avformat/hlsenc: Add WebVtt support in hls

Signed-off-by: Anshul Maheshwari <er.anshul.maheshwari@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
This commit is contained in:
Anshul Maheshwari 2015-06-04 13:25:52 +05:30 committed by Michael Niedermayer
parent ec7f04c13a
commit 379ddc6af4

View File

@ -42,6 +42,7 @@
typedef struct HLSSegment {
char filename[1024];
char sub_filename[1024];
double duration; /* in seconds */
int64_t pos;
int64_t size;
@ -67,8 +68,10 @@ typedef struct HLSContext {
int64_t sequence;
int64_t start_sequence;
AVOutputFormat *oformat;
AVOutputFormat *vtt_oformat;
AVFormatContext *avf;
AVFormatContext *vtt_avf;
float time; // Set by a private option.
int max_nb_segments; // Set by a private option.
@ -79,6 +82,7 @@ typedef struct HLSContext {
int allowcache;
int64_t recording_time;
int has_video;
int has_subtitle;
int64_t start_pts;
int64_t end_pts;
double duration; // last segment duration computed so far, in seconds
@ -92,8 +96,12 @@ typedef struct HLSContext {
HLSSegment *old_segments;
char *basename;
char *vtt_basename;
char *vtt_m3u8_name;
char *baseurl;
char *format_options_str;
char *vtt_format_options_str;
char *subtitle_filename;
AVDictionary *format_options;
char *key_info_file;
@ -101,14 +109,16 @@ typedef struct HLSContext {
char key_uri[LINE_BUFFER_SIZE + 1];
char key_string[KEYSIZE*2 + 1];
char iv_string[KEYSIZE*2 + 1];
AVDictionary *vtt_format_options;
} HLSContext;
static int hls_delete_old_segments(HLSContext *hls) {
HLSSegment *segment, *previous_segment = NULL;
float playlist_duration = 0.0f;
int ret = 0, path_size;
char *dirname = NULL, *p, *path;
int ret = 0, path_size, sub_path_size;
char *dirname = NULL, *p, *path, *sub_path;
segment = hls->segments;
while (segment) {
@ -150,13 +160,28 @@ static int hls_delete_old_segments(HLSContext *hls) {
ret = AVERROR(ENOMEM);
goto fail;
}
sub_path_size = strlen(dirname) + strlen(segment->sub_filename) + 1;
sub_path = av_malloc(sub_path_size);
if (!sub_path) {
ret = AVERROR(ENOMEM);
goto fail;
}
av_strlcpy(path, dirname, path_size);
av_strlcat(path, segment->filename, path_size);
if (unlink(path) < 0) {
av_log(hls, AV_LOG_ERROR, "failed to delete old segment %s: %s\n",
path, strerror(errno));
}
av_strlcpy(sub_path, dirname, sub_path_size);
av_strlcat(sub_path, segment->sub_filename, sub_path_size);
if (unlink(sub_path) < 0) {
av_log(hls, AV_LOG_ERROR, "failed to delete old segment %s: %s\n",
sub_path, strerror(errno));
}
av_free(path);
av_free(sub_path);
previous_segment = segment;
segment = previous_segment->next;
av_free(previous_segment);
@ -226,6 +251,7 @@ static int hls_mux_init(AVFormatContext *s)
{
HLSContext *hls = s->priv_data;
AVFormatContext *oc;
AVFormatContext *vtt_oc;
int i, ret;
ret = avformat_alloc_output_context2(&hls->avf, hls->oformat, NULL, NULL);
@ -238,9 +264,24 @@ static int hls_mux_init(AVFormatContext *s)
oc->max_delay = s->max_delay;
av_dict_copy(&oc->metadata, s->metadata, 0);
if(hls->vtt_oformat) {
ret = avformat_alloc_output_context2(&hls->vtt_avf, hls->vtt_oformat, NULL, NULL);
if (ret < 0)
return ret;
vtt_oc = hls->vtt_avf;
vtt_oc->oformat = hls->vtt_oformat;
av_dict_copy(&vtt_oc->metadata, s->metadata, 0);
}
for (i = 0; i < s->nb_streams; i++) {
AVStream *st;
if (!(st = avformat_new_stream(oc, NULL)))
AVFormatContext *loc;
if (s->streams[i]->codec->codec_type == AVMEDIA_TYPE_SUBTITLE)
loc = vtt_oc;
else
loc = oc;
if (!(st = avformat_new_stream(loc, NULL)))
return AVERROR(ENOMEM);
avcodec_copy_context(st->codec, s->streams[i]->codec);
st->sample_aspect_ratio = s->streams[i]->sample_aspect_ratio;
@ -263,6 +304,9 @@ static int hls_append_segment(HLSContext *hls, double duration, int64_t pos,
av_strlcpy(en->filename, av_basename(hls->avf->filename), sizeof(en->filename));
if(hls->has_subtitle)
av_strlcpy(en->sub_filename, av_basename(hls->vtt_avf->filename), sizeof(en->sub_filename));
en->duration = duration;
en->pos = pos;
en->size = size;
@ -317,6 +361,7 @@ static int hls_window(AVFormatContext *s, int last)
int target_duration = 0;
int ret = 0;
AVIOContext *out = NULL;
AVIOContext *sub_out = NULL;
char temp_filename[1024];
int64_t sequence = FFMAX(hls->start_sequence, hls->sequence - hls->nb_entries);
int version = hls->flags & HLS_SINGLE_FILE ? 4 : 3;
@ -380,8 +425,39 @@ static int hls_window(AVFormatContext *s, int last)
if (last && (hls->flags & HLS_OMIT_ENDLIST)==0)
avio_printf(out, "#EXT-X-ENDLIST\n");
if( hls->vtt_m3u8_name ) {
if ((ret = avio_open2(&sub_out, hls->vtt_m3u8_name, AVIO_FLAG_WRITE,
&s->interrupt_callback, NULL)) < 0)
goto fail;
avio_printf(sub_out, "#EXTM3U\n");
avio_printf(sub_out, "#EXT-X-VERSION:%d\n", version);
if (hls->allowcache == 0 || hls->allowcache == 1) {
avio_printf(sub_out, "#EXT-X-ALLOW-CACHE:%s\n", hls->allowcache == 0 ? "NO" : "YES");
}
avio_printf(sub_out, "#EXT-X-TARGETDURATION:%d\n", target_duration);
avio_printf(sub_out, "#EXT-X-MEDIA-SEQUENCE:%"PRId64"\n", sequence);
av_log(s, AV_LOG_VERBOSE, "EXT-X-MEDIA-SEQUENCE:%"PRId64"\n",
sequence);
for (en = hls->segments; en; en = en->next) {
avio_printf(sub_out, "#EXTINF:%f,\n", en->duration);
if (hls->flags & HLS_SINGLE_FILE)
avio_printf(sub_out, "#EXT-X-BYTERANGE:%"PRIi64"@%"PRIi64"\n",
en->size, en->pos);
if (hls->baseurl)
avio_printf(sub_out, "%s", hls->baseurl);
avio_printf(sub_out, "%s\n", en->sub_filename);
}
if (last)
avio_printf(sub_out, "#EXT-X-ENDLIST\n");
}
fail:
avio_closep(&out);
avio_closep(&sub_out);
if (ret >= 0 && use_rename)
ff_rename(temp_filename, s->filename, s);
return ret;
@ -391,19 +467,31 @@ static int hls_start(AVFormatContext *s)
{
HLSContext *c = s->priv_data;
AVFormatContext *oc = c->avf;
AVFormatContext *vtt_oc = c->vtt_avf;
AVDictionary *options = NULL;
char *filename, iv_string[KEYSIZE*2 + 1];
int err = 0;
if (c->flags & HLS_SINGLE_FILE)
if (c->flags & HLS_SINGLE_FILE) {
av_strlcpy(oc->filename, c->basename,
sizeof(oc->filename));
else
if (c->vtt_basename)
av_strlcpy(vtt_oc->filename, c->vtt_basename,
sizeof(vtt_oc->filename));
} else {
if (av_get_frame_filename(oc->filename, sizeof(oc->filename),
c->basename, c->wrap ? c->sequence % c->wrap : c->sequence) < 0) {
av_log(oc, AV_LOG_ERROR, "Invalid segment filename template '%s'\n", c->basename);
return AVERROR(EINVAL);
}
if( c->vtt_basename) {
if (av_get_frame_filename(vtt_oc->filename, sizeof(vtt_oc->filename),
c->vtt_basename, c->wrap ? c->sequence % c->wrap : c->sequence) < 0) {
av_log(vtt_oc, AV_LOG_ERROR, "Invalid segment filename template '%s'\n", c->vtt_basename);
return AVERROR(EINVAL);
}
}
}
c->number++;
if (c->key_info_file) {
@ -433,10 +521,18 @@ static int hls_start(AVFormatContext *s)
if ((err = avio_open2(&oc->pb, oc->filename, AVIO_FLAG_WRITE,
&s->interrupt_callback, NULL)) < 0)
return err;
if (c->vtt_basename) {
if ((err = avio_open2(&vtt_oc->pb, vtt_oc->filename, AVIO_FLAG_WRITE,
&s->interrupt_callback, NULL)) < 0)
return err;
}
if (oc->oformat->priv_class && oc->priv_data)
av_opt_set(oc->priv_data, "mpegts_flags", "resend_headers", 0);
if (c->vtt_basename)
avformat_write_header(vtt_oc,NULL);
return 0;
}
@ -446,8 +542,10 @@ static int hls_write_header(AVFormatContext *s)
int ret, i;
char *p;
const char *pattern = "%d.ts";
const char *vtt_pattern = "%d.vtt";
AVDictionary *options = NULL;
int basename_size;
int vtt_basename_size;
hls->sequence = hls->start_sequence;
hls->recording_time = hls->time * AV_TIME_BASE;
@ -461,9 +559,12 @@ static int hls_write_header(AVFormatContext *s)
}
}
for (i = 0; i < s->nb_streams; i++)
for (i = 0; i < s->nb_streams; i++) {
hls->has_video +=
s->streams[i]->codec->codec_type == AVMEDIA_TYPE_VIDEO;
hls->has_subtitle +=
s->streams[i]->codec->codec_type == AVMEDIA_TYPE_SUBTITLE;
}
if (hls->has_video > 1)
av_log(s, AV_LOG_WARNING,
@ -477,6 +578,14 @@ static int hls_write_header(AVFormatContext *s)
goto fail;
}
if(hls->has_subtitle) {
hls->vtt_oformat = av_guess_format("webvtt", NULL, NULL);
if (!hls->oformat) {
ret = AVERROR_MUXER_NOT_FOUND;
goto fail;
}
}
if (hls->segment_filename) {
hls->basename = av_strdup(hls->segment_filename);
if (!hls->basename) {
@ -502,6 +611,35 @@ static int hls_write_header(AVFormatContext *s)
av_strlcat(hls->basename, pattern, basename_size);
}
if(hls->has_subtitle) {
if (hls->flags & HLS_SINGLE_FILE)
vtt_pattern = ".vtt";
vtt_basename_size = strlen(s->filename) + strlen(vtt_pattern) + 1;
hls->vtt_basename = av_malloc(vtt_basename_size);
if (!hls->vtt_basename) {
ret = AVERROR(ENOMEM);
goto fail;
}
hls->vtt_m3u8_name = av_malloc(vtt_basename_size);
if (!hls->vtt_m3u8_name ) {
ret = AVERROR(ENOMEM);
goto fail;
}
av_strlcpy(hls->vtt_basename, s->filename, vtt_basename_size);
p = strrchr(hls->vtt_basename, '.');
if (p)
*p = '\0';
if( hls->subtitle_filename ) {
strcpy(hls->vtt_m3u8_name, hls->subtitle_filename);
} else {
strcpy(hls->vtt_m3u8_name, hls->vtt_basename);
av_strlcat(hls->vtt_m3u8_name, "_vtt.m3u8", vtt_basename_size);
}
av_strlcat(hls->vtt_basename, vtt_pattern, vtt_basename_size);
}
if ((ret = hls_mux_init(s)) < 0)
goto fail;
@ -515,10 +653,19 @@ static int hls_write_header(AVFormatContext *s)
ret = AVERROR(EINVAL);
goto fail;
}
av_assert0(s->nb_streams == hls->avf->nb_streams);
//av_assert0(s->nb_streams == hls->avf->nb_streams);
for (i = 0; i < s->nb_streams; i++) {
AVStream *inner_st = hls->avf->streams[i];
AVStream *inner_st;
AVStream *outer_st = s->streams[i];
if (outer_st->codec->codec_type != AVMEDIA_TYPE_SUBTITLE)
inner_st = hls->avf->streams[i];
else if (hls->vtt_avf)
inner_st = hls->vtt_avf->streams[0];
else {
/* We have subtitle stream, when user dont want */
inner_st = NULL;
continue;
}
avpriv_set_pts_info(outer_st, inner_st->pts_wrap_bits, inner_st->time_base.num, inner_st->time_base.den);
}
fail:
@ -526,8 +673,12 @@ fail:
av_dict_free(&options);
if (ret < 0) {
av_freep(&hls->basename);
av_freep(&hls->vtt_basename);
if (hls->avf)
avformat_free_context(hls->avf);
if (hls->vtt_avf)
avformat_free_context(hls->vtt_avf);
}
return ret;
}
@ -535,12 +686,20 @@ fail:
static int hls_write_packet(AVFormatContext *s, AVPacket *pkt)
{
HLSContext *hls = s->priv_data;
AVFormatContext *oc = hls->avf;
AVFormatContext *oc = NULL;
AVStream *st = s->streams[pkt->stream_index];
int64_t end_pts = hls->recording_time * hls->number;
int is_ref_pkt = 1;
int ret, can_split = 1;
int stream_index = 0;
if( st->codec->codec_type == AVMEDIA_TYPE_SUBTITLE ) {
oc = hls->vtt_avf;
stream_index = 0;
} else {
oc = hls->avf;
stream_index = pkt->stream_index;
}
if (hls->start_pts == AV_NOPTS_VALUE) {
hls->start_pts = pkt->pts;
hls->end_pts = pkt->pts;
@ -579,6 +738,8 @@ static int hls_write_packet(AVFormatContext *s, AVPacket *pkt)
hls->number++;
} else {
avio_closep(&oc->pb);
if (hls->vtt_avf)
avio_close(hls->vtt_avf->pb);
ret = hls_start(s);
}
@ -586,13 +747,16 @@ static int hls_write_packet(AVFormatContext *s, AVPacket *pkt)
if (ret < 0)
return ret;
if( st->codec->codec_type == AVMEDIA_TYPE_SUBTITLE )
oc = hls->vtt_avf;
else
oc = hls->avf;
if ((ret = hls_window(s, 0)) < 0)
return ret;
}
ret = ff_write_chained(oc, pkt->stream_index, pkt, s, 0);
ret = ff_write_chained(oc, stream_index, pkt, s, 0);
return ret;
}
@ -601,6 +765,7 @@ static int hls_write_trailer(struct AVFormatContext *s)
{
HLSContext *hls = s->priv_data;
AVFormatContext *oc = hls->avf;
AVFormatContext *vtt_oc = hls->vtt_avf;
av_write_trailer(oc);
if (oc->pb) {
@ -608,8 +773,22 @@ static int hls_write_trailer(struct AVFormatContext *s)
avio_closep(&oc->pb);
hls_append_segment(hls, hls->duration, hls->start_pos, hls->size);
}
if (vtt_oc) {
if (vtt_oc->pb)
av_write_trailer(vtt_oc);
hls->size = avio_tell(hls->vtt_avf->pb) - hls->start_pos;
avio_closep(&vtt_oc->pb);
}
av_freep(&hls->basename);
avformat_free_context(oc);
if (vtt_oc) {
av_freep(&hls->vtt_basename);
av_freep(&hls->vtt_m3u8_name);
avformat_free_context(vtt_oc);
}
hls->avf = NULL;
hls_window(s, 1);
@ -625,11 +804,13 @@ static const AVOption options[] = {
{"hls_time", "set segment length in seconds", OFFSET(time), AV_OPT_TYPE_FLOAT, {.dbl = 2}, 0, FLT_MAX, E},
{"hls_list_size", "set maximum number of playlist entries", OFFSET(max_nb_segments), AV_OPT_TYPE_INT, {.i64 = 5}, 0, INT_MAX, E},
{"hls_ts_options","set hls mpegts list of options for the container format used for hls", OFFSET(format_options_str), AV_OPT_TYPE_STRING, {.str = NULL}, 0, 0, E},
{"hls_vtt_options","set hls vtt list of options for the container format used for hls", OFFSET(vtt_format_options_str), AV_OPT_TYPE_STRING, {.str = NULL}, 0, 0, E},
{"hls_wrap", "set number after which the index wraps", OFFSET(wrap), AV_OPT_TYPE_INT, {.i64 = 0}, 0, INT_MAX, E},
{"hls_allow_cache", "explicitly set whether the client MAY (1) or MUST NOT (0) cache media segments", OFFSET(allowcache), AV_OPT_TYPE_INT, {.i64 = -1}, INT_MIN, INT_MAX, E},
{"hls_base_url", "url to prepend to each playlist entry", OFFSET(baseurl), AV_OPT_TYPE_STRING, {.str = NULL}, 0, 0, E},
{"hls_segment_filename", "filename template for segment files", OFFSET(segment_filename), AV_OPT_TYPE_STRING, {.str = NULL}, 0, 0, E},
{"hls_key_info_file", "file with key URI and key file path", OFFSET(key_info_file), AV_OPT_TYPE_STRING, {.str = NULL}, 0, 0, E},
{"hls_subtitle_path", "set path of hls subtitles", OFFSET(subtitle_filename), AV_OPT_TYPE_STRING, {.str = NULL}, 0, 0, E},
{"hls_flags", "set flags affecting HLS playlist and media file generation", OFFSET(flags), AV_OPT_TYPE_FLAGS, {.i64 = 0 }, 0, UINT_MAX, E, "flags"},
{"single_file", "generate a single media file indexed with byte ranges", 0, AV_OPT_TYPE_CONST, {.i64 = HLS_SINGLE_FILE }, 0, UINT_MAX, E, "flags"},
{"delete_segments", "delete segment files that are no longer part of the playlist", 0, AV_OPT_TYPE_CONST, {.i64 = HLS_DELETE_SEGMENTS }, 0, UINT_MAX, E, "flags"},
@ -655,6 +836,7 @@ AVOutputFormat ff_hls_muxer = {
.priv_data_size = sizeof(HLSContext),
.audio_codec = AV_CODEC_ID_AAC,
.video_codec = AV_CODEC_ID_H264,
.subtitle_codec = AV_CODEC_ID_WEBVTT,
.flags = AVFMT_NOFILE | AVFMT_ALLOW_FLUSH,
.write_header = hls_write_header,
.write_packet = hls_write_packet,