FFmpeg  5.0
demuxing_decoding.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2012 Stefano Sabatini
3  *
4  * Permission is hereby granted, free of charge, to any person obtaining a copy
5  * of this software and associated documentation files (the "Software"), to deal
6  * in the Software without restriction, including without limitation the rights
7  * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
8  * copies of the Software, and to permit persons to whom the Software is
9  * furnished to do so, subject to the following conditions:
10  *
11  * The above copyright notice and this permission notice shall be included in
12  * all copies or substantial portions of the Software.
13  *
14  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
15  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
16  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
17  * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
18  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
19  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
20  * THE SOFTWARE.
21  */
22 
23 /**
24  * @file
25  * Demuxing and decoding example.
26  *
27  * Show how to use the libavformat and libavcodec API to demux and
28  * decode audio and video data.
29  * @example demuxing_decoding.c
30  */
31 
32 #include <libavutil/imgutils.h>
33 #include <libavutil/samplefmt.h>
34 #include <libavutil/timestamp.h>
35 #include <libavcodec/avcodec.h>
36 #include <libavformat/avformat.h>
37 
38 static AVFormatContext *fmt_ctx = NULL;
40 static int width, height;
41 static enum AVPixelFormat pix_fmt;
42 static AVStream *video_stream = NULL, *audio_stream = NULL;
43 static const char *src_filename = NULL;
44 static const char *video_dst_filename = NULL;
45 static const char *audio_dst_filename = NULL;
46 static FILE *video_dst_file = NULL;
47 static FILE *audio_dst_file = NULL;
48 
49 static uint8_t *video_dst_data[4] = {NULL};
50 static int video_dst_linesize[4];
51 static int video_dst_bufsize;
52 
53 static int video_stream_idx = -1, audio_stream_idx = -1;
54 static AVFrame *frame = NULL;
55 static AVPacket *pkt = NULL;
56 static int video_frame_count = 0;
57 static int audio_frame_count = 0;
58 
60 {
61  if (frame->width != width || frame->height != height ||
62  frame->format != pix_fmt) {
63  /* To handle this change, one could call av_image_alloc again and
64  * decode the following frames into another rawvideo file. */
65  fprintf(stderr, "Error: Width, height and pixel format have to be "
66  "constant in a rawvideo file, but the width, height or "
67  "pixel format of the input video changed:\n"
68  "old: width = %d, height = %d, format = %s\n"
69  "new: width = %d, height = %d, format = %s\n",
73  return -1;
74  }
75 
76  printf("video_frame n:%d coded_n:%d\n",
78 
79  /* copy decoded frame to destination buffer:
80  * this is required since rawvideo expects non aligned data */
82  (const uint8_t **)(frame->data), frame->linesize,
83  pix_fmt, width, height);
84 
85  /* write to rawvideo file */
87  return 0;
88 }
89 
91 {
92  size_t unpadded_linesize = frame->nb_samples * av_get_bytes_per_sample(frame->format);
93  printf("audio_frame n:%d nb_samples:%d pts:%s\n",
96 
97  /* Write the raw audio data samples of the first plane. This works
98  * fine for packed formats (e.g. AV_SAMPLE_FMT_S16). However,
99  * most audio decoders output planar audio, which uses a separate
100  * plane of audio samples for each channel (e.g. AV_SAMPLE_FMT_S16P).
101  * In other words, this code will write only the first audio channel
102  * in these cases.
103  * You should use libswresample or libavfilter to convert the frame
104  * to packed data. */
105  fwrite(frame->extended_data[0], 1, unpadded_linesize, audio_dst_file);
106 
107  return 0;
108 }
109 
110 static int decode_packet(AVCodecContext *dec, const AVPacket *pkt)
111 {
112  int ret = 0;
113 
114  // submit the packet to the decoder
115  ret = avcodec_send_packet(dec, pkt);
116  if (ret < 0) {
117  fprintf(stderr, "Error submitting a packet for decoding (%s)\n", av_err2str(ret));
118  return ret;
119  }
120 
121  // get all the available frames from the decoder
122  while (ret >= 0) {
123  ret = avcodec_receive_frame(dec, frame);
124  if (ret < 0) {
125  // those two return values are special and mean there is no output
126  // frame available, but there were no errors during decoding
127  if (ret == AVERROR_EOF || ret == AVERROR(EAGAIN))
128  return 0;
129 
130  fprintf(stderr, "Error during decoding (%s)\n", av_err2str(ret));
131  return ret;
132  }
133 
134  // write the frame data to output file
135  if (dec->codec->type == AVMEDIA_TYPE_VIDEO)
136  ret = output_video_frame(frame);
137  else
138  ret = output_audio_frame(frame);
139 
141  if (ret < 0)
142  return ret;
143  }
144 
145  return 0;
146 }
147 
148 static int open_codec_context(int *stream_idx,
150 {
151  int ret, stream_index;
152  AVStream *st;
153  const AVCodec *dec = NULL;
154 
155  ret = av_find_best_stream(fmt_ctx, type, -1, -1, NULL, 0);
156  if (ret < 0) {
157  fprintf(stderr, "Could not find %s stream in input file '%s'\n",
159  return ret;
160  } else {
161  stream_index = ret;
162  st = fmt_ctx->streams[stream_index];
163 
164  /* find decoder for the stream */
166  if (!dec) {
167  fprintf(stderr, "Failed to find %s codec\n",
169  return AVERROR(EINVAL);
170  }
171 
172  /* Allocate a codec context for the decoder */
174  if (!*dec_ctx) {
175  fprintf(stderr, "Failed to allocate the %s codec context\n",
177  return AVERROR(ENOMEM);
178  }
179 
180  /* Copy codec parameters from input stream to output codec context */
181  if ((ret = avcodec_parameters_to_context(*dec_ctx, st->codecpar)) < 0) {
182  fprintf(stderr, "Failed to copy %s codec parameters to decoder context\n",
184  return ret;
185  }
186 
187  /* Init the decoders */
188  if ((ret = avcodec_open2(*dec_ctx, dec, NULL)) < 0) {
189  fprintf(stderr, "Failed to open %s codec\n",
191  return ret;
192  }
193  *stream_idx = stream_index;
194  }
195 
196  return 0;
197 }
198 
199 static int get_format_from_sample_fmt(const char **fmt,
200  enum AVSampleFormat sample_fmt)
201 {
202  int i;
203  struct sample_fmt_entry {
204  enum AVSampleFormat sample_fmt; const char *fmt_be, *fmt_le;
205  } sample_fmt_entries[] = {
206  { AV_SAMPLE_FMT_U8, "u8", "u8" },
207  { AV_SAMPLE_FMT_S16, "s16be", "s16le" },
208  { AV_SAMPLE_FMT_S32, "s32be", "s32le" },
209  { AV_SAMPLE_FMT_FLT, "f32be", "f32le" },
210  { AV_SAMPLE_FMT_DBL, "f64be", "f64le" },
211  };
212  *fmt = NULL;
213 
214  for (i = 0; i < FF_ARRAY_ELEMS(sample_fmt_entries); i++) {
215  struct sample_fmt_entry *entry = &sample_fmt_entries[i];
216  if (sample_fmt == entry->sample_fmt) {
217  *fmt = AV_NE(entry->fmt_be, entry->fmt_le);
218  return 0;
219  }
220  }
221 
222  fprintf(stderr,
223  "sample format %s is not supported as output format\n",
224  av_get_sample_fmt_name(sample_fmt));
225  return -1;
226 }
227 
228 int main (int argc, char **argv)
229 {
230  int ret = 0;
231 
232  if (argc != 4) {
233  fprintf(stderr, "usage: %s input_file video_output_file audio_output_file\n"
234  "API example program to show how to read frames from an input file.\n"
235  "This program reads frames from a file, decodes them, and writes decoded\n"
236  "video frames to a rawvideo file named video_output_file, and decoded\n"
237  "audio frames to a rawaudio file named audio_output_file.\n",
238  argv[0]);
239  exit(1);
240  }
241  src_filename = argv[1];
242  video_dst_filename = argv[2];
243  audio_dst_filename = argv[3];
244 
245  /* open input file, and allocate format context */
246  if (avformat_open_input(&fmt_ctx, src_filename, NULL, NULL) < 0) {
247  fprintf(stderr, "Could not open source file %s\n", src_filename);
248  exit(1);
249  }
250 
251  /* retrieve stream information */
252  if (avformat_find_stream_info(fmt_ctx, NULL) < 0) {
253  fprintf(stderr, "Could not find stream information\n");
254  exit(1);
255  }
256 
259 
260  video_dst_file = fopen(video_dst_filename, "wb");
261  if (!video_dst_file) {
262  fprintf(stderr, "Could not open destination file %s\n", video_dst_filename);
263  ret = 1;
264  goto end;
265  }
266 
267  /* allocate image where the decoded image will be put */
272  width, height, pix_fmt, 1);
273  if (ret < 0) {
274  fprintf(stderr, "Could not allocate raw video buffer\n");
275  goto end;
276  }
277  video_dst_bufsize = ret;
278  }
279 
282  audio_dst_file = fopen(audio_dst_filename, "wb");
283  if (!audio_dst_file) {
284  fprintf(stderr, "Could not open destination file %s\n", audio_dst_filename);
285  ret = 1;
286  goto end;
287  }
288  }
289 
290  /* dump input information to stderr */
292 
293  if (!audio_stream && !video_stream) {
294  fprintf(stderr, "Could not find audio or video stream in the input, aborting\n");
295  ret = 1;
296  goto end;
297  }
298 
299  frame = av_frame_alloc();
300  if (!frame) {
301  fprintf(stderr, "Could not allocate frame\n");
302  ret = AVERROR(ENOMEM);
303  goto end;
304  }
305 
306  pkt = av_packet_alloc();
307  if (!pkt) {
308  fprintf(stderr, "Could not allocate packet\n");
309  ret = AVERROR(ENOMEM);
310  goto end;
311  }
312 
313  if (video_stream)
314  printf("Demuxing video from file '%s' into '%s'\n", src_filename, video_dst_filename);
315  if (audio_stream)
316  printf("Demuxing audio from file '%s' into '%s'\n", src_filename, audio_dst_filename);
317 
318  /* read frames from the file */
319  while (av_read_frame(fmt_ctx, pkt) >= 0) {
320  // check if the packet belongs to a stream we are interested in, otherwise
321  // skip it
324  else if (pkt->stream_index == audio_stream_idx)
327  if (ret < 0)
328  break;
329  }
330 
331  /* flush the decoders */
332  if (video_dec_ctx)
334  if (audio_dec_ctx)
336 
337  printf("Demuxing succeeded.\n");
338 
339  if (video_stream) {
340  printf("Play the output video file with the command:\n"
341  "ffplay -f rawvideo -pix_fmt %s -video_size %dx%d %s\n",
344  }
345 
346  if (audio_stream) {
348  int n_channels = audio_dec_ctx->channels;
349  const char *fmt;
350 
351  if (av_sample_fmt_is_planar(sfmt)) {
352  const char *packed = av_get_sample_fmt_name(sfmt);
353  printf("Warning: the sample format the decoder produced is planar "
354  "(%s). This example will output the first channel only.\n",
355  packed ? packed : "?");
356  sfmt = av_get_packed_sample_fmt(sfmt);
357  n_channels = 1;
358  }
359 
360  if ((ret = get_format_from_sample_fmt(&fmt, sfmt)) < 0)
361  goto end;
362 
363  printf("Play the output audio file with the command:\n"
364  "ffplay -f %s -ac %d -ar %d %s\n",
365  fmt, n_channels, audio_dec_ctx->sample_rate,
367  }
368 
369 end:
373  if (video_dst_file)
374  fclose(video_dst_file);
375  if (audio_dst_file)
376  fclose(audio_dst_file);
380 
381  return ret < 0;
382 }
Libavcodec external API header.
Main libavformat public API header.
static uint8_t * video_dst_data[4]
static int audio_stream_idx
static int width
static int output_audio_frame(AVFrame *frame)
static AVCodecContext * video_dec_ctx
static AVPacket * pkt
int main(int argc, char **argv)
static AVFormatContext * fmt_ctx
static int audio_frame_count
static int get_format_from_sample_fmt(const char **fmt, enum AVSampleFormat sample_fmt)
static int output_video_frame(AVFrame *frame)
static enum AVPixelFormat pix_fmt
static const char * video_dst_filename
static int video_dst_bufsize
static int video_dst_linesize[4]
static AVStream * video_stream
static int decode_packet(AVCodecContext *dec, const AVPacket *pkt)
static const char * audio_dst_filename
static AVStream * audio_stream
static int open_codec_context(int *stream_idx, AVCodecContext **dec_ctx, AVFormatContext *fmt_ctx, enum AVMediaType type)
static FILE * video_dst_file
static int video_frame_count
static int height
static const char * src_filename
static AVFrame * frame
static AVCodecContext * audio_dec_ctx
static FILE * audio_dst_file
static int video_stream_idx
static AVCodecContext * dec_ctx
int avcodec_open2(AVCodecContext *avctx, const AVCodec *codec, AVDictionary **options)
Initialize the AVCodecContext to use the given AVCodec.
const AVCodec * avcodec_find_decoder(enum AVCodecID id)
Find a registered decoder with a matching codec ID.
int avcodec_parameters_to_context(AVCodecContext *codec, const AVCodecParameters *par)
Fill the codec context based on the values from the supplied codec parameters.
AVCodecContext * avcodec_alloc_context3(const AVCodec *codec)
Allocate an AVCodecContext and set its fields to default values.
void avcodec_free_context(AVCodecContext **avctx)
Free the codec context and everything associated with it and write NULL to the provided pointer.
int avcodec_receive_frame(AVCodecContext *avctx, AVFrame *frame)
Return decoded output data from a decoder.
int avcodec_send_packet(AVCodecContext *avctx, const AVPacket *avpkt)
Supply raw packet data as input to a decoder.
void av_packet_free(AVPacket **pkt)
Free the packet, if the packet is reference counted, it will be unreferenced first.
void av_packet_unref(AVPacket *pkt)
Wipe the packet.
AVPacket * av_packet_alloc(void)
Allocate an AVPacket and set its fields to default values.
int av_read_frame(AVFormatContext *s, AVPacket *pkt)
Return the next frame of a stream.
int av_find_best_stream(AVFormatContext *ic, enum AVMediaType type, int wanted_stream_nb, int related_stream, const AVCodec **decoder_ret, int flags)
Find the "best" stream in the file.
int avformat_open_input(AVFormatContext **ps, const char *url, const AVInputFormat *fmt, AVDictionary **options)
Open an input stream and read the header.
int avformat_find_stream_info(AVFormatContext *ic, AVDictionary **options)
Read packets of a media file to get stream information.
void avformat_close_input(AVFormatContext **s)
Close an opened input AVFormatContext.
void av_dump_format(AVFormatContext *ic, int index, const char *url, int is_output)
Print detailed information about the input or output format, such as duration, bitrate,...
#define AVERROR_EOF
End of file.
Definition: error.h:57
#define av_err2str(errnum)
Convenience macro, the return value should be used only directly in function arguments but never stan...
Definition: error.h:121
#define AVERROR(e)
Definition: error.h:45
void av_frame_unref(AVFrame *frame)
Unreference all the buffers referenced by frame and reset the frame fields.
void av_frame_free(AVFrame **frame)
Free the frame and any dynamically allocated objects in it, e.g.
AVFrame * av_frame_alloc(void)
Allocate an AVFrame and set its fields to default values.
void av_free(void *ptr)
Free a memory block which has been allocated with a function of av_malloc() or av_realloc() family.
AVMediaType
Definition: avutil.h:199
const char * av_get_media_type_string(enum AVMediaType media_type)
Return a string describing the media_type enum, NULL if media_type is unknown.
@ AVMEDIA_TYPE_AUDIO
Definition: avutil.h:202
@ AVMEDIA_TYPE_VIDEO
Definition: avutil.h:201
void av_image_copy(uint8_t *dst_data[4], int dst_linesizes[4], const uint8_t *src_data[4], const int src_linesizes[4], enum AVPixelFormat pix_fmt, int width, int height)
Copy image in src_data to dst_data.
int av_image_alloc(uint8_t *pointers[4], int linesizes[4], int w, int h, enum AVPixelFormat pix_fmt, int align)
Allocate an image with size w and h and pixel format pix_fmt, and fill pointers and linesizes accordi...
int av_sample_fmt_is_planar(enum AVSampleFormat sample_fmt)
Check if the sample format is planar.
int av_get_bytes_per_sample(enum AVSampleFormat sample_fmt)
Return number of bytes per sample.
const char * av_get_sample_fmt_name(enum AVSampleFormat sample_fmt)
Return the name of sample_fmt, or NULL if sample_fmt is not recognized.
enum AVSampleFormat av_get_packed_sample_fmt(enum AVSampleFormat sample_fmt)
Get the packed alternative form of the given sample format.
AVSampleFormat
Audio sample formats.
Definition: samplefmt.h:58
@ AV_SAMPLE_FMT_FLT
float
Definition: samplefmt.h:63
@ AV_SAMPLE_FMT_S32
signed 32 bits
Definition: samplefmt.h:62
@ AV_SAMPLE_FMT_U8
unsigned 8 bits
Definition: samplefmt.h:60
@ AV_SAMPLE_FMT_DBL
double
Definition: samplefmt.h:64
@ AV_SAMPLE_FMT_S16
signed 16 bits
Definition: samplefmt.h:61
misc image utilities
#define AV_NE(be, le)
Definition: macros.h:33
#define FF_ARRAY_ELEMS(a)
Definition: macros.h:53
const char * av_get_pix_fmt_name(enum AVPixelFormat pix_fmt)
Return the short name for a pixel format, NULL in case pix_fmt is unknown.
AVPixelFormat
Pixel format.
Definition: pixfmt.h:64
main external API structure.
Definition: avcodec.h:383
enum AVPixelFormat pix_fmt
Pixel format, see AV_PIX_FMT_xxx.
Definition: avcodec.h:593
int width
picture width / height.
Definition: avcodec.h:556
enum AVSampleFormat sample_fmt
audio sample format
Definition: avcodec.h:1000
const struct AVCodec * codec
Definition: avcodec.h:392
int sample_rate
samples per second
Definition: avcodec.h:992
AVRational time_base
This is the fundamental unit of time (in seconds) in terms of which frame timestamps are represented.
Definition: avcodec.h:506
int channels
number of audio channels
Definition: avcodec.h:993
enum AVCodecID codec_id
Specific type of the encoded data (the codec used).
Definition: codec_par.h:60
AVCodec.
Definition: codec.h:202
enum AVMediaType type
Definition: codec.h:215
Format I/O context.
Definition: avformat.h:1200
AVStream ** streams
A list of all streams in the file.
Definition: avformat.h:1268
This structure describes decoded (raw) audio or video data.
Definition: frame.h:317
int coded_picture_number
picture number in bitstream order
Definition: frame.h:444
int nb_samples
number of audio samples (per channel) described by this frame
Definition: frame.h:397
int64_t pts
Presentation timestamp in time_base units (time when frame should be shown to user).
Definition: frame.h:424
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
Definition: frame.h:338
int width
Definition: frame.h:389
int height
Definition: frame.h:389
int linesize[AV_NUM_DATA_POINTERS]
For video, a positive or negative value, which is typically indicating the size in bytes of each pict...
Definition: frame.h:362
int format
format of the frame, -1 if unknown or unset Values correspond to enum AVPixelFormat for video frames,...
Definition: frame.h:404
uint8_t ** extended_data
pointers to the data planes/channels.
Definition: frame.h:378
This structure stores compressed data.
Definition: packet.h:350
int stream_index
Definition: packet.h:375
Stream structure.
Definition: avformat.h:935
AVCodecParameters * codecpar
Codec parameters associated with this stream.
Definition: avformat.h:1095
timestamp utils, mostly useful for debugging/logging purposes
#define av_ts2timestr(ts, tb)
Convenience macro, the return value should be used only directly in function arguments but never stan...
Definition: timestamp.h:76