File size: 4,257 Bytes
d1ed09d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
"""
Binary graphical composition operators

See https://www.cairographics.org/operators/; more could easily be added from there.
"""

from __future__ import annotations

import numba as nb
import numpy as np
import os

image_operators = ('over', 'add', 'saturate', 'source')
array_operators = ('add_arr', 'max_arr', 'min_arr', 'source_arr')
__all__ = ('composite_op_lookup', 'validate_operator') + image_operators + array_operators


def validate_operator(how, is_image):
    name = how if is_image else how + '_arr'
    if is_image:
        if name not in image_operators:
            raise ValueError('Operator %r not one of the supported image operators: %s'
                            % (how, ', '.join(repr(el) for el in image_operators)))
    elif name not in array_operators:
        raise ValueError('Operator %r not one of the supported array operators: %s'
                        % (how, ', '.join(repr(el[:-4]) for el in array_operators)))


@nb.jit('(uint32,)', nopython=True, nogil=True, cache=True)
def extract_scaled(x):
    """Extract components as float64 values in [0.0, 1.0]"""
    r = np.float64(( x        & 255) / 255)
    g = np.float64(((x >>  8) & 255) / 255)
    b = np.float64(((x >> 16) & 255) / 255)
    a = np.float64(((x >> 24) & 255) / 255)
    return r, g, b, a


@nb.jit('(float64, float64, float64, float64)', nopython=True,
        nogil=True, cache=True)
def combine_scaled(r, g, b, a):
    """Combine components in [0, 1] to rgba uint32"""
    r2 = min(255, np.uint32(r * 255))
    g2 = min(255, np.uint32(g * 255))
    b2 = min(255, np.uint32(b * 255))
    a2 = min(255, np.uint32(a * 255))
    return np.uint32((a2 << 24) | (b2 << 16) | (g2 << 8) | r2)


jit_enabled = os.environ.get('NUMBA_DISABLE_JIT', '0') == '0'


if jit_enabled:
    extract_scaled.disable_compile()
    combine_scaled.disable_compile()

# Lookup table for storing compositing operators by function name
composite_op_lookup = {}


def operator(f):
    """Define and register a new image composite operator"""

    if jit_enabled:
        f2 = nb.vectorize(f)
        f2._compile_for_argtys((nb.types.uint32, nb.types.uint32))
        f2._frozen = True
    else:
        f2 = np.vectorize(f)

    composite_op_lookup[f.__name__] = f2
    return f2


@operator
def source(src, dst):
    if src & 0xff000000:
        return src
    else:
        return dst


@operator
def over(src, dst):
    sr, sg, sb, sa = extract_scaled(src)
    dr, dg, db, da = extract_scaled(dst)

    factor = 1 - sa
    a = sa + da * factor
    if a == 0:
        return np.uint32(0)
    r = (sr * sa + dr * da * factor)/a
    g = (sg * sa + dg * da * factor)/a
    b = (sb * sa + db * da * factor)/a
    return combine_scaled(r, g, b, a)


@operator
def add(src, dst):
    sr, sg, sb, sa = extract_scaled(src)
    dr, dg, db, da = extract_scaled(dst)

    a = min(1, sa + da)
    if a == 0:
        return np.uint32(0)
    r = (sr * sa + dr * da)/a
    g = (sg * sa + dg * da)/a
    b = (sb * sa + db * da)/a
    return combine_scaled(r, g, b, a)


@operator
def saturate(src, dst):
    sr, sg, sb, sa = extract_scaled(src)
    dr, dg, db, da = extract_scaled(dst)

    a = min(1, sa + da)
    if a == 0:
        return np.uint32(0)
    factor = min(sa, 1 - da)
    r = (factor * sr + dr * da)/a
    g = (factor * sg + dg * da)/a
    b = (factor * sb + db * da)/a
    return combine_scaled(r, g, b, a)



def arr_operator(f):
    """Define and register a new array composite operator"""

    if jit_enabled:
        f2 = nb.vectorize(f)
        f2._compile_for_argtys(
           (nb.types.int32, nb.types.int32))
        f2._compile_for_argtys(
           (nb.types.int64, nb.types.int64))
        f2._compile_for_argtys(
            (nb.types.float32, nb.types.float32))
        f2._compile_for_argtys(
            (nb.types.float64, nb.types.float64))
        f2._frozen = True
    else:
        f2 = np.vectorize(f)

    composite_op_lookup[f.__name__] = f2
    return f2


@arr_operator
def source_arr(src, dst):
    if src:
        return src
    else:
        return dst

@arr_operator
def add_arr(src, dst):
    return src + dst

@arr_operator
def max_arr(src, dst):
    return max([src,  dst])

@arr_operator
def min_arr(src, dst):
    return min([src,  dst])