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.
195 lines
5.6 KiB
195 lines
5.6 KiB
/* |
|
* 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 <wavpack/wavpack.h> |
|
#include <string.h> |
|
|
|
#include "libavutil/attributes.h" |
|
#include "libavutil/opt.h" |
|
#include "libavutil/samplefmt.h" |
|
|
|
#include "audio_frame_queue.h" |
|
#include "avcodec.h" |
|
#include "internal.h" |
|
|
|
#define WV_DEFAULT_BLOCK_SIZE 32768 |
|
|
|
typedef struct LibWavpackContext { |
|
const AVClass *class; |
|
WavpackContext *wv; |
|
AudioFrameQueue afq; |
|
|
|
AVPacket *pkt; |
|
int user_size; |
|
|
|
int got_output; |
|
} LibWavpackContext; |
|
|
|
static int wavpack_encode_frame(AVCodecContext *avctx, AVPacket *pkt, |
|
const AVFrame *frame, int *got_output) |
|
{ |
|
LibWavpackContext *s = avctx->priv_data; |
|
int ret; |
|
|
|
s->got_output = 0; |
|
s->pkt = pkt; |
|
s->user_size = pkt->size; |
|
|
|
if (frame) { |
|
ret = ff_af_queue_add(&s->afq, frame); |
|
if (ret < 0) |
|
return ret; |
|
|
|
ret = WavpackPackSamples(s->wv, (int32_t*)frame->data[0], frame->nb_samples); |
|
if (!ret) { |
|
av_log(avctx, AV_LOG_ERROR, "Error encoding a frame: %s\n", |
|
WavpackGetErrorMessage(s->wv)); |
|
return AVERROR_UNKNOWN; |
|
} |
|
} |
|
|
|
if (!s->got_output && |
|
(!frame || frame->nb_samples < avctx->frame_size)) { |
|
ret = WavpackFlushSamples(s->wv); |
|
if (!ret) { |
|
av_log(avctx, AV_LOG_ERROR, "Error flushing the encoder: %s\n", |
|
WavpackGetErrorMessage(s->wv)); |
|
return AVERROR_UNKNOWN; |
|
} |
|
} |
|
|
|
if (s->got_output) { |
|
ff_af_queue_remove(&s->afq, avctx->frame_size, &pkt->pts, &pkt->duration); |
|
*got_output = 1; |
|
} |
|
|
|
return 0; |
|
} |
|
|
|
static int encode_callback(void *id, void *data, int32_t count) |
|
{ |
|
AVCodecContext *avctx = id; |
|
LibWavpackContext *s = avctx->priv_data; |
|
int ret, offset = s->pkt->size; |
|
|
|
if (s->user_size) { |
|
if (s->user_size - count < s->pkt->size) { |
|
av_log(avctx, AV_LOG_ERROR, "Provided packet too small.\n"); |
|
return 0; |
|
} |
|
s->pkt->size += count; |
|
} else { |
|
ret = av_grow_packet(s->pkt, count); |
|
if (ret < 0) { |
|
av_log(avctx, AV_LOG_ERROR, "Error allocating output packet.\n"); |
|
return 0; |
|
} |
|
} |
|
|
|
memcpy(s->pkt->data + offset, data, count); |
|
|
|
s->got_output = 1; |
|
|
|
return 1; |
|
} |
|
|
|
static av_cold int wavpack_encode_init(AVCodecContext *avctx) |
|
{ |
|
LibWavpackContext *s = avctx->priv_data; |
|
WavpackConfig config = { 0 }; |
|
int ret; |
|
|
|
s->wv = WavpackOpenFileOutput(encode_callback, avctx, NULL); |
|
if (!s->wv) { |
|
av_log(avctx, AV_LOG_ERROR, "Error allocating the encoder.\n"); |
|
return AVERROR(ENOMEM); |
|
} |
|
|
|
if (!avctx->frame_size) |
|
avctx->frame_size = WV_DEFAULT_BLOCK_SIZE; |
|
|
|
config.bytes_per_sample = 4; |
|
config.bits_per_sample = 32; |
|
config.block_samples = avctx->frame_size; |
|
config.channel_mask = avctx->channel_layout; |
|
config.num_channels = avctx->channels; |
|
config.sample_rate = avctx->sample_rate; |
|
|
|
if (avctx->compression_level != FF_COMPRESSION_DEFAULT) { |
|
if (avctx->compression_level >= 3) { |
|
config.flags |= CONFIG_VERY_HIGH_FLAG; |
|
|
|
if (avctx->compression_level >= 8) |
|
config.xmode = 6; |
|
else if (avctx->compression_level >= 7) |
|
config.xmode = 5; |
|
else if (avctx->compression_level >= 6) |
|
config.xmode = 4; |
|
else if (avctx->compression_level >= 5) |
|
config.xmode = 3; |
|
else if (avctx->compression_level >= 4) |
|
config.xmode = 2; |
|
} else if (avctx->compression_level >= 2) |
|
config.flags |= CONFIG_HIGH_FLAG; |
|
else if (avctx->compression_level < 1) |
|
config.flags |= CONFIG_FAST_FLAG; |
|
} |
|
|
|
ret = WavpackSetConfiguration(s->wv, &config, -1); |
|
if (!ret) |
|
goto fail; |
|
|
|
ret = WavpackPackInit(s->wv); |
|
if (!ret) |
|
goto fail; |
|
|
|
ff_af_queue_init(avctx, &s->afq); |
|
|
|
return 0; |
|
|
|
fail: |
|
av_log(avctx, AV_LOG_ERROR, "Error configuring the encoder: %s.\n", |
|
WavpackGetErrorMessage(s->wv)); |
|
WavpackCloseFile(s->wv); |
|
return AVERROR_UNKNOWN; |
|
} |
|
|
|
static av_cold int wavpack_encode_close(AVCodecContext *avctx) |
|
{ |
|
LibWavpackContext *s = avctx->priv_data; |
|
|
|
WavpackCloseFile(s->wv); |
|
|
|
ff_af_queue_close(&s->afq); |
|
|
|
return 0; |
|
} |
|
|
|
AVCodec ff_libwavpack_encoder = { |
|
.name = "libwavpack", |
|
.type = AVMEDIA_TYPE_AUDIO, |
|
.id = AV_CODEC_ID_WAVPACK, |
|
.priv_data_size = sizeof(LibWavpackContext), |
|
.init = wavpack_encode_init, |
|
.encode2 = wavpack_encode_frame, |
|
.close = wavpack_encode_close, |
|
.capabilities = AV_CODEC_CAP_DELAY | AV_CODEC_CAP_SMALL_LAST_FRAME, |
|
.sample_fmts = (const enum AVSampleFormat[]){ AV_SAMPLE_FMT_S32, |
|
AV_SAMPLE_FMT_NONE }, |
|
.wrapper_name = "libwavpack", |
|
};
|
|
|