FFmpeg
transcode_aac.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2013-2022 Andreas Unterweger
3  *
4  * This file is part of FFmpeg.
5  *
6  * FFmpeg is free software; you can redistribute it and/or
7  * modify it under the terms of the GNU Lesser General Public
8  * License as published by the Free Software Foundation; either
9  * version 2.1 of the License, or (at your option) any later version.
10  *
11  * FFmpeg is distributed in the hope that it will be useful,
12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14  * Lesser General Public License for more details.
15  *
16  * You should have received a copy of the GNU Lesser General Public
17  * License along with FFmpeg; if not, write to the Free Software
18  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
19  */
20 
21 /**
22  * @file audio transcoding to MPEG/AAC API usage example
23  * @example transcode_aac.c
24  *
25  * Convert an input audio file to AAC in an MP4 container. Formats other than
26  * MP4 are supported based on the output file extension.
27  * @author Andreas Unterweger (dustsigns@gmail.com)
28  */
29 
30 #include <stdio.h>
31 
32 #include <libavutil/mem.h>
33 #include <libavformat/avformat.h>
34 #include <libavformat/avio.h>
35 
36 #include <libavcodec/avcodec.h>
37 
38 #include <libavutil/audio_fifo.h>
39 #include <libavutil/avassert.h>
40 #include <libavutil/avstring.h>
42 #include <libavutil/frame.h>
43 #include <libavutil/opt.h>
44 
46 
47 /* The output bit rate in bit/s */
48 #define OUTPUT_BIT_RATE 96000
49 /* The number of output channels */
50 #define OUTPUT_CHANNELS 2
51 
52 /**
53  * Open an input file and the required decoder.
54  * @param filename File to be opened
55  * @param[out] input_format_context Format context of opened file
56  * @param[out] input_codec_context Codec context of opened file
57  * @return Error code (0 if successful)
58  */
59 static int open_input_file(const char *filename,
60  AVFormatContext **input_format_context,
61  AVCodecContext **input_codec_context)
62 {
63  AVCodecContext *avctx;
64  const AVCodec *input_codec;
65  const AVStream *stream;
66  int error;
67 
68  /* Open the input file to read from it. */
69  if ((error = avformat_open_input(input_format_context, filename, NULL,
70  NULL)) < 0) {
71  fprintf(stderr, "Could not open input file '%s' (error '%s')\n",
72  filename, av_err2str(error));
73  *input_format_context = NULL;
74  return error;
75  }
76 
77  /* Get information on the input file (number of streams etc.). */
78  if ((error = avformat_find_stream_info(*input_format_context, NULL)) < 0) {
79  fprintf(stderr, "Could not open find stream info (error '%s')\n",
80  av_err2str(error));
81  avformat_close_input(input_format_context);
82  return error;
83  }
84 
85  /* Make sure that there is only one stream in the input file. */
86  if ((*input_format_context)->nb_streams != 1) {
87  fprintf(stderr, "Expected one audio input stream, but found %d\n",
88  (*input_format_context)->nb_streams);
89  avformat_close_input(input_format_context);
90  return AVERROR_EXIT;
91  }
92 
93  stream = (*input_format_context)->streams[0];
94 
95  /* Find a decoder for the audio stream. */
96  if (!(input_codec = avcodec_find_decoder(stream->codecpar->codec_id))) {
97  fprintf(stderr, "Could not find input codec\n");
98  avformat_close_input(input_format_context);
99  return AVERROR_EXIT;
100  }
101 
102  /* Allocate a new decoding context. */
103  avctx = avcodec_alloc_context3(input_codec);
104  if (!avctx) {
105  fprintf(stderr, "Could not allocate a decoding context\n");
106  avformat_close_input(input_format_context);
107  return AVERROR(ENOMEM);
108  }
109 
110  /* Initialize the stream parameters with demuxer information. */
111  error = avcodec_parameters_to_context(avctx, stream->codecpar);
112  if (error < 0) {
113  avformat_close_input(input_format_context);
114  avcodec_free_context(&avctx);
115  return error;
116  }
117 
118  /* Open the decoder for the audio stream to use it later. */
119  if ((error = avcodec_open2(avctx, input_codec, NULL)) < 0) {
120  fprintf(stderr, "Could not open input codec (error '%s')\n",
121  av_err2str(error));
122  avcodec_free_context(&avctx);
123  avformat_close_input(input_format_context);
124  return error;
125  }
126 
127  /* Set the packet timebase for the decoder. */
128  avctx->pkt_timebase = stream->time_base;
129 
130  /* Save the decoder context for easier access later. */
131  *input_codec_context = avctx;
132 
133  return 0;
134 }
135 
136 /**
137  * Open an output file and the required encoder.
138  * Also set some basic encoder parameters.
139  * Some of these parameters are based on the input file's parameters.
140  * @param filename File to be opened
141  * @param input_codec_context Codec context of input file
142  * @param[out] output_format_context Format context of output file
143  * @param[out] output_codec_context Codec context of output file
144  * @return Error code (0 if successful)
145  */
146 static int open_output_file(const char *filename,
147  AVCodecContext *input_codec_context,
148  AVFormatContext **output_format_context,
149  AVCodecContext **output_codec_context)
150 {
151  AVCodecContext *avctx = NULL;
152  AVIOContext *output_io_context = NULL;
153  AVStream *stream = NULL;
154  const AVCodec *output_codec = NULL;
155  int error;
156 
157  /* Open the output file to write to it. */
158  if ((error = avio_open(&output_io_context, filename,
159  AVIO_FLAG_WRITE)) < 0) {
160  fprintf(stderr, "Could not open output file '%s' (error '%s')\n",
161  filename, av_err2str(error));
162  return error;
163  }
164 
165  /* Create a new format context for the output container format. */
166  if (!(*output_format_context = avformat_alloc_context())) {
167  fprintf(stderr, "Could not allocate output format context\n");
168  return AVERROR(ENOMEM);
169  }
170 
171  /* Associate the output file (pointer) with the container format context. */
172  (*output_format_context)->pb = output_io_context;
173 
174  /* Guess the desired container format based on the file extension. */
175  if (!((*output_format_context)->oformat = av_guess_format(NULL, filename,
176  NULL))) {
177  fprintf(stderr, "Could not find output file format\n");
178  goto cleanup;
179  }
180 
181  if (!((*output_format_context)->url = av_strdup(filename))) {
182  fprintf(stderr, "Could not allocate url.\n");
183  error = AVERROR(ENOMEM);
184  goto cleanup;
185  }
186 
187  /* Find the encoder to be used by its name. */
188  if (!(output_codec = avcodec_find_encoder(AV_CODEC_ID_AAC))) {
189  fprintf(stderr, "Could not find an AAC encoder.\n");
190  goto cleanup;
191  }
192 
193  /* Create a new audio stream in the output file container. */
194  if (!(stream = avformat_new_stream(*output_format_context, NULL))) {
195  fprintf(stderr, "Could not create new stream\n");
196  error = AVERROR(ENOMEM);
197  goto cleanup;
198  }
199 
200  avctx = avcodec_alloc_context3(output_codec);
201  if (!avctx) {
202  fprintf(stderr, "Could not allocate an encoding context\n");
203  error = AVERROR(ENOMEM);
204  goto cleanup;
205  }
206 
207  /* Set the basic encoder parameters.
208  * The input file's sample rate is used to avoid a sample rate conversion. */
210  avctx->sample_rate = input_codec_context->sample_rate;
211  avctx->sample_fmt = output_codec->sample_fmts[0];
212  avctx->bit_rate = OUTPUT_BIT_RATE;
213 
214  /* Set the sample rate for the container. */
215  stream->time_base.den = input_codec_context->sample_rate;
216  stream->time_base.num = 1;
217 
218  /* Some container formats (like MP4) require global headers to be present.
219  * Mark the encoder so that it behaves accordingly. */
220  if ((*output_format_context)->oformat->flags & AVFMT_GLOBALHEADER)
222 
223  /* Open the encoder for the audio stream to use it later. */
224  if ((error = avcodec_open2(avctx, output_codec, NULL)) < 0) {
225  fprintf(stderr, "Could not open output codec (error '%s')\n",
226  av_err2str(error));
227  goto cleanup;
228  }
229 
231  if (error < 0) {
232  fprintf(stderr, "Could not initialize stream parameters\n");
233  goto cleanup;
234  }
235 
236  /* Save the encoder context for easier access later. */
237  *output_codec_context = avctx;
238 
239  return 0;
240 
241 cleanup:
242  avcodec_free_context(&avctx);
243  avio_closep(&(*output_format_context)->pb);
244  avformat_free_context(*output_format_context);
245  *output_format_context = NULL;
246  return error < 0 ? error : AVERROR_EXIT;
247 }
248 
249 /**
250  * Initialize one data packet for reading or writing.
251  * @param[out] packet Packet to be initialized
252  * @return Error code (0 if successful)
253  */
254 static int init_packet(AVPacket **packet)
255 {
256  if (!(*packet = av_packet_alloc())) {
257  fprintf(stderr, "Could not allocate packet\n");
258  return AVERROR(ENOMEM);
259  }
260  return 0;
261 }
262 
263 /**
264  * Initialize one audio frame for reading from the input file.
265  * @param[out] frame Frame to be initialized
266  * @return Error code (0 if successful)
267  */
269 {
270  if (!(*frame = av_frame_alloc())) {
271  fprintf(stderr, "Could not allocate input frame\n");
272  return AVERROR(ENOMEM);
273  }
274  return 0;
275 }
276 
277 /**
278  * Initialize the audio resampler based on the input and output codec settings.
279  * If the input and output sample formats differ, a conversion is required
280  * libswresample takes care of this, but requires initialization.
281  * @param input_codec_context Codec context of the input file
282  * @param output_codec_context Codec context of the output file
283  * @param[out] resample_context Resample context for the required conversion
284  * @return Error code (0 if successful)
285  */
286 static int init_resampler(AVCodecContext *input_codec_context,
287  AVCodecContext *output_codec_context,
288  SwrContext **resample_context)
289 {
290  int error;
291 
292  /*
293  * Create a resampler context for the conversion.
294  * Set the conversion parameters.
295  */
296  error = swr_alloc_set_opts2(resample_context,
297  &output_codec_context->ch_layout,
298  output_codec_context->sample_fmt,
299  output_codec_context->sample_rate,
300  &input_codec_context->ch_layout,
301  input_codec_context->sample_fmt,
302  input_codec_context->sample_rate,
303  0, NULL);
304  if (error < 0) {
305  fprintf(stderr, "Could not allocate resample context\n");
306  return error;
307  }
308  /*
309  * Perform a sanity check so that the number of converted samples is
310  * not greater than the number of samples to be converted.
311  * If the sample rates differ, this case has to be handled differently
312  */
313  av_assert0(output_codec_context->sample_rate == input_codec_context->sample_rate);
314 
315  /* Open the resampler with the specified parameters. */
316  if ((error = swr_init(*resample_context)) < 0) {
317  fprintf(stderr, "Could not open resample context\n");
318  swr_free(resample_context);
319  return error;
320  }
321  return 0;
322 }
323 
324 /**
325  * Initialize a FIFO buffer for the audio samples to be encoded.
326  * @param[out] fifo Sample buffer
327  * @param output_codec_context Codec context of the output file
328  * @return Error code (0 if successful)
329  */
330 static int init_fifo(AVAudioFifo **fifo, AVCodecContext *output_codec_context)
331 {
332  /* Create the FIFO buffer based on the specified output sample format. */
333  if (!(*fifo = av_audio_fifo_alloc(output_codec_context->sample_fmt,
334  output_codec_context->ch_layout.nb_channels, 1))) {
335  fprintf(stderr, "Could not allocate FIFO\n");
336  return AVERROR(ENOMEM);
337  }
338  return 0;
339 }
340 
341 /**
342  * Write the header of the output file container.
343  * @param output_format_context Format context of the output file
344  * @return Error code (0 if successful)
345  */
346 static int write_output_file_header(AVFormatContext *output_format_context)
347 {
348  int error;
349  if ((error = avformat_write_header(output_format_context, NULL)) < 0) {
350  fprintf(stderr, "Could not write output file header (error '%s')\n",
351  av_err2str(error));
352  return error;
353  }
354  return 0;
355 }
356 
357 /**
358  * Decode one audio frame from the input file.
359  * @param frame Audio frame to be decoded
360  * @param input_format_context Format context of the input file
361  * @param input_codec_context Codec context of the input file
362  * @param[out] data_present Indicates whether data has been decoded
363  * @param[out] finished Indicates whether the end of file has
364  * been reached and all data has been
365  * decoded. If this flag is false, there
366  * is more data to be decoded, i.e., this
367  * function has to be called again.
368  * @return Error code (0 if successful)
369  */
371  AVFormatContext *input_format_context,
372  AVCodecContext *input_codec_context,
373  int *data_present, int *finished)
374 {
375  /* Packet used for temporary storage. */
376  AVPacket *input_packet;
377  int error;
378 
379  error = init_packet(&input_packet);
380  if (error < 0)
381  return error;
382 
383  *data_present = 0;
384  *finished = 0;
385  /* Read one audio frame from the input file into a temporary packet. */
386  if ((error = av_read_frame(input_format_context, input_packet)) < 0) {
387  /* If we are at the end of the file, flush the decoder below. */
388  if (error == AVERROR_EOF)
389  *finished = 1;
390  else {
391  fprintf(stderr, "Could not read frame (error '%s')\n",
392  av_err2str(error));
393  goto cleanup;
394  }
395  }
396 
397  /* Send the audio frame stored in the temporary packet to the decoder.
398  * The input audio stream decoder is used to do this. */
399  if ((error = avcodec_send_packet(input_codec_context, input_packet)) < 0) {
400  fprintf(stderr, "Could not send packet for decoding (error '%s')\n",
401  av_err2str(error));
402  goto cleanup;
403  }
404 
405  /* Receive one frame from the decoder. */
406  error = avcodec_receive_frame(input_codec_context, frame);
407  /* If the decoder asks for more data to be able to decode a frame,
408  * return indicating that no data is present. */
409  if (error == AVERROR(EAGAIN)) {
410  error = 0;
411  goto cleanup;
412  /* If the end of the input file is reached, stop decoding. */
413  } else if (error == AVERROR_EOF) {
414  *finished = 1;
415  error = 0;
416  goto cleanup;
417  } else if (error < 0) {
418  fprintf(stderr, "Could not decode frame (error '%s')\n",
419  av_err2str(error));
420  goto cleanup;
421  /* Default case: Return decoded data. */
422  } else {
423  *data_present = 1;
424  goto cleanup;
425  }
426 
427 cleanup:
428  av_packet_free(&input_packet);
429  return error;
430 }
431 
432 /**
433  * Initialize a temporary storage for the specified number of audio samples.
434  * The conversion requires temporary storage due to the different format.
435  * The number of audio samples to be allocated is specified in frame_size.
436  * @param[out] converted_input_samples Array of converted samples. The
437  * dimensions are reference, channel
438  * (for multi-channel audio), sample.
439  * @param output_codec_context Codec context of the output file
440  * @param frame_size Number of samples to be converted in
441  * each round
442  * @return Error code (0 if successful)
443  */
444 static int init_converted_samples(uint8_t ***converted_input_samples,
445  AVCodecContext *output_codec_context,
446  int frame_size)
447 {
448  int error;
449 
450  /* Allocate as many pointers as there are audio channels.
451  * Each pointer will point to the audio samples of the corresponding
452  * channels (although it may be NULL for interleaved formats).
453  * Allocate memory for the samples of all channels in one consecutive
454  * block for convenience. */
455  if ((error = av_samples_alloc_array_and_samples(converted_input_samples, NULL,
456  output_codec_context->ch_layout.nb_channels,
457  frame_size,
458  output_codec_context->sample_fmt, 0)) < 0) {
459  fprintf(stderr,
460  "Could not allocate converted input samples (error '%s')\n",
461  av_err2str(error));
462  return error;
463  }
464  return 0;
465 }
466 
467 /**
468  * Convert the input audio samples into the output sample format.
469  * The conversion happens on a per-frame basis, the size of which is
470  * specified by frame_size.
471  * @param input_data Samples to be decoded. The dimensions are
472  * channel (for multi-channel audio), sample.
473  * @param[out] converted_data Converted samples. The dimensions are channel
474  * (for multi-channel audio), sample.
475  * @param frame_size Number of samples to be converted
476  * @param resample_context Resample context for the conversion
477  * @return Error code (0 if successful)
478  */
479 static int convert_samples(const uint8_t **input_data,
480  uint8_t **converted_data, const int frame_size,
481  SwrContext *resample_context)
482 {
483  int error;
484 
485  /* Convert the samples using the resampler. */
486  if ((error = swr_convert(resample_context,
487  converted_data, frame_size,
488  input_data , frame_size)) < 0) {
489  fprintf(stderr, "Could not convert input samples (error '%s')\n",
490  av_err2str(error));
491  return error;
492  }
493 
494  return 0;
495 }
496 
497 /**
498  * Add converted input audio samples to the FIFO buffer for later processing.
499  * @param fifo Buffer to add the samples to
500  * @param converted_input_samples Samples to be added. The dimensions are channel
501  * (for multi-channel audio), sample.
502  * @param frame_size Number of samples to be converted
503  * @return Error code (0 if successful)
504  */
506  uint8_t **converted_input_samples,
507  const int frame_size)
508 {
509  int error;
510 
511  /* Make the FIFO as large as it needs to be to hold both,
512  * the old and the new samples. */
513  if ((error = av_audio_fifo_realloc(fifo, av_audio_fifo_size(fifo) + frame_size)) < 0) {
514  fprintf(stderr, "Could not reallocate FIFO\n");
515  return error;
516  }
517 
518  /* Store the new samples in the FIFO buffer. */
519  if (av_audio_fifo_write(fifo, (void **)converted_input_samples,
520  frame_size) < frame_size) {
521  fprintf(stderr, "Could not write data to FIFO\n");
522  return AVERROR_EXIT;
523  }
524  return 0;
525 }
526 
527 /**
528  * Read one audio frame from the input file, decode, convert and store
529  * it in the FIFO buffer.
530  * @param fifo Buffer used for temporary storage
531  * @param input_format_context Format context of the input file
532  * @param input_codec_context Codec context of the input file
533  * @param output_codec_context Codec context of the output file
534  * @param resampler_context Resample context for the conversion
535  * @param[out] finished Indicates whether the end of file has
536  * been reached and all data has been
537  * decoded. If this flag is false,
538  * there is more data to be decoded,
539  * i.e., this function has to be called
540  * again.
541  * @return Error code (0 if successful)
542  */
544  AVFormatContext *input_format_context,
545  AVCodecContext *input_codec_context,
546  AVCodecContext *output_codec_context,
547  SwrContext *resampler_context,
548  int *finished)
549 {
550  /* Temporary storage of the input samples of the frame read from the file. */
551  AVFrame *input_frame = NULL;
552  /* Temporary storage for the converted input samples. */
553  uint8_t **converted_input_samples = NULL;
554  int data_present;
555  int ret = AVERROR_EXIT;
556 
557  /* Initialize temporary storage for one input frame. */
558  if (init_input_frame(&input_frame))
559  goto cleanup;
560  /* Decode one frame worth of audio samples. */
561  if (decode_audio_frame(input_frame, input_format_context,
562  input_codec_context, &data_present, finished))
563  goto cleanup;
564  /* If we are at the end of the file and there are no more samples
565  * in the decoder which are delayed, we are actually finished.
566  * This must not be treated as an error. */
567  if (*finished) {
568  ret = 0;
569  goto cleanup;
570  }
571  /* If there is decoded data, convert and store it. */
572  if (data_present) {
573  /* Initialize the temporary storage for the converted input samples. */
574  if (init_converted_samples(&converted_input_samples, output_codec_context,
575  input_frame->nb_samples))
576  goto cleanup;
577 
578  /* Convert the input samples to the desired output sample format.
579  * This requires a temporary storage provided by converted_input_samples. */
580  if (convert_samples((const uint8_t**)input_frame->extended_data, converted_input_samples,
581  input_frame->nb_samples, resampler_context))
582  goto cleanup;
583 
584  /* Add the converted input samples to the FIFO buffer for later processing. */
585  if (add_samples_to_fifo(fifo, converted_input_samples,
586  input_frame->nb_samples))
587  goto cleanup;
588  ret = 0;
589  }
590  ret = 0;
591 
592 cleanup:
593  if (converted_input_samples)
594  av_freep(&converted_input_samples[0]);
595  av_freep(&converted_input_samples);
596  av_frame_free(&input_frame);
597 
598  return ret;
599 }
600 
601 /**
602  * Initialize one input frame for writing to the output file.
603  * The frame will be exactly frame_size samples large.
604  * @param[out] frame Frame to be initialized
605  * @param output_codec_context Codec context of the output file
606  * @param frame_size Size of the frame
607  * @return Error code (0 if successful)
608  */
610  AVCodecContext *output_codec_context,
611  int frame_size)
612 {
613  int error;
614 
615  /* Create a new frame to store the audio samples. */
616  if (!(*frame = av_frame_alloc())) {
617  fprintf(stderr, "Could not allocate output frame\n");
618  return AVERROR_EXIT;
619  }
620 
621  /* Set the frame's parameters, especially its size and format.
622  * av_frame_get_buffer needs this to allocate memory for the
623  * audio samples of the frame.
624  * Default channel layouts based on the number of channels
625  * are assumed for simplicity. */
626  (*frame)->nb_samples = frame_size;
627  av_channel_layout_copy(&(*frame)->ch_layout, &output_codec_context->ch_layout);
628  (*frame)->format = output_codec_context->sample_fmt;
629  (*frame)->sample_rate = output_codec_context->sample_rate;
630 
631  /* Allocate the samples of the created frame. This call will make
632  * sure that the audio frame can hold as many samples as specified. */
633  if ((error = av_frame_get_buffer(*frame, 0)) < 0) {
634  fprintf(stderr, "Could not allocate output frame samples (error '%s')\n",
635  av_err2str(error));
637  return error;
638  }
639 
640  return 0;
641 }
642 
643 /* Global timestamp for the audio frames. */
644 static int64_t pts = 0;
645 
646 /**
647  * Encode one frame worth of audio to the output file.
648  * @param frame Samples to be encoded
649  * @param output_format_context Format context of the output file
650  * @param output_codec_context Codec context of the output file
651  * @param[out] data_present Indicates whether data has been
652  * encoded
653  * @return Error code (0 if successful)
654  */
656  AVFormatContext *output_format_context,
657  AVCodecContext *output_codec_context,
658  int *data_present)
659 {
660  /* Packet used for temporary storage. */
662  int error;
663 
665  if (error < 0)
666  return error;
667 
668  /* Set a timestamp based on the sample rate for the container. */
669  if (frame) {
670  frame->pts = pts;
671  pts += frame->nb_samples;
672  }
673 
674  *data_present = 0;
675  /* Send the audio frame stored in the temporary packet to the encoder.
676  * The output audio stream encoder is used to do this. */
677  error = avcodec_send_frame(output_codec_context, frame);
678  /* Check for errors, but proceed with fetching encoded samples if the
679  * encoder signals that it has nothing more to encode. */
680  if (error < 0 && error != AVERROR_EOF) {
681  fprintf(stderr, "Could not send packet for encoding (error '%s')\n",
682  av_err2str(error));
683  goto cleanup;
684  }
685 
686  /* Receive one encoded frame from the encoder. */
687  error = avcodec_receive_packet(output_codec_context, output_packet);
688  /* If the encoder asks for more data to be able to provide an
689  * encoded frame, return indicating that no data is present. */
690  if (error == AVERROR(EAGAIN)) {
691  error = 0;
692  goto cleanup;
693  /* If the last frame has been encoded, stop encoding. */
694  } else if (error == AVERROR_EOF) {
695  error = 0;
696  goto cleanup;
697  } else if (error < 0) {
698  fprintf(stderr, "Could not encode frame (error '%s')\n",
699  av_err2str(error));
700  goto cleanup;
701  /* Default case: Return encoded data. */
702  } else {
703  *data_present = 1;
704  }
705 
706  /* Write one audio frame from the temporary packet to the output file. */
707  if (*data_present &&
708  (error = av_write_frame(output_format_context, output_packet)) < 0) {
709  fprintf(stderr, "Could not write frame (error '%s')\n",
710  av_err2str(error));
711  goto cleanup;
712  }
713 
714 cleanup:
716  return error;
717 }
718 
719 /**
720  * Load one audio frame from the FIFO buffer, encode and write it to the
721  * output file.
722  * @param fifo Buffer used for temporary storage
723  * @param output_format_context Format context of the output file
724  * @param output_codec_context Codec context of the output file
725  * @return Error code (0 if successful)
726  */
728  AVFormatContext *output_format_context,
729  AVCodecContext *output_codec_context)
730 {
731  /* Temporary storage of the output samples of the frame written to the file. */
733  /* Use the maximum number of possible samples per frame.
734  * If there is less than the maximum possible frame size in the FIFO
735  * buffer use this number. Otherwise, use the maximum possible frame size. */
736  const int frame_size = FFMIN(av_audio_fifo_size(fifo),
737  output_codec_context->frame_size);
738  int data_written;
739 
740  /* Initialize temporary storage for one output frame. */
741  if (init_output_frame(&output_frame, output_codec_context, frame_size))
742  return AVERROR_EXIT;
743 
744  /* Read as many samples from the FIFO buffer as required to fill the frame.
745  * The samples are stored in the frame temporarily. */
746  if (av_audio_fifo_read(fifo, (void **)output_frame->data, frame_size) < frame_size) {
747  fprintf(stderr, "Could not read data from FIFO\n");
749  return AVERROR_EXIT;
750  }
751 
752  /* Encode one frame worth of audio samples. */
753  if (encode_audio_frame(output_frame, output_format_context,
754  output_codec_context, &data_written)) {
756  return AVERROR_EXIT;
757  }
759  return 0;
760 }
761 
762 /**
763  * Write the trailer of the output file container.
764  * @param output_format_context Format context of the output file
765  * @return Error code (0 if successful)
766  */
767 static int write_output_file_trailer(AVFormatContext *output_format_context)
768 {
769  int error;
770  if ((error = av_write_trailer(output_format_context)) < 0) {
771  fprintf(stderr, "Could not write output file trailer (error '%s')\n",
772  av_err2str(error));
773  return error;
774  }
775  return 0;
776 }
777 
778 int main(int argc, char **argv)
779 {
780  AVFormatContext *input_format_context = NULL, *output_format_context = NULL;
781  AVCodecContext *input_codec_context = NULL, *output_codec_context = NULL;
782  SwrContext *resample_context = NULL;
783  AVAudioFifo *fifo = NULL;
784  int ret = AVERROR_EXIT;
785 
786  if (argc != 3) {
787  fprintf(stderr, "Usage: %s <input file> <output file>\n", argv[0]);
788  exit(1);
789  }
790 
791  /* Open the input file for reading. */
792  if (open_input_file(argv[1], &input_format_context,
793  &input_codec_context))
794  goto cleanup;
795  /* Open the output file for writing. */
796  if (open_output_file(argv[2], input_codec_context,
797  &output_format_context, &output_codec_context))
798  goto cleanup;
799  /* Initialize the resampler to be able to convert audio sample formats. */
800  if (init_resampler(input_codec_context, output_codec_context,
801  &resample_context))
802  goto cleanup;
803  /* Initialize the FIFO buffer to store audio samples to be encoded. */
804  if (init_fifo(&fifo, output_codec_context))
805  goto cleanup;
806  /* Write the header of the output file container. */
807  if (write_output_file_header(output_format_context))
808  goto cleanup;
809 
810  /* Loop as long as we have input samples to read or output samples
811  * to write; abort as soon as we have neither. */
812  while (1) {
813  /* Use the encoder's desired frame size for processing. */
814  const int output_frame_size = output_codec_context->frame_size;
815  int finished = 0;
816 
817  /* Make sure that there is one frame worth of samples in the FIFO
818  * buffer so that the encoder can do its work.
819  * Since the decoder's and the encoder's frame size may differ, we
820  * need to FIFO buffer to store as many frames worth of input samples
821  * that they make up at least one frame worth of output samples. */
822  while (av_audio_fifo_size(fifo) < output_frame_size) {
823  /* Decode one frame worth of audio samples, convert it to the
824  * output sample format and put it into the FIFO buffer. */
825  if (read_decode_convert_and_store(fifo, input_format_context,
826  input_codec_context,
827  output_codec_context,
828  resample_context, &finished))
829  goto cleanup;
830 
831  /* If we are at the end of the input file, we continue
832  * encoding the remaining audio samples to the output file. */
833  if (finished)
834  break;
835  }
836 
837  /* If we have enough samples for the encoder, we encode them.
838  * At the end of the file, we pass the remaining samples to
839  * the encoder. */
840  while (av_audio_fifo_size(fifo) >= output_frame_size ||
841  (finished && av_audio_fifo_size(fifo) > 0))
842  /* Take one frame worth of audio samples from the FIFO buffer,
843  * encode it and write it to the output file. */
844  if (load_encode_and_write(fifo, output_format_context,
845  output_codec_context))
846  goto cleanup;
847 
848  /* If we are at the end of the input file and have encoded
849  * all remaining samples, we can exit this loop and finish. */
850  if (finished) {
851  int data_written;
852  /* Flush the encoder as it may have delayed frames. */
853  do {
854  if (encode_audio_frame(NULL, output_format_context,
855  output_codec_context, &data_written))
856  goto cleanup;
857  } while (data_written);
858  break;
859  }
860  }
861 
862  /* Write the trailer of the output file container. */
863  if (write_output_file_trailer(output_format_context))
864  goto cleanup;
865  ret = 0;
866 
867 cleanup:
868  if (fifo)
869  av_audio_fifo_free(fifo);
870  swr_free(&resample_context);
871  if (output_codec_context)
872  avcodec_free_context(&output_codec_context);
873  if (output_format_context) {
874  avio_closep(&output_format_context->pb);
875  avformat_free_context(output_format_context);
876  }
877  if (input_codec_context)
878  avcodec_free_context(&input_codec_context);
879  if (input_format_context)
880  avformat_close_input(&input_format_context);
881 
882  return ret;
883 }
error
static void error(const char *err)
Definition: target_bsf_fuzzer.c:32
av_audio_fifo_free
void av_audio_fifo_free(AVAudioFifo *af)
Free an AVAudioFifo.
Definition: audio_fifo.c:48
AVCodecContext::frame_size
int frame_size
Number of samples per channel in an audio frame.
Definition: avcodec.h:1083
AVCodec
AVCodec.
Definition: codec.h:187
load_encode_and_write
static int load_encode_and_write(AVAudioFifo *fifo, AVFormatContext *output_format_context, AVCodecContext *output_codec_context)
Load one audio frame from the FIFO buffer, encode and write it to the output file.
Definition: transcode_aac.c:727
avcodec_receive_packet
int avcodec_receive_packet(AVCodecContext *avctx, AVPacket *avpkt)
Read encoded data from the encoder.
Definition: encode.c:541
AVERROR
Filter the word “frame” indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel sample they are references to shared objects When the negotiation mechanism computes the intersection of the formats supported at each end of a all references to both lists are replaced with a reference to the intersection And when a single format is eventually chosen for a link amongst the remaining all references to the list are updated That means that if a filter requires that its input and output have the same format amongst a supported all it has to do is use a reference to the same list of formats query_formats can leave some formats unset and return AVERROR(EAGAIN) to cause the negotiation mechanism toagain later. That can be used by filters with complex requirements to use the format negotiated on one link to set the formats supported on another. Frame references ownership and permissions
opt.h
open_input_file
static int open_input_file(const char *filename, AVFormatContext **input_format_context, AVCodecContext **input_codec_context)
Open an input file and the required decoder.
Definition: transcode_aac.c:59
av_frame_get_buffer
int av_frame_get_buffer(AVFrame *frame, int align)
Allocate new buffer(s) for audio or video data.
Definition: frame.c:292
AVCodecContext::sample_rate
int sample_rate
samples per second
Definition: avcodec.h:1056
avformat_new_stream
AVStream * avformat_new_stream(AVFormatContext *s, const struct AVCodec *c)
Add a new stream to a media file.
AVERROR_EOF
#define AVERROR_EOF
End of file.
Definition: error.h:57
av_audio_fifo_write
int av_audio_fifo_write(AVAudioFifo *af, void *const *data, int nb_samples)
Write data to an AVAudioFifo.
Definition: audio_fifo.c:119
av_audio_fifo_realloc
int av_audio_fifo_realloc(AVAudioFifo *af, int nb_samples)
Reallocate an AVAudioFifo.
Definition: audio_fifo.c:99
int64_t
long long int64_t
Definition: coverity.c:34
init_fifo
static int init_fifo(AVAudioFifo **fifo, AVCodecContext *output_codec_context)
Initialize a FIFO buffer for the audio samples to be encoded.
Definition: transcode_aac.c:330
av_frame_free
void av_frame_free(AVFrame **frame)
Free the frame and any dynamically allocated objects in it, e.g.
Definition: frame.c:162
avcodec_find_encoder
const AVCodec * avcodec_find_encoder(enum AVCodecID id)
Find a registered encoder with a matching codec ID.
Definition: allcodecs.c:1006
AVFrame
This structure describes decoded (raw) audio or video data.
Definition: frame.h:389
cleanup
static av_cold void cleanup(FlashSV2Context *s)
Definition: flashsv2enc.c:130
write_output_file_header
static int write_output_file_header(AVFormatContext *output_format_context)
Write the header of the output file container.
Definition: transcode_aac.c:346
open_output_file
static int open_output_file(const char *filename, AVCodecContext *input_codec_context, AVFormatContext **output_format_context, AVCodecContext **output_codec_context)
Open an output file and the required encoder.
Definition: transcode_aac.c:146
avio_open
int avio_open(AVIOContext **s, const char *filename, int flags)
Create and initialize a AVIOContext for accessing the resource indicated by url.
Definition: avio.c:497
av_read_frame
int av_read_frame(AVFormatContext *s, AVPacket *pkt)
Return the next frame of a stream.
Definition: demux.c:1547
AVChannelLayout::nb_channels
int nb_channels
Number of channels in this layout.
Definition: channel_layout.h:321
av_packet_free
void av_packet_free(AVPacket **pkt)
Free the packet, if the packet is reference counted, it will be unreferenced first.
Definition: packet.c:74
AV_CODEC_FLAG_GLOBAL_HEADER
#define AV_CODEC_FLAG_GLOBAL_HEADER
Place global headers in extradata instead of every keyframe.
Definition: avcodec.h:338
avformat_close_input
void avformat_close_input(AVFormatContext **s)
Close an opened input AVFormatContext.
Definition: demux.c:366
AVAudioFifo
Context for an Audio FIFO Buffer.
Definition: audio_fifo.c:37
AVCodecContext::ch_layout
AVChannelLayout ch_layout
Audio channel layout.
Definition: avcodec.h:1071
output_packet
static int output_packet(AVFormatContext *ctx, int flush)
Definition: mpegenc.c:1010
AVCodecContext::flags
int flags
AV_CODEC_FLAG_*.
Definition: avcodec.h:508
pts
static int64_t pts
Definition: transcode_aac.c:644
AVRational::num
int num
Numerator.
Definition: rational.h:59
swr_convert
int attribute_align_arg swr_convert(struct SwrContext *s, uint8_t *const *out_arg, int out_count, const uint8_t *const *in_arg, int in_count)
Convert audio.
Definition: swresample.c:719
av_frame_alloc
AVFrame * av_frame_alloc(void)
Allocate an AVFrame and set its fields to default values.
Definition: frame.c:150
avassert.h
swr_init
av_cold int swr_init(struct SwrContext *s)
Initialize context after user parameters have been set.
Definition: swresample.c:140
avformat_open_input
int avformat_open_input(AVFormatContext **ps, const char *url, const AVInputFormat *fmt, AVDictionary **options)
Open an input stream and read the header.
Definition: demux.c:216
avcodec_alloc_context3
AVCodecContext * avcodec_alloc_context3(const AVCodec *codec)
Allocate an AVCodecContext and set its fields to default values.
Definition: options.c:149
add_samples_to_fifo
static int add_samples_to_fifo(AVAudioFifo *fifo, uint8_t **converted_input_samples, const int frame_size)
Add converted input audio samples to the FIFO buffer for later processing.
Definition: transcode_aac.c:505
frame_size
int frame_size
Definition: mxfenc.c:2429
decode_audio_frame
static int decode_audio_frame(AVFrame *frame, AVFormatContext *input_format_context, AVCodecContext *input_codec_context, int *data_present, int *finished)
Decode one audio frame from the input file.
Definition: transcode_aac.c:370
avcodec_receive_frame
int attribute_align_arg avcodec_receive_frame(AVCodecContext *avctx, AVFrame *frame)
Return decoded output data from a decoder or encoder (when the AV_CODEC_FLAG_RECON_FRAME flag is used...
Definition: avcodec.c:713
av_assert0
#define av_assert0(cond)
assert() equivalent, that is always enabled.
Definition: avassert.h:40
AVIO_FLAG_WRITE
#define AVIO_FLAG_WRITE
write-only
Definition: avio.h:618
SwrContext
The libswresample context.
Definition: swresample_internal.h:95
avformat_write_header
av_warn_unused_result int avformat_write_header(AVFormatContext *s, AVDictionary **options)
Allocate the stream private data and write the stream header to an output media file.
Definition: mux.c:488
AVFormatContext
Format I/O context.
Definition: avformat.h:1300
AVStream::codecpar
AVCodecParameters * codecpar
Codec parameters associated with this stream.
Definition: avformat.h:771
avcodec_parameters_to_context
int avcodec_parameters_to_context(AVCodecContext *codec, const struct AVCodecParameters *par)
Fill the codec context based on the values from the supplied codec parameters.
AVStream::time_base
AVRational time_base
This is the fundamental unit of time (in seconds) in terms of which frame timestamps are represented.
Definition: avformat.h:787
NULL
#define NULL
Definition: coverity.c:32
avcodec_free_context
void avcodec_free_context(AVCodecContext **avctx)
Free the codec context and everything associated with it and write NULL to the provided pointer.
Definition: options.c:164
av_audio_fifo_alloc
AVAudioFifo * av_audio_fifo_alloc(enum AVSampleFormat sample_fmt, int channels, int nb_samples)
Allocate an AVAudioFifo.
Definition: audio_fifo.c:62
read_decode_convert_and_store
static int read_decode_convert_and_store(AVAudioFifo *fifo, AVFormatContext *input_format_context, AVCodecContext *input_codec_context, AVCodecContext *output_codec_context, SwrContext *resampler_context, int *finished)
Read one audio frame from the input file, decode, convert and store it in the FIFO buffer.
Definition: transcode_aac.c:543
AVCodecContext::bit_rate
int64_t bit_rate
the average bitrate
Definition: avcodec.h:501
OUTPUT_BIT_RATE
#define OUTPUT_BIT_RATE
Definition: transcode_aac.c:48
avcodec_open2
int attribute_align_arg avcodec_open2(AVCodecContext *avctx, const AVCodec *codec, AVDictionary **options)
Initialize the AVCodecContext to use the given AVCodec.
Definition: avcodec.c:143
av_write_frame
int av_write_frame(AVFormatContext *s, AVPacket *pkt)
Write a packet to an output media file.
Definition: mux.c:1236
init_output_frame
static int init_output_frame(AVFrame **frame, AVCodecContext *output_codec_context, int frame_size)
Initialize one input frame for writing to the output file.
Definition: transcode_aac.c:609
swresample.h
avcodec_find_decoder
const AVCodec * avcodec_find_decoder(enum AVCodecID id)
Find a registered decoder with a matching codec ID.
Definition: allcodecs.c:1011
AV_CODEC_ID_AAC
@ AV_CODEC_ID_AAC
Definition: codec_id.h:448
init_input_frame
static int init_input_frame(AVFrame **frame)
Initialize one audio frame for reading from the input file.
Definition: transcode_aac.c:268
avformat_find_stream_info
int avformat_find_stream_info(AVFormatContext *ic, AVDictionary **options)
Read packets of a media file to get stream information.
Definition: demux.c:2525
AVIOContext
Bytestream IO Context.
Definition: avio.h:160
swr_alloc_set_opts2
int swr_alloc_set_opts2(struct SwrContext **ps, const AVChannelLayout *out_ch_layout, enum AVSampleFormat out_sample_fmt, int out_sample_rate, const AVChannelLayout *in_ch_layout, enum AVSampleFormat in_sample_fmt, int in_sample_rate, int log_offset, void *log_ctx)
Allocate SwrContext if needed and set/reset common parameters.
Definition: swresample.c:40
avformat_alloc_context
AVFormatContext * avformat_alloc_context(void)
Allocate an AVFormatContext.
Definition: options.c:162
av_err2str
#define av_err2str(errnum)
Convenience macro, the return value should be used only directly in function arguments but never stan...
Definition: error.h:122
AVCodecContext::sample_fmt
enum AVSampleFormat sample_fmt
audio sample format
Definition: avcodec.h:1063
AVCodecContext::pkt_timebase
AVRational pkt_timebase
Timebase in which pkt_dts/pts and AVPacket.dts/pts are expressed.
Definition: avcodec.h:557
av_audio_fifo_read
int av_audio_fifo_read(AVAudioFifo *af, void *const *data, int nb_samples)
Read data from an AVAudioFifo.
Definition: audio_fifo.c:175
encode_audio_frame
static int encode_audio_frame(AVFrame *frame, AVFormatContext *output_format_context, AVCodecContext *output_codec_context, int *data_present)
Encode one frame worth of audio to the output file.
Definition: transcode_aac.c:655
main
int main(int argc, char **argv)
Definition: transcode_aac.c:778
avio.h
swr_free
av_cold void swr_free(SwrContext **ss)
Free the given SwrContext and set the pointer to NULL.
Definition: swresample.c:121
init_packet
static int init_packet(AVPacket **packet)
Initialize one data packet for reading or writing.
Definition: transcode_aac.c:254
frame.h
OUTPUT_CHANNELS
#define OUTPUT_CHANNELS
Definition: transcode_aac.c:50
output_frame
static int output_frame(H264Context *h, AVFrame *dst, H264Picture *srcp)
Definition: h264dec.c:877
av_packet_alloc
AVPacket * av_packet_alloc(void)
Allocate an AVPacket and set its fields to default values.
Definition: packet.c:63
av_audio_fifo_size
int av_audio_fifo_size(AVAudioFifo *af)
Get the current number of samples in the AVAudioFifo available for reading.
Definition: audio_fifo.c:222
init_resampler
static int init_resampler(AVCodecContext *input_codec_context, AVCodecContext *output_codec_context, SwrContext **resample_context)
Initialize the audio resampler based on the input and output codec settings.
Definition: transcode_aac.c:286
av_channel_layout_default
void av_channel_layout_default(AVChannelLayout *ch_layout, int nb_channels)
Get the default channel layout for a given number of channels.
Definition: channel_layout.c:834
avcodec_send_packet
int avcodec_send_packet(AVCodecContext *avctx, const AVPacket *avpkt)
Supply raw packet data as input to a decoder.
Definition: decode.c:728
av_write_trailer
int av_write_trailer(AVFormatContext *s)
Write the stream trailer to an output media file and free the file private data.
Definition: mux.c:1298
AVFrame::nb_samples
int nb_samples
number of audio samples (per channel) described by this frame
Definition: frame.h:469
AVFMT_GLOBALHEADER
#define AVFMT_GLOBALHEADER
Format wants global header.
Definition: avformat.h:478
convert_samples
static int convert_samples(const uint8_t **input_data, uint8_t **converted_data, const int frame_size, SwrContext *resample_context)
Convert the input audio samples into the output sample format.
Definition: transcode_aac.c:479
AVFrame::extended_data
uint8_t ** extended_data
pointers to the data planes/channels.
Definition: frame.h:450
FFMIN
#define FFMIN(a, b)
Definition: macros.h:49
init_converted_samples
static int init_converted_samples(uint8_t ***converted_input_samples, AVCodecContext *output_codec_context, int frame_size)
Initialize a temporary storage for the specified number of audio samples.
Definition: transcode_aac.c:444
audio_fifo.h
avcodec_send_frame
int avcodec_send_frame(AVCodecContext *avctx, const AVFrame *frame)
Supply a raw video or audio frame to the encoder.
Definition: encode.c:508
AVCodec::sample_fmts
attribute_deprecated enum AVSampleFormat * sample_fmts
Definition: codec.h:219
avcodec.h
ret
ret
Definition: filter_design.txt:187
AVStream
Stream structure.
Definition: avformat.h:748
frame
these buffered frames must be flushed immediately if a new input produces new the filter must not call request_frame to get more It must just process the frame or queue it The task of requesting more frames is left to the filter s request_frame method or the application If a filter has several the filter must be ready for frames arriving randomly on any input any filter with several inputs will most likely require some kind of queuing mechanism It is perfectly acceptable to have a limited queue and to drop frames when the inputs are too unbalanced request_frame For filters that do not use the this method is called when a frame is wanted on an output For a it should directly call filter_frame on the corresponding output For a if there are queued frames already one of these frames should be pushed If the filter should request a frame on one of its repeatedly until at least one frame has been pushed Return or at least make progress towards producing a frame
Definition: filter_design.txt:264
avformat.h
AVCodecContext
main external API structure.
Definition: avcodec.h:451
channel_layout.h
AVRational::den
int den
Denominator.
Definition: rational.h:60
avformat_free_context
void avformat_free_context(AVFormatContext *s)
Free an AVFormatContext and all its streams.
Definition: avformat.c:149
av_channel_layout_copy
int av_channel_layout_copy(AVChannelLayout *dst, const AVChannelLayout *src)
Make a copy of a channel layout.
Definition: channel_layout.c:444
av_strdup
char * av_strdup(const char *s)
Duplicate a string.
Definition: mem.c:272
av_guess_format
const AVOutputFormat * av_guess_format(const char *short_name, const char *filename, const char *mime_type)
Return the output format in the list of registered output formats which best matches the provided par...
Definition: format.c:79
mem.h
av_samples_alloc_array_and_samples
int av_samples_alloc_array_and_samples(uint8_t ***audio_data, int *linesize, int nb_channels, int nb_samples, enum AVSampleFormat sample_fmt, int align)
Allocate a data pointers array, samples buffer for nb_samples samples, and fill data pointers and lin...
Definition: samplefmt.c:207
avcodec_parameters_from_context
int avcodec_parameters_from_context(struct AVCodecParameters *par, const AVCodecContext *codec)
Fill the parameters struct based on the values from the supplied codec context.
Definition: codec_par.c:137
AVCodecParameters::codec_id
enum AVCodecID codec_id
Specific type of the encoded data (the codec used).
Definition: codec_par.h:55
AVPacket
This structure stores compressed data.
Definition: packet.h:516
avio_closep
int avio_closep(AVIOContext **s)
Close the resource accessed by the AVIOContext *s, free it and set the pointer pointing to it to NULL...
Definition: avio.c:649
av_freep
#define av_freep(p)
Definition: tableprint_vlc.h:34
AVERROR_EXIT
#define AVERROR_EXIT
Immediate exit was requested; the called function should not be restarted.
Definition: error.h:58
avstring.h
input_data
static void input_data(MLPEncodeContext *ctx, MLPSubstream *s, uint8_t **const samples, int nb_samples)
Wrapper function for inputting data in two different bit-depths.
Definition: mlpenc.c:1224
write_output_file_trailer
static int write_output_file_trailer(AVFormatContext *output_format_context)
Write the trailer of the output file container.
Definition: transcode_aac.c:767