mirror of
https://git.ffmpeg.org/ffmpeg.git
synced 2024-12-27 18:02:11 +00:00
30327865f3
In Smooth Streaming, the fragments are addressed by time, and the manifest only stores one list of time offests for all streams, so all streams need to have identical fragment offsets. Warn if this isn't the case, so that the user can fix the files instead of getting failures at runtime when the fragments can't be found. Signed-off-by: Martin Storsjö <martin@martin.st>
555 lines
18 KiB
C
555 lines
18 KiB
C
/*
|
|
* Copyright (c) 2012 Martin Storsjo
|
|
*
|
|
* This file is part of Libav.
|
|
*
|
|
* Libav 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.
|
|
*
|
|
* Libav 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 Libav; if not, write to the Free Software
|
|
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
|
|
*/
|
|
|
|
/*
|
|
* To create a simple file for smooth streaming:
|
|
* avconv <normal input/transcoding options> -movflags frag_keyframe foo.ismv
|
|
* ismindex -n foo foo.ismv
|
|
* This step creates foo.ism and foo.ismc that is required by IIS for
|
|
* serving it.
|
|
*
|
|
* To pre-split files for serving as static files by a web server without
|
|
* any extra server support, create the ismv file as above, and split it:
|
|
* ismindex -split foo.ismv
|
|
* This step creates a file Manifest and directories QualityLevel(...),
|
|
* that can be read directly by a smooth streaming player.
|
|
*/
|
|
|
|
#include <stdio.h>
|
|
#include <string.h>
|
|
#include <sys/stat.h>
|
|
#ifdef _WIN32
|
|
#include <io.h>
|
|
#define mkdir(a, b) mkdir(a)
|
|
#endif
|
|
|
|
#include "libavformat/avformat.h"
|
|
#include "libavutil/intreadwrite.h"
|
|
#include "libavutil/mathematics.h"
|
|
|
|
static int usage(const char *argv0, int ret)
|
|
{
|
|
fprintf(stderr, "%s [-split] [-n basename] file1 [file2] ...\n", argv0);
|
|
return ret;
|
|
}
|
|
|
|
struct MoofOffset {
|
|
int64_t time;
|
|
int64_t offset;
|
|
int duration;
|
|
};
|
|
|
|
struct VideoFile {
|
|
const char *name;
|
|
int64_t duration;
|
|
int bitrate;
|
|
int track_id;
|
|
int is_audio, is_video;
|
|
int width, height;
|
|
int chunks;
|
|
int sample_rate, channels;
|
|
uint8_t *codec_private;
|
|
int codec_private_size;
|
|
struct MoofOffset *offsets;
|
|
int timescale;
|
|
const char *fourcc;
|
|
int blocksize;
|
|
int tag;
|
|
};
|
|
|
|
struct VideoFiles {
|
|
int nb_files;
|
|
int64_t duration;
|
|
struct VideoFile **files;
|
|
int video_file, audio_file;
|
|
int nb_video_files, nb_audio_files;
|
|
};
|
|
|
|
static int copy_tag(AVIOContext *in, AVIOContext *out, int32_t tag_name)
|
|
{
|
|
int32_t size, tag;
|
|
|
|
size = avio_rb32(in);
|
|
tag = avio_rb32(in);
|
|
avio_wb32(out, size);
|
|
avio_wb32(out, tag);
|
|
if (tag != tag_name)
|
|
return -1;
|
|
size -= 8;
|
|
while (size > 0) {
|
|
char buf[1024];
|
|
int len = FFMIN(sizeof(buf), size);
|
|
if (avio_read(in, buf, len) != len)
|
|
break;
|
|
avio_write(out, buf, len);
|
|
size -= len;
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
static int write_fragment(const char *filename, AVIOContext *in)
|
|
{
|
|
AVIOContext *out = NULL;
|
|
int ret;
|
|
|
|
if ((ret = avio_open2(&out, filename, AVIO_FLAG_WRITE, NULL, NULL)) < 0)
|
|
return ret;
|
|
copy_tag(in, out, MKBETAG('m', 'o', 'o', 'f'));
|
|
copy_tag(in, out, MKBETAG('m', 'd', 'a', 't'));
|
|
|
|
avio_flush(out);
|
|
avio_close(out);
|
|
|
|
return ret;
|
|
}
|
|
|
|
static int write_fragments(struct VideoFiles *files, int start_index,
|
|
AVIOContext *in)
|
|
{
|
|
char dirname[100], filename[500];
|
|
int i, j;
|
|
|
|
for (i = start_index; i < files->nb_files; i++) {
|
|
struct VideoFile *vf = files->files[i];
|
|
const char *type = vf->is_video ? "video" : "audio";
|
|
snprintf(dirname, sizeof(dirname), "QualityLevels(%d)", vf->bitrate);
|
|
mkdir(dirname, 0777);
|
|
for (j = 0; j < vf->chunks; j++) {
|
|
snprintf(filename, sizeof(filename), "%s/Fragments(%s=%"PRId64")",
|
|
dirname, type, vf->offsets[j].time);
|
|
avio_seek(in, vf->offsets[j].offset, SEEK_SET);
|
|
write_fragment(filename, in);
|
|
}
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
static int read_tfra(struct VideoFiles *files, int start_index, AVIOContext *f)
|
|
{
|
|
int ret = AVERROR_EOF, track_id;
|
|
int version, fieldlength, i, j;
|
|
int64_t pos = avio_tell(f);
|
|
uint32_t size = avio_rb32(f);
|
|
struct VideoFile *vf = NULL;
|
|
|
|
if (avio_rb32(f) != MKBETAG('t', 'f', 'r', 'a'))
|
|
goto fail;
|
|
version = avio_r8(f);
|
|
avio_rb24(f);
|
|
track_id = avio_rb32(f); /* track id */
|
|
for (i = start_index; i < files->nb_files && !vf; i++)
|
|
if (files->files[i]->track_id == track_id)
|
|
vf = files->files[i];
|
|
if (!vf) {
|
|
/* Ok, continue parsing the next atom */
|
|
ret = 0;
|
|
goto fail;
|
|
}
|
|
fieldlength = avio_rb32(f);
|
|
vf->chunks = avio_rb32(f);
|
|
vf->offsets = av_mallocz(sizeof(*vf->offsets) * vf->chunks);
|
|
if (!vf->offsets) {
|
|
ret = AVERROR(ENOMEM);
|
|
goto fail;
|
|
}
|
|
for (i = 0; i < vf->chunks; i++) {
|
|
if (version == 1) {
|
|
vf->offsets[i].time = avio_rb64(f);
|
|
vf->offsets[i].offset = avio_rb64(f);
|
|
} else {
|
|
vf->offsets[i].time = avio_rb32(f);
|
|
vf->offsets[i].offset = avio_rb32(f);
|
|
}
|
|
for (j = 0; j < ((fieldlength >> 4) & 3) + 1; j++)
|
|
avio_r8(f);
|
|
for (j = 0; j < ((fieldlength >> 2) & 3) + 1; j++)
|
|
avio_r8(f);
|
|
for (j = 0; j < ((fieldlength >> 0) & 3) + 1; j++)
|
|
avio_r8(f);
|
|
if (i > 0)
|
|
vf->offsets[i - 1].duration = vf->offsets[i].time -
|
|
vf->offsets[i - 1].time;
|
|
}
|
|
if (vf->chunks > 0)
|
|
vf->offsets[vf->chunks - 1].duration = vf->duration -
|
|
vf->offsets[vf->chunks - 1].time;
|
|
ret = 0;
|
|
|
|
fail:
|
|
avio_seek(f, pos + size, SEEK_SET);
|
|
return ret;
|
|
}
|
|
|
|
static int read_mfra(struct VideoFiles *files, int start_index,
|
|
const char *file, int split)
|
|
{
|
|
int err = 0;
|
|
AVIOContext *f = NULL;
|
|
int32_t mfra_size;
|
|
|
|
if ((err = avio_open2(&f, file, AVIO_FLAG_READ, NULL, NULL)) < 0)
|
|
goto fail;
|
|
avio_seek(f, avio_size(f) - 4, SEEK_SET);
|
|
mfra_size = avio_rb32(f);
|
|
avio_seek(f, -mfra_size, SEEK_CUR);
|
|
if (avio_rb32(f) != mfra_size) {
|
|
err = AVERROR_INVALIDDATA;
|
|
goto fail;
|
|
}
|
|
if (avio_rb32(f) != MKBETAG('m', 'f', 'r', 'a')) {
|
|
err = AVERROR_INVALIDDATA;
|
|
goto fail;
|
|
}
|
|
while (!read_tfra(files, start_index, f)) {
|
|
/* Empty */
|
|
}
|
|
|
|
if (split)
|
|
write_fragments(files, start_index, f);
|
|
|
|
fail:
|
|
if (f)
|
|
avio_close(f);
|
|
if (err)
|
|
fprintf(stderr, "Unable to read the MFRA atom in %s\n", file);
|
|
return err;
|
|
}
|
|
|
|
static int get_private_data(struct VideoFile *vf, AVCodecContext *codec)
|
|
{
|
|
vf->codec_private_size = codec->extradata_size;
|
|
vf->codec_private = av_mallocz(codec->extradata_size);
|
|
if (!vf->codec_private)
|
|
return AVERROR(ENOMEM);
|
|
memcpy(vf->codec_private, codec->extradata, codec->extradata_size);
|
|
return 0;
|
|
}
|
|
|
|
static int get_video_private_data(struct VideoFile *vf, AVCodecContext *codec)
|
|
{
|
|
AVIOContext *io = NULL;
|
|
uint16_t sps_size, pps_size;
|
|
int err = AVERROR(EINVAL);
|
|
|
|
if (codec->codec_id == CODEC_ID_VC1)
|
|
return get_private_data(vf, codec);
|
|
|
|
avio_open_dyn_buf(&io);
|
|
if (codec->extradata_size < 11 || codec->extradata[0] != 1)
|
|
goto fail;
|
|
sps_size = AV_RB16(&codec->extradata[6]);
|
|
if (11 + sps_size > codec->extradata_size)
|
|
goto fail;
|
|
avio_wb32(io, 0x00000001);
|
|
avio_write(io, &codec->extradata[8], sps_size);
|
|
pps_size = AV_RB16(&codec->extradata[9 + sps_size]);
|
|
if (11 + sps_size + pps_size > codec->extradata_size)
|
|
goto fail;
|
|
avio_wb32(io, 0x00000001);
|
|
avio_write(io, &codec->extradata[11 + sps_size], pps_size);
|
|
err = 0;
|
|
|
|
fail:
|
|
vf->codec_private_size = avio_close_dyn_buf(io, &vf->codec_private);
|
|
return err;
|
|
}
|
|
|
|
static int handle_file(struct VideoFiles *files, const char *file, int split)
|
|
{
|
|
AVFormatContext *ctx = NULL;
|
|
int err = 0, i, orig_files = files->nb_files;
|
|
char errbuf[50], *ptr;
|
|
struct VideoFile *vf;
|
|
|
|
err = avformat_open_input(&ctx, file, NULL, NULL);
|
|
if (err < 0) {
|
|
av_strerror(err, errbuf, sizeof(errbuf));
|
|
fprintf(stderr, "Unable to open %s: %s\n", file, errbuf);
|
|
return 1;
|
|
}
|
|
|
|
err = avformat_find_stream_info(ctx, NULL);
|
|
if (err < 0) {
|
|
av_strerror(err, errbuf, sizeof(errbuf));
|
|
fprintf(stderr, "Unable to identify %s: %s\n", file, errbuf);
|
|
goto fail;
|
|
}
|
|
|
|
if (ctx->nb_streams < 1) {
|
|
fprintf(stderr, "No streams found in %s\n", file);
|
|
goto fail;
|
|
}
|
|
if (!files->duration)
|
|
files->duration = ctx->duration;
|
|
|
|
for (i = 0; i < ctx->nb_streams; i++) {
|
|
AVStream *st = ctx->streams[i];
|
|
vf = av_mallocz(sizeof(*vf));
|
|
files->files = av_realloc(files->files,
|
|
sizeof(*files->files) * (files->nb_files + 1));
|
|
files->files[files->nb_files] = vf;
|
|
|
|
vf->name = file;
|
|
if ((ptr = strrchr(file, '/')) != NULL)
|
|
vf->name = ptr + 1;
|
|
|
|
vf->bitrate = st->codec->bit_rate;
|
|
vf->track_id = st->id;
|
|
vf->timescale = st->time_base.den;
|
|
vf->duration = av_rescale_rnd(ctx->duration, vf->timescale,
|
|
AV_TIME_BASE, AV_ROUND_UP);
|
|
vf->is_audio = st->codec->codec_type == AVMEDIA_TYPE_AUDIO;
|
|
vf->is_video = st->codec->codec_type == AVMEDIA_TYPE_VIDEO;
|
|
|
|
if (!vf->is_audio && !vf->is_video) {
|
|
fprintf(stderr,
|
|
"Track %d in %s is neither video nor audio, skipping\n",
|
|
vf->track_id, file);
|
|
av_freep(&files->files[files->nb_files]);
|
|
continue;
|
|
}
|
|
|
|
if (vf->is_audio) {
|
|
if (files->audio_file < 0)
|
|
files->audio_file = files->nb_files;
|
|
files->nb_audio_files++;
|
|
vf->channels = st->codec->channels;
|
|
vf->sample_rate = st->codec->sample_rate;
|
|
if (st->codec->codec_id == CODEC_ID_AAC) {
|
|
vf->fourcc = "AACL";
|
|
vf->tag = 255;
|
|
vf->blocksize = 4;
|
|
} else if (st->codec->codec_id == CODEC_ID_WMAPRO) {
|
|
vf->fourcc = "WMAP";
|
|
vf->tag = st->codec->codec_tag;
|
|
vf->blocksize = st->codec->block_align;
|
|
}
|
|
get_private_data(vf, st->codec);
|
|
}
|
|
if (vf->is_video) {
|
|
if (files->video_file < 0)
|
|
files->video_file = files->nb_files;
|
|
files->nb_video_files++;
|
|
vf->width = st->codec->width;
|
|
vf->height = st->codec->height;
|
|
if (st->codec->codec_id == CODEC_ID_H264)
|
|
vf->fourcc = "H264";
|
|
else if (st->codec->codec_id == CODEC_ID_VC1)
|
|
vf->fourcc = "WVC1";
|
|
get_video_private_data(vf, st->codec);
|
|
}
|
|
|
|
files->nb_files++;
|
|
}
|
|
|
|
avformat_close_input(&ctx);
|
|
|
|
err = read_mfra(files, orig_files, file, split);
|
|
|
|
fail:
|
|
if (ctx)
|
|
avformat_close_input(&ctx);
|
|
return err;
|
|
}
|
|
|
|
static void output_server_manifest(struct VideoFiles *files,
|
|
const char *basename)
|
|
{
|
|
char filename[1000];
|
|
FILE *out;
|
|
int i;
|
|
|
|
snprintf(filename, sizeof(filename), "%s.ism", basename);
|
|
out = fopen(filename, "w");
|
|
if (!out) {
|
|
perror(filename);
|
|
return;
|
|
}
|
|
fprintf(out, "<?xml version=\"1.0\" encoding=\"utf-8\"?>\n");
|
|
fprintf(out, "<smil xmlns=\"http://www.w3.org/2001/SMIL20/Language\">\n");
|
|
fprintf(out, "\t<head>\n");
|
|
fprintf(out, "\t\t<meta name=\"clientManifestRelativePath\" "
|
|
"content=\"%s.ismc\" />\n", basename);
|
|
fprintf(out, "\t</head>\n");
|
|
fprintf(out, "\t<body>\n");
|
|
fprintf(out, "\t\t<switch>\n");
|
|
for (i = 0; i < files->nb_files; i++) {
|
|
struct VideoFile *vf = files->files[i];
|
|
const char *type = vf->is_video ? "video" : "audio";
|
|
fprintf(out, "\t\t\t<%s src=\"%s\" systemBitrate=\"%d\">\n",
|
|
type, vf->name, vf->bitrate);
|
|
fprintf(out, "\t\t\t\t<param name=\"trackID\" value=\"%d\" "
|
|
"valueType=\"data\" />\n", vf->track_id);
|
|
fprintf(out, "\t\t\t</%s>\n", type);
|
|
}
|
|
fprintf(out, "\t\t</switch>\n");
|
|
fprintf(out, "\t</body>\n");
|
|
fprintf(out, "</smil>\n");
|
|
fclose(out);
|
|
}
|
|
|
|
static void output_client_manifest(struct VideoFiles *files,
|
|
const char *basename, int split)
|
|
{
|
|
char filename[1000];
|
|
FILE *out;
|
|
int i, j;
|
|
|
|
if (split)
|
|
snprintf(filename, sizeof(filename), "Manifest");
|
|
else
|
|
snprintf(filename, sizeof(filename), "%s.ismc", basename);
|
|
out = fopen(filename, "w");
|
|
if (!out) {
|
|
perror(filename);
|
|
return;
|
|
}
|
|
fprintf(out, "<?xml version=\"1.0\" encoding=\"utf-8\"?>\n");
|
|
fprintf(out, "<SmoothStreamingMedia MajorVersion=\"2\" MinorVersion=\"0\" "
|
|
"Duration=\"%"PRId64 "\">\n", files->duration * 10);
|
|
if (files->video_file >= 0) {
|
|
struct VideoFile *vf = files->files[files->video_file];
|
|
struct VideoFile *first_vf = vf;
|
|
int index = 0;
|
|
fprintf(out,
|
|
"\t<StreamIndex Type=\"video\" QualityLevels=\"%d\" "
|
|
"Chunks=\"%d\" "
|
|
"Url=\"QualityLevels({bitrate})/Fragments(video={start time})\">\n",
|
|
files->nb_video_files, vf->chunks);
|
|
for (i = 0; i < files->nb_files; i++) {
|
|
vf = files->files[i];
|
|
if (!vf->is_video)
|
|
continue;
|
|
fprintf(out,
|
|
"\t\t<QualityLevel Index=\"%d\" Bitrate=\"%d\" "
|
|
"FourCC=\"%s\" MaxWidth=\"%d\" MaxHeight=\"%d\" "
|
|
"CodecPrivateData=\"",
|
|
index, vf->bitrate, vf->fourcc, vf->width, vf->height);
|
|
for (j = 0; j < vf->codec_private_size; j++)
|
|
fprintf(out, "%02X", vf->codec_private[j]);
|
|
fprintf(out, "\" />\n");
|
|
index++;
|
|
if (vf->chunks != first_vf->chunks)
|
|
fprintf(stderr, "Mismatched number of video chunks in %s and %s\n",
|
|
vf->name, first_vf->name);
|
|
}
|
|
vf = first_vf;
|
|
for (i = 0; i < vf->chunks; i++) {
|
|
for (j = files->video_file + 1; j < files->nb_files; j++) {
|
|
if (files->files[j]->is_video &&
|
|
vf->offsets[i].duration != files->files[j]->offsets[i].duration)
|
|
fprintf(stderr, "Mismatched duration of video chunk %d in %s and %s\n",
|
|
i, vf->name, files->files[j]->name);
|
|
}
|
|
fprintf(out, "\t\t<c n=\"%d\" d=\"%d\" />\n", i,
|
|
vf->offsets[i].duration);
|
|
}
|
|
fprintf(out, "\t</StreamIndex>\n");
|
|
}
|
|
if (files->audio_file >= 0) {
|
|
struct VideoFile *vf = files->files[files->audio_file];
|
|
struct VideoFile *first_vf = vf;
|
|
int index = 0;
|
|
fprintf(out,
|
|
"\t<StreamIndex Type=\"audio\" QualityLevels=\"%d\" "
|
|
"Chunks=\"%d\" "
|
|
"Url=\"QualityLevels({bitrate})/Fragments(audio={start time})\">\n",
|
|
files->nb_audio_files, vf->chunks);
|
|
for (i = 0; i < files->nb_files; i++) {
|
|
vf = files->files[i];
|
|
if (!vf->is_audio)
|
|
continue;
|
|
fprintf(out,
|
|
"\t\t<QualityLevel Index=\"%d\" Bitrate=\"%d\" "
|
|
"FourCC=\"%s\" SamplingRate=\"%d\" Channels=\"%d\" "
|
|
"BitsPerSample=\"16\" PacketSize=\"%d\" "
|
|
"AudioTag=\"%d\" CodecPrivateData=\"",
|
|
index, vf->bitrate, vf->fourcc, vf->sample_rate,
|
|
vf->channels, vf->blocksize, vf->tag);
|
|
for (j = 0; j < vf->codec_private_size; j++)
|
|
fprintf(out, "%02X", vf->codec_private[j]);
|
|
fprintf(out, "\" />\n");
|
|
index++;
|
|
if (vf->chunks != first_vf->chunks)
|
|
fprintf(stderr, "Mismatched number of audio chunks in %s and %s\n",
|
|
vf->name, first_vf->name);
|
|
}
|
|
vf = first_vf;
|
|
for (i = 0; i < vf->chunks; i++) {
|
|
for (j = files->audio_file + 1; j < files->nb_files; j++) {
|
|
if (files->files[j]->is_audio &&
|
|
vf->offsets[i].duration != files->files[j]->offsets[i].duration)
|
|
fprintf(stderr, "Mismatched duration of audio chunk %d in %s and %s\n",
|
|
i, vf->name, files->files[j]->name);
|
|
}
|
|
fprintf(out, "\t\t<c n=\"%d\" d=\"%d\" />\n",
|
|
i, vf->offsets[i].duration);
|
|
}
|
|
fprintf(out, "\t</StreamIndex>\n");
|
|
}
|
|
fprintf(out, "</SmoothStreamingMedia>\n");
|
|
fclose(out);
|
|
}
|
|
|
|
static void clean_files(struct VideoFiles *files)
|
|
{
|
|
int i;
|
|
for (i = 0; i < files->nb_files; i++) {
|
|
av_freep(&files->files[i]->codec_private);
|
|
av_freep(&files->files[i]->offsets);
|
|
av_freep(&files->files[i]);
|
|
}
|
|
av_freep(&files->files);
|
|
files->nb_files = 0;
|
|
}
|
|
|
|
int main(int argc, char **argv)
|
|
{
|
|
const char *basename = NULL;
|
|
int split = 0, i;
|
|
struct VideoFiles vf = { 0, .video_file = -1, .audio_file = -1 };
|
|
|
|
av_register_all();
|
|
|
|
for (i = 1; i < argc; i++) {
|
|
if (!strcmp(argv[i], "-n")) {
|
|
basename = argv[i + 1];
|
|
i++;
|
|
} else if (!strcmp(argv[i], "-split")) {
|
|
split = 1;
|
|
} else if (argv[i][0] == '-') {
|
|
return usage(argv[0], 1);
|
|
} else {
|
|
if (handle_file(&vf, argv[i], split))
|
|
return 1;
|
|
}
|
|
}
|
|
if (!vf.nb_files || (!basename && !split))
|
|
return usage(argv[0], 1);
|
|
|
|
if (!split)
|
|
output_server_manifest(&vf, basename);
|
|
output_client_manifest(&vf, basename, split);
|
|
|
|
clean_files(&vf);
|
|
|
|
return 0;
|
|
}
|