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
|
#include "audio_stream.h"
#include <c10/util/Logging.h>
#include <limits>
#include "util.h"
namespace ffmpeg {
namespace {
static int get_nb_channels(const AVFrame* frame, const AVCodecContext* codec) {
#if LIBAVUTIL_VERSION_INT >= AV_VERSION_INT(57, 28, 100)
return frame ? frame->ch_layout.nb_channels : codec->ch_layout.nb_channels;
#else
return frame ? frame->channels : codec->channels;
#endif
}
bool operator==(const AudioFormat& x, const AVFrame& y) {
return x.samples == static_cast<size_t>(y.sample_rate) &&
x.channels == static_cast<size_t>(get_nb_channels(&y, nullptr)) &&
x.format == y.format;
}
bool operator==(const AudioFormat& x, const AVCodecContext& y) {
return x.samples == static_cast<size_t>(y.sample_rate) &&
x.channels == static_cast<size_t>(get_nb_channels(nullptr, &y)) &&
x.format == y.sample_fmt;
}
AudioFormat& toAudioFormat(AudioFormat& x, const AVFrame& y) {
x.samples = y.sample_rate;
x.channels = get_nb_channels(&y, nullptr);
x.format = y.format;
return x;
}
AudioFormat& toAudioFormat(AudioFormat& x, const AVCodecContext& y) {
x.samples = y.sample_rate;
x.channels = get_nb_channels(nullptr, &y);
x.format = y.sample_fmt;
return x;
}
} // namespace
AudioStream::AudioStream(
AVFormatContext* inputCtx,
int index,
bool convertPtsToWallTime,
const AudioFormat& format)
: Stream(
inputCtx,
MediaFormat::makeMediaFormat(format, index),
convertPtsToWallTime,
0) {}
AudioStream::~AudioStream() {
if (sampler_) {
sampler_->shutdown();
sampler_.reset();
}
}
int AudioStream::initFormat() {
// set output format
if (format_.format.audio.samples == 0) {
format_.format.audio.samples = codecCtx_->sample_rate;
}
#if LIBAVUTIL_VERSION_INT >= AV_VERSION_INT(57, 28, 100)
if (format_.format.audio.channels == 0) {
format_.format.audio.channels = codecCtx_->ch_layout.nb_channels;
}
#else
if (format_.format.audio.channels == 0) {
format_.format.audio.channels = codecCtx_->channels;
}
#endif
if (format_.format.audio.format == AV_SAMPLE_FMT_NONE) {
format_.format.audio.format = codecCtx_->sample_fmt;
}
return format_.format.audio.samples != 0 &&
format_.format.audio.channels != 0 &&
format_.format.audio.format != AV_SAMPLE_FMT_NONE
? 0
: -1;
}
// copies audio sample bytes via swr_convert call in audio_sampler.cpp
int AudioStream::copyFrameBytes(ByteStorage* out, bool flush) {
if (!sampler_) {
sampler_ = std::make_unique<AudioSampler>(codecCtx_);
}
// check if input format gets changed
if (flush ? !(sampler_->getInputFormat().audio == *codecCtx_)
: !(sampler_->getInputFormat().audio == *frame_)) {
// - reinit sampler
SamplerParameters params;
params.type = format_.type;
params.out = format_.format;
params.in = FormatUnion();
flush ? toAudioFormat(params.in.audio, *codecCtx_)
: toAudioFormat(params.in.audio, *frame_);
if (!sampler_->init(params)) {
return -1;
}
VLOG(1) << "Set input audio sampler format"
<< ", samples: " << params.in.audio.samples
<< ", channels: " << params.in.audio.channels
<< ", format: " << params.in.audio.format
<< " : output audio sampler format"
<< ", samples: " << format_.format.audio.samples
<< ", channels: " << format_.format.audio.channels
<< ", format: " << format_.format.audio.format;
}
// calls to a sampler that converts the audio samples and copies them to the
// out buffer via ffmpeg::swr_convert
return sampler_->sample(flush ? nullptr : frame_, out);
}
} // namespace ffmpeg
|
