2 * Copyright (c) 2012 Stefano Sabatini
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:
11 * The above copyright notice and this permission notice shall be included in
12 * all copies or substantial portions of the Software.
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
25 * Demuxing and decoding example.
27 * Show how to use the libavformat and libavcodec API to demux and
28 * decode audio and video data.
29 * @example demuxing_decoding.c
32 #include <libavutil/imgutils.h>
33 #include <libavutil/samplefmt.h>
34 #include <libavutil/timestamp.h>
35 #include <libavformat/avformat.h>
37 static AVFormatContext *fmt_ctx = NULL;
38 static AVCodecContext *video_dec_ctx = NULL, *audio_dec_ctx;
39 static int width, height;
40 static enum AVPixelFormat pix_fmt;
41 static AVStream *video_stream = NULL, *audio_stream = NULL;
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;
48 static uint8_t *video_dst_data[4] = {NULL};
49 static int video_dst_linesize[4];
50 static int video_dst_bufsize;
52 static int video_stream_idx = -1, audio_stream_idx = -1;
53 static AVFrame *frame = NULL;
55 static int video_frame_count = 0;
56 static int audio_frame_count = 0;
58 /* Enable or disable frame reference counting. You are not supposed to support
59 * both paths in your application but pick the one most appropriate to your
60 * needs. Look for the use of refcount in this example to see what are the
61 * differences of API usage between them. */
62 static int refcount = 0;
64 static int decode_packet(int *got_frame, int cached)
67 int decoded = pkt.size;
71 if (pkt.stream_index == video_stream_idx) {
72 /* decode video frame */
73 ret = avcodec_decode_video2(video_dec_ctx, frame, got_frame, &pkt);
75 fprintf(stderr, "Error decoding video frame (%s)\n", av_err2str(ret));
81 if (frame->width != width || frame->height != height ||
82 frame->format != pix_fmt) {
83 /* To handle this change, one could call av_image_alloc again and
84 * decode the following frames into another rawvideo file. */
85 fprintf(stderr, "Error: Width, height and pixel format have to be "
86 "constant in a rawvideo file, but the width, height or "
87 "pixel format of the input video changed:\n"
88 "old: width = %d, height = %d, format = %s\n"
89 "new: width = %d, height = %d, format = %s\n",
90 width, height, av_get_pix_fmt_name(pix_fmt),
91 frame->width, frame->height,
92 av_get_pix_fmt_name(frame->format));
96 printf("video_frame%s n:%d coded_n:%d\n",
97 cached ? "(cached)" : "",
98 video_frame_count++, frame->coded_picture_number);
100 /* copy decoded frame to destination buffer:
101 * this is required since rawvideo expects non aligned data */
102 av_image_copy(video_dst_data, video_dst_linesize,
103 (const uint8_t **)(frame->data), frame->linesize,
104 pix_fmt, width, height);
106 /* write to rawvideo file */
107 fwrite(video_dst_data[0], 1, video_dst_bufsize, video_dst_file);
109 } else if (pkt.stream_index == audio_stream_idx) {
110 /* decode audio frame */
111 ret = avcodec_decode_audio4(audio_dec_ctx, frame, got_frame, &pkt);
113 fprintf(stderr, "Error decoding audio frame (%s)\n", av_err2str(ret));
116 /* Some audio decoders decode only part of the packet, and have to be
117 * called again with the remainder of the packet data.
118 * Sample: fate-suite/lossless-audio/luckynight-partial.shn
119 * Also, some decoders might over-read the packet. */
120 decoded = FFMIN(ret, pkt.size);
123 size_t unpadded_linesize = frame->nb_samples * av_get_bytes_per_sample(frame->format);
124 printf("audio_frame%s n:%d nb_samples:%d pts:%s\n",
125 cached ? "(cached)" : "",
126 audio_frame_count++, frame->nb_samples,
127 av_ts2timestr(frame->pts, &audio_dec_ctx->time_base));
129 /* Write the raw audio data samples of the first plane. This works
130 * fine for packed formats (e.g. AV_SAMPLE_FMT_S16). However,
131 * most audio decoders output planar audio, which uses a separate
132 * plane of audio samples for each channel (e.g. AV_SAMPLE_FMT_S16P).
133 * In other words, this code will write only the first audio channel
135 * You should use libswresample or libavfilter to convert the frame
137 fwrite(frame->extended_data[0], 1, unpadded_linesize, audio_dst_file);
141 /* If we use frame reference counting, we own the data and need
142 * to de-reference it when we don't use it anymore */
143 if (*got_frame && refcount)
144 av_frame_unref(frame);
149 static int open_codec_context(int *stream_idx,
150 AVCodecContext **dec_ctx, AVFormatContext *fmt_ctx, enum AVMediaType type)
152 int ret, stream_index;
155 AVDictionary *opts = NULL;
157 ret = av_find_best_stream(fmt_ctx, type, -1, -1, NULL, 0);
159 fprintf(stderr, "Could not find %s stream in input file '%s'\n",
160 av_get_media_type_string(type), src_filename);
164 st = fmt_ctx->streams[stream_index];
166 /* find decoder for the stream */
167 dec = avcodec_find_decoder(st->codecpar->codec_id);
169 fprintf(stderr, "Failed to find %s codec\n",
170 av_get_media_type_string(type));
171 return AVERROR(EINVAL);
174 /* Allocate a codec context for the decoder */
175 *dec_ctx = avcodec_alloc_context3(dec);
177 fprintf(stderr, "Failed to allocate the %s codec context\n",
178 av_get_media_type_string(type));
179 return AVERROR(ENOMEM);
182 /* Copy codec parameters from input stream to output codec context */
183 if ((ret = avcodec_parameters_to_context(*dec_ctx, st->codecpar)) < 0) {
184 fprintf(stderr, "Failed to copy %s codec parameters to decoder context\n",
185 av_get_media_type_string(type));
189 /* Init the decoders, with or without reference counting */
190 av_dict_set(&opts, "refcounted_frames", refcount ? "1" : "0", 0);
191 if ((ret = avcodec_open2(*dec_ctx, dec, &opts)) < 0) {
192 fprintf(stderr, "Failed to open %s codec\n",
193 av_get_media_type_string(type));
196 *stream_idx = stream_index;
202 static int get_format_from_sample_fmt(const char **fmt,
203 enum AVSampleFormat sample_fmt)
206 struct sample_fmt_entry {
207 enum AVSampleFormat sample_fmt; const char *fmt_be, *fmt_le;
208 } sample_fmt_entries[] = {
209 { AV_SAMPLE_FMT_U8, "u8", "u8" },
210 { AV_SAMPLE_FMT_S16, "s16be", "s16le" },
211 { AV_SAMPLE_FMT_S32, "s32be", "s32le" },
212 { AV_SAMPLE_FMT_FLT, "f32be", "f32le" },
213 { AV_SAMPLE_FMT_DBL, "f64be", "f64le" },
217 for (i = 0; i < FF_ARRAY_ELEMS(sample_fmt_entries); i++) {
218 struct sample_fmt_entry *entry = &sample_fmt_entries[i];
219 if (sample_fmt == entry->sample_fmt) {
220 *fmt = AV_NE(entry->fmt_be, entry->fmt_le);
226 "sample format %s is not supported as output format\n",
227 av_get_sample_fmt_name(sample_fmt));
231 int main (int argc, char **argv)
233 int ret = 0, got_frame;
235 if (argc != 4 && argc != 5) {
236 fprintf(stderr, "usage: %s [-refcount] input_file video_output_file audio_output_file\n"
237 "API example program to show how to read frames from an input file.\n"
238 "This program reads frames from a file, decodes them, and writes decoded\n"
239 "video frames to a rawvideo file named video_output_file, and decoded\n"
240 "audio frames to a rawaudio file named audio_output_file.\n\n"
241 "If the -refcount option is specified, the program use the\n"
242 "reference counting frame system which allows keeping a copy of\n"
243 "the data for longer than one decode call.\n"
247 if (argc == 5 && !strcmp(argv[1], "-refcount")) {
251 src_filename = argv[1];
252 video_dst_filename = argv[2];
253 audio_dst_filename = argv[3];
255 /* register all formats and codecs */
258 /* open input file, and allocate format context */
259 if (avformat_open_input(&fmt_ctx, src_filename, NULL, NULL) < 0) {
260 fprintf(stderr, "Could not open source file %s\n", src_filename);
264 /* retrieve stream information */
265 if (avformat_find_stream_info(fmt_ctx, NULL) < 0) {
266 fprintf(stderr, "Could not find stream information\n");
270 if (open_codec_context(&video_stream_idx, &video_dec_ctx, fmt_ctx, AVMEDIA_TYPE_VIDEO) >= 0) {
271 video_stream = fmt_ctx->streams[video_stream_idx];
273 video_dst_file = fopen(video_dst_filename, "wb");
274 if (!video_dst_file) {
275 fprintf(stderr, "Could not open destination file %s\n", video_dst_filename);
280 /* allocate image where the decoded image will be put */
281 width = video_dec_ctx->width;
282 height = video_dec_ctx->height;
283 pix_fmt = video_dec_ctx->pix_fmt;
284 ret = av_image_alloc(video_dst_data, video_dst_linesize,
285 width, height, pix_fmt, 1);
287 fprintf(stderr, "Could not allocate raw video buffer\n");
290 video_dst_bufsize = ret;
293 if (open_codec_context(&audio_stream_idx, &audio_dec_ctx, fmt_ctx, AVMEDIA_TYPE_AUDIO) >= 0) {
294 audio_stream = fmt_ctx->streams[audio_stream_idx];
295 audio_dst_file = fopen(audio_dst_filename, "wb");
296 if (!audio_dst_file) {
297 fprintf(stderr, "Could not open destination file %s\n", audio_dst_filename);
303 /* dump input information to stderr */
304 av_dump_format(fmt_ctx, 0, src_filename, 0);
306 if (!audio_stream && !video_stream) {
307 fprintf(stderr, "Could not find audio or video stream in the input, aborting\n");
312 frame = av_frame_alloc();
314 fprintf(stderr, "Could not allocate frame\n");
315 ret = AVERROR(ENOMEM);
319 /* initialize packet, set data to NULL, let the demuxer fill it */
320 av_init_packet(&pkt);
325 printf("Demuxing video from file '%s' into '%s'\n", src_filename, video_dst_filename);
327 printf("Demuxing audio from file '%s' into '%s'\n", src_filename, audio_dst_filename);
329 /* read frames from the file */
330 while (av_read_frame(fmt_ctx, &pkt) >= 0) {
331 AVPacket orig_pkt = pkt;
333 ret = decode_packet(&got_frame, 0);
338 } while (pkt.size > 0);
339 av_packet_unref(&orig_pkt);
342 /* flush cached frames */
346 decode_packet(&got_frame, 1);
349 printf("Demuxing succeeded.\n");
352 printf("Play the output video file with the command:\n"
353 "ffplay -f rawvideo -pix_fmt %s -video_size %dx%d %s\n",
354 av_get_pix_fmt_name(pix_fmt), width, height,
359 enum AVSampleFormat sfmt = audio_dec_ctx->sample_fmt;
360 int n_channels = audio_dec_ctx->channels;
363 if (av_sample_fmt_is_planar(sfmt)) {
364 const char *packed = av_get_sample_fmt_name(sfmt);
365 printf("Warning: the sample format the decoder produced is planar "
366 "(%s). This example will output the first channel only.\n",
367 packed ? packed : "?");
368 sfmt = av_get_packed_sample_fmt(sfmt);
372 if ((ret = get_format_from_sample_fmt(&fmt, sfmt)) < 0)
375 printf("Play the output audio file with the command:\n"
376 "ffplay -f %s -ac %d -ar %d %s\n",
377 fmt, n_channels, audio_dec_ctx->sample_rate,
382 avcodec_free_context(&video_dec_ctx);
383 avcodec_free_context(&audio_dec_ctx);
384 avformat_close_input(&fmt_ctx);
386 fclose(video_dst_file);
388 fclose(audio_dst_file);
389 av_frame_free(&frame);
390 av_free(video_dst_data[0]);