FFmpeg
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 <libavformat/avformat.h>
36 
39 static int width, height;
40 static enum AVPixelFormat pix_fmt;
42 static const char *src_filename = NULL;
43 static const char *video_dst_filename = NULL;
44 static const char *audio_dst_filename = NULL;
45 static FILE *video_dst_file = NULL;
46 static FILE *audio_dst_file = NULL;
47 
48 static uint8_t *video_dst_data[4] = {NULL};
49 static int video_dst_linesize[4];
50 static int video_dst_bufsize;
51 
52 static int video_stream_idx = -1, audio_stream_idx = -1;
53 static AVFrame *frame = NULL;
54 static AVPacket pkt;
55 static int video_frame_count = 0;
56 static int audio_frame_count = 0;
57 
58 static int output_video_frame(AVFrame *frame)
59 {
60  if (frame->width != width || frame->height != height ||
61  frame->format != pix_fmt) {
62  /* To handle this change, one could call av_image_alloc again and
63  * decode the following frames into another rawvideo file. */
64  fprintf(stderr, "Error: Width, height and pixel format have to be "
65  "constant in a rawvideo file, but the width, height or "
66  "pixel format of the input video changed:\n"
67  "old: width = %d, height = %d, format = %s\n"
68  "new: width = %d, height = %d, format = %s\n",
70  frame->width, frame->height,
71  av_get_pix_fmt_name(frame->format));
72  return -1;
73  }
74 
75  printf("video_frame n:%d coded_n:%d\n",
77 
78  /* copy decoded frame to destination buffer:
79  * this is required since rawvideo expects non aligned data */
81  (const uint8_t **)(frame->data), frame->linesize,
82  pix_fmt, width, height);
83 
84  /* write to rawvideo file */
86  return 0;
87 }
88 
89 static int output_audio_frame(AVFrame *frame)
90 {
91  size_t unpadded_linesize = frame->nb_samples * av_get_bytes_per_sample(frame->format);
92  printf("audio_frame n:%d nb_samples:%d pts:%s\n",
93  audio_frame_count++, frame->nb_samples,
95 
96  /* Write the raw audio data samples of the first plane. This works
97  * fine for packed formats (e.g. AV_SAMPLE_FMT_S16). However,
98  * most audio decoders output planar audio, which uses a separate
99  * plane of audio samples for each channel (e.g. AV_SAMPLE_FMT_S16P).
100  * In other words, this code will write only the first audio channel
101  * in these cases.
102  * You should use libswresample or libavfilter to convert the frame
103  * to packed data. */
104  fwrite(frame->extended_data[0], 1, unpadded_linesize, audio_dst_file);
105 
106  return 0;
107 }
108 
109 static int decode_packet(AVCodecContext *dec, const AVPacket *pkt)
110 {
111  int ret = 0;
112 
113  // submit the packet to the decoder
114  ret = avcodec_send_packet(dec, pkt);
115  if (ret < 0) {
116  fprintf(stderr, "Error submitting a packet for decoding (%s)\n", av_err2str(ret));
117  return ret;
118  }
119 
120  // get all the available frames from the decoder
121  while (ret >= 0) {
122  ret = avcodec_receive_frame(dec, frame);
123  if (ret < 0) {
124  // those two return values are special and mean there is no output
125  // frame available, but there were no errors during decoding
126  if (ret == AVERROR_EOF || ret == AVERROR(EAGAIN))
127  return 0;
128 
129  fprintf(stderr, "Error during decoding (%s)\n", av_err2str(ret));
130  return ret;
131  }
132 
133  // write the frame data to output file
134  if (dec->codec->type == AVMEDIA_TYPE_VIDEO)
135  ret = output_video_frame(frame);
136  else
137  ret = output_audio_frame(frame);
138 
139  av_frame_unref(frame);
140  if (ret < 0)
141  return ret;
142  }
143 
144  return 0;
145 }
146 
147 static int open_codec_context(int *stream_idx,
149 {
150  int ret, stream_index;
151  AVStream *st;
152  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 */
173  *dec_ctx = avcodec_alloc_context3(dec);
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, &opts)) < 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 
257  if (open_codec_context(&video_stream_idx, &video_dec_ctx, fmt_ctx, AVMEDIA_TYPE_VIDEO) >= 0) {
258  video_stream = fmt_ctx->streams[video_stream_idx];
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 */
268  width = video_dec_ctx->width;
269  height = video_dec_ctx->height;
270  pix_fmt = video_dec_ctx->pix_fmt;
272  width, height, pix_fmt, 1);
273  if (ret < 0) {
274  fprintf(stderr, "Could not allocate raw video buffer\n");
275  goto end;
276  }
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 */
291  av_dump_format(fmt_ctx, 0, src_filename, 0);
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  /* initialize packet, set data to NULL, let the demuxer fill it */
307  av_init_packet(&pkt);
308  pkt.data = NULL;
309  pkt.size = 0;
310 
311  if (video_stream)
312  printf("Demuxing video from file '%s' into '%s'\n", src_filename, video_dst_filename);
313  if (audio_stream)
314  printf("Demuxing audio from file '%s' into '%s'\n", src_filename, audio_dst_filename);
315 
316  /* read frames from the file */
317  while (av_read_frame(fmt_ctx, &pkt) >= 0) {
318  // check if the packet belongs to a stream we are interested in, otherwise
319  // skip it
320  if (pkt.stream_index == video_stream_idx)
321  ret = decode_packet(video_dec_ctx, &pkt);
322  else if (pkt.stream_index == audio_stream_idx)
323  ret = decode_packet(audio_dec_ctx, &pkt);
324  av_packet_unref(&pkt);
325  if (ret < 0)
326  break;
327  }
328 
329  /* flush the decoders */
330  if (video_dec_ctx)
331  decode_packet(video_dec_ctx, NULL);
332  if (audio_dec_ctx)
334 
335  printf("Demuxing succeeded.\n");
336 
337  if (video_stream) {
338  printf("Play the output video file with the command:\n"
339  "ffplay -f rawvideo -pix_fmt %s -video_size %dx%d %s\n",
342  }
343 
344  if (audio_stream) {
346  int n_channels = audio_dec_ctx->channels;
347  const char *fmt;
348 
349  if (av_sample_fmt_is_planar(sfmt)) {
350  const char *packed = av_get_sample_fmt_name(sfmt);
351  printf("Warning: the sample format the decoder produced is planar "
352  "(%s). This example will output the first channel only.\n",
353  packed ? packed : "?");
354  sfmt = av_get_packed_sample_fmt(sfmt);
355  n_channels = 1;
356  }
357 
358  if ((ret = get_format_from_sample_fmt(&fmt, sfmt)) < 0)
359  goto end;
360 
361  printf("Play the output audio file with the command:\n"
362  "ffplay -f %s -ac %d -ar %d %s\n",
363  fmt, n_channels, audio_dec_ctx->sample_rate,
365  }
366 
367 end:
368  avcodec_free_context(&video_dec_ctx);
370  avformat_close_input(&fmt_ctx);
371  if (video_dst_file)
372  fclose(video_dst_file);
373  if (audio_dst_file)
374  fclose(audio_dst_file);
375  av_frame_free(&frame);
377 
378  return ret < 0;
379 }
static int output_audio_frame(AVFrame *frame)
#define NULL
Definition: coverity.c:32
const struct AVCodec * codec
Definition: avcodec.h:535
static enum AVPixelFormat pix_fmt
This structure describes decoded (raw) audio or video data.
Definition: frame.h:308
misc image utilities
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...
Definition: imgutils.c:216
static AVFormatContext * fmt_ctx
enum AVCodecID codec_id
Specific type of the encoded data (the codec used).
Definition: codec_par.h:60
int size
Definition: packet.h:364
static int audio_frame_count
enum AVPixelFormat pix_fmt
Pixel format, see AV_PIX_FMT_xxx.
Definition: avcodec.h:736
static int open_codec_context(int *stream_idx, AVCodecContext **dec_ctx, AVFormatContext *fmt_ctx, enum AVMediaType type)
static AVStream * video_stream
GLint GLenum type
Definition: opengl_enc.c:104
int avformat_open_input(AVFormatContext **ps, const char *url, ff_const59 AVInputFormat *fmt, AVDictionary **options)
Open an input stream and read the header.
Definition: utils.c:500
static int output_video_frame(AVFrame *frame)
enum AVMediaType type
Definition: codec.h:203
static AVPacket pkt
static int audio_stream_idx
AVCodec.
Definition: codec.h:190
AVRational time_base
This is the fundamental unit of time (in seconds) in terms of which frame timestamps are represented...
Definition: avcodec.h:649
Format I/O context.
Definition: avformat.h:1239
static FILE * audio_dst_file
enum AVSampleFormat sample_fmt
audio sample format
Definition: avcodec.h:1194
uint8_t
AV_SAMPLE_FMT_U8
AVFrame * av_frame_alloc(void)
Allocate an AVFrame and set its fields to default values.
Definition: frame.c:190
static int video_stream_idx
timestamp utils, mostly useful for debugging/logging purposes
static av_cold int end(AVCodecContext *avctx)
Definition: avrndec.c:92
#define AV_NE(be, le)
Definition: common.h:50
int64_t pts
Presentation timestamp in time_base units (time when frame should be shown to user).
Definition: frame.h:401
AVStream ** streams
A list of all streams in the file.
Definition: avformat.h:1307
int avcodec_parameters_to_context(AVCodecContext *codec, const AVCodecParameters *par)
Fill the codec context based on the values from the supplied codec parameters.
Definition: utils.c:2146
static AVFrame * frame
uint8_t * data
Definition: packet.h:363
#define AVERROR_EOF
End of file.
Definition: error.h:55
static uint8_t * video_dst_data[4]
signed 32 bits
Definition: samplefmt.h:62
int av_sample_fmt_is_planar(enum AVSampleFormat sample_fmt)
Check if the sample format is planar.
Definition: samplefmt.c:112
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, streams, container, programs, metadata, side data, codec and time base.
Definition: dump.c:640
int av_find_best_stream(AVFormatContext *ic, enum AVMediaType type, int wanted_stream_nb, int related_stream, AVCodec **decoder_ret, int flags)
Find the "best" stream in the file.
Definition: utils.c:4176
int width
Definition: frame.h:366
static int video_dst_linesize[4]
static const char * audio_dst_filename
static AVStream * audio_stream
void av_frame_free(AVFrame **frame)
Free the frame and any dynamically allocated objects in it, e.g.
Definition: frame.c:203
#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
static AVCodecContext * video_dec_ctx
int avcodec_receive_frame(AVCodecContext *avctx, AVFrame *frame)
Return decoded output data from a decoder.
Definition: decode.c:669
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.
Definition: samplefmt.c:49
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.
Definition: imgutils.c:422
AVDictionary * opts
Definition: movenc.c:50
AVCodecContext * avcodec_alloc_context3(const AVCodec *codec)
Allocate an AVCodecContext and set its fields to default values.
Definition: options.c:172
int width
picture width / height.
Definition: avcodec.h:699
#define av_err2str(errnum)
Convenience macro, the return value should be used only directly in function arguments but never stan...
Definition: error.h:119
static const char * video_dst_filename
#define FF_ARRAY_ELEMS(a)
Stream structure.
Definition: avformat.h:880
int format
format of the frame, -1 if unknown or unset Values correspond to enum AVPixelFormat for video frames...
Definition: frame.h:381
int avcodec_send_packet(AVCodecContext *avctx, const AVPacket *avpkt)
Supply raw packet data as input to a decoder.
Definition: decode.c:606
int coded_picture_number
picture number in bitstream order
Definition: frame.h:422
AVSampleFormat
Audio sample formats.
Definition: samplefmt.h:58
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:187
int sample_rate
samples per second
Definition: avcodec.h:1186
int linesize[AV_NUM_DATA_POINTERS]
For video, size in bytes of each picture line.
Definition: frame.h:339
int main(int argc, char **argv)
main external API structure.
Definition: avcodec.h:526
void av_packet_unref(AVPacket *pkt)
Wipe the packet.
Definition: avpacket.c:606
static int decode_packet(AVCodecContext *dec, const AVPacket *pkt)
static int get_format_from_sample_fmt(const char **fmt, enum AVSampleFormat sample_fmt)
AVMediaType
Definition: avutil.h:199
static const char * src_filename
int avcodec_open2(AVCodecContext *avctx, const AVCodec *codec, AVDictionary **options)
Initialize the AVCodecContext to use the given AVCodec.
Definition: utils.c:550
int av_read_frame(AVFormatContext *s, AVPacket *pkt)
Return the next frame of a stream.
Definition: utils.c:1712
static AVCodecContext * audio_dec_ctx
AVCodec * avcodec_find_decoder(enum AVCodecID id)
Find a registered decoder with a matching codec ID.
Definition: allcodecs.c:937
static int video_frame_count
void av_frame_unref(AVFrame *frame)
Unreference all the buffers referenced by frame and reset the frame fields.
Definition: frame.c:566
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.
Definition: utils.c:76
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
Definition: frame.h:322
int av_get_bytes_per_sample(enum AVSampleFormat sample_fmt)
Return number of bytes per sample.
Definition: samplefmt.c:106
Main libavformat public API header.
enum AVSampleFormat av_get_packed_sample_fmt(enum AVSampleFormat sample_fmt)
Get the packed alternative form of the given sample format.
Definition: samplefmt.c:75
int avformat_find_stream_info(AVFormatContext *ic, AVDictionary **options)
Read packets of a media file to get stream information.
Definition: utils.c:3547
signed 16 bits
Definition: samplefmt.h:61
static AVCodecContext * dec_ctx
static FILE * video_dst_file
void av_init_packet(AVPacket *pkt)
Initialize optional fields of a packet with default values.
Definition: avpacket.c:35
void avformat_close_input(AVFormatContext **s)
Close an opened input AVFormatContext.
Definition: utils.c:4425
#define av_free(p)
int channels
number of audio channels
Definition: avcodec.h:1187
printf("static const uint8_t my_array[100] = {\n")
int height
Definition: frame.h:366
static int video_dst_bufsize
AVCodecParameters * codecpar
Codec parameters associated with this stream.
Definition: avformat.h:1045
static int height
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.
Definition: pixdesc.c:2489
int stream_index
Definition: packet.h:365
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
uint8_t ** extended_data
pointers to the data planes/channels.
Definition: frame.h:355
AVPixelFormat
Pixel format.
Definition: pixfmt.h:64
This structure stores compressed data.
Definition: packet.h:340
int nb_samples
number of audio samples (per channel) described by this frame
Definition: frame.h:374
int i
Definition: input.c:407
static int width