[FFmpeg-devel,v2,1/3] avformat/hlsenc:addition of #EXT-X-MEDIA tag and AUDIO attribute

Submitted by kjeyapal@akamai.com on Nov. 24, 2017, 9:34 a.m.

Details

Message ID 1511516067-19633-1-git-send-email-kjeyapal@akamai.com
State New
Headers show

Commit Message

kjeyapal@akamai.com Nov. 24, 2017, 9:34 a.m.
From: Vishwanath Dixit <vdixit@akamai.com>

---
 doc/muxers.texi      | 12 ++++++++++
 libavformat/hlsenc.c | 63 ++++++++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 73 insertions(+), 2 deletions(-)

Patch hide | download patch | download mbox

diff --git a/doc/muxers.texi b/doc/muxers.texi
index 9d9ca31..7e6d594 100644
--- a/doc/muxers.texi
+++ b/doc/muxers.texi
@@ -831,6 +831,18 @@  be a video only stream with video bitrate 1000k, the second variant stream will
 be an audio only stream with bitrate 64k and the third variant stream will be a
 video only stream with bitrate 256k. Here, three media playlist with file names
 out_1.m3u8, out_2.m3u8 and out_3.m3u8 will be created.
+@example
+ffmpeg -re -i in.ts -b:a:0 32k -b:a:1 64k -b:v:0 1000k -b:v:1 3000k  \
+  -map 0:a -map 0:a -map 0:v -map 0:v -f hls \
+  -var_stream_map "a:0,agroup:aud_low a:1,agroup:aud_high v:0,agroup:aud_low v:1,agroup:aud_high" \
+  -master_pl_name master.m3u8 \
+  http://example.com/live/out.m3u8
+@end example
+This example creates two audio only and two video only variant streams. In
+addition to the #EXT-X-STREAM-INF tag for each variant stream in the master
+playlist, #EXT-X-MEDIA tag is also added for the two audio only variant streams
+and they are mapped to the two video only variant streams with audio group names
+'aud_low' and 'aud_high'.
 
 By default, a single hls variant containing all the encoded streams is created.
 
diff --git a/libavformat/hlsenc.c b/libavformat/hlsenc.c
index 30ccf73..58d7c58 100644
--- a/libavformat/hlsenc.c
+++ b/libavformat/hlsenc.c
@@ -147,6 +147,7 @@  typedef struct VariantStream {
     AVStream **streams;
     unsigned int nb_streams;
     int m3u8_created; /* status of media play-list creation */
+    char *agroup; /* audio group name */
     char *baseurl;
 } VariantStream;
 
@@ -1069,7 +1070,7 @@  static int create_master_playlist(AVFormatContext *s,
                                   VariantStream * const input_vs)
 {
     HLSContext *hls = s->priv_data;
-    VariantStream *vs;
+    VariantStream *vs, *temp_vs;
     AVStream *vid_st, *aud_st;
     AVIOContext *master_pb = 0;
     AVDictionary *options = NULL;
@@ -1104,6 +1105,34 @@  static int create_master_playlist(AVFormatContext *s,
     avio_printf(master_pb, "#EXTM3U\n");
     avio_printf(master_pb, "#EXT-X-VERSION:%d\n", hls->version);
 
+    /* For audio only variant streams add #EXT-X-MEDIA tag with attributes*/
+    for (i = 0; i < hls->nb_varstreams; i++) {
+        vs = &(hls->var_streams[i]);
+
+        if (vs->has_video || vs->has_subtitle || !vs->agroup)
+            continue;
+
+        m3u8_name_size = strlen(vs->m3u8_name) + 1;
+        m3u8_rel_name = av_malloc(m3u8_name_size);
+        if (!m3u8_rel_name) {
+            ret = AVERROR(ENOMEM);
+            goto fail;
+        }
+        av_strlcpy(m3u8_rel_name, vs->m3u8_name, m3u8_name_size);
+        ret = get_relative_url(hls->master_m3u8_url, vs->m3u8_name,
+                               m3u8_rel_name, m3u8_name_size);
+        if (ret < 0) {
+            av_log(s, AV_LOG_ERROR, "Unable to find relative URL\n");
+            goto fail;
+        }
+
+        avio_printf(master_pb, "#EXT-X-MEDIA:TYPE=AUDIO,GROUP-ID=\"group_%s\"",
+                vs->agroup);
+        avio_printf(master_pb, ",NAME=\"audio_0\",DEFAULT=YES,URI=\"%s\"\n",
+                m3u8_rel_name);
+        av_freep(&m3u8_rel_name);
+    }
+
     /* For variant streams with video add #EXT-X-STREAM-INF tag with attributes*/
     for (i = 0; i < hls->nb_varstreams; i++) {
         vs = &(hls->var_streams[i]);
@@ -1136,6 +1165,25 @@  static int create_master_playlist(AVFormatContext *s,
             continue;
         }
 
+        /**
+         * Traverse through the list of audio only rendition streams and find
+         * the rendition which has highest bitrate in the same audio group
+         */
+        if (vs->agroup) {
+            for (j = 0; j < hls->nb_varstreams; j++) {
+                temp_vs = &(hls->var_streams[j]);
+                if (!temp_vs->has_video && !temp_vs->has_subtitle &&
+                    temp_vs->agroup &&
+                    !strcmp(temp_vs->agroup, vs->agroup)) {
+                    if (!aud_st)
+                        aud_st = temp_vs->streams[0];
+                    if (temp_vs->streams[0]->codecpar->bit_rate >
+                            aud_st->codecpar->bit_rate)
+                        aud_st = temp_vs->streams[0];
+                }
+            }
+        }
+
         bandwidth = 0;
         if (vid_st)
             bandwidth += vid_st->codecpar->bit_rate;
@@ -1154,6 +1202,10 @@  static int create_master_playlist(AVFormatContext *s,
         if (vid_st && vid_st->codecpar->width > 0 && vid_st->codecpar->height > 0)
             avio_printf(master_pb, ",RESOLUTION=%dx%d", vid_st->codecpar->width,
                     vid_st->codecpar->height);
+
+        if (vs->agroup && aud_st)
+            avio_printf(master_pb, ",AUDIO=\"group_%s\"", vs->agroup);
+
         avio_printf(master_pb, "\n%s\n\n", m3u8_rel_name);
 
         av_freep(&m3u8_rel_name);
@@ -1546,6 +1598,7 @@  static int parse_variant_stream_mapstring(AVFormatContext *s)
     /**
      * Expected format for var_stream_map string is as below:
      * "a:0,v:0 a:1,v:1"
+     * "a:0,agroup:a0 a:1,agroup:a1 v:0,agroup:a0  v:1,agroup:a1"
      * This string specifies how to group the audio, video and subtitle streams
      * into different variant streams. The variant stream groups are separated
      * by space.
@@ -1554,6 +1607,7 @@  static int parse_variant_stream_mapstring(AVFormatContext *s)
      * respectively. Allowed values are 0 to 9 digits (limited just based on
      * practical usage)
      *
+     * agroup: is key to specify audio group. A string can be given as value.
      */
     p = av_strdup(hls->var_stream_map);
     q = p;
@@ -1592,7 +1646,10 @@  static int parse_variant_stream_mapstring(AVFormatContext *s)
         while (keyval = av_strtok(varstr, ",", &saveptr2)) {
             varstr = NULL;
 
-            if (av_strstart(keyval, "v:", &val)) {
+            if (av_strstart(keyval, "agroup:", &val)) {
+                vs->agroup = av_strdup(val);
+                continue;
+            } else if (av_strstart(keyval, "v:", &val)) {
                 codec_type = AVMEDIA_TYPE_VIDEO;
             } else if (av_strstart(keyval, "a:", &val)) {
                 codec_type = AVMEDIA_TYPE_AUDIO;
@@ -2050,6 +2107,7 @@  fail:
             av_freep(&vs->m3u8_name);
             av_freep(&vs->vtt_m3u8_name);
             av_freep(&vs->streams);
+            av_freep(&vs->agroup);
             av_freep(&vs->baseurl);
             if (vs->avf)
                 avformat_free_context(vs->avf);
@@ -2289,6 +2347,7 @@  static int hls_write_trailer(struct AVFormatContext *s)
     av_free(old_filename);
     av_freep(&vs->m3u8_name);
     av_freep(&vs->streams);
+    av_freep(&vs->agroup);
     av_freep(&vs->baseurl);
     }