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
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267

//! `MixedSinker<Yuv411p>` integration tests — covers the full output
//! set (rgb / rgba / luma / luma_u16 / hsv) plus Strategy A
//! (RGB + RGBA both attached → run RGB kernel once, fan out via
//! `expand_rgb_to_rgba_row`).
//!
//! 4:1:1 is DV-NTSC legacy with quarter-width chroma. FFmpeg's
//! `AV_PIX_FMT_YUV411P` defines chroma row width as
//! `width.div_ceil(4)`, so non-4-aligned widths are accepted; the
//! scalar tail handles the partial 1..3-pixel final chroma group.

use super::*;

/// Build a solid 4:1:1 frame with the given Y / U / V byte values.
/// Uses contiguous strides (`y_stride = w`, `u_stride = v_stride =
/// w.div_ceil(4)` per FFmpeg's `AV_PIX_FMT_YUV411P` ceil-shift rule).
fn solid_yuv411p_frame(
  width: u32,
  height: u32,
  y: u8,
  u: u8,
  v: u8,
) -> (std::vec::Vec<u8>, std::vec::Vec<u8>, std::vec::Vec<u8>) {
  let w = width as usize;
  let h = height as usize;
  let cw = w.div_ceil(4);
  (
    std::vec![y; w * h],
    std::vec![u; cw * h],
    std::vec![v; cw * h],
  )
}

#[test]
#[cfg_attr(
  miri,
  ignore = "SIMD-dispatched row kernels use intrinsics unsupported by Miri"
)]
fn yuv411p_luma_only_copies_y_plane() {
  let (yp, up, vp) = solid_yuv411p_frame(16, 8, 42, 128, 128);
  let src = Yuv411pFrame::new(&yp, &up, &vp, 16, 8, 16, 4, 4);

  let mut luma = std::vec![0u8; 16 * 8];
  let mut sink = MixedSinker::<Yuv411p>::new(16, 8)
    .with_luma(&mut luma)
    .unwrap();
  yuv411p_to(&src, true, ColorMatrix::Bt601, &mut sink).unwrap();

  assert!(luma.iter().all(|&y| y == 42), "luma should be solid 42");
}

#[test]
#[cfg_attr(
  miri,
  ignore = "SIMD-dispatched row kernels use intrinsics unsupported by Miri"
)]
fn yuv411p_luma_u16_zero_extends_y_plane() {
  let (yp, up, vp) = solid_yuv411p_frame(16, 8, 200, 128, 128);
  let src = Yuv411pFrame::new(&yp, &up, &vp, 16, 8, 16, 4, 4);

  let mut luma_u16 = std::vec![0u16; 16 * 8];
  let mut sink = MixedSinker::<Yuv411p>::new(16, 8)
    .with_luma_u16(&mut luma_u16)
    .unwrap();
  yuv411p_to(&src, true, ColorMatrix::Bt601, &mut sink).unwrap();

  // 8-bit Y zero-extends into u16.
  assert!(luma_u16.iter().all(|&y| y == 200));
}

#[test]
#[cfg_attr(
  miri,
  ignore = "SIMD-dispatched row kernels use intrinsics unsupported by Miri"
)]
fn yuv411p_rgb_only_converts_gray_to_gray() {
  // Neutral chroma → gray RGB; solid Y=128 → ~128 in every RGB byte.
  let (yp, up, vp) = solid_yuv411p_frame(16, 8, 128, 128, 128);
  let src = Yuv411pFrame::new(&yp, &up, &vp, 16, 8, 16, 4, 4);

  let mut rgb = std::vec![0u8; 16 * 8 * 3];
  let mut sink = MixedSinker::<Yuv411p>::new(16, 8)
    .with_rgb(&mut rgb)
    .unwrap();
  yuv411p_to(&src, true, ColorMatrix::Bt601, &mut sink).unwrap();

  for px in rgb.chunks(3) {
    assert!(px[0].abs_diff(128) <= 1);
    assert_eq!(px[0], px[1]);
    assert_eq!(px[1], px[2]);
  }
}

#[test]
#[cfg_attr(
  miri,
  ignore = "SIMD-dispatched row kernels use intrinsics unsupported by Miri"
)]
fn yuv411p_rgba_only_with_opaque_alpha() {
  let (yp, up, vp) = solid_yuv411p_frame(16, 8, 200, 128, 128);
  let src = Yuv411pFrame::new(&yp, &up, &vp, 16, 8, 16, 4, 4);

  let mut rgba = std::vec![0u8; 16 * 8 * 4];
  let mut sink = MixedSinker::<Yuv411p>::new(16, 8)
    .with_rgba(&mut rgba)
    .unwrap();
  yuv411p_to(&src, true, ColorMatrix::Bt601, &mut sink).unwrap();

  for px in rgba.chunks(4) {
    assert!(px[0].abs_diff(200) <= 1);
    assert_eq!(px[3], 0xFF, "alpha must be opaque");
  }
}

#[test]
#[cfg_attr(
  miri,
  ignore = "SIMD-dispatched row kernels use intrinsics unsupported by Miri"
)]
fn yuv411p_strategy_a_rgb_and_rgba_match_byte_for_byte() {
  // Strategy A: when both RGB and RGBA are attached, the RGB kernel
  // runs once and `expand_rgb_to_rgba_row` fans out into RGBA.
  // The first three bytes per pixel must match the dedicated
  // RGB-only output, with alpha = 0xFF.
  let w: u32 = 16;
  let h: u32 = 4;
  let (yp, up, vp) = solid_yuv411p_frame(w, h, 180, 60, 200);
  let src = Yuv411pFrame::new(&yp, &up, &vp, w, h, w, w / 4, w / 4);

  let ws = w as usize;
  let hs = h as usize;

  let mut rgb_only = std::vec![0u8; ws * hs * 3];
  let mut sink_rgb = MixedSinker::<Yuv411p>::new(ws, hs)
    .with_rgb(&mut rgb_only)
    .unwrap();
  yuv411p_to(&src, true, ColorMatrix::Bt601, &mut sink_rgb).unwrap();

  let mut rgb_combo = std::vec![0u8; ws * hs * 3];
  let mut rgba_combo = std::vec![0u8; ws * hs * 4];
  let mut sink_combo = MixedSinker::<Yuv411p>::new(ws, hs)
    .with_rgb(&mut rgb_combo)
    .unwrap()
    .with_rgba(&mut rgba_combo)
    .unwrap();
  yuv411p_to(&src, true, ColorMatrix::Bt601, &mut sink_combo).unwrap();

  assert_eq!(rgb_only, rgb_combo, "RGB-only and combo RGB must match");
  for px in 0..(ws * hs) {
    assert_eq!(rgba_combo[px * 4], rgb_only[px * 3]);
    assert_eq!(rgba_combo[px * 4 + 1], rgb_only[px * 3 + 1]);
    assert_eq!(rgba_combo[px * 4 + 2], rgb_only[px * 3 + 2]);
    assert_eq!(rgba_combo[px * 4 + 3], 0xFF);
  }
}

#[test]
#[cfg_attr(
  miri,
  ignore = "SIMD-dispatched row kernels use intrinsics unsupported by Miri"
)]
fn yuv411p_hsv_only_allocates_scratch_and_produces_gray_hsv() {
  // Neutral gray → H=0, S=0, V=~128. No RGB buffer provided.
  let (yp, up, vp) = solid_yuv411p_frame(16, 8, 128, 128, 128);
  let src = Yuv411pFrame::new(&yp, &up, &vp, 16, 8, 16, 4, 4);

  let mut h = std::vec![0xFFu8; 16 * 8];
  let mut s = std::vec![0xFFu8; 16 * 8];
  let mut v = std::vec![0xFFu8; 16 * 8];
  let mut sink = MixedSinker::<Yuv411p>::new(16, 8)
    .with_hsv(&mut h, &mut s, &mut v)
    .unwrap();
  yuv411p_to(&src, true, ColorMatrix::Bt601, &mut sink).unwrap();

  assert!(h.iter().all(|&b| b == 0));
  assert!(s.iter().all(|&b| b == 0));
  assert!(v.iter().all(|&b| b.abs_diff(128) <= 1));
}

#[test]
fn yuv411p_accepts_non_4_aligned_widths_via_div_ceil() {
  // FFmpeg `AV_PIX_FMT_YUV411P` defines chroma width as
  // `width.div_ceil(4)`. Width=15 → chroma_width=4 (3 full 4-pixel
  // groups + 1 partial 3-pixel group sharing the last chroma sample).
  // Both frame construction and `begin_frame` must accept it.
  let yp = std::vec![0u8; 15 * 4];
  let up = std::vec![128u8; 4 * 4];
  let vp = std::vec![128u8; 4 * 4];
  let frame = Yuv411pFrame::try_new(&yp, &up, &vp, 15, 4, 15, 4, 4)
    .expect("width=15 valid: chroma row carries 4 samples per FFmpeg semantics");
  assert_eq!(frame.width(), 15);

  let mut sink: MixedSinker<'_, Yuv411p> = MixedSinker::new(15usize, 4usize);
  <MixedSinker<'_, Yuv411p> as crate::PixelSink>::begin_frame(&mut sink, 15, 4)
    .expect("begin_frame should accept non-4-aligned width");
}

#[test]
#[cfg_attr(
  miri,
  ignore = "SIMD-dispatched row kernels use intrinsics unsupported by Miri"
)]
fn yuv411p_simd_matches_scalar_with_random_yuv() {
  // End-to-end SIMD vs scalar parity for the sinker pipeline.
  let w: u32 = 64;
  let h: u32 = 8;
  let ws = w as usize;
  let hs = h as usize;

  let mut yp = std::vec![0u8; ws * hs];
  let mut up = std::vec![0u8; (ws / 4) * hs];
  let mut vp = std::vec![0u8; (ws / 4) * hs];
  pseudo_random_u8(&mut yp, 0x1111);
  pseudo_random_u8(&mut up, 0x2222);
  pseudo_random_u8(&mut vp, 0x3333);
  let src = Yuv411pFrame::new(&yp, &up, &vp, w, h, w, w / 4, w / 4);

  for &matrix in &[ColorMatrix::Bt601, ColorMatrix::Bt709, ColorMatrix::YCgCo] {
    for full_range in [true, false] {
      let mut rgb_simd = std::vec![0u8; ws * hs * 3];
      let mut rgb_scalar = std::vec![0u8; ws * hs * 3];

      let mut s_simd = MixedSinker::<Yuv411p>::new(ws, hs)
        .with_rgb(&mut rgb_simd)
        .unwrap();
      yuv411p_to(&src, full_range, matrix, &mut s_simd).unwrap();

      let mut s_scalar = MixedSinker::<Yuv411p>::new(ws, hs)
        .with_rgb(&mut rgb_scalar)
        .unwrap();
      s_scalar.set_simd(false);
      yuv411p_to(&src, full_range, matrix, &mut s_scalar).unwrap();

      assert_eq!(
        rgb_simd, rgb_scalar,
        "SIMD vs scalar diverges at matrix={matrix:?} full_range={full_range}"
      );
    }
  }
}

#[test]
fn yuv411p_luma_u16_buffer_too_short_returns_err() {
  // Regression: validation must measure `buf.len()` in u16 elements,
  // not bytes. A buffer one element short of `width x height` u16s
  // must be rejected — this would have slipped through if the check
  // had compared `buf.len()` (u16 count) against a byte count.
  let mut buf = std::vec![0u16; 16 * 8 - 1];
  let err = MixedSinker::<Yuv411p>::new(16, 8)
    .with_luma_u16(&mut buf)
    .err()
    .unwrap();
  assert_eq!(
    err,
    MixedSinkerError::InsufficientLumaU16Buffer(InsufficientBuffer::new(128, 127))
  );
}

#[test]
fn yuv411p_luma_u16_buffer_exactly_sized_accepts() {
  // Companion to the negative test above: an exactly-sized buffer
  // (`width x height` u16 elements) must be accepted. Pins down the
  // boundary condition so the negative test can't pass for the wrong
  // reason (e.g. an off-by-one in the opposite direction).
  let mut buf = std::vec![0u16; 16 * 8];
  let result = MixedSinker::<Yuv411p>::new(16, 8).with_luma_u16(&mut buf);
  assert!(result.is_ok());
}