ffmpeg/libavformat/webmdashenc.c

556 lines
21 KiB
C

/*
* WebM DASH Manifest XML muxer
* Copyright (c) 2014 Vignesh Venkatasubramanian
*
* This file is part of FFmpeg.
*
* FFmpeg is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* FFmpeg is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
/*
* WebM DASH Specification:
* https://sites.google.com/a/webmproject.org/wiki/adaptive-streaming/webm-dash-specification
* ISO DASH Specification:
* http://standards.iso.org/ittf/PubliclyAvailableStandards/c065274_ISO_IEC_23009-1_2014.zip
*/
#include <float.h>
#include <stdint.h>
#include <string.h>
#include <time.h>
#include "avformat.h"
#include "matroska.h"
#include "mux.h"
#include "libavutil/avstring.h"
#include "libavutil/dict.h"
#include "libavutil/mem.h"
#include "libavutil/opt.h"
#include "libavutil/time_internal.h"
#include "libavcodec/codec_desc.h"
typedef struct AdaptationSet {
char id[10];
int *streams;
int nb_streams;
} AdaptationSet;
typedef struct WebMDashMuxContext {
const AVClass *class;
char *adaptation_sets;
AdaptationSet *as;
int nb_as;
int representation_id;
int is_live;
int chunk_start_index;
int chunk_duration;
char *utc_timing_url;
double time_shift_buffer_depth;
int minimum_update_period;
} WebMDashMuxContext;
static const char *get_codec_name(int codec_id)
{
return avcodec_descriptor_get(codec_id)->name;
}
static double get_duration(AVFormatContext *s)
{
int i = 0;
double max = 0.0;
for (i = 0; i < s->nb_streams; i++) {
AVDictionaryEntry *duration = av_dict_get(s->streams[i]->metadata,
DURATION, NULL, 0);
if (!duration || atof(duration->value) < 0) continue;
if (atof(duration->value) > max) max = atof(duration->value);
}
return max / 1000;
}
static int write_header(AVFormatContext *s)
{
WebMDashMuxContext *w = s->priv_data;
AVIOContext *pb = s->pb;
double min_buffer_time = 1.0;
avio_printf(pb, "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n");
avio_printf(pb, "<MPD\n");
avio_printf(pb, " xmlns:xsi=\"http://www.w3.org/2001/XMLSchema-instance\"\n");
avio_printf(pb, " xmlns=\"urn:mpeg:DASH:schema:MPD:2011\"\n");
avio_printf(pb, " xsi:schemaLocation=\"urn:mpeg:DASH:schema:MPD:2011\"\n");
avio_printf(pb, " type=\"%s\"\n", w->is_live ? "dynamic" : "static");
if (!w->is_live) {
avio_printf(pb, " mediaPresentationDuration=\"PT%gS\"\n",
get_duration(s));
}
avio_printf(pb, " minBufferTime=\"PT%gS\"\n", min_buffer_time);
avio_printf(pb, " profiles=\"%s\"%s",
w->is_live ? "urn:mpeg:dash:profile:isoff-live:2011" : "urn:mpeg:dash:profile:webm-on-demand:2012",
w->is_live ? "\n" : ">\n");
if (w->is_live) {
time_t local_time = time(NULL);
struct tm gmt_buffer;
struct tm *gmt = gmtime_r(&local_time, &gmt_buffer);
char gmt_iso[21];
if (!strftime(gmt_iso, 21, "%Y-%m-%dT%H:%M:%SZ", gmt)) {
return AVERROR_UNKNOWN;
}
if (s->flags & AVFMT_FLAG_BITEXACT) {
av_strlcpy(gmt_iso, "", 1);
}
avio_printf(pb, " availabilityStartTime=\"%s\"\n", gmt_iso);
avio_printf(pb, " timeShiftBufferDepth=\"PT%gS\"\n", w->time_shift_buffer_depth);
avio_printf(pb, " minimumUpdatePeriod=\"PT%dS\"", w->minimum_update_period);
avio_printf(pb, ">\n");
if (w->utc_timing_url) {
avio_printf(pb, "<UTCTiming\n");
avio_printf(pb, " schemeIdUri=\"urn:mpeg:dash:utc:http-iso:2014\"\n");
avio_printf(pb, " value=\"%s\"/>\n", w->utc_timing_url);
}
}
return 0;
}
static void write_footer(AVFormatContext *s)
{
avio_printf(s->pb, "</MPD>\n");
}
static int subsegment_alignment(AVFormatContext *s, const AdaptationSet *as)
{
int i;
AVDictionaryEntry *gold = av_dict_get(s->streams[as->streams[0]]->metadata,
CUE_TIMESTAMPS, NULL, 0);
if (!gold) return 0;
for (i = 1; i < as->nb_streams; i++) {
AVDictionaryEntry *ts = av_dict_get(s->streams[as->streams[i]]->metadata,
CUE_TIMESTAMPS, NULL, 0);
if (!ts || !av_strstart(ts->value, gold->value, NULL)) return 0;
}
return 1;
}
static int bitstream_switching(AVFormatContext *s, const AdaptationSet *as)
{
int i;
const AVStream *gold_st = s->streams[as->streams[0]];
AVDictionaryEntry *gold_track_num = av_dict_get(gold_st->metadata,
TRACK_NUMBER, NULL, 0);
AVCodecParameters *gold_par = gold_st->codecpar;
if (!gold_track_num) return 0;
for (i = 1; i < as->nb_streams; i++) {
const AVStream *st = s->streams[as->streams[i]];
AVDictionaryEntry *track_num = av_dict_get(st->metadata,
TRACK_NUMBER, NULL, 0);
AVCodecParameters *par = st->codecpar;
if (!track_num ||
!av_strstart(track_num->value, gold_track_num->value, NULL) ||
gold_par->codec_id != par->codec_id ||
gold_par->extradata_size != par->extradata_size ||
(par->extradata_size > 0 &&
memcmp(gold_par->extradata, par->extradata, par->extradata_size))) {
return 0;
}
}
return 1;
}
/*
* Writes a Representation within an Adaptation Set. Returns 0 on success and
* < 0 on failure.
*/
static int write_representation(AVFormatContext *s, AVStream *st, char *id,
int output_width, int output_height,
int output_sample_rate)
{
WebMDashMuxContext *w = s->priv_data;
AVIOContext *pb = s->pb;
const AVCodecParameters *par = st->codecpar;
AVDictionaryEntry *bandwidth = av_dict_get(st->metadata, BANDWIDTH, NULL, 0);
const char *bandwidth_str;
avio_printf(pb, "<Representation id=\"%s\"", id);
if (bandwidth) {
bandwidth_str = bandwidth->value;
} else if (w->is_live) {
// if bandwidth for live was not provided, use a default
bandwidth_str = (par->codec_type == AVMEDIA_TYPE_AUDIO) ? "128000" : "1000000";
} else {
return AVERROR(EINVAL);
}
avio_printf(pb, " bandwidth=\"%s\"", bandwidth_str);
if (par->codec_type == AVMEDIA_TYPE_VIDEO && output_width)
avio_printf(pb, " width=\"%d\"", par->width);
if (par->codec_type == AVMEDIA_TYPE_VIDEO && output_height)
avio_printf(pb, " height=\"%d\"", par->height);
if (par->codec_type == AVMEDIA_TYPE_AUDIO && output_sample_rate)
avio_printf(pb, " audioSamplingRate=\"%d\"", par->sample_rate);
if (w->is_live) {
// For live streams, Codec and Mime Type always go in the Representation tag.
avio_printf(pb, " codecs=\"%s\"", get_codec_name(par->codec_id));
avio_printf(pb, " mimeType=\"%s/webm\"",
par->codec_type == AVMEDIA_TYPE_VIDEO ? "video" : "audio");
// For live streams, subsegments always start with key frames. So this
// is always 1.
avio_printf(pb, " startsWithSAP=\"1\"");
avio_printf(pb, ">");
} else {
AVDictionaryEntry *irange = av_dict_get(st->metadata, INITIALIZATION_RANGE, NULL, 0);
AVDictionaryEntry *cues_start = av_dict_get(st->metadata, CUES_START, NULL, 0);
AVDictionaryEntry *cues_end = av_dict_get(st->metadata, CUES_END, NULL, 0);
AVDictionaryEntry *filename = av_dict_get(st->metadata, FILENAME, NULL, 0);
if (!irange || !cues_start || !cues_end || !filename)
return AVERROR(EINVAL);
avio_printf(pb, ">\n");
avio_printf(pb, "<BaseURL>%s</BaseURL>\n", filename->value);
avio_printf(pb, "<SegmentBase\n");
avio_printf(pb, " indexRange=\"%s-%s\">\n", cues_start->value, cues_end->value);
avio_printf(pb, "<Initialization\n");
avio_printf(pb, " range=\"0-%s\" />\n", irange->value);
avio_printf(pb, "</SegmentBase>\n");
}
avio_printf(pb, "</Representation>\n");
return 0;
}
/*
* Checks if width of all streams are the same. Returns 1 if true, 0 otherwise.
*/
static int check_matching_width(AVFormatContext *s, const AdaptationSet *as)
{
int first_width, i;
if (as->nb_streams < 2) return 1;
first_width = s->streams[as->streams[0]]->codecpar->width;
for (i = 1; i < as->nb_streams; i++)
if (first_width != s->streams[as->streams[i]]->codecpar->width)
return 0;
return 1;
}
/*
* Checks if height of all streams are the same. Returns 1 if true, 0 otherwise.
*/
static int check_matching_height(AVFormatContext *s, const AdaptationSet *as)
{
int first_height, i;
if (as->nb_streams < 2) return 1;
first_height = s->streams[as->streams[0]]->codecpar->height;
for (i = 1; i < as->nb_streams; i++)
if (first_height != s->streams[as->streams[i]]->codecpar->height)
return 0;
return 1;
}
/*
* Checks if sample rate of all streams are the same. Returns 1 if true, 0 otherwise.
*/
static int check_matching_sample_rate(AVFormatContext *s, const AdaptationSet *as)
{
int first_sample_rate, i;
if (as->nb_streams < 2) return 1;
first_sample_rate = s->streams[as->streams[0]]->codecpar->sample_rate;
for (i = 1; i < as->nb_streams; i++)
if (first_sample_rate != s->streams[as->streams[i]]->codecpar->sample_rate)
return 0;
return 1;
}
static void free_adaptation_sets(AVFormatContext *s)
{
WebMDashMuxContext *w = s->priv_data;
int i;
for (i = 0; i < w->nb_as; i++) {
av_freep(&w->as[i].streams);
}
av_freep(&w->as);
w->nb_as = 0;
}
/*
* Parses a live header filename and returns the position of the '_' and '.'
* delimiting <file_description> and <representation_id>.
*
* Name of the header file should conform to the following pattern:
* <file_description>_<representation_id>.hdr where <file_description> can be
* anything. The chunks should be named according to the following pattern:
* <file_description>_<representation_id>_<chunk_number>.chk
*/
static int split_filename(char *filename, char **underscore_pos,
char **period_pos)
{
*underscore_pos = strrchr(filename, '_');
if (!*underscore_pos)
return AVERROR(EINVAL);
*period_pos = strchr(*underscore_pos, '.');
if (!*period_pos)
return AVERROR(EINVAL);
return 0;
}
/*
* Writes an Adaptation Set. Returns 0 on success and < 0 on failure.
*/
static int write_adaptation_set(AVFormatContext *s, int as_index)
{
WebMDashMuxContext *w = s->priv_data;
AdaptationSet *as = &w->as[as_index];
const AVStream *st = s->streams[as->streams[0]];
AVCodecParameters *par = st->codecpar;
AVDictionaryEntry *lang;
AVIOContext *pb = s->pb;
int i;
static const char boolean[2][6] = { "false", "true" };
int subsegmentStartsWithSAP = 1;
// Width, Height and Sample Rate will go in the AdaptationSet tag if they
// are the same for all contained Representations. otherwise, they will go
// on their respective Representation tag. For live streams, they always go
// in the Representation tag.
int width_in_as = 1, height_in_as = 1, sample_rate_in_as = 1;
if (par->codec_type == AVMEDIA_TYPE_VIDEO) {
width_in_as = !w->is_live && check_matching_width (s, as);
height_in_as = !w->is_live && check_matching_height(s, as);
} else {
sample_rate_in_as = !w->is_live && check_matching_sample_rate(s, as);
}
avio_printf(pb, "<AdaptationSet id=\"%s\"", as->id);
avio_printf(pb, " mimeType=\"%s/webm\"",
par->codec_type == AVMEDIA_TYPE_VIDEO ? "video" : "audio");
avio_printf(pb, " codecs=\"%s\"", get_codec_name(par->codec_id));
lang = av_dict_get(st->metadata, "language", NULL, 0);
if (lang)
avio_printf(pb, " lang=\"%s\"", lang->value);
if (par->codec_type == AVMEDIA_TYPE_VIDEO && width_in_as)
avio_printf(pb, " width=\"%d\"", par->width);
if (par->codec_type == AVMEDIA_TYPE_VIDEO && height_in_as)
avio_printf(pb, " height=\"%d\"", par->height);
if (par->codec_type == AVMEDIA_TYPE_AUDIO && sample_rate_in_as)
avio_printf(pb, " audioSamplingRate=\"%d\"", par->sample_rate);
avio_printf(pb, " bitstreamSwitching=\"%s\"",
boolean[bitstream_switching(s, as)]);
avio_printf(pb, " subsegmentAlignment=\"%s\"",
boolean[w->is_live || subsegment_alignment(s, as)]);
for (i = 0; i < as->nb_streams; i++) {
AVDictionaryEntry *kf = av_dict_get(s->streams[as->streams[i]]->metadata,
CLUSTER_KEYFRAME, NULL, 0);
if (!w->is_live && (!kf || !strncmp(kf->value, "0", 1))) subsegmentStartsWithSAP = 0;
}
avio_printf(pb, " subsegmentStartsWithSAP=\"%d\"", subsegmentStartsWithSAP);
avio_printf(pb, ">\n");
if (w->is_live) {
AVDictionaryEntry *filename =
av_dict_get(st->metadata, FILENAME, NULL, 0);
char *underscore_pos, *period_pos;
int ret;
if (!filename)
return AVERROR(EINVAL);
ret = split_filename(filename->value, &underscore_pos, &period_pos);
if (ret) return ret;
*underscore_pos = '\0';
avio_printf(pb, "<ContentComponent id=\"1\" type=\"%s\"/>\n",
par->codec_type == AVMEDIA_TYPE_VIDEO ? "video" : "audio");
avio_printf(pb, "<SegmentTemplate");
avio_printf(pb, " timescale=\"1000\"");
avio_printf(pb, " duration=\"%d\"", w->chunk_duration);
avio_printf(pb, " media=\"%s_$RepresentationID$_$Number$.chk\"",
filename->value);
avio_printf(pb, " startNumber=\"%d\"", w->chunk_start_index);
avio_printf(pb, " initialization=\"%s_$RepresentationID$.hdr\"",
filename->value);
avio_printf(pb, "/>\n");
*underscore_pos = '_';
}
for (i = 0; i < as->nb_streams; i++) {
char buf[25], *representation_id = buf, *underscore_pos, *period_pos;
AVStream *st = s->streams[as->streams[i]];
int ret;
if (w->is_live) {
AVDictionaryEntry *filename =
av_dict_get(st->metadata, FILENAME, NULL, 0);
if (!filename)
return AVERROR(EINVAL);
ret = split_filename(filename->value, &underscore_pos, &period_pos);
if (ret < 0)
return ret;
representation_id = underscore_pos + 1;
*period_pos = '\0';
} else {
snprintf(buf, sizeof(buf), "%d", w->representation_id++);
}
ret = write_representation(s, st, representation_id, !width_in_as,
!height_in_as, !sample_rate_in_as);
if (ret) return ret;
if (w->is_live)
*period_pos = '.';
}
avio_printf(s->pb, "</AdaptationSet>\n");
return 0;
}
static int parse_adaptation_sets(AVFormatContext *s)
{
WebMDashMuxContext *w = s->priv_data;
char *p = w->adaptation_sets;
char *q;
enum { new_set, parsed_id, parsing_streams } state;
if (!w->adaptation_sets) {
av_log(s, AV_LOG_ERROR, "The 'adaptation_sets' option must be set.\n");
return AVERROR(EINVAL);
}
// syntax id=0,streams=0,1,2 id=1,streams=3,4 and so on
state = new_set;
while (1) {
if (*p == '\0') {
if (state == new_set)
break;
else
return AVERROR(EINVAL);
} else if (state == new_set && *p == ' ') {
p++;
continue;
} else if (state == new_set && !strncmp(p, "id=", 3)) {
void *mem = av_realloc(w->as, sizeof(*w->as) * (w->nb_as + 1));
const char *comma;
if (mem == NULL)
return AVERROR(ENOMEM);
w->as = mem;
++w->nb_as;
w->as[w->nb_as - 1].nb_streams = 0;
w->as[w->nb_as - 1].streams = NULL;
p += 3; // consume "id="
q = w->as[w->nb_as - 1].id;
comma = strchr(p, ',');
if (!comma || comma - p >= sizeof(w->as[w->nb_as - 1].id)) {
av_log(s, AV_LOG_ERROR, "'id' in 'adaptation_sets' is malformed.\n");
return AVERROR(EINVAL);
}
while (*p != ',') *q++ = *p++;
*q = 0;
p++;
state = parsed_id;
} else if (state == parsed_id && !strncmp(p, "streams=", 8)) {
p += 8; // consume "streams="
state = parsing_streams;
} else if (state == parsing_streams) {
struct AdaptationSet *as = &w->as[w->nb_as - 1];
int64_t num;
int ret = av_reallocp_array(&as->streams, ++as->nb_streams,
sizeof(*as->streams));
if (ret < 0)
return ret;
num = strtoll(p, &q, 10);
if (!av_isdigit(*p) || (*q != ' ' && *q != '\0' && *q != ',') ||
num < 0 || num >= s->nb_streams) {
av_log(s, AV_LOG_ERROR, "Invalid value for 'streams' in adapation_sets.\n");
return AVERROR(EINVAL);
}
as->streams[as->nb_streams - 1] = num;
if (*q == '\0') break;
if (*q == ' ') state = new_set;
p = ++q;
} else {
return -1;
}
}
return 0;
}
static int webm_dash_manifest_write_header(AVFormatContext *s)
{
int i;
double start = 0.0;
int ret;
WebMDashMuxContext *w = s->priv_data;
for (unsigned i = 0; i < s->nb_streams; i++) {
enum AVCodecID codec_id = s->streams[i]->codecpar->codec_id;
if (codec_id != AV_CODEC_ID_VP8 && codec_id != AV_CODEC_ID_VP9 &&
codec_id != AV_CODEC_ID_AV1 && codec_id != AV_CODEC_ID_VORBIS &&
codec_id != AV_CODEC_ID_OPUS)
return AVERROR(EINVAL);
}
ret = parse_adaptation_sets(s);
if (ret < 0) {
goto fail;
}
ret = write_header(s);
if (ret < 0) {
goto fail;
}
avio_printf(s->pb, "<Period id=\"0\"");
avio_printf(s->pb, " start=\"PT%gS\"", start);
if (!w->is_live) {
avio_printf(s->pb, " duration=\"PT%gS\"", get_duration(s));
}
avio_printf(s->pb, " >\n");
for (i = 0; i < w->nb_as; i++) {
ret = write_adaptation_set(s, i);
if (ret < 0) {
goto fail;
}
}
avio_printf(s->pb, "</Period>\n");
write_footer(s);
fail:
free_adaptation_sets(s);
return ret < 0 ? ret : 0;
}
static int webm_dash_manifest_write_packet(AVFormatContext *s, AVPacket *pkt)
{
return AVERROR_EOF;
}
#define OFFSET(x) offsetof(WebMDashMuxContext, x)
static const AVOption options[] = {
{ "adaptation_sets", "Adaptation sets. Syntax: id=0,streams=0,1,2 id=1,streams=3,4 and so on", OFFSET(adaptation_sets), AV_OPT_TYPE_STRING, { 0 }, 0, 0, AV_OPT_FLAG_ENCODING_PARAM },
{ "live", "create a live stream manifest", OFFSET(is_live), AV_OPT_TYPE_BOOL, {.i64 = 0}, 0, 1, AV_OPT_FLAG_ENCODING_PARAM },
{ "chunk_start_index", "start index of the chunk", OFFSET(chunk_start_index), AV_OPT_TYPE_INT, {.i64 = 0}, 0, INT_MAX, AV_OPT_FLAG_ENCODING_PARAM },
{ "chunk_duration_ms", "duration of each chunk (in milliseconds)", OFFSET(chunk_duration), AV_OPT_TYPE_INT, {.i64 = 1000}, 0, INT_MAX, AV_OPT_FLAG_ENCODING_PARAM },
{ "utc_timing_url", "URL of the page that will return the UTC timestamp in ISO format", OFFSET(utc_timing_url), AV_OPT_TYPE_STRING, { 0 }, 0, 0, AV_OPT_FLAG_ENCODING_PARAM },
{ "time_shift_buffer_depth", "Smallest time (in seconds) shifting buffer for which any Representation is guaranteed to be available.", OFFSET(time_shift_buffer_depth), AV_OPT_TYPE_DOUBLE, { .dbl = 60.0 }, 1.0, DBL_MAX, AV_OPT_FLAG_ENCODING_PARAM },
{ "minimum_update_period", "Minimum Update Period (in seconds) of the manifest.", OFFSET(minimum_update_period), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, INT_MAX, AV_OPT_FLAG_ENCODING_PARAM },
{ NULL },
};
static const AVClass webm_dash_class = {
.class_name = "WebM DASH Manifest muxer",
.item_name = av_default_item_name,
.option = options,
.version = LIBAVUTIL_VERSION_INT,
};
const FFOutputFormat ff_webm_dash_manifest_muxer = {
.p.name = "webm_dash_manifest",
.p.long_name = NULL_IF_CONFIG_SMALL("WebM DASH Manifest"),
.p.mime_type = "application/xml",
.p.extensions = "xml",
.priv_data_size = sizeof(WebMDashMuxContext),
.write_header = webm_dash_manifest_write_header,
.write_packet = webm_dash_manifest_write_packet,
.p.priv_class = &webm_dash_class,
};