-
Notifications
You must be signed in to change notification settings - Fork 98
Expand file tree
/
Copy pathapi2-decode-encode-audio.cpp
More file actions
333 lines (270 loc) · 10.1 KB
/
Copy pathapi2-decode-encode-audio.cpp
File metadata and controls
333 lines (270 loc) · 10.1 KB
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
#include <iostream>
#include <set>
#include <map>
#include <memory>
#include <functional>
#include "avcpp/av.h"
#include "avcpp/ffmpeg.h"
#include "avcpp/codec.h"
#include "avcpp/packet.h"
#include "avcpp/videorescaler.h"
#include "avcpp/audioresampler.h"
#include "avcpp/avutils.h"
// API2
#include "avcpp/format.h"
#include "avcpp/formatcontext.h"
#include "avcpp/codec.h"
#include "avcpp/codeccontext.h"
using namespace std;
using namespace av;
int main(int argc, char **argv)
{
if (argc < 3)
return 1;
av::init();
av::setFFmpegLoggingLevel(AV_LOG_TRACE);
string uri (argv[1]);
string out (argv[2]);
ssize_t audioStream = -1;
AudioDecoderContext adec;
Stream ast;
error_code ec;
int count = 0;
{
//
// INPUT
//
FormatContext ictx;
ictx.openInput(uri, ec);
if (ec) {
cerr << "Can't open input\n";
return 1;
}
ictx.findStreamInfo();
for (size_t i = 0; i < ictx.streamsCount(); ++i) {
auto st = ictx.stream(i);
if (st.isAudio()) {
audioStream = i;
ast = st;
break;
}
}
cerr << audioStream << endl;
if (ast.isNull()) {
cerr << "Audio stream not found\n";
return 1;
}
if (ast.isValid()) {
adec = AudioDecoderContext(ast);
//Codec codec = findDecodingCodec(adec.raw()->codec_id);
//adec.setCodec(codec);
//adec.setRefCountedFrames(true);
adec.open(ec);
if (ec) {
cerr << "Can't open codec\n";
return 1;
}
}
//
// OUTPUT
//
OutputFormat ofmt;
FormatContext octx;
ofmt = av::guessOutputFormat(out, out);
clog << "Output format: " << ofmt.name() << " / " << ofmt.longName() << '\n';
octx.setFormat(ofmt);
Codec ocodec = av::findEncodingCodec(ofmt, false);
AudioEncoderContext enc (ocodec);
clog << ocodec.name() << " / " << ocodec.longName() << ", audio: " << (ocodec.type()==AVMEDIA_TYPE_AUDIO) << '\n';
auto sampleFmts = ocodec.supportedSampleFormats();
auto sampleRates = ocodec.supportedSamplerates();
auto layouts = ocodec.supportedChannelLayouts();
clog << "Supported sample formats:\n";
for (const auto &fmt : sampleFmts) {
clog << " " << av_get_sample_fmt_name(fmt) << '\n';
}
clog << "Supported sample rates:\n";
for (const auto &rate : sampleRates) {
clog << " " << rate << '\n';
}
clog << "Supported sample layouts:\n";
for (const auto &lay : layouts) {
char buf[128] = {0};
// FIXME: Add AVChannelLayout API
#if AVCPP_API_NEW_CHANNEL_LAYOUT
AVChannelLayout layout{};
av_channel_layout_from_mask(&layout, lay);
av_channel_layout_describe(&layout, buf, sizeof(buf));
#else
av_get_channel_layout_string(buf,
sizeof(buf),
av_get_channel_layout_nb_channels(lay),
lay);
#endif
clog << " " << buf << '\n';
}
//return 0;
// Settings
#if 1
enc.setSampleRate(48000);
enc.setSampleFormat(sampleFmts[0]);
// Layout
//enc.setChannelLayout(adec.channelLayout());
enc.setChannelLayout(AV_CH_LAYOUT_STEREO);
//enc.setChannelLayout(AV_CH_LAYOUT_MONO);
enc.setTimeBase(Rational(1, enc.sampleRate()));
enc.setBitRate(adec.bitRate());
#else
enc.setSampleRate(adec.sampleRate());
enc.setSampleFormat(adec.sampleFormat());
enc.setChannelLayout(adec.channelLayout());
enc.setTimeBase(adec.timeBase());
enc.setBitRate(adec.bitRate());
#endif
enc.open(ec);
if (ec) {
cerr << "Can't open encoder\n";
return 1;
}
Stream ost = octx.addStream(enc);
octx.openOutput(out, ec);
if (ec) {
cerr << "Can't open output\n";
return 1;
}
clog << "Encoder frame size: " << enc.frameSize() << '\n';
octx.dump();
octx.writeHeader();
octx.flush();
//
// RESAMPLER
//
AudioResampler resampler(enc.channelLayout(), enc.sampleRate(), enc.sampleFormat(),
adec.channelLayout(), adec.sampleRate(), adec.sampleFormat());
//
// PROCESS
//
while (true) {
Packet pkt = ictx.readPacket(ec);
if (ec)
{
clog << "Packet reading error: " << ec << ", " << ec.message() << endl;
break;
}
if (pkt && pkt.streamIndex() != audioStream) {
continue;
}
clog << "Read packet: isNull=" << (bool)!pkt << ", " << pkt.pts() << "(nopts:" << pkt.pts().isNoPts() << ")" << " / " << pkt.pts().seconds() << " / " << pkt.timeBase() << " / st: " << pkt.streamIndex() << endl;
#if 0
if (pkt.pts() == av::NoPts && pkt.timeBase() == Rational())
{
clog << "Skip invalid timestamp packet: data=" << (void*)pkt.data()
<< ", size=" << pkt.size()
<< ", flags=" << pkt.flags() << " (corrupt:" << (pkt.flags() & AV_PKT_FLAG_CORRUPT) << ";key:" << (pkt.flags() & AV_PKT_FLAG_KEY) << ")"
<< ", side_data=" << (void*)pkt.raw()->side_data
<< ", side_data_count=" << pkt.raw()->side_data_elems
<< endl;
//continue;
}
#endif
auto samples = adec.decode(pkt, ec);
count++;
//if (count > 200)
// break;
if (ec) {
cerr << "Decode error: " << ec << ", " << ec.message() << endl;
return 1;
} else if (!samples) {
cerr << "Empty samples set\n";
//if (!pkt) // decoder flushed here
// break;
//continue;
}
clog << " Samples [in]: " << samples.samplesCount()
<< ", ch: " << samples.channelsCount()
<< ", freq: " << samples.sampleRate()
<< ", name: " << samples.channelsLayoutString()
<< ", pts: " << samples.pts().seconds()
<< ", ref=" << samples.isReferenced() << ":" << samples.refCount()
<< endl;
// Empty samples set should not be pushed to the resampler, but it is valid case for the
// end of reading: during samples empty, some cached data can be stored at the resampler
// internal buffer, so we should consume it.
if (samples)
{
resampler.push(samples, ec);
if (ec) {
clog << "Resampler push error: " << ec << ", text: " << ec.message() << endl;
continue;
}
}
// Pop resampler data
bool getAll = !samples;
while (true) {
AudioSamples ouSamples(enc.sampleFormat(),
enc.frameSize(),
enc.channelLayout(),
enc.sampleRate());
// Resample:
bool hasFrame = resampler.pop(ouSamples, getAll, ec);
if (ec) {
clog << "Resampling status: " << ec << ", text: " << ec.message() << endl;
break;
} else if (!hasFrame) {
break;
} else
clog << " Samples [ou]: " << ouSamples.samplesCount()
<< ", ch: " << ouSamples.channelsCount()
<< ", freq: " << ouSamples.sampleRate()
<< ", name: " << ouSamples.channelsLayoutString()
<< ", pts: " << ouSamples.pts().seconds()
<< ", ref=" << ouSamples.isReferenced() << ":" << ouSamples.refCount()
<< endl;
// ENCODE
ouSamples.setStreamIndex(0);
ouSamples.setTimeBase(enc.timeBase());
Packet opkt = enc.encode(ouSamples, ec);
if (ec) {
cerr << "Encoding error: " << ec << ", " << ec.message() << endl;
return 1;
} else if (!opkt) {
//cerr << "Empty packet\n";
continue;
}
opkt.setStreamIndex(0);
clog << "Write packet: pts=" << opkt.pts() << ", dts=" << opkt.dts() << " / " << opkt.pts().seconds() << " / " << opkt.timeBase() << " / st: " << opkt.streamIndex() << endl;
octx.writePacket(opkt, ec);
if (ec) {
cerr << "Error write packet: " << ec << ", " << ec.message() << endl;
return 1;
}
}
// For the first packets samples can be empty: decoder caching
if (!pkt && !samples)
break;
}
//
// Is resampler flushed?
//
cerr << "Delay: " << resampler.delay() << endl;
//
// Flush encoder queue
//
clog << "Flush encoder:\n";
while (true) {
AudioSamples null(nullptr);
Packet opkt = enc.encode(null, ec);
if (ec || !opkt)
break;
opkt.setStreamIndex(0);
clog << "Write packet: pts=" << opkt.pts() << ", dts=" << opkt.dts() << " / " << opkt.pts().seconds() << " / " << opkt.timeBase() << " / st: " << opkt.streamIndex() << endl;
octx.writePacket(opkt, ec);
if (ec) {
cerr << "Error write packet: " << ec << ", " << ec.message() << endl;
return 1;
}
}
octx.flush();
octx.writeTrailer();
}
}