|
|
|
/*
|
|
|
|
* Common code for the RTP depacketization of MPEG-4 formats.
|
|
|
|
* Copyright (c) 2010 Fabrice Bellard
|
|
|
|
* Romain Degez
|
|
|
|
*
|
|
|
|
* 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
|
|
|
|
*/
|
|
|
|
|
|
|
|
/**
|
|
|
|
* @file
|
|
|
|
* @brief MPEG-4 / RTP Code
|
|
|
|
* @author Fabrice Bellard
|
|
|
|
* @author Romain Degez
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include "rtpdec_formats.h"
|
|
|
|
#include "internal.h"
|
|
|
|
#include "libavutil/attributes.h"
|
|
|
|
#include "libavutil/avstring.h"
|
|
|
|
#include "libavcodec/get_bits.h"
|
|
|
|
|
|
|
|
#define MAX_AAC_HBR_FRAME_SIZE 8191
|
|
|
|
|
|
|
|
/** Structure listing useful vars to parse RTP packet payload */
|
|
|
|
struct PayloadContext {
|
|
|
|
int sizelength;
|
|
|
|
int indexlength;
|
|
|
|
int indexdeltalength;
|
|
|
|
int profile_level_id;
|
|
|
|
int streamtype;
|
|
|
|
int objecttype;
|
|
|
|
char *mode;
|
|
|
|
|
|
|
|
/** mpeg 4 AU headers */
|
|
|
|
struct AUHeaders {
|
|
|
|
int size;
|
|
|
|
int index;
|
|
|
|
int cts_flag;
|
|
|
|
int cts;
|
|
|
|
int dts_flag;
|
|
|
|
int dts;
|
|
|
|
int rap_flag;
|
|
|
|
int streamstate;
|
|
|
|
} *au_headers;
|
|
|
|
int au_headers_allocated;
|
|
|
|
int nb_au_headers;
|
|
|
|
int au_headers_length_bytes;
|
|
|
|
int cur_au_index;
|
|
|
|
|
|
|
|
uint8_t buf[FFMAX(RTP_MAX_PACKET_LENGTH, MAX_AAC_HBR_FRAME_SIZE)];
|
|
|
|
int buf_pos, buf_size;
|
|
|
|
uint32_t timestamp;
|
|
|
|
};
|
|
|
|
|
|
|
|
typedef struct AttrNameMap {
|
|
|
|
const char *str;
|
|
|
|
uint16_t type;
|
|
|
|
uint32_t offset;
|
|
|
|
} AttrNameMap;
|
|
|
|
|
|
|
|
/* All known fmtp parameters and the corresponding RTPAttrTypeEnum */
|
|
|
|
#define ATTR_NAME_TYPE_INT 0
|
|
|
|
#define ATTR_NAME_TYPE_STR 1
|
|
|
|
static const AttrNameMap attr_names[] = {
|
|
|
|
{ "SizeLength", ATTR_NAME_TYPE_INT,
|
|
|
|
offsetof(PayloadContext, sizelength) },
|
|
|
|
{ "IndexLength", ATTR_NAME_TYPE_INT,
|
|
|
|
offsetof(PayloadContext, indexlength) },
|
|
|
|
{ "IndexDeltaLength", ATTR_NAME_TYPE_INT,
|
|
|
|
offsetof(PayloadContext, indexdeltalength) },
|
|
|
|
{ "profile-level-id", ATTR_NAME_TYPE_INT,
|
|
|
|
offsetof(PayloadContext, profile_level_id) },
|
|
|
|
{ "StreamType", ATTR_NAME_TYPE_INT,
|
|
|
|
offsetof(PayloadContext, streamtype) },
|
|
|
|
{ "mode", ATTR_NAME_TYPE_STR,
|
|
|
|
offsetof(PayloadContext, mode) },
|
|
|
|
{ NULL, -1, -1 },
|
|
|
|
};
|
|
|
|
|
|
|
|
static void close_context(PayloadContext *data)
|
|
|
|
{
|
|
|
|
av_freep(&data->au_headers);
|
|
|
|
av_freep(&data->mode);
|
|
|
|
}
|
|
|
|
|
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.
11 years ago
|
|
|
static int parse_fmtp_config(AVCodecParameters *par, char *value)
|
|
|
|
{
|
|
|
|
/* decode the hexa encoded parameter */
|
|
|
|
int len = ff_hex_to_data(NULL, value);
|
|
|
|
av_freep(&par->extradata);
|
|
|
|
if (ff_alloc_extradata(par, len))
|
|
|
|
return AVERROR(ENOMEM);
|
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.
11 years ago
|
|
|
ff_hex_to_data(par->extradata, value);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int rtp_parse_mp4_au(PayloadContext *data, const uint8_t *buf, int len)
|
|
|
|
{
|
|
|
|
int au_headers_length, au_header_size, i;
|
|
|
|
GetBitContext getbitcontext;
|
|
|
|
|
|
|
|
if (len < 2)
|
|
|
|
return AVERROR_INVALIDDATA;
|
|
|
|
|
|
|
|
/* decode the first 2 bytes where the AUHeader sections are stored
|
|
|
|
length in bits */
|
|
|
|
au_headers_length = AV_RB16(buf);
|
|
|
|
|
|
|
|
if (au_headers_length > RTP_MAX_PACKET_LENGTH)
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
data->au_headers_length_bytes = (au_headers_length + 7) / 8;
|
|
|
|
|
|
|
|
/* skip AU headers length section (2 bytes) */
|
|
|
|
buf += 2;
|
|
|
|
len -= 2;
|
|
|
|
|
|
|
|
if (len < data->au_headers_length_bytes)
|
|
|
|
return AVERROR_INVALIDDATA;
|
|
|
|
|
|
|
|
init_get_bits(&getbitcontext, buf, data->au_headers_length_bytes * 8);
|
|
|
|
|
|
|
|
/* XXX: Wrong if optional additional sections are present (cts, dts etc...) */
|
|
|
|
au_header_size = data->sizelength + data->indexlength;
|
|
|
|
if (au_header_size <= 0 || (au_headers_length % au_header_size != 0))
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
data->nb_au_headers = au_headers_length / au_header_size;
|
|
|
|
if (!data->au_headers || data->au_headers_allocated < data->nb_au_headers) {
|
|
|
|
av_free(data->au_headers);
|
|
|
|
data->au_headers = av_malloc(sizeof(struct AUHeaders) * data->nb_au_headers);
|
|
|
|
if (!data->au_headers)
|
|
|
|
return AVERROR(ENOMEM);
|
|
|
|
data->au_headers_allocated = data->nb_au_headers;
|
|
|
|
}
|
|
|
|
|
|
|
|
for (i = 0; i < data->nb_au_headers; ++i) {
|
|
|
|
data->au_headers[i].size = get_bits_long(&getbitcontext, data->sizelength);
|
|
|
|
data->au_headers[i].index = get_bits_long(&getbitcontext, data->indexlength);
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
/* Follows RFC 3640 */
|
|
|
|
static int aac_parse_packet(AVFormatContext *ctx, PayloadContext *data,
|
|
|
|
AVStream *st, AVPacket *pkt, uint32_t *timestamp,
|
|
|
|
const uint8_t *buf, int len, uint16_t seq,
|
|
|
|
int flags)
|
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
|
|
|
|
if (!buf) {
|
|
|
|
if (data->cur_au_index > data->nb_au_headers) {
|
|
|
|
av_log(ctx, AV_LOG_ERROR, "Invalid parser state\n");
|
|
|
|
return AVERROR_INVALIDDATA;
|
|
|
|
}
|
|
|
|
if (data->buf_size - data->buf_pos < data->au_headers[data->cur_au_index].size) {
|
|
|
|
av_log(ctx, AV_LOG_ERROR, "Invalid AU size\n");
|
|
|
|
return AVERROR_INVALIDDATA;
|
|
|
|
}
|
|
|
|
if ((ret = av_new_packet(pkt, data->au_headers[data->cur_au_index].size)) < 0) {
|
|
|
|
av_log(ctx, AV_LOG_ERROR, "Out of memory\n");
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
memcpy(pkt->data, &data->buf[data->buf_pos], data->au_headers[data->cur_au_index].size);
|
|
|
|
data->buf_pos += data->au_headers[data->cur_au_index].size;
|
|
|
|
pkt->stream_index = st->index;
|
|
|
|
data->cur_au_index++;
|
|
|
|
|
|
|
|
if (data->cur_au_index == data->nb_au_headers) {
|
|
|
|
data->buf_pos = 0;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (rtp_parse_mp4_au(data, buf, len)) {
|
|
|
|
av_log(ctx, AV_LOG_ERROR, "Error parsing AU headers\n");
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
buf += data->au_headers_length_bytes + 2;
|
|
|
|
len -= data->au_headers_length_bytes + 2;
|
|
|
|
if (data->nb_au_headers == 1 && len < data->au_headers[0].size) {
|
|
|
|
/* Packet is fragmented */
|
|
|
|
|
|
|
|
if (!data->buf_pos) {
|
|
|
|
if (data->au_headers[0].size > MAX_AAC_HBR_FRAME_SIZE) {
|
|
|
|
av_log(ctx, AV_LOG_ERROR, "Invalid AU size\n");
|
|
|
|
return AVERROR_INVALIDDATA;
|
|
|
|
}
|
|
|
|
|
|
|
|
data->buf_size = data->au_headers[0].size;
|
|
|
|
data->timestamp = *timestamp;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (data->timestamp != *timestamp ||
|
|
|
|
data->au_headers[0].size != data->buf_size ||
|
|
|
|
data->buf_pos + len > MAX_AAC_HBR_FRAME_SIZE) {
|
|
|
|
data->buf_pos = 0;
|
|
|
|
data->buf_size = 0;
|
|
|
|
av_log(ctx, AV_LOG_ERROR, "Invalid packet received\n");
|
|
|
|
return AVERROR_INVALIDDATA;
|
|
|
|
}
|
|
|
|
|
|
|
|
memcpy(&data->buf[data->buf_pos], buf, len);
|
|
|
|
data->buf_pos += len;
|
|
|
|
|
|
|
|
if (!(flags & RTP_FLAG_MARKER))
|
|
|
|
return AVERROR(EAGAIN);
|
|
|
|
|
|
|
|
if (data->buf_pos != data->buf_size) {
|
|
|
|
data->buf_pos = 0;
|
|
|
|
av_log(ctx, AV_LOG_ERROR, "Missed some packets, discarding frame\n");
|
|
|
|
return AVERROR_INVALIDDATA;
|
|
|
|
}
|
|
|
|
|
|
|
|
data->buf_pos = 0;
|
|
|
|
ret = av_new_packet(pkt, data->buf_size);
|
|
|
|
if (ret < 0) {
|
|
|
|
av_log(ctx, AV_LOG_ERROR, "Out of memory\n");
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
pkt->stream_index = st->index;
|
|
|
|
|
|
|
|
memcpy(pkt->data, data->buf, data->buf_size);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (len < data->au_headers[0].size) {
|
|
|
|
av_log(ctx, AV_LOG_ERROR, "First AU larger than packet size\n");
|
|
|
|
return AVERROR_INVALIDDATA;
|
|
|
|
}
|
|
|
|
if ((ret = av_new_packet(pkt, data->au_headers[0].size)) < 0) {
|
|
|
|
av_log(ctx, AV_LOG_ERROR, "Out of memory\n");
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
memcpy(pkt->data, buf, data->au_headers[0].size);
|
|
|
|
len -= data->au_headers[0].size;
|
|
|
|
buf += data->au_headers[0].size;
|
|
|
|
pkt->stream_index = st->index;
|
|
|
|
|
|
|
|
if (len > 0 && data->nb_au_headers > 1) {
|
|
|
|
data->buf_size = FFMIN(len, sizeof(data->buf));
|
|
|
|
memcpy(data->buf, buf, data->buf_size);
|
|
|
|
data->cur_au_index = 1;
|
|
|
|
data->buf_pos = 0;
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int parse_fmtp(AVFormatContext *s,
|
|
|
|
AVStream *stream, PayloadContext *data,
|
|
|
|
const char *attr, const char *value)
|
|
|
|
{
|
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.
11 years ago
|
|
|
AVCodecParameters *par = stream->codecpar;
|
|
|
|
int res, i;
|
|
|
|
|
|
|
|
if (!strcmp(attr, "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.
11 years ago
|
|
|
res = parse_fmtp_config(par, value);
|
|
|
|
|
|
|
|
if (res < 0)
|
|
|
|
return res;
|
|
|
|
}
|
|
|
|
|
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.
11 years ago
|
|
|
if (par->codec_id == AV_CODEC_ID_AAC) {
|
|
|
|
/* Looking for a known attribute */
|
|
|
|
for (i = 0; attr_names[i].str; ++i) {
|
|
|
|
if (!av_strcasecmp(attr, attr_names[i].str)) {
|
|
|
|
if (attr_names[i].type == ATTR_NAME_TYPE_INT) {
|
|
|
|
*(int *)((char *)data+
|
|
|
|
attr_names[i].offset) = atoi(value);
|
|
|
|
} else if (attr_names[i].type == ATTR_NAME_TYPE_STR)
|
|
|
|
*(char **)((char *)data+
|
|
|
|
attr_names[i].offset) = av_strdup(value);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int parse_sdp_line(AVFormatContext *s, int st_index,
|
|
|
|
PayloadContext *data, const char *line)
|
|
|
|
{
|
|
|
|
const char *p;
|
|
|
|
|
|
|
|
if (st_index < 0)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
if (av_strstart(line, "fmtp:", &p))
|
|
|
|
return ff_parse_fmtp(s, s->streams[st_index], data, p, parse_fmtp);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
RTPDynamicProtocolHandler ff_mp4v_es_dynamic_handler = {
|
|
|
|
.enc_name = "MP4V-ES",
|
|
|
|
.codec_type = AVMEDIA_TYPE_VIDEO,
|
|
|
|
.codec_id = AV_CODEC_ID_MPEG4,
|
|
|
|
.need_parsing = AVSTREAM_PARSE_FULL,
|
|
|
|
.priv_data_size = sizeof(PayloadContext),
|
|
|
|
.parse_sdp_a_line = parse_sdp_line,
|
|
|
|
};
|
|
|
|
|
|
|
|
RTPDynamicProtocolHandler ff_mpeg4_generic_dynamic_handler = {
|
|
|
|
.enc_name = "mpeg4-generic",
|
|
|
|
.codec_type = AVMEDIA_TYPE_AUDIO,
|
|
|
|
.codec_id = AV_CODEC_ID_AAC,
|
|
|
|
.priv_data_size = sizeof(PayloadContext),
|
|
|
|
.parse_sdp_a_line = parse_sdp_line,
|
|
|
|
.close = close_context,
|
|
|
|
.parse_packet = aac_parse_packet,
|
|
|
|
};
|