ffmpeg/libavformat/sdp.c

924 lines
32 KiB
C
Raw Permalink Normal View History

/*
* copyright (c) 2007 Luca Abeni
*
* 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
*/
#include "config_components.h"
#include <string.h>
#include "libavutil/avstring.h"
#include "libavutil/base64.h"
#include "libavutil/dict.h"
#include "libavutil/mem.h"
#include "libavutil/parseutils.h"
#include "libavutil/opt.h"
#include "libavcodec/xiph.h"
2011-05-18 15:21:48 +02:00
#include "libavcodec/mpeg4audio.h"
#include "avformat.h"
#include "internal.h"
#include "avc.h"
#include "hevc.h"
#include "rtp.h"
#include "version.h"
#if CONFIG_NETWORK
#include "network.h"
#endif
#if CONFIG_RTP_MUXER
#define MAX_EXTRADATA_SIZE ((INT_MAX - 10) / 2)
struct sdp_session_level {
int sdp_version; /**< protocol version (currently 0) */
int id; /**< session ID */
int version; /**< session version */
int start_time; /**< session start time (NTP time, in seconds),
or 0 in case of permanent session */
int end_time; /**< session end time (NTP time, in seconds),
or 0 if the session is not bounded */
int ttl; /**< TTL, in case of multicast stream */
const char *user; /**< username of the session's creator */
const char *src_addr; /**< IP address of the machine from which the session was created */
const char *src_type; /**< address type of src_addr */
const char *dst_addr; /**< destination IP address (can be multicast) */
const char *dst_type; /**< destination IP address type */
const char *name; /**< session name (can be an empty string) */
};
static void sdp_write_address(char *buff, int size, const char *dest_addr,
const char *dest_type, int ttl)
{
if (dest_addr) {
if (!dest_type)
dest_type = "IP4";
if (ttl > 0 && !strcmp(dest_type, "IP4")) {
/* The TTL should only be specified for IPv4 multicast addresses,
* not for IPv6. */
av_strlcatf(buff, size, "c=IN %s %s/%d\r\n", dest_type, dest_addr, ttl);
} else {
av_strlcatf(buff, size, "c=IN %s %s\r\n", dest_type, dest_addr);
}
}
}
static void sdp_write_header(char *buff, int size, struct sdp_session_level *s)
{
av_strlcatf(buff, size, "v=%d\r\n"
"o=- %d %d IN %s %s\r\n"
"s=%s\r\n",
s->sdp_version,
s->id, s->version, s->src_type, s->src_addr,
s->name);
sdp_write_address(buff, size, s->dst_addr, s->dst_type, s->ttl);
av_strlcatf(buff, size, "t=%d %d\r\n"
"a=tool:libavformat " AV_STRINGIFY(LIBAVFORMAT_VERSION) "\r\n",
s->start_time, s->end_time);
}
#if CONFIG_NETWORK
static int resolve_destination(char *dest_addr, int size, char *type,
int type_size)
{
struct addrinfo hints = { 0 }, *ai;
int is_multicast;
av_strlcpy(type, "IP4", type_size);
if (!dest_addr[0])
return 0;
/* Resolve the destination, since it must be written
* as a numeric IP address in the SDP. */
if (getaddrinfo(dest_addr, NULL, &hints, &ai))
return 0;
getnameinfo(ai->ai_addr, ai->ai_addrlen, dest_addr, size,
NULL, 0, NI_NUMERICHOST);
#ifdef AF_INET6
if (ai->ai_family == AF_INET6)
av_strlcpy(type, "IP6", type_size);
#endif
is_multicast = ff_is_multicast_address(ai->ai_addr);
freeaddrinfo(ai);
return is_multicast;
}
#else
static int resolve_destination(char *dest_addr, int size, char *type,
int type_size)
{
return 0;
}
#endif
static int sdp_get_address(char *dest_addr, int size, int *ttl, const char *url)
{
int port;
const char *p;
char proto[32];
av_url_split(proto, sizeof(proto), NULL, 0, dest_addr, size, &port, NULL, 0, url);
*ttl = 0;
if (strcmp(proto, "rtp") && strcmp(proto, "srtp")) {
/* The url isn't for the actual rtp sessions,
* don't parse out anything else than the destination.
*/
return 0;
}
p = strchr(url, '?');
if (p) {
char buff[64];
if (av_find_info_tag(buff, sizeof(buff), "ttl", p)) {
*ttl = strtol(buff, NULL, 10);
} else {
*ttl = 5;
}
}
return port;
}
#define MAX_PSET_SIZE 1024
static int extradata2psets(AVFormatContext *s, const AVCodecParameters *par,
char **out)
{
char *psets, *p;
const uint8_t *r;
static const char pset_string[] = "; sprop-parameter-sets=";
static const char profile_string[] = "; profile-level-id=";
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
uint8_t *extradata = par->extradata;
int extradata_size = par->extradata_size;
uint8_t *tmpbuf = NULL;
const uint8_t *sps = NULL, *sps_end;
*out = NULL;
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
if (par->extradata_size > MAX_EXTRADATA_SIZE) {
av_log(s, AV_LOG_ERROR, "Too much extradata!\n");
return AVERROR_INVALIDDATA;
}
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
if (par->extradata[0] == 1) {
int ret = ff_avc_write_annexb_extradata(par->extradata, &extradata,
&extradata_size);
if (ret < 0)
return ret;
tmpbuf = extradata;
}
psets = av_mallocz(MAX_PSET_SIZE);
if (!psets) {
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
av_log(s, AV_LOG_ERROR, "Cannot allocate memory for the parameter sets.\n");
av_free(tmpbuf);
return AVERROR(ENOMEM);
}
memcpy(psets, pset_string, strlen(pset_string));
p = psets + strlen(pset_string);
r = ff_avc_find_startcode(extradata, extradata + extradata_size);
while (r < extradata + extradata_size) {
const uint8_t *r1;
uint8_t nal_type;
while (!*(r++));
nal_type = *r & 0x1f;
r1 = ff_avc_find_startcode(r, extradata + extradata_size);
if (nal_type != 7 && nal_type != 8) { /* Only output SPS and PPS */
r = r1;
continue;
}
if (p != (psets + strlen(pset_string))) {
*p = ',';
p++;
}
if (!sps) {
sps = r;
sps_end = r1;
}
if (!av_base64_encode(p, MAX_PSET_SIZE - (p - psets), r, r1 - r)) {
av_log(s, AV_LOG_ERROR, "Cannot Base64-encode %"PTRDIFF_SPECIFIER" %"PTRDIFF_SPECIFIER"!\n",
MAX_PSET_SIZE - (p - psets), r1 - r);
av_free(psets);
av_free(tmpbuf);
return AVERROR_INVALIDDATA;
}
p += strlen(p);
r = r1;
}
if (sps && sps_end - sps >= 4 && p - psets <= MAX_PSET_SIZE - strlen(profile_string) - 7) {
memcpy(p, profile_string, strlen(profile_string));
p += strlen(p);
ff_data_to_hex(p, sps + 1, 3, 0);
}
av_free(tmpbuf);
*out = psets;
return 0;
}
static int extradata2psets_hevc(const AVCodecParameters *par, char **out)
{
char *psets;
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
uint8_t *extradata = par->extradata;
int extradata_size = par->extradata_size;
uint8_t *tmpbuf = NULL;
int ps_pos[3] = { 0 };
static const char * const ps_names[3] = { "vps", "sps", "pps" };
int num_arrays, num_nalus;
int pos, i, j, ret = 0;
*out = NULL;
// Convert to hvcc format. Since we need to group multiple NALUs of
// the same type, and we might need to convert from one format to the
// other anyway, we get away with a little less work by using the hvcc
// format.
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
if (par->extradata[0] != 1) {
AVIOContext *pb;
ret = avio_open_dyn_buf(&pb);
if (ret < 0)
return ret;
ret = ff_isom_write_hvcc(pb, par->extradata, par->extradata_size, 0);
if (ret < 0) {
avio_close_dyn_buf(pb, &tmpbuf);
goto err;
}
extradata_size = avio_close_dyn_buf(pb, &extradata);
tmpbuf = extradata;
}
if (extradata_size < 23)
goto err;
num_arrays = extradata[22];
pos = 23;
for (i = 0; i < num_arrays; i++) {
int num_nalus, nalu_type;
if (pos + 3 > extradata_size)
goto err;
nalu_type = extradata[pos] & 0x3f;
// Not including libavcodec/hevc.h to avoid confusion between
// NAL_* with the same name for both H.264 and HEVC.
if (nalu_type == 32) // VPS
ps_pos[0] = pos;
else if (nalu_type == 33) // SPS
ps_pos[1] = pos;
else if (nalu_type == 34) // PPS
ps_pos[2] = pos;
num_nalus = AV_RB16(&extradata[pos + 1]);
pos += 3;
for (j = 0; j < num_nalus; j++) {
int len;
if (pos + 2 > extradata_size)
goto err;
len = AV_RB16(&extradata[pos]);
pos += 2;
if (pos + len > extradata_size)
goto err;
pos += len;
}
}
if (!ps_pos[0] || !ps_pos[1] || !ps_pos[2])
goto err;
psets = av_mallocz(MAX_PSET_SIZE);
if (!psets) {
ret = AVERROR(ENOMEM);
goto err;
}
psets[0] = '\0';
for (i = 0; i < 3; i++) {
pos = ps_pos[i];
if (i > 0)
av_strlcat(psets, "; ", MAX_PSET_SIZE);
av_strlcatf(psets, MAX_PSET_SIZE, "sprop-%s=", ps_names[i]);
// Skipping boundary checks in the input here; we've already traversed
// the whole hvcc structure above without issues
num_nalus = AV_RB16(&extradata[pos + 1]);
pos += 3;
for (j = 0; j < num_nalus; j++) {
int len = AV_RB16(&extradata[pos]);
int strpos;
pos += 2;
if (j > 0)
av_strlcat(psets, ",", MAX_PSET_SIZE);
strpos = strlen(psets);
if (!av_base64_encode(psets + strpos, MAX_PSET_SIZE - strpos,
&extradata[pos], len)) {
av_free(psets);
goto err;
}
pos += len;
}
}
av_free(tmpbuf);
*out = psets;
return 0;
err:
if (ret >= 0)
ret = AVERROR_INVALIDDATA;
av_free(tmpbuf);
return ret;
}
static int extradata2config(AVFormatContext *s, const AVCodecParameters *par,
char **out)
{
char *config;
*out = NULL;
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
if (par->extradata_size > MAX_EXTRADATA_SIZE) {
av_log(s, AV_LOG_ERROR, "Too much extradata!\n");
return AVERROR_INVALIDDATA;
}
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
config = av_malloc(10 + par->extradata_size * 2);
if (!config) {
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
av_log(s, AV_LOG_ERROR, "Cannot allocate memory for the config info.\n");
return AVERROR(ENOMEM);
}
memcpy(config, "; config=", 9);
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
ff_data_to_hex(config + 9, par->extradata, par->extradata_size, 0);
*out = config;
return 0;
}
static int xiph_extradata2config(AVFormatContext *s, const AVCodecParameters *par,
char **out)
{
uint8_t *config;
char *encoded_config;
const uint8_t *header_start[3];
int headers_len, header_len[3], config_len;
int first_header_size, ret;
*out = NULL;
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
switch (par->codec_id) {
case AV_CODEC_ID_THEORA:
first_header_size = 42;
break;
case AV_CODEC_ID_VORBIS:
first_header_size = 30;
break;
default:
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
av_log(s, AV_LOG_ERROR, "Unsupported Xiph codec ID\n");
return AVERROR(ENOSYS);
}
ret = avpriv_split_xiph_headers(par->extradata, par->extradata_size,
first_header_size, header_start,
header_len);
if (ret < 0) {
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
av_log(s, AV_LOG_ERROR, "Extradata corrupt.\n");
return ret;
}
headers_len = header_len[0] + header_len[2];
config_len = 4 + // count
3 + // ident
2 + // packet size
1 + // header count
2 + // header size
headers_len; // and the rest
config = av_malloc(config_len);
if (!config)
goto xiph_fail;
encoded_config = av_malloc(AV_BASE64_SIZE(config_len));
if (!encoded_config) {
av_free(config);
goto xiph_fail;
}
config[0] = config[1] = config[2] = 0;
config[3] = 1;
config[4] = (RTP_XIPH_IDENT >> 16) & 0xff;
config[5] = (RTP_XIPH_IDENT >> 8) & 0xff;
config[6] = (RTP_XIPH_IDENT ) & 0xff;
config[7] = (headers_len >> 8) & 0xff;
config[8] = headers_len & 0xff;
config[9] = 2;
config[10] = header_len[0];
config[11] = 0; // size of comment header; nonexistent
memcpy(config + 12, header_start[0], header_len[0]);
memcpy(config + 12 + header_len[0], header_start[2], header_len[2]);
av_base64_encode(encoded_config, AV_BASE64_SIZE(config_len),
config, config_len);
av_free(config);
*out = encoded_config;
return 0;
xiph_fail:
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
av_log(s, AV_LOG_ERROR,
"Not enough memory for configuration string\n");
return AVERROR(ENOMEM);
}
static int latm_context2profilelevel(const AVCodecParameters *par)
{
2011-05-18 15:21:48 +02:00
/* MP4A-LATM
* The RTP payload format specification is described in RFC 3016
* The encoding specifications are provided in ISO/IEC 14496-3 */
int profile_level = 0x2B;
/* TODO: AAC Profile only supports AAC LC Object Type.
* Different Object Types should implement different Profile Levels */
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
if (par->sample_rate <= 24000) {
if (par->ch_layout.nb_channels <= 2)
2011-05-18 15:21:48 +02:00
profile_level = 0x28; // AAC Profile, Level 1
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
} else if (par->sample_rate <= 48000) {
if (par->ch_layout.nb_channels <= 2) {
2011-05-18 15:21:48 +02:00
profile_level = 0x29; // AAC Profile, Level 2
} else if (par->ch_layout.nb_channels <= 5) {
2011-05-18 15:21:48 +02:00
profile_level = 0x2A; // AAC Profile, Level 4
}
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
} else if (par->sample_rate <= 96000) {
if (par->ch_layout.nb_channels <= 5) {
2011-05-18 15:21:48 +02:00
profile_level = 0x2B; // AAC Profile, Level 5
}
}
return profile_level;
}
static int latm_context2config(AVFormatContext *s, const AVCodecParameters *par,
char **out)
{
2011-05-18 15:21:48 +02:00
/* MP4A-LATM
* The RTP payload format specification is described in RFC 3016
* The encoding specifications are provided in ISO/IEC 14496-3 */
uint8_t config_byte[6];
int rate_index;
char *config;
*out = NULL;
2011-05-18 15:21:48 +02:00
for (rate_index = 0; rate_index < 16; rate_index++)
if (ff_mpeg4audio_sample_rates[rate_index] == par->sample_rate)
2011-05-18 15:21:48 +02:00
break;
if (rate_index == 16) {
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
av_log(s, AV_LOG_ERROR, "Unsupported sample rate\n");
return AVERROR(ENOSYS);
2011-05-18 15:21:48 +02:00
}
config_byte[0] = 0x40;
config_byte[1] = 0;
config_byte[2] = 0x20 | rate_index;
config_byte[3] = par->ch_layout.nb_channels << 4;
2011-05-18 15:21:48 +02:00
config_byte[4] = 0x3f;
config_byte[5] = 0xc0;
config = av_malloc(6*2+1);
if (!config) {
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
av_log(s, AV_LOG_ERROR, "Cannot allocate memory for the config info.\n");
return AVERROR(ENOMEM);
2011-05-18 15:21:48 +02:00
}
ff_data_to_hex(config, config_byte, 6, 1);
*out = config;
return 0;
2011-05-18 15:21:48 +02:00
}
static int sdp_write_media_attributes(char *buff, int size, const AVStream *st,
int payload_type, AVFormatContext *fmt)
{
char *config = NULL;
const AVCodecParameters *p = st->codecpar;
int ret = 0;
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
switch (p->codec_id) {
case AV_CODEC_ID_DIRAC:
av_strlcatf(buff, size, "a=rtpmap:%d VC2/90000\r\n", payload_type);
break;
case AV_CODEC_ID_H264: {
int mode = 1;
if (fmt && fmt->oformat && fmt->oformat->priv_class &&
av_opt_flag_is_set(fmt->priv_data, "rtpflags", "h264_mode0"))
mode = 0;
if (p->extradata_size) {
ret = extradata2psets(fmt, p, &config);
if (ret < 0)
return ret;
}
av_strlcatf(buff, size, "a=rtpmap:%d H264/90000\r\n"
"a=fmtp:%d packetization-mode=%d%s\r\n",
payload_type,
payload_type, mode, config ? config : "");
break;
}
case AV_CODEC_ID_H261:
{
const char *pic_fmt = NULL;
/* only QCIF and CIF are specified as supported in RFC 4587 */
if (p->width == 176 && p->height == 144)
pic_fmt = "QCIF=1";
else if (p->width == 352 && p->height == 288)
pic_fmt = "CIF=1";
if (payload_type >= RTP_PT_PRIVATE)
av_strlcatf(buff, size, "a=rtpmap:%d H261/90000\r\n", payload_type);
if (pic_fmt)
av_strlcatf(buff, size, "a=fmtp:%d %s\r\n", payload_type, pic_fmt);
break;
}
case AV_CODEC_ID_H263:
case AV_CODEC_ID_H263P:
/* a=framesize is required by 3GPP TS 26.234 (PSS). It
* actually specifies the maximum video size, but we only know
* the current size. This is required for playback on Android
* stagefright and on Samsung bada. */
if (!fmt || !fmt->oformat->priv_class ||
!av_opt_flag_is_set(fmt->priv_data, "rtpflags", "rfc2190") ||
p->codec_id == AV_CODEC_ID_H263P)
av_strlcatf(buff, size, "a=rtpmap:%d H263-2000/90000\r\n"
"a=framesize:%d %d-%d\r\n",
payload_type,
payload_type, p->width, p->height);
break;
case AV_CODEC_ID_HEVC:
if (p->extradata_size) {
ret = extradata2psets_hevc(p, &config);
if (ret < 0)
return ret;
}
av_strlcatf(buff, size, "a=rtpmap:%d H265/90000\r\n", payload_type);
if (config)
av_strlcatf(buff, size, "a=fmtp:%d %s\r\n",
payload_type, config);
break;
case AV_CODEC_ID_MPEG4:
if (p->extradata_size) {
ret = extradata2config(fmt, p, &config);
if (ret < 0)
return ret;
}
av_strlcatf(buff, size, "a=rtpmap:%d MP4V-ES/90000\r\n"
"a=fmtp:%d profile-level-id=1%s\r\n",
payload_type,
payload_type, config ? config : "");
break;
case AV_CODEC_ID_AAC:
if (fmt && fmt->oformat && fmt->oformat->priv_class &&
av_opt_flag_is_set(fmt->priv_data, "rtpflags", "latm")) {
ret = latm_context2config(fmt, p, &config);
if (ret < 0)
return ret;
av_strlcatf(buff, size, "a=rtpmap:%d MP4A-LATM/%d/%d\r\n"
"a=fmtp:%d profile-level-id=%d;cpresent=0;config=%s\r\n",
payload_type, p->sample_rate, p->ch_layout.nb_channels,
payload_type, latm_context2profilelevel(p), config);
} else {
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
if (p->extradata_size) {
ret = extradata2config(fmt, p, &config);
if (ret < 0)
return ret;
2011-05-18 15:21:48 +02:00
} else {
/* FIXME: maybe we can forge config information based on the
* codec parameters...
*/
av_log(fmt, AV_LOG_ERROR, "AAC with no global headers is currently not supported.\n");
return AVERROR(ENOSYS);
}
av_strlcatf(buff, size, "a=rtpmap:%d MPEG4-GENERIC/%d/%d\r\n"
"a=fmtp:%d profile-level-id=1;"
"mode=AAC-hbr;sizelength=13;indexlength=3;"
"indexdeltalength=3%s\r\n",
payload_type, p->sample_rate, p->ch_layout.nb_channels,
payload_type, config);
}
break;
case AV_CODEC_ID_PCM_S16BE:
if (payload_type >= RTP_PT_PRIVATE)
av_strlcatf(buff, size, "a=rtpmap:%d L16/%d/%d\r\n",
payload_type,
p->sample_rate, p->ch_layout.nb_channels);
break;
case AV_CODEC_ID_PCM_S24BE:
if (payload_type >= RTP_PT_PRIVATE)
av_strlcatf(buff, size, "a=rtpmap:%d L24/%d/%d\r\n",
payload_type,
p->sample_rate, p->ch_layout.nb_channels);
break;
case AV_CODEC_ID_PCM_MULAW:
if (payload_type >= RTP_PT_PRIVATE)
av_strlcatf(buff, size, "a=rtpmap:%d PCMU/%d/%d\r\n",
payload_type,
p->sample_rate, p->ch_layout.nb_channels);
break;
case AV_CODEC_ID_PCM_ALAW:
if (payload_type >= RTP_PT_PRIVATE)
av_strlcatf(buff, size, "a=rtpmap:%d PCMA/%d/%d\r\n",
payload_type,
p->sample_rate, p->ch_layout.nb_channels);
break;
case AV_CODEC_ID_AMR_NB:
av_strlcatf(buff, size, "a=rtpmap:%d AMR/%d/%d\r\n"
"a=fmtp:%d octet-align=1\r\n",
payload_type, p->sample_rate, p->ch_layout.nb_channels,
payload_type);
break;
case AV_CODEC_ID_AMR_WB:
av_strlcatf(buff, size, "a=rtpmap:%d AMR-WB/%d/%d\r\n"
"a=fmtp:%d octet-align=1\r\n",
payload_type, p->sample_rate, p->ch_layout.nb_channels,
payload_type);
break;
case AV_CODEC_ID_VORBIS:
if (p->extradata_size)
ret = xiph_extradata2config(fmt, p, &config);
else {
av_log(fmt, AV_LOG_ERROR, "Vorbis configuration info missing\n");
ret = AVERROR_INVALIDDATA;
}
if (ret < 0)
return ret;
av_strlcatf(buff, size, "a=rtpmap:%d vorbis/%d/%d\r\n"
"a=fmtp:%d configuration=%s\r\n",
payload_type, p->sample_rate, p->ch_layout.nb_channels,
payload_type, config);
break;
case AV_CODEC_ID_THEORA: {
const char *pix_fmt;
switch (p->format) {
case AV_PIX_FMT_YUV420P:
pix_fmt = "YCbCr-4:2:0";
2012-09-23 16:17:41 +02:00
break;
case AV_PIX_FMT_YUV422P:
pix_fmt = "YCbCr-4:2:2";
break;
case AV_PIX_FMT_YUV444P:
pix_fmt = "YCbCr-4:4:4";
break;
default:
av_log(fmt, AV_LOG_ERROR, "Unsupported pixel format.\n");
return AVERROR(ENOSYS);
}
if (p->extradata_size)
ret = xiph_extradata2config(fmt, p, &config);
else {
av_log(fmt, AV_LOG_ERROR, "Theora configuration info missing\n");
ret = AVERROR_INVALIDDATA;
}
if (ret < 0)
return ret;
av_strlcatf(buff, size, "a=rtpmap:%d theora/90000\r\n"
"a=fmtp:%d delivery-method=inline; "
"width=%d; height=%d; sampling=%s; "
"configuration=%s\r\n",
payload_type, payload_type,
p->width, p->height, pix_fmt, config);
break;
}
case AV_CODEC_ID_BITPACKED:
case AV_CODEC_ID_RAWVIDEO: {
const char *pix_fmt;
int bit_depth = 8;
switch (p->format) {
case AV_PIX_FMT_UYVY422:
pix_fmt = "YCbCr-4:2:2";
break;
case AV_PIX_FMT_YUV422P10:
pix_fmt = "YCbCr-4:2:2";
bit_depth = 10;
break;
case AV_PIX_FMT_YUV420P:
pix_fmt = "YCbCr-4:2:0";
break;
case AV_PIX_FMT_RGB24:
pix_fmt = "RGB";
break;
case AV_PIX_FMT_BGR24:
pix_fmt = "BGR";
break;
default:
av_log(fmt, AV_LOG_ERROR, "Unsupported pixel format.\n");
return AVERROR(ENOSYS);
}
av_strlcatf(buff, size, "a=rtpmap:%d raw/90000\r\n"
"a=fmtp:%d sampling=%s; "
"width=%d; height=%d; "
"depth=%d",
payload_type, payload_type,
pix_fmt, p->width, p->height, bit_depth);
if (p->field_order != AV_FIELD_PROGRESSIVE)
av_strlcatf(buff, size, "; interlace");
av_strlcatf(buff, size, "\r\n");
break;
}
case AV_CODEC_ID_VP8:
av_strlcatf(buff, size, "a=rtpmap:%d VP8/90000\r\n",
payload_type);
break;
case AV_CODEC_ID_VP9:
av_strlcatf(buff, size, "a=rtpmap:%d VP9/90000\r\n",
payload_type);
break;
case AV_CODEC_ID_MJPEG:
if (payload_type >= RTP_PT_PRIVATE)
av_strlcatf(buff, size, "a=rtpmap:%d JPEG/90000\r\n",
payload_type);
break;
case AV_CODEC_ID_ADPCM_G722:
if (payload_type >= RTP_PT_PRIVATE)
av_strlcatf(buff, size, "a=rtpmap:%d G722/%d/%d\r\n",
payload_type,
8000, p->ch_layout.nb_channels);
break;
case AV_CODEC_ID_ADPCM_G726: {
if (payload_type >= RTP_PT_PRIVATE)
av_strlcatf(buff, size, "a=rtpmap:%d AAL2-G726-%d/%d\r\n",
payload_type,
p->bits_per_coded_sample*8,
p->sample_rate);
break;
}
case AV_CODEC_ID_ADPCM_G726LE: {
if (payload_type >= RTP_PT_PRIVATE)
av_strlcatf(buff, size, "a=rtpmap:%d G726-%d/%d\r\n",
payload_type,
p->bits_per_coded_sample*8,
p->sample_rate);
break;
}
case AV_CODEC_ID_ILBC:
av_strlcatf(buff, size, "a=rtpmap:%d iLBC/%d\r\n"
"a=fmtp:%d mode=%d\r\n",
payload_type, p->sample_rate,
payload_type, p->block_align == 38 ? 20 : 30);
break;
case AV_CODEC_ID_SPEEX:
av_strlcatf(buff, size, "a=rtpmap:%d speex/%d\r\n",
payload_type, p->sample_rate);
break;
case AV_CODEC_ID_OPUS:
/* The opus RTP draft says that all opus streams MUST be declared
as stereo, to avoid negotiation failures. The actual number of
channels can change on a packet-by-packet basis. The number of
channels a receiver prefers to receive or a sender plans to send
can be declared via fmtp parameters (both default to mono), but
receivers MUST be able to receive and process stereo packets. */
av_strlcatf(buff, size, "a=rtpmap:%d opus/48000/2\r\n",
payload_type);
if (p->ch_layout.nb_channels == 2) {
av_strlcatf(buff, size, "a=fmtp:%d sprop-stereo=1\r\n",
payload_type);
}
break;
default:
/* Nothing special to do here... */
break;
}
av_free(config);
return 0;
}
int ff_sdp_write_media(char *buff, int size, const AVStream *st, int idx,
const char *dest_addr, const char *dest_type,
int port, int ttl, AVFormatContext *fmt)
{
const AVCodecParameters *p = st->codecpar;
const char *type;
int payload_type;
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
payload_type = ff_rtp_get_payload_type(fmt, st->codecpar, idx);
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
switch (p->codec_type) {
case AVMEDIA_TYPE_VIDEO : type = "video" ; break;
case AVMEDIA_TYPE_AUDIO : type = "audio" ; break;
case AVMEDIA_TYPE_SUBTITLE: type = "text" ; break;
default : type = "application"; break;
}
av_strlcatf(buff, size, "m=%s %d RTP/AVP %d\r\n", type, port, payload_type);
sdp_write_address(buff, size, dest_addr, dest_type, ttl);
lavf: replace AVStream.codec with AVStream.codecpar Currently, AVStream contains an embedded AVCodecContext instance, which is used by demuxers to export stream parameters to the caller and by muxers to receive stream parameters from the caller. It is also used internally as the codec context that is passed to parsers. In addition, it is also widely used by the callers as the decoding (when demuxer) or encoding (when muxing) context, though this has been officially discouraged since Libav 11. There are multiple important problems with this approach: - the fields in AVCodecContext are in general one of * stream parameters * codec options * codec state However, it's not clear which ones are which. It is consequently unclear which fields are a demuxer allowed to set or a muxer allowed to read. This leads to erratic behaviour depending on whether decoding or encoding is being performed or not (and whether it uses the AVStream embedded codec context). - various synchronization issues arising from the fact that the same context is used by several different APIs (muxers/demuxers, parsers, bitstream filters and encoders/decoders) simultaneously, with there being no clear rules for who can modify what and the different processes being typically delayed with respect to each other. - avformat_find_stream_info() making it necessary to support opening and closing a single codec context multiple times, thus complicating the semantics of freeing various allocated objects in the codec context. Those problems are resolved by replacing the AVStream embedded codec context with a newly added AVCodecParameters instance, which stores only the stream parameters exported by the demuxers or read by the muxers.
2014-06-18 20:42:52 +02:00
if (p->bit_rate) {
av_strlcatf(buff, size, "b=AS:%"PRId64"\r\n", p->bit_rate / 1000);
}
return sdp_write_media_attributes(buff, size, st, payload_type, fmt);
}
int av_sdp_create(AVFormatContext *ac[], int n_files, char *buf, int size)
{
AVDictionaryEntry *title = av_dict_get(ac[0]->metadata, "title", NULL, 0);
struct sdp_session_level s = { 0 };
int i, j, port, ttl, is_multicast, index = 0;
char dst[32], dst_type[5];
memset(buf, 0, size);
s.user = "-";
s.src_addr = "127.0.0.1"; /* FIXME: Properly set this */
s.src_type = "IP4";
s.name = title ? title->value : "No Name";
port = 0;
ttl = 0;
if (n_files == 1) {
port = sdp_get_address(dst, sizeof(dst), &ttl, ac[0]->url ? ac[0]->url : "");
is_multicast = resolve_destination(dst, sizeof(dst), dst_type,
sizeof(dst_type));
if (!is_multicast)
ttl = 0;
if (dst[0]) {
s.dst_addr = dst;
s.dst_type = dst_type;
s.ttl = ttl;
if (!strcmp(dst_type, "IP6")) {
s.src_addr = "::1";
s.src_type = "IP6";
}
}
}
sdp_write_header(buf, size, &s);
dst[0] = 0;
for (i = 0; i < n_files; i++) {
if (n_files != 1) {
port = sdp_get_address(dst, sizeof(dst), &ttl, ac[i]->url ? ac[i]->url : "");
is_multicast = resolve_destination(dst, sizeof(dst), dst_type,
sizeof(dst_type));
if (!is_multicast)
ttl = 0;
}
for (j = 0; j < ac[i]->nb_streams; j++) {
int ret = ff_sdp_write_media(buf, size, ac[i]->streams[j], index++,
dst[0] ? dst : NULL, dst_type,
(port > 0) ? port + j * 2 : 0,
ttl, ac[i]);
if (ret < 0)
return ret;
if (port <= 0) {
av_strlcatf(buf, size,
"a=control:streamid=%d\r\n", i + j);
}
if (ac[i]->pb && ac[i]->pb->av_class) {
uint8_t *crypto_suite = NULL, *crypto_params = NULL;
av_opt_get(ac[i]->pb, "srtp_out_suite", AV_OPT_SEARCH_CHILDREN,
&crypto_suite);
av_opt_get(ac[i]->pb, "srtp_out_params", AV_OPT_SEARCH_CHILDREN,
&crypto_params);
if (crypto_suite && crypto_suite[0])
av_strlcatf(buf, size,
"a=crypto:1 %s inline:%s\r\n",
crypto_suite, crypto_params);
av_free(crypto_suite);
av_free(crypto_params);
}
}
}
return 0;
}
#else
int av_sdp_create(AVFormatContext *ac[], int n_files, char *buf, int size)
{
return AVERROR(ENOSYS);
}
int ff_sdp_write_media(char *buff, int size, const AVStream *st, int idx,
const char *dest_addr, const char *dest_type,
int port, int ttl, AVFormatContext *fmt)
{
return AVERROR(ENOSYS);
}
#endif