FazBrowse GitHub Viewer
|
Trending
|
URL:
|
Home
Tools:
[Download Repo ZIP]
[View Raw Code]
[Original HTTPS Page]
client-sdk-cpp/src/video_frame.cpp at main · siddimore/client-sdk-cpp · GitHub
siddimore
/
client-sdk-cpp
Public
forked from
livekit/client-sdk-cpp
Notifications
You must be signed in to change notification settings
Fork
0
Star
0
Code
Pull requests
0
Actions
Projects
Security and quality
0
Insights
Additional navigation options
Code
Pull requests
Actions
Projects
Security and quality
Insights
Expand file tree
Breadcrumbs
client-sdk-cpp
/
src
/
video_frame.cpp
Copy path
More file actions
More file actions
Latest commit
History
History
History
370 lines (311 loc) · 11.5 KB
Breadcrumbs
client-sdk-cpp
/
src
/
video_frame.cpp
Copy path
File metadata and controls
370 lines (311 loc) · 11.5 KB
Raw
Copy raw file
Download raw file
Open symbols panel
Edit and raw actions
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
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
/*
* Copyright 2026 LiveKit
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#
include
"
livekit/video_frame.h
"
#
include
<
cstddef
>
#
include
<
cstdint
>
#
include
<
stdexcept
>
#
include
<
vector
>
#
include
"
livekit/ffi_handle.h
"
#
include
"
lk_log.h
"
#
include
"
video_utils.h
"
namespace
livekit
{
namespace
{
//
Compute total buffer size in bytes for (width, height, type).
std::
size_t
computeBufferSize
(
int
width,
int
height, VideoBufferType type) {
if
(width <=
0
|| height <=
0
) {
throw
std::invalid_argument
(
"
VideoFrame: width and height must be positive
"
);
}
const
auto
w =
static_cast
<std::
size_t
>(width);
const
auto
h =
static_cast
<std::
size_t
>(height);
switch
(type) {
case
VideoBufferType::
ARGB
:
case
VideoBufferType::
ABGR
:
case
VideoBufferType::
RGBA
:
case
VideoBufferType::
BGRA
:
//
4 bytes per pixel
return
w * h *
4
;
case
VideoBufferType::
RGB24
:
case
VideoBufferType::
I444
:
//
3 bytes per pixel (RGB24: packed; I444: Y+U+V all full resolution)
return
w * h *
3
;
case
VideoBufferType::
I420
:
//
Y (1 byte) + U (1 byte) + V (1 byte)
case
VideoBufferType::
NV12
: {
//
Y (1 byte) + UV interleaved (2 bytes per chroma sample)
const
std::
size_t
chroma_w = (w +
1
) /
2
;
const
std::
size_t
chroma_h = (h +
1
) /
2
;
return
w * h + chroma_w * chroma_h *
2
;
}
case
VideoBufferType::
I010
: {
//
16 bits per sample in memory
//
Y: 2 bytes per sample, U & V: 2 bytes per sample
const
std::
size_t
chroma_w = (w +
1
) /
2
;
const
std::
size_t
chroma_h = (h +
1
) /
2
;
return
w * h *
2
+ chroma_w * chroma_h *
4
;
}
case
VideoBufferType::
I420A
: {
//
Y full, U & V 2x2, plus alpha full res
const
std::
size_t
chroma_w = (w +
1
) /
2
;
const
std::
size_t
chroma_h = (h +
1
) /
2
;
//
Y + A are full resolution, U + V subsampled
return
w * h *
2
+ chroma_w * chroma_h *
2
;
}
case
VideoBufferType::
I422
: {
//
Y full, U & V subsampled horizontally only
const
std::
size_t
chroma_w = (w +
1
) /
2
;
return
w * h + chroma_w * h *
2
;
}
default
:
throw
std::runtime_error
(
"
VideoFrame: unsupported VideoBufferType
"
);
}
}
//
Compute plane layout for (base_ptr, width, height, type)
std::vector<VideoPlaneInfo>
computePlaneInfos
(
uintptr_t
base,
int
width,
int
height, VideoBufferType type) {
std::vector<VideoPlaneInfo> planes;
if
(!base || width <=
0
|| height <=
0
) {
LK_LOG_WARN
(
"
VideoFrame: invalid planeInfos input (ptr={}, w={}, h={})
"
, base, width, height);
return
planes;
}
const
auto
w =
static_cast
<
uint32_t
>(width);
const
auto
h =
static_cast
<
uint32_t
>(height);
auto
pushPlane = [&](
uintptr_t
ptr,
uint32_t
stride,
uint32_t
size) {
const
VideoPlaneInfo info{ptr, stride, size};
planes.
push_back
(info);
};
switch
(type) {
case
VideoBufferType::
ARGB
:
case
VideoBufferType::
ABGR
:
case
VideoBufferType::
RGBA
:
case
VideoBufferType::
BGRA
: {
const
uint32_t
stride = w *
4
;
const
uint32_t
size = stride * h;
pushPlane
(base, stride, size);
break
;
}
case
VideoBufferType::
RGB24
: {
const
uint32_t
stride = w *
3
;
const
uint32_t
size = stride * h;
pushPlane
(base, stride, size);
break
;
}
case
VideoBufferType::
I420
: {
const
uint32_t
chroma_w = (w +
1
) /
2
;
const
uint32_t
chroma_h = (h +
1
) /
2
;
//
Y
const
uint32_t
y_stride = w;
const
uint32_t
y_size = w * h;
const
uintptr_t
y_ptr = base;
pushPlane
(y_ptr, y_stride, y_size);
//
U
const
uint32_t
u_stride = chroma_w;
const
uint32_t
u_size = chroma_w * chroma_h;
const
uintptr_t
u_ptr = y_ptr + y_size;
pushPlane
(u_ptr, u_stride, u_size);
//
V
const
uint32_t
v_stride = chroma_w;
const
uint32_t
v_size = chroma_w * chroma_h;
const
uintptr_t
v_ptr = u_ptr + u_size;
pushPlane
(v_ptr, v_stride, v_size);
break
;
}
case
VideoBufferType::
I420A
: {
const
uint32_t
chroma_w = (w +
1
) /
2
;
const
uint32_t
chroma_h = (h +
1
) /
2
;
//
Y
const
uint32_t
y_stride = w;
const
uint32_t
y_size = w * h;
const
uintptr_t
y_ptr = base;
pushPlane
(y_ptr, y_stride, y_size);
//
U
const
uint32_t
u_stride = chroma_w;
const
uint32_t
u_size = chroma_w * chroma_h;
const
uintptr_t
u_ptr = y_ptr + y_size;
pushPlane
(u_ptr, u_stride, u_size);
//
V
const
uint32_t
v_stride = chroma_w;
const
uint32_t
v_size = chroma_w * chroma_h;
const
uintptr_t
v_ptr = u_ptr + u_size;
pushPlane
(v_ptr, v_stride, v_size);
//
A (full res)
const
uint32_t
a_stride = w;
const
uint32_t
a_size = w * h;
const
uintptr_t
a_ptr = v_ptr + v_size;
pushPlane
(a_ptr, a_stride, a_size);
break
;
}
case
VideoBufferType::
I422
: {
const
uint32_t
chroma_w = (w +
1
) /
2
;
//
Y
const
uint32_t
y_stride = w;
const
uint32_t
y_size = w * h;
const
uintptr_t
y_ptr = base;
pushPlane
(y_ptr, y_stride, y_size);
//
U
const
uint32_t
u_stride = chroma_w;
const
uint32_t
u_size = chroma_w * h;
const
uintptr_t
u_ptr = y_ptr + y_size;
pushPlane
(u_ptr, u_stride, u_size);
//
V
const
uint32_t
v_stride = chroma_w;
const
uint32_t
v_size = chroma_w * h;
const
uintptr_t
v_ptr = u_ptr + u_size;
pushPlane
(v_ptr, v_stride, v_size);
break
;
}
case
VideoBufferType::
I444
: {
//
All planes full-res
const
uint32_t
y_stride = w;
const
uint32_t
y_size = w * h;
const
uintptr_t
y_ptr = base;
pushPlane
(y_ptr, y_stride, y_size);
const
uint32_t
u_stride = w;
const
uint32_t
u_size = w * h;
const
uintptr_t
u_ptr = y_ptr + y_size;
pushPlane
(u_ptr, u_stride, u_size);
const
uint32_t
v_stride = w;
const
uint32_t
v_size = w * h;
const
uintptr_t
v_ptr = u_ptr + u_size;
pushPlane
(v_ptr, v_stride, v_size);
break
;
}
case
VideoBufferType::
I010
: {
//
16-bit per sample
const
uint32_t
chroma_w = (w +
1
) /
2
;
const
uint32_t
chroma_h = (h +
1
) /
2
;
//
Y
const
uint32_t
y_stride = w *
2
;
const
uint32_t
y_size = w * h *
2
;
const
uintptr_t
y_ptr = base;
pushPlane
(y_ptr, y_stride, y_size);
//
U
const
uint32_t
u_stride = chroma_w *
2
;
const
uint32_t
u_size = chroma_w * chroma_h *
2
;
const
uintptr_t
u_ptr = y_ptr + y_size;
pushPlane
(u_ptr, u_stride, u_size);
//
V
const
uint32_t
v_stride = chroma_w *
2
;
const
uint32_t
v_size = chroma_w * chroma_h *
2
;
const
uintptr_t
v_ptr = u_ptr + u_size;
pushPlane
(v_ptr, v_stride, v_size);
break
;
}
case
VideoBufferType::
NV12
: {
const
uint32_t
chroma_w = (w +
1
) /
2
;
const
uint32_t
chroma_h = (h +
1
) /
2
;
//
Y
const
uint32_t
y_stride = w;
const
uint32_t
y_size = w * h;
const
uintptr_t
y_ptr = base;
pushPlane
(y_ptr, y_stride, y_size);
//
UV interleaved
const
uint32_t
uv_stride = chroma_w *
2
;
const
uint32_t
uv_size = chroma_w * chroma_h *
2
;
const
uintptr_t
uv_ptr = y_ptr + y_size;
pushPlane
(uv_ptr, uv_stride, uv_size);
break
;
}
default
:
//
Unknown or unsupported -> no planes
break
;
}
return
planes;
}
}
//
namespace
//
----------------------------------------------------------------------------
//
VideoFrame implementation
//
----------------------------------------------------------------------------
VideoFrame::VideoFrame
() : width_{
0
}, height_{
0
}, type_{VideoBufferType::
BGRA
}, data_{} {}
VideoFrame::VideoFrame
(
int
width,
int
height, VideoBufferType type, std::vector<std::
uint8_t
> data)
: width_(width), height_(height), type_(type), data_(std::move(data)) {
const
std::
size_t
expected =
computeBufferSize
(width_, height_, type_);
if
(data_.
size
() < expected) {
throw
std::invalid_argument
(
"
VideoFrame: provided data is too small for
"
"
the specified format and size
"
);
}
}
VideoFrame
VideoFrame::create
(
int
width,
int
height, VideoBufferType type) {
const
std::
size_t
size =
computeBufferSize
(width, height, type);
std::vector<std::
uint8_t
>
buffer
(size,
0
);
return
VideoFrame
(width, height, type,
std::move
(buffer));
}
std::vector<VideoPlaneInfo>
VideoFrame::planeInfos
()
const
{
if
(data_.
empty
()) {
return
{};
}
const
uintptr_t
base =
reinterpret_cast
<
uintptr_t
>(data_.
data
());
return
computePlaneInfos
(base, width_, height_, type_);
}
VideoFrame
VideoFrame::convert
(VideoBufferType dst,
bool
flip_y)
const
{
//
Fast path: same format, no flip -> just clone the buffer.
//
We still return a *new* VideoFrame, never `*this`, so copy-ctor
//
being deleted is not a problem.
if
(dst == type_ && !flip_y) {
LK_LOG_WARN
(
"
VideoFrame::convert: converting to the same format
"
);
//
copy pixel data
std::vector<std::
uint8_t
> buf = data_;
return
VideoFrame
(width_, height_, type_,
std::move
(buf));
}
//
General path: delegate to the FFI-based conversion helper.
//
This returns a brand new VideoFrame (move-constructed / elided).
return
convertViaFfi
(*
this
, dst, flip_y);
}
VideoFrame
VideoFrame::fromOwnedInfo
(
const
proto::OwnedVideoBuffer& owned) {
const
auto
& info = owned.
info
();
const
int
width =
static_cast
<
int
>(info.
width
());
const
int
height =
static_cast
<
int
>(info.
height
());
const
VideoBufferType type =
fromProto
(info.
type
());
std::vector<std::
uint8_t
> buffer;
if
(info.
components_size
() >
0
) {
//
Multi-plane (e.g. I420, NV12, etc.). We pack planes back-to-back.
std::
size_t
total_size =
0
;
for
(
const
auto
& comp : info.
components
()) {
total_size +=
static_cast
<std::
size_t
>(comp.
size
());
}
buffer.
resize
(total_size);
std::
size_t
offset =
0
;
for
(
const
auto
& comp : info.
components
()) {
const
auto
sz =
static_cast
<std::
size_t
>(comp.
size
());
const
auto
* src_ptr =
//
NOLINTNEXTLINE(performance-no-int-to-ptr)
reinterpret_cast
<
const
std::
uint8_t
*>(comp.
data_ptr
());
std::memcpy
(buffer.
data
() + offset, src_ptr, sz);
offset += sz;
}
}
else
{
//
Packed format: treat top-level data_ptr as a single contiguous buffer.
const
auto
* src_ptr =
//
NOLINTNEXTLINE(performance-no-int-to-ptr)
reinterpret_cast
<
const
std::
uint8_t
*>(info.
data_ptr
());
std::
size_t
total_size =
0
;
if
(info.
has_stride
()) {
//
Use stride * height as total size (includes per-row padding if any).
total_size =
static_cast
<std::
size_t
>(info.
stride
()) *
static_cast
<std::
size_t
>(height);
}
else
{
//
Use our generic buffer-size helper (width/height/type).
total_size =
computeBufferSize
(width, height, type);
}
buffer.
resize
(total_size);
std::memcpy
(buffer.
data
(), src_ptr, total_size);
}
//
Release the FFI-owned buffer after copying the data.
{
const
FfiHandle
owned_handle
(
static_cast
<std::
uintptr_t
>(owned.
handle
().
id
()));
//
owned_handle destroyed at end of scope → native buffer disposed.
}
return
VideoFrame
(width, height, type,
std::move
(buffer));
}
}
//
namespace livekit
Back
|
FazBrowse Home
|
New Git URL