colconv 0.1.0

SIMD-dispatched color-conversion kernels covering the FFmpeg AVPixelFormat space, with a Sink-based API so consumers pick which derived outputs (RGB / Luma / HSV / custom) they want without paying for the ones they don't.
Documentation 
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

//! Scalar reference kernels for the Tier 3 packed YUV 4:2:2 (8-bit)
//! sources: YUYV422 / UYVY422 / YVYU422.
//!
//! All three layouts share a single per-pixel-pair pipeline: extract
//! `Y0, Y1, U, V` from a 4-byte block, run the same Q15 chroma /
//! Y / channel math as the planar 4:2:0 / 4:2:2 path, and store
//! 6 bytes of RGB (or 8 bytes of RGBA when `ALPHA = true`).
//!
//! The two const generics select **byte positions** at compile time:
//! - `Y_LSB = true`  → bytes `[Y0, c0, Y1, c1]` (YUYV / YVYU layout).
//! - `Y_LSB = false` → bytes `[c0, Y0, c1, Y1]` (UYVY layout).
//! - `SWAP_UV = false` → `c0 = U, c1 = V` (YUYV / UYVY layout).
//! - `SWAP_UV = true`  → `c0 = V, c1 = U` (YVYU layout).
//!
//! The fourth corner `<Y_LSB=false, SWAP_UV=true>` would be VYUY422
//! (not in FFmpeg) and is never instantiated.

use super::*;

/// Generic packed YUV 4:2:2 → RGB / RGBA row kernel. Three formats
/// share this template via the `Y_LSB` and `SWAP_UV` const generics.
///
/// `packed.len() >= 2 * width`. `width` must be even. `out` length is
/// `3 * width` bytes for `ALPHA = false`, `4 * width` for `ALPHA = true`.
#[cfg_attr(not(tarpaulin), inline(always))]
pub(crate) fn yuv422_packed_to_rgb_or_rgba_row<const Y_LSB: bool, const SWAP_UV: bool>(
  packed: &[u8],
  out: &mut [u8],
  width: usize,
  matrix: ColorMatrix,
  full_range: bool,
  alpha: bool,
) {
  debug_assert_eq!(width & 1, 0, "packed YUV 4:2:2 requires even width");
  debug_assert!(packed.len() >= width * 2, "packed row too short");

  // Byte positions inside each 4-byte (2-pixel) block:
  // Y_LSB = true  → `[Y0, c0, Y1, c1]`
  // Y_LSB = false → `[c0, Y0, c1, Y1]`
  let (y0_idx, y1_idx, c0_idx, c1_idx) = if Y_LSB { (0, 2, 1, 3) } else { (1, 3, 0, 2) };
  // c0 / c1 are U / V when SWAP_UV = false, V / U when true.
  let (u_idx, v_idx) = if SWAP_UV {
    (c1_idx, c0_idx)
  } else {
    (c0_idx, c1_idx)
  };

  let bpp: usize = if alpha { 4 } else { 3 };
  debug_assert!(out.len() >= width * bpp, "out row too short for {bpp}bpp");

  let coeffs = Coefficients::for_matrix(matrix);
  let (y_off, y_scale, c_scale) = range_params_n::<8, 8>(full_range);

  // Round-to-nearest Q15.
  const RND: i32 = 1 << 14;

  let mut x = 0;
  while x < width {
    let block = (x / 2) * 4;
    let y0 = packed[block + y0_idx] as i32;
    let y1 = packed[block + y1_idx] as i32;
    let u = packed[block + u_idx] as i32;
    let v = packed[block + v_idx] as i32;

    let u_d = ((u - 128) * c_scale + RND) >> 15;
    let v_d = ((v - 128) * c_scale + RND) >> 15;

    let r_chroma = (coeffs.r_u() * u_d + coeffs.r_v() * v_d + RND) >> 15;
    let g_chroma = (coeffs.g_u() * u_d + coeffs.g_v() * v_d + RND) >> 15;
    let b_chroma = (coeffs.b_u() * u_d + coeffs.b_v() * v_d + RND) >> 15;

    // Pixel x.
    let y0_s = ((y0 - y_off) * y_scale + RND) >> 15;
    out[x * bpp] = clamp_u8(y0_s + r_chroma);
    out[x * bpp + 1] = clamp_u8(y0_s + g_chroma);
    out[x * bpp + 2] = clamp_u8(y0_s + b_chroma);
    if alpha {
      out[x * bpp + 3] = 0xFF;
    }

    // Pixel x+1 shares chroma.
    let y1_s = ((y1 - y_off) * y_scale + RND) >> 15;
    out[(x + 1) * bpp] = clamp_u8(y1_s + r_chroma);
    out[(x + 1) * bpp + 1] = clamp_u8(y1_s + g_chroma);
    out[(x + 1) * bpp + 2] = clamp_u8(y1_s + b_chroma);
    if alpha {
      out[(x + 1) * bpp + 3] = 0xFF;
    }

    x += 2;
  }
}

/// Scalar YUYV422 → packed RGB. Byte order `Y0, U0, Y1, V0` per
/// 4-byte / 2-pixel block.
#[cfg_attr(not(tarpaulin), inline(always))]
pub(crate) fn yuyv422_to_rgb_row(
  packed: &[u8],
  rgb_out: &mut [u8],
  width: usize,
  matrix: ColorMatrix,
  full_range: bool,
) {
  yuv422_packed_to_rgb_or_rgba_row::<true, false>(
    packed, rgb_out, width, matrix, full_range, false,
  );
}

/// Scalar YUYV422 → packed RGBA (alpha = `0xFF`).
#[cfg_attr(not(tarpaulin), inline(always))]
pub(crate) fn yuyv422_to_rgba_row(
  packed: &[u8],
  rgba_out: &mut [u8],
  width: usize,
  matrix: ColorMatrix,
  full_range: bool,
) {
  yuv422_packed_to_rgb_or_rgba_row::<true, false>(
    packed, rgba_out, width, matrix, full_range, true,
  );
}

/// Scalar UYVY422 → packed RGB. Byte order `U0, Y0, V0, Y1` per
/// 4-byte / 2-pixel block.
#[cfg_attr(not(tarpaulin), inline(always))]
pub(crate) fn uyvy422_to_rgb_row(
  packed: &[u8],
  rgb_out: &mut [u8],
  width: usize,
  matrix: ColorMatrix,
  full_range: bool,
) {
  yuv422_packed_to_rgb_or_rgba_row::<false, false>(
    packed, rgb_out, width, matrix, full_range, false,
  );
}

/// Scalar UYVY422 → packed RGBA (alpha = `0xFF`).
#[cfg_attr(not(tarpaulin), inline(always))]
pub(crate) fn uyvy422_to_rgba_row(
  packed: &[u8],
  rgba_out: &mut [u8],
  width: usize,
  matrix: ColorMatrix,
  full_range: bool,
) {
  yuv422_packed_to_rgb_or_rgba_row::<false, false>(
    packed, rgba_out, width, matrix, full_range, true,
  );
}

/// Scalar YVYU422 → packed RGB. Byte order `Y0, V0, Y1, U0` per
/// 4-byte / 2-pixel block (UV swapped relative to YUYV).
#[cfg_attr(not(tarpaulin), inline(always))]
pub(crate) fn yvyu422_to_rgb_row(
  packed: &[u8],
  rgb_out: &mut [u8],
  width: usize,
  matrix: ColorMatrix,
  full_range: bool,
) {
  yuv422_packed_to_rgb_or_rgba_row::<true, true>(packed, rgb_out, width, matrix, full_range, false);
}

/// Scalar YVYU422 → packed RGBA (alpha = `0xFF`).
#[cfg_attr(not(tarpaulin), inline(always))]
pub(crate) fn yvyu422_to_rgba_row(
  packed: &[u8],
  rgba_out: &mut [u8],
  width: usize,
  matrix: ColorMatrix,
  full_range: bool,
) {
  yuv422_packed_to_rgb_or_rgba_row::<true, true>(packed, rgba_out, width, matrix, full_range, true);
}

/// Copies the Y bytes from a packed YUV 4:2:2 row directly into a
/// `width`-byte luma plane. Avoids the YUV→RGB pipeline entirely
/// when only luma is needed.
#[cfg_attr(not(tarpaulin), inline(always))]
pub(crate) fn yuv422_packed_to_luma_row<const Y_LSB: bool>(
  packed: &[u8],
  luma_out: &mut [u8],
  width: usize,
) {
  debug_assert_eq!(width & 1, 0, "packed YUV 4:2:2 requires even width");
  debug_assert!(packed.len() >= width * 2, "packed row too short");
  debug_assert!(luma_out.len() >= width, "luma row too short");

  let (y0_idx, y1_idx) = if Y_LSB { (0, 2) } else { (1, 3) };
  let mut x = 0;
  while x < width {
    let block = (x / 2) * 4;
    luma_out[x] = packed[block + y0_idx];
    luma_out[x + 1] = packed[block + y1_idx];
    x += 2;
  }
}

/// Scalar YUYV422 luma extraction.
#[cfg_attr(not(tarpaulin), inline(always))]
pub(crate) fn yuyv422_to_luma_row(packed: &[u8], luma_out: &mut [u8], width: usize) {
  yuv422_packed_to_luma_row::<true>(packed, luma_out, width);
}

/// Scalar UYVY422 luma extraction.
#[cfg_attr(not(tarpaulin), inline(always))]
pub(crate) fn uyvy422_to_luma_row(packed: &[u8], luma_out: &mut [u8], width: usize) {
  yuv422_packed_to_luma_row::<false>(packed, luma_out, width);
}

/// Scalar YVYU422 luma extraction (Y positions identical to YUYV).
#[cfg_attr(not(tarpaulin), inline(always))]
pub(crate) fn yvyu422_to_luma_row(packed: &[u8], luma_out: &mut [u8], width: usize) {
  yuv422_packed_to_luma_row::<true>(packed, luma_out, width);
}

/// Extract Y as u16 (zero-extended) from `Yuyv422` packed `[Y, U, Y, V]` layout.
#[cfg_attr(not(any(feature = "std", feature = "alloc")), allow(dead_code))]
#[cfg_attr(not(tarpaulin), inline(always))]
pub(crate) fn yuyv422_to_luma_u16_row(packed: &[u8], out: &mut [u16], width: usize) {
  debug_assert!(packed.len() >= width * 2, "packed too short");
  debug_assert!(out.len() >= width, "out too short");
  for x in 0..width {
    out[x] = packed[x * 2] as u16;
  }
}

/// Extract Y as u16 from `Uyvy422` packed `[U, Y, V, Y]` layout (Y at offset 1).
#[cfg_attr(not(any(feature = "std", feature = "alloc")), allow(dead_code))]
#[cfg_attr(not(tarpaulin), inline(always))]
pub(crate) fn uyvy422_to_luma_u16_row(packed: &[u8], out: &mut [u16], width: usize) {
  debug_assert!(packed.len() >= width * 2, "packed too short");
  debug_assert!(out.len() >= width, "out too short");
  for x in 0..width {
    out[x] = packed[x * 2 + 1] as u16;
  }
}

/// Extract Y as u16 from `Yvyu422` packed `[Y, V, Y, U]` layout.
#[cfg_attr(not(any(feature = "std", feature = "alloc")), allow(dead_code))]
#[cfg_attr(not(tarpaulin), inline(always))]
pub(crate) fn yvyu422_to_luma_u16_row(packed: &[u8], out: &mut [u16], width: usize) {
  debug_assert!(packed.len() >= width * 2, "packed too short");
  debug_assert!(out.len() >= width, "out too short");
  for x in 0..width {
    out[x] = packed[x * 2] as u16;
  }
}