examples/demuxing: add audio decoding/demuxing

doc/examples/demuxing.c

@@ -25,75 +25,185 @@
  * libavformat demuxing API use example.
  *
  * Show how to use the libavformat and libavcodec API to demux and
- * decode video data.
+ * decode audio and video data.
  */
 
 #include <libavutil/imgutils.h>
+#include <libavutil/samplefmt.h>
 #include <libavutil/timestamp.h>
 #include <libavformat/avformat.h>
 
 static AVFormatContext *fmt_ctx = NULL;
-static AVCodecContext *dec_ctx = NULL;
-static AVCodec *dec = NULL;
-static AVStream *stream = NULL;
+static AVCodecContext *video_dec_ctx = NULL, *audio_dec_ctx;
+static AVStream *video_stream = NULL, *audio_stream = NULL;
 static const char *src_filename = NULL;
-static const char *dst_filename = NULL;
-static FILE *dst_file = NULL;
-static uint8_t *dst_data[4] = {NULL};
-static int dst_linesize[4];
-static int dst_bufsize;
-static int stream_idx;
+static const char *video_dst_filename = NULL;
+static const char *audio_dst_filename = NULL;
+static FILE *video_dst_file = NULL;
+static FILE *audio_dst_file = NULL;
+
+static uint8_t *video_dst_data[4] = {NULL};
+static int      video_dst_linesize[4];
+static int video_dst_bufsize;
+
+static uint8_t **audio_dst_data = NULL;
+static int       audio_dst_linesize;
+static int audio_dst_bufsize;
+
+static int video_stream_idx = -1, audio_stream_idx = -1;
 static AVFrame *frame = NULL;
 static AVPacket pkt;
-static int frame_count = 0;
+static int video_frame_count = 0;
+static int audio_frame_count = 0;
 
 static int decode_packet(int *got_frame, int cached)
 {
-    int ret;
+    int ret = 0;
+
+    if (pkt.stream_index == video_stream_idx) {
+        /* decode video frame */
+        ret = avcodec_decode_video2(video_dec_ctx, frame, got_frame, &pkt);
+        if (ret < 0) {
+            fprintf(stderr, "Error decoding video frame\n");
+            return ret;
+        }
+
+        if (*got_frame) {
+            printf("video_frame%s n:%d coded_n:%d pts:%s\n",
+                   cached ? "(cached)" : "",
+                   video_frame_count++, frame->coded_picture_number,
+                   av_ts2timestr(frame->pts, &video_dec_ctx->time_base));
+
+            /* copy decoded frame to destination buffer:
+             * this is required since rawvideo expects non aligned data */
+            av_image_copy(video_dst_data, video_dst_linesize,
+                          (const uint8_t **)(frame->data), frame->linesize,
+                          video_dec_ctx->pix_fmt, video_dec_ctx->width, video_dec_ctx->height);
+
+            /* write to rawvideo file */
+            fwrite(video_dst_data[0], 1, video_dst_bufsize, video_dst_file);
+        }
+    } else if (pkt.stream_index == audio_stream_idx) {
+        /* decode audio frame */
+        ret = avcodec_decode_audio4(audio_dec_ctx, frame, got_frame, &pkt);
+        if (ret < 0) {
+            fprintf(stderr, "Error decoding audio frame\n");
+            return ret;
+        }
+
+        if (*got_frame) {
+            printf("audio_frame%s n:%d nb_samples:%d pts:%s\n",
+                   cached ? "(cached)" : "",
+                   audio_frame_count++, frame->nb_samples,
+                   av_ts2timestr(frame->pts, &audio_dec_ctx->time_base));
+
+            ret = av_samples_alloc(audio_dst_data, &audio_dst_linesize, frame->channels,
+                                   frame->nb_samples, frame->format, 1);
+            if (ret < 0) {
+                fprintf(stderr, "Could not allocate audio buffer\n");
+                return AVERROR(ENOMEM);
+            }
+
+            /* TODO: extend return code of the av_samples_* functions so that this call is not needed */
+            audio_dst_bufsize =
+                av_samples_get_buffer_size(NULL, frame->channels,
+                                           frame->nb_samples, frame->format, 1);
+
+            /* copy audio data to destination buffer:
+             * this is required since rawaudio expects non aligned data */
+            av_samples_copy(audio_dst_data, frame->data, 0, 0,
+                            frame->nb_samples, frame->channels, frame->format);
 
-    if (pkt.stream_index != stream_idx)
-        return 0;
+            /* write to rawaudio file */
+            fwrite(audio_dst_data[0], 1, audio_dst_bufsize, audio_dst_file);
+            av_freep(&audio_dst_data[0]);
+        }
+    }
+
+    return ret;
+}
+
+static int open_codec_context(int *stream_idx,
+                              AVFormatContext *fmt_ctx, enum AVMediaType type)
+{
+    int ret;
+    AVStream *st;
+    AVCodecContext *dec_ctx = NULL;
+    AVCodec *dec = NULL;
 
-    /* decode video frame */
-    ret = avcodec_decode_video2(dec_ctx, frame, got_frame, &pkt);
+    ret = av_find_best_stream(fmt_ctx, type, -1, -1, NULL, 0);
     if (ret < 0) {
-        fprintf(stderr, "Error decoding video frame\n");
+        fprintf(stderr, "Could not find %s stream in input file '%s'\n",
+                av_get_media_type_string(type), src_filename);
         return ret;
+    } else {
+        *stream_idx = ret;
+        st = fmt_ctx->streams[*stream_idx];
+
+        /* find decoder for the stream */
+        dec_ctx = st->codec;
+        dec = avcodec_find_decoder(dec_ctx->codec_id);
+        if (!dec) {
+            fprintf(stderr, "Failed to find %s codec\n",
+                    av_get_media_type_string(type));
+            return ret;
+        }
+
+        if ((ret = avcodec_open2(dec_ctx, dec, NULL)) < 0) {
+            fprintf(stderr, "Failed to open %s codec\n",
+                    av_get_media_type_string(type));
+            return ret;
+        }
     }
 
-    if (*got_frame) {
-        printf("frame%s n:%d coded_n:%d pts:%s\n",
-               cached ? "(cached)" : "",
-               frame_count++, frame->coded_picture_number,
-               av_ts2timestr(frame->pts, &dec_ctx->time_base));
+    return 0;
+}
 
-        /* copy decoded frame to destination buffer:
-         * this is required since rawvideo expect non aligned data */
-        av_image_copy(dst_data, dst_linesize,
-                      (const uint8_t **)(frame->data), frame->linesize,
-                      dec_ctx->pix_fmt, dec_ctx->width, dec_ctx->height);
+static int get_format_from_sample_fmt(const char **fmt,
+                                      enum AVSampleFormat sample_fmt)
+{
+    int i;
+    struct sample_fmt_entry {
+        enum AVSampleFormat sample_fmt; const char *fmt_be, *fmt_le;
+    } sample_fmt_entries[] = {
+        { AV_SAMPLE_FMT_U8,  "u8",    "u8"    },
+        { AV_SAMPLE_FMT_S16, "s16be", "s16le" },
+        { AV_SAMPLE_FMT_S32, "s32be", "s32le" },
+        { AV_SAMPLE_FMT_FLT, "f32be", "f32le" },
+        { AV_SAMPLE_FMT_DBL, "f64be", "f64le" },
+    };
+    *fmt = NULL;
 
-        /* write to rawvideo file */
-        fwrite(dst_data[0], 1, dst_bufsize, dst_file);
+    for (i = 0; i < FF_ARRAY_ELEMS(sample_fmt_entries); i++) {
+        struct sample_fmt_entry *entry = &sample_fmt_entries[i];
+        if (sample_fmt == entry->sample_fmt) {
+            *fmt = AV_NE(entry->fmt_be, entry->fmt_le);
+            return 0;
+        }
     }
 
-    return ret;
+    fprintf(stderr,
+            "sample format %s is not supported as output format\n",
+            av_get_sample_fmt_name(sample_fmt));
+    return -1;
 }
 
 int main (int argc, char **argv)
 {
-    int ret, got_frame;
+    int ret = 0, got_frame;
 
-    if (argc != 3) {
-        fprintf(stderr, "usage: %s input_file output_file\n"
+    if (argc != 4) {
+        fprintf(stderr, "usage: %s input_file video_output_file audio_output_file\n"
                 "API example program to show how to read frames from an input file.\n"
-                "This program reads frames from a file, decode them, and write them "
-                "to a rawvideo file named like output_file."
+                "This program reads frames from a file, decodes them, and writes decoded\n"
+                "video frames to a rawvideo file named video_output_file, and decoded\n"
+                "audio frames to a rawaudio file named audio_output_file.\n"
                 "\n", argv[0]);
         exit(1);
     }
     src_filename = argv[1];
-    dst_filename = argv[2];
+    video_dst_filename = argv[2];
+    audio_dst_filename = argv[3];
 
     /* register all formats and codecs */
     av_register_all();
@@ -110,60 +220,75 @@ int main (int argc, char **argv)
         exit(1);
     }
 
-    ret = av_find_best_stream(fmt_ctx, AVMEDIA_TYPE_VIDEO, -1, -1, NULL, 0);
-    if (ret < 0) {
-        fprintf(stderr, "Could not find video stream in file\n");
-        goto end;
-    }
-    stream_idx = ret;
-    stream = fmt_ctx->streams[stream_idx];
-
-    /* find decoder for the stream */
-    dec_ctx = stream->codec;
-    dec = avcodec_find_decoder(dec_ctx->codec_id);
-    if (!dec) {
-        fprintf(stderr, "Failed to find any codec\n");
-        ret = 1;
-        goto end;
-    }
+    if (open_codec_context(&video_stream_idx, fmt_ctx, AVMEDIA_TYPE_VIDEO) >= 0) {
+        video_stream = fmt_ctx->streams[video_stream_idx];
+        video_dec_ctx = video_stream->codec;
 
-    if ((ret = avcodec_open2(dec_ctx, dec, NULL)) < 0) {
-        fprintf(stderr, "Failed to open codec\n");
-        goto end;
+        video_dst_file = fopen(video_dst_filename, "wb");
+        if (!video_dst_file) {
+            fprintf(stderr, "Could not open destination file %s\n", video_dst_filename);
+            ret = 1;
+            goto end;
+        }
+
+        /* allocate image where the decoded image will be put */
+        ret = av_image_alloc(video_dst_data, video_dst_linesize,
+                             video_dec_ctx->width, video_dec_ctx->height,
+                             video_dec_ctx->pix_fmt, 1);
+        if (ret < 0) {
+            fprintf(stderr, "Could not allocate raw video buffer\n");
+            goto end;
+        }
+        video_dst_bufsize = ret;
     }
 
     /* dump input information to stderr */
     av_dump_format(fmt_ctx, 0, src_filename, 0);
 
-    dst_file = fopen(dst_filename, "wb");
-    if (!dst_file) {
-        fprintf(stderr, "Could not open destination file %s\n", dst_filename);
-        ret = 1;
-        goto end;
+    if (open_codec_context(&audio_stream_idx, fmt_ctx, AVMEDIA_TYPE_AUDIO) >= 0) {
+        int nb_planes;
+
+        audio_stream = fmt_ctx->streams[audio_stream_idx];
+        audio_dec_ctx = audio_stream->codec;
+        audio_dst_file = fopen(audio_dst_filename, "wb");
+        if (!audio_dst_file) {
+            fprintf(stderr, "Could not open destination file %s\n", video_dst_filename);
+            ret = 1;
+            goto end;
+        }
+
+        nb_planes = av_sample_fmt_is_planar(audio_dec_ctx->sample_fmt) ?
+            audio_dec_ctx->channels : 1;
+        audio_dst_data = av_mallocz(sizeof(uint8_t *) * nb_planes);
+        if (!audio_dst_data) {
+            fprintf(stderr, "Could not allocate audio data buffers\n");
+            ret = AVERROR(ENOMEM);
+            goto end;
+        }
     }
 
-    frame = avcodec_alloc_frame();
-    if (!frame) {
-        fprintf(stderr, "Could not allocate video frame\n");
+    if (!audio_stream && !video_stream) {
+        fprintf(stderr, "Could not find audio or video stream in the input, aborting\n");
         ret = 1;
         goto end;
     }
 
-    /* allocate image where the decoded image will be put */
-    ret = av_image_alloc(dst_data, dst_linesize,
-                         dec_ctx->width, dec_ctx->height, dec_ctx->pix_fmt, 1);
-    if (ret < 0) {
-        fprintf(stderr, "Could not alloc raw video buffer\n");
+    frame = avcodec_alloc_frame();
+    if (!frame) {
+        fprintf(stderr, "Could not allocate frame\n");
+        ret = AVERROR(ENOMEM);
         goto end;
     }
-    dst_bufsize = ret;
 
     /* initialize packet, set data to NULL, let the demuxer fill it */
     av_init_packet(&pkt);
-    pkt.size = 0;
     pkt.data = NULL;
+    pkt.size = 0;
 
-    printf("Demuxing file '%s' to '%s'\n", src_filename, dst_filename);
+    if (video_stream)
+        printf("Demuxing video from file '%s' into '%s'\n", src_filename, video_dst_filename);
+    if (audio_stream)
+        printf("Demuxing video from file '%s' into '%s'\n", src_filename, audio_dst_filename);
 
     /* read frames from the file */
     while (av_read_frame(fmt_ctx, &pkt) >= 0)
@@ -176,18 +301,39 @@ int main (int argc, char **argv)
         decode_packet(&got_frame, 1);
     } while (got_frame);
 
-    printf("Demuxing succeeded. Play the output file with the command:\n"
-           "ffplay -f rawvideo -pix_fmt %s -video_size %dx%d %s\n",
-           av_get_pix_fmt_name(dec_ctx->pix_fmt), dec_ctx->width, dec_ctx->height,
-           dst_filename);
+    printf("Demuxing succeeded.\n");
+
+    if (video_stream) {
+        printf("Play the output video file with the command:\n"
+               "ffplay -f rawvideo -pix_fmt %s -video_size %dx%d %s\n",
+               av_get_pix_fmt_name(video_dec_ctx->pix_fmt), video_dec_ctx->width, video_dec_ctx->height,
+               video_dst_filename);
+    }
+
+    if (audio_stream) {
+        const char *fmt;
+
+        if ((ret = get_format_from_sample_fmt(&fmt, audio_dec_ctx->sample_fmt) < 0))
+            goto end;
+        printf("Play the output audio file with the command:\n"
+               "ffplay -f %s -ac %d -ar %d %s\n",
+               fmt, audio_dec_ctx->channels, audio_dec_ctx->sample_rate,
+               audio_dst_filename);
+    }
 
 end:
-    avcodec_close(dec_ctx);
+    if (video_dec_ctx)
+        avcodec_close(video_dec_ctx);
+    if (audio_dec_ctx)
+        avcodec_close(audio_dec_ctx);
     avformat_close_input(&fmt_ctx);
-    if (dst_file)
-        fclose(dst_file);
+    if (video_dst_file)
+        fclose(video_dst_file);
+    if (audio_dst_file)
+        fclose(audio_dst_file);
     av_free(frame);
-    av_free(dst_data[0]);
+    av_free(video_dst_data[0]);
+    av_free(audio_dst_data);
 
     return ret < 0;
 }