mirror of https://github.com/FFmpeg/FFmpeg.git
You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
360 lines
12 KiB
360 lines
12 KiB
/* |
|
* copyright (c) 2002 Mark Hills <mark@pogo.org.uk> |
|
* |
|
* 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 |
|
*/ |
|
|
|
/** |
|
* @file |
|
* Vorbis encoding support via libvorbisenc. |
|
* @author Mark Hills <mark@pogo.org.uk> |
|
*/ |
|
|
|
#include <vorbis/vorbisenc.h> |
|
|
|
#include "libavutil/fifo.h" |
|
#include "libavutil/opt.h" |
|
#include "avcodec.h" |
|
#include "audio_frame_queue.h" |
|
#include "bytestream.h" |
|
#include "internal.h" |
|
#include "vorbis.h" |
|
#include "vorbis_parser.h" |
|
|
|
#undef NDEBUG |
|
#include <assert.h> |
|
|
|
/* Number of samples the user should send in each call. |
|
* This value is used because it is the LCD of all possible frame sizes, so |
|
* an output packet will always start at the same point as one of the input |
|
* packets. |
|
*/ |
|
#define LIBVORBIS_FRAME_SIZE 64 |
|
|
|
#define BUFFER_SIZE (1024 * 64) |
|
|
|
typedef struct LibvorbisContext { |
|
AVClass *av_class; /**< class for AVOptions */ |
|
vorbis_info vi; /**< vorbis_info used during init */ |
|
vorbis_dsp_state vd; /**< DSP state used for analysis */ |
|
vorbis_block vb; /**< vorbis_block used for analysis */ |
|
AVFifoBuffer *pkt_fifo; /**< output packet buffer */ |
|
int eof; /**< end-of-file flag */ |
|
int dsp_initialized; /**< vd has been initialized */ |
|
vorbis_comment vc; /**< VorbisComment info */ |
|
ogg_packet op; /**< ogg packet */ |
|
double iblock; /**< impulse block bias option */ |
|
AVVorbisParseContext *vp; /**< parse context to get durations */ |
|
AudioFrameQueue afq; /**< frame queue for timestamps */ |
|
} LibvorbisContext; |
|
|
|
static const AVOption options[] = { |
|
{ "iblock", "Sets the impulse block bias", offsetof(LibvorbisContext, iblock), AV_OPT_TYPE_DOUBLE, { .dbl = 0 }, -15, 0, AV_OPT_FLAG_AUDIO_PARAM | AV_OPT_FLAG_ENCODING_PARAM }, |
|
{ NULL } |
|
}; |
|
|
|
static const AVCodecDefault defaults[] = { |
|
{ "b", "0" }, |
|
{ NULL }, |
|
}; |
|
|
|
static const AVClass class = { |
|
.class_name = "libvorbis", |
|
.item_name = av_default_item_name, |
|
.option = options, |
|
.version = LIBAVUTIL_VERSION_INT, |
|
}; |
|
|
|
|
|
static int vorbis_error_to_averror(int ov_err) |
|
{ |
|
switch (ov_err) { |
|
case OV_EFAULT: return AVERROR_BUG; |
|
case OV_EINVAL: return AVERROR(EINVAL); |
|
case OV_EIMPL: return AVERROR(EINVAL); |
|
default: return AVERROR_UNKNOWN; |
|
} |
|
} |
|
|
|
static av_cold int libvorbis_setup(vorbis_info *vi, AVCodecContext *avctx) |
|
{ |
|
LibvorbisContext *s = avctx->priv_data; |
|
double cfreq; |
|
int ret; |
|
|
|
if (avctx->flags & AV_CODEC_FLAG_QSCALE || !avctx->bit_rate) { |
|
/* variable bitrate |
|
* NOTE: we use the oggenc range of -1 to 10 for global_quality for |
|
* user convenience, but libvorbis uses -0.1 to 1.0. |
|
*/ |
|
float q = avctx->global_quality / (float)FF_QP2LAMBDA; |
|
/* default to 3 if the user did not set quality or bitrate */ |
|
if (!(avctx->flags & AV_CODEC_FLAG_QSCALE)) |
|
q = 3.0; |
|
if ((ret = vorbis_encode_setup_vbr(vi, avctx->channels, |
|
avctx->sample_rate, |
|
q / 10.0))) |
|
goto error; |
|
} else { |
|
int minrate = avctx->rc_min_rate > 0 ? avctx->rc_min_rate : -1; |
|
int maxrate = avctx->rc_max_rate > 0 ? avctx->rc_max_rate : -1; |
|
|
|
/* average bitrate */ |
|
if ((ret = vorbis_encode_setup_managed(vi, avctx->channels, |
|
avctx->sample_rate, maxrate, |
|
avctx->bit_rate, minrate))) |
|
goto error; |
|
|
|
/* variable bitrate by estimate, disable slow rate management */ |
|
if (minrate == -1 && maxrate == -1) |
|
if ((ret = vorbis_encode_ctl(vi, OV_ECTL_RATEMANAGE2_SET, NULL))) |
|
goto error; |
|
} |
|
|
|
/* cutoff frequency */ |
|
if (avctx->cutoff > 0) { |
|
cfreq = avctx->cutoff / 1000.0; |
|
if ((ret = vorbis_encode_ctl(vi, OV_ECTL_LOWPASS_SET, &cfreq))) |
|
goto error; |
|
} |
|
|
|
/* impulse block bias */ |
|
if (s->iblock) { |
|
if ((ret = vorbis_encode_ctl(vi, OV_ECTL_IBLOCK_SET, &s->iblock))) |
|
goto error; |
|
} |
|
|
|
if ((ret = vorbis_encode_setup_init(vi))) |
|
goto error; |
|
|
|
return 0; |
|
error: |
|
return vorbis_error_to_averror(ret); |
|
} |
|
|
|
/* How many bytes are needed for a buffer of length 'l' */ |
|
static int xiph_len(int l) |
|
{ |
|
return 1 + l / 255 + l; |
|
} |
|
|
|
static av_cold int libvorbis_encode_close(AVCodecContext *avctx) |
|
{ |
|
LibvorbisContext *s = avctx->priv_data; |
|
|
|
/* notify vorbisenc this is EOF */ |
|
if (s->dsp_initialized) |
|
vorbis_analysis_wrote(&s->vd, 0); |
|
|
|
vorbis_block_clear(&s->vb); |
|
vorbis_dsp_clear(&s->vd); |
|
vorbis_info_clear(&s->vi); |
|
|
|
av_fifo_free(s->pkt_fifo); |
|
ff_af_queue_close(&s->afq); |
|
av_freep(&avctx->extradata); |
|
|
|
av_vorbis_parse_free(&s->vp); |
|
|
|
return 0; |
|
} |
|
|
|
static av_cold int libvorbis_encode_init(AVCodecContext *avctx) |
|
{ |
|
LibvorbisContext *s = avctx->priv_data; |
|
ogg_packet header, header_comm, header_code; |
|
uint8_t *p; |
|
unsigned int offset; |
|
int ret; |
|
|
|
vorbis_info_init(&s->vi); |
|
if ((ret = libvorbis_setup(&s->vi, avctx))) { |
|
av_log(avctx, AV_LOG_ERROR, "encoder setup failed\n"); |
|
goto error; |
|
} |
|
if ((ret = vorbis_analysis_init(&s->vd, &s->vi))) { |
|
av_log(avctx, AV_LOG_ERROR, "analysis init failed\n"); |
|
ret = vorbis_error_to_averror(ret); |
|
goto error; |
|
} |
|
s->dsp_initialized = 1; |
|
if ((ret = vorbis_block_init(&s->vd, &s->vb))) { |
|
av_log(avctx, AV_LOG_ERROR, "dsp init failed\n"); |
|
ret = vorbis_error_to_averror(ret); |
|
goto error; |
|
} |
|
|
|
vorbis_comment_init(&s->vc); |
|
vorbis_comment_add_tag(&s->vc, "encoder", LIBAVCODEC_IDENT); |
|
|
|
if ((ret = vorbis_analysis_headerout(&s->vd, &s->vc, &header, &header_comm, |
|
&header_code))) { |
|
ret = vorbis_error_to_averror(ret); |
|
goto error; |
|
} |
|
|
|
avctx->extradata_size = 1 + xiph_len(header.bytes) + |
|
xiph_len(header_comm.bytes) + |
|
header_code.bytes; |
|
p = avctx->extradata = av_malloc(avctx->extradata_size + |
|
AV_INPUT_BUFFER_PADDING_SIZE); |
|
if (!p) { |
|
ret = AVERROR(ENOMEM); |
|
goto error; |
|
} |
|
p[0] = 2; |
|
offset = 1; |
|
offset += av_xiphlacing(&p[offset], header.bytes); |
|
offset += av_xiphlacing(&p[offset], header_comm.bytes); |
|
memcpy(&p[offset], header.packet, header.bytes); |
|
offset += header.bytes; |
|
memcpy(&p[offset], header_comm.packet, header_comm.bytes); |
|
offset += header_comm.bytes; |
|
memcpy(&p[offset], header_code.packet, header_code.bytes); |
|
offset += header_code.bytes; |
|
assert(offset == avctx->extradata_size); |
|
|
|
s->vp = av_vorbis_parse_init(avctx->extradata, avctx->extradata_size); |
|
if (!s->vp) { |
|
av_log(avctx, AV_LOG_ERROR, "invalid extradata\n"); |
|
return ret; |
|
} |
|
|
|
vorbis_comment_clear(&s->vc); |
|
|
|
avctx->frame_size = LIBVORBIS_FRAME_SIZE; |
|
ff_af_queue_init(avctx, &s->afq); |
|
|
|
s->pkt_fifo = av_fifo_alloc(BUFFER_SIZE); |
|
if (!s->pkt_fifo) { |
|
ret = AVERROR(ENOMEM); |
|
goto error; |
|
} |
|
|
|
return 0; |
|
error: |
|
libvorbis_encode_close(avctx); |
|
return ret; |
|
} |
|
|
|
static int libvorbis_encode_frame(AVCodecContext *avctx, AVPacket *avpkt, |
|
const AVFrame *frame, int *got_packet_ptr) |
|
{ |
|
LibvorbisContext *s = avctx->priv_data; |
|
ogg_packet op; |
|
int ret, duration; |
|
|
|
/* send samples to libvorbis */ |
|
if (frame) { |
|
const int samples = frame->nb_samples; |
|
float **buffer; |
|
int c, channels = s->vi.channels; |
|
|
|
buffer = vorbis_analysis_buffer(&s->vd, samples); |
|
for (c = 0; c < channels; c++) { |
|
int co = (channels > 8) ? c : |
|
ff_vorbis_encoding_channel_layout_offsets[channels - 1][c]; |
|
memcpy(buffer[c], frame->extended_data[co], |
|
samples * sizeof(*buffer[c])); |
|
} |
|
if ((ret = vorbis_analysis_wrote(&s->vd, samples)) < 0) { |
|
av_log(avctx, AV_LOG_ERROR, "error in vorbis_analysis_wrote()\n"); |
|
return vorbis_error_to_averror(ret); |
|
} |
|
if ((ret = ff_af_queue_add(&s->afq, frame)) < 0) |
|
return ret; |
|
} else { |
|
if (!s->eof) |
|
if ((ret = vorbis_analysis_wrote(&s->vd, 0)) < 0) { |
|
av_log(avctx, AV_LOG_ERROR, "error in vorbis_analysis_wrote()\n"); |
|
return vorbis_error_to_averror(ret); |
|
} |
|
s->eof = 1; |
|
} |
|
|
|
/* retrieve available packets from libvorbis */ |
|
while ((ret = vorbis_analysis_blockout(&s->vd, &s->vb)) == 1) { |
|
if ((ret = vorbis_analysis(&s->vb, NULL)) < 0) |
|
break; |
|
if ((ret = vorbis_bitrate_addblock(&s->vb)) < 0) |
|
break; |
|
|
|
/* add any available packets to the output packet buffer */ |
|
while ((ret = vorbis_bitrate_flushpacket(&s->vd, &op)) == 1) { |
|
if (av_fifo_space(s->pkt_fifo) < sizeof(ogg_packet) + op.bytes) { |
|
av_log(avctx, AV_LOG_ERROR, "packet buffer is too small"); |
|
return AVERROR_BUG; |
|
} |
|
av_fifo_generic_write(s->pkt_fifo, &op, sizeof(ogg_packet), NULL); |
|
av_fifo_generic_write(s->pkt_fifo, op.packet, op.bytes, NULL); |
|
} |
|
if (ret < 0) { |
|
av_log(avctx, AV_LOG_ERROR, "error getting available packets\n"); |
|
break; |
|
} |
|
} |
|
if (ret < 0) { |
|
av_log(avctx, AV_LOG_ERROR, "error getting available packets\n"); |
|
return vorbis_error_to_averror(ret); |
|
} |
|
|
|
/* check for available packets */ |
|
if (av_fifo_size(s->pkt_fifo) < sizeof(ogg_packet)) |
|
return 0; |
|
|
|
av_fifo_generic_read(s->pkt_fifo, &op, sizeof(ogg_packet), NULL); |
|
|
|
if ((ret = ff_alloc_packet(avpkt, op.bytes))) { |
|
av_log(avctx, AV_LOG_ERROR, "Error getting output packet\n"); |
|
return ret; |
|
} |
|
av_fifo_generic_read(s->pkt_fifo, avpkt->data, op.bytes, NULL); |
|
|
|
avpkt->pts = ff_samples_to_time_base(avctx, op.granulepos); |
|
|
|
duration = av_vorbis_parse_frame(s->vp, avpkt->data, avpkt->size); |
|
if (duration > 0) { |
|
/* we do not know encoder delay until we get the first packet from |
|
* libvorbis, so we have to update the AudioFrameQueue counts */ |
|
if (!avctx->initial_padding) { |
|
avctx->initial_padding = duration; |
|
s->afq.remaining_delay += duration; |
|
s->afq.remaining_samples += duration; |
|
} |
|
ff_af_queue_remove(&s->afq, duration, &avpkt->pts, &avpkt->duration); |
|
} |
|
|
|
*got_packet_ptr = 1; |
|
return 0; |
|
} |
|
|
|
AVCodec ff_libvorbis_encoder = { |
|
.name = "libvorbis", |
|
.long_name = NULL_IF_CONFIG_SMALL("libvorbis Vorbis"), |
|
.type = AVMEDIA_TYPE_AUDIO, |
|
.id = AV_CODEC_ID_VORBIS, |
|
.priv_data_size = sizeof(LibvorbisContext), |
|
.init = libvorbis_encode_init, |
|
.encode2 = libvorbis_encode_frame, |
|
.close = libvorbis_encode_close, |
|
.capabilities = AV_CODEC_CAP_DELAY, |
|
.sample_fmts = (const enum AVSampleFormat[]) { AV_SAMPLE_FMT_FLTP, |
|
AV_SAMPLE_FMT_NONE }, |
|
.priv_class = &class, |
|
.defaults = defaults, |
|
.wrapper_name = "libvorbis", |
|
};
|
|
|