    Sdf: SignedDistanceField<Vec3, Distance>,
{
    fn evaluate(&self, p: Vec3) -> Normal<Vec3> {
        let k = Vec2::new(1.0, -1.0);
        (k.xyy() * *self.sdf.evaluate(p + k.xyy() * self.epsilon)
            + k.yyx() * *self.sdf.evaluate(p + k.yyx() * self.epsilon)
            + k.yxy() * *self.sdf.evaluate(p + k.yxy() * self.epsilon)
            + k.xxx() * *self.sdf.evaluate(p + k.xxx() * self.epsilon))
        .into()
    }
}

pub type TetrahedronNormal<Sdf> = Normalize<TetrahedronDerivative<Sdf>>;

#[derive(Debug, Default, Copy, Clone, PartialEq, PartialOrd, Field)]
pub struct CentralDiffDerivative<Sdf> {
    pub sdf: Sdf,
    pub epsilon: f32,
}

impl<Sdf> SignedDistanceField<f32, Normal<f32>> for CentralDiffDerivative<Sdf>
where
    Sdf: SignedDistanceField<f32, Distance>,
{
    fn evaluate(&self, p: f32) -> Normal<f32> {
        Normal(*self.sdf.evaluate(p + self.epsilon) - *self.sdf.evaluate(p - self.epsilon))
    }
}

impl<Sdf> SignedDistanceField<Vec2, Normal<Vec2>> for CentralDiffDerivative<Sdf>
where
    Sdf: SignedDistanceField<Vec2, Distance>,
{
    fn evaluate(&self, p: Vec2) -> Normal<Vec2> {
        (Vec2::new(
            *self.sdf.evaluate(Vec2::new(p.x + self.epsilon, p.y))
                - *self.sdf.evaluate(Vec2::new(p.x - self.epsilon, p.y)),
            *self.sdf.evaluate(Vec2::new(p.x, p.y + self.epsilon))
                - *self.sdf.evaluate(Vec2::new(p.x, p.y - self.epsilon)),
        ))
        .into()
    }
}

impl<Sdf> SignedDistanceField<Vec3, Normal<Vec3>> for CentralDiffDerivative<Sdf>
where
    Sdf: SignedDistanceField<Vec3, Distance>,
{
    fn evaluate(&self, p: Vec3) -> Normal<Vec3> {
        (Vec3::new(
            *self.sdf.evaluate(Vec3::new(p.x + self.epsilon, p.y, p.z))
                - *self.sdf.evaluate(Vec3::new(p.x - self.epsilon, p.y, p.z)),
            *self.sdf.evaluate(Vec3::new(p.x, p.y + self.epsilon, p.z))
                - *self.sdf.evaluate(Vec3::new(p.x, p.y - self.epsilon, p.z)),
            *self.sdf.evaluate(Vec3::new(p.x, p.y, p.z + self.epsilon))
                - *self.sdf.evaluate(Vec3::new(p.x, p.y, p.z - self.epsilon)),
        ))
        .into()
    }
}

pub type CentralDiffNormal<Sdf> = Normalize<CentralDiffDerivative<Sdf>>;

impl<Sdf> CentralDiffNormal<Sdf> {
    pub fn sdf(&mut self) -> &mut Sdf {
        &mut self.target.sdf
    }

    pub fn epsilon(&mut self) -> &mut f32 {
        &mut self.target.epsilon
    }
}

impl<Sdf> CentralDiffNormal<Sdf> {
    pub fn new(sdf: Sdf, epsilon: f32) -> Self {
        CentralDiffNormal {
            target: CentralDiffDerivative { sdf, epsilon },
            op: default(),
        }
    }
}