1/* 2 * Copyright (c) 2002 Mark Hills <mark@pogo.org.uk> 3 * 4 * This file is part of FFmpeg. 5 * 6 * FFmpeg is free software; you can redistribute it and/or 7 * modify it under the terms of the GNU Lesser General Public 8 * License as published by the Free Software Foundation; either 9 * version 2.1 of the License, or (at your option) any later version. 10 * 11 * FFmpeg is distributed in the hope that it will be useful, 12 * but WITHOUT ANY WARRANTY; without even the implied warranty of 13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 14 * Lesser General Public License for more details. 15 * 16 * You should have received a copy of the GNU Lesser General Public 17 * License along with FFmpeg; if not, write to the Free Software 18 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA 19 */ 20 21#include <vorbis/vorbisenc.h> 22 23#include "libavutil/avassert.h" 24#include "libavutil/channel_layout.h" 25#include "libavutil/fifo.h" 26#include "libavutil/opt.h" 27#include "avcodec.h" 28#include "audio_frame_queue.h" 29#include "codec_internal.h" 30#include "encode.h" 31#include "internal.h" 32#include "version.h" 33#include "vorbis.h" 34#include "vorbis_parser.h" 35 36 37/* Number of samples the user should send in each call. 38 * This value is used because it is the LCD of all possible frame sizes, so 39 * an output packet will always start at the same point as one of the input 40 * packets. 41 */ 42#define LIBVORBIS_FRAME_SIZE 64 43 44#define BUFFER_SIZE (1024 * 64) 45 46typedef struct LibvorbisEncContext { 47 AVClass *av_class; /**< class for AVOptions */ 48 vorbis_info vi; /**< vorbis_info used during init */ 49 vorbis_dsp_state vd; /**< DSP state used for analysis */ 50 vorbis_block vb; /**< vorbis_block used for analysis */ 51 AVFifo *pkt_fifo; /**< output packet buffer */ 52 int eof; /**< end-of-file flag */ 53 int dsp_initialized; /**< vd has been initialized */ 54 vorbis_comment vc; /**< VorbisComment info */ 55 double iblock; /**< impulse block bias option */ 56 AVVorbisParseContext *vp; /**< parse context to get durations */ 57 AudioFrameQueue afq; /**< frame queue for timestamps */ 58} LibvorbisEncContext; 59 60static const AVOption options[] = { 61 { "iblock", "Sets the impulse block bias", offsetof(LibvorbisEncContext, iblock), AV_OPT_TYPE_DOUBLE, { .dbl = 0 }, -15, 0, AV_OPT_FLAG_AUDIO_PARAM | AV_OPT_FLAG_ENCODING_PARAM }, 62 { NULL } 63}; 64 65static const FFCodecDefault defaults[] = { 66 { "b", "0" }, 67 { NULL }, 68}; 69 70static const AVClass vorbis_class = { 71 .class_name = "libvorbis", 72 .item_name = av_default_item_name, 73 .option = options, 74 .version = LIBAVUTIL_VERSION_INT, 75}; 76 77static const uint8_t vorbis_encoding_channel_layout_offsets[8][8] = { 78 { 0 }, 79 { 0, 1 }, 80 { 0, 2, 1 }, 81 { 0, 1, 2, 3 }, 82 { 0, 2, 1, 3, 4 }, 83 { 0, 2, 1, 4, 5, 3 }, 84 { 0, 2, 1, 5, 6, 4, 3 }, 85 { 0, 2, 1, 6, 7, 4, 5, 3 }, 86}; 87 88static int vorbis_error_to_averror(int ov_err) 89{ 90 switch (ov_err) { 91 case OV_EFAULT: return AVERROR_BUG; 92 case OV_EINVAL: return AVERROR(EINVAL); 93 case OV_EIMPL: return AVERROR(EINVAL); 94 default: return AVERROR_UNKNOWN; 95 } 96} 97 98static av_cold int libvorbis_setup(vorbis_info *vi, AVCodecContext *avctx) 99{ 100 LibvorbisEncContext *s = avctx->priv_data; 101 int channels = avctx->ch_layout.nb_channels; 102 double cfreq; 103 int ret; 104 105 if (avctx->flags & AV_CODEC_FLAG_QSCALE || !avctx->bit_rate) { 106 /* variable bitrate 107 * NOTE: we use the oggenc range of -1 to 10 for global_quality for 108 * user convenience, but libvorbis uses -0.1 to 1.0. 109 */ 110 float q = avctx->global_quality / (float)FF_QP2LAMBDA; 111 /* default to 3 if the user did not set quality or bitrate */ 112 if (!(avctx->flags & AV_CODEC_FLAG_QSCALE)) 113 q = 3.0; 114 if ((ret = vorbis_encode_setup_vbr(vi, channels, 115 avctx->sample_rate, 116 q / 10.0))) 117 goto error; 118 } else { 119 int minrate = avctx->rc_min_rate > 0 ? avctx->rc_min_rate : -1; 120 int maxrate = avctx->rc_max_rate > 0 ? avctx->rc_max_rate : -1; 121 122 /* average bitrate */ 123 if ((ret = vorbis_encode_setup_managed(vi, channels, 124 avctx->sample_rate, maxrate, 125 avctx->bit_rate, minrate))) 126 goto error; 127 128 /* variable bitrate by estimate, disable slow rate management */ 129 if (minrate == -1 && maxrate == -1) 130 if ((ret = vorbis_encode_ctl(vi, OV_ECTL_RATEMANAGE2_SET, NULL))) 131 goto error; /* should not happen */ 132 } 133 134 /* cutoff frequency */ 135 if (avctx->cutoff > 0) { 136 cfreq = avctx->cutoff / 1000.0; 137 if ((ret = vorbis_encode_ctl(vi, OV_ECTL_LOWPASS_SET, &cfreq))) 138 goto error; /* should not happen */ 139 } 140 141 /* impulse block bias */ 142 if (s->iblock) { 143 if ((ret = vorbis_encode_ctl(vi, OV_ECTL_IBLOCK_SET, &s->iblock))) 144 goto error; 145 } 146 147 if ((channels == 3 && 148 av_channel_layout_compare(&avctx->ch_layout, &(AVChannelLayout)AV_CHANNEL_LAYOUT_SURROUND)) || 149 (channels == 4 && 150 av_channel_layout_compare(&avctx->ch_layout, &(AVChannelLayout)AV_CHANNEL_LAYOUT_2_2) && 151 av_channel_layout_compare(&avctx->ch_layout, &(AVChannelLayout)AV_CHANNEL_LAYOUT_QUAD)) || 152 (channels == 5 && 153 av_channel_layout_compare(&avctx->ch_layout, &(AVChannelLayout)AV_CHANNEL_LAYOUT_5POINT0) && 154 av_channel_layout_compare(&avctx->ch_layout, &(AVChannelLayout)AV_CHANNEL_LAYOUT_5POINT0_BACK)) || 155 (channels == 6 && 156 av_channel_layout_compare(&avctx->ch_layout, &(AVChannelLayout)AV_CHANNEL_LAYOUT_5POINT1) && 157 av_channel_layout_compare(&avctx->ch_layout, &(AVChannelLayout)AV_CHANNEL_LAYOUT_5POINT1_BACK)) || 158 (channels == 7 && 159 av_channel_layout_compare(&avctx->ch_layout, &(AVChannelLayout)AV_CHANNEL_LAYOUT_6POINT1)) || 160 (channels == 8 && 161 av_channel_layout_compare(&avctx->ch_layout, &(AVChannelLayout)AV_CHANNEL_LAYOUT_7POINT1))) { 162 if (avctx->ch_layout.order != AV_CHANNEL_ORDER_UNSPEC) { 163 char name[32]; 164 av_channel_layout_describe(&avctx->ch_layout, name, sizeof(name)); 165 av_log(avctx, AV_LOG_ERROR, "%s not supported by Vorbis: " 166 "output stream will have incorrect " 167 "channel layout.\n", name); 168 } else { 169 av_log(avctx, AV_LOG_WARNING, "No channel layout specified. The encoder " 170 "will use Vorbis channel layout for " 171 "%d channels.\n", channels); 172 } 173 } 174 175 if ((ret = vorbis_encode_setup_init(vi))) 176 goto error; 177 178 return 0; 179error: 180 return vorbis_error_to_averror(ret); 181} 182 183/* How many bytes are needed for a buffer of length 'l' */ 184static int xiph_len(int l) 185{ 186 return 1 + l / 255 + l; 187} 188 189static av_cold int libvorbis_encode_close(AVCodecContext *avctx) 190{ 191 LibvorbisEncContext *s = avctx->priv_data; 192 193 /* notify vorbisenc this is EOF */ 194 if (s->dsp_initialized) 195 vorbis_analysis_wrote(&s->vd, 0); 196 197 vorbis_block_clear(&s->vb); 198 vorbis_dsp_clear(&s->vd); 199 vorbis_info_clear(&s->vi); 200 201 av_fifo_freep2(&s->pkt_fifo); 202 ff_af_queue_close(&s->afq); 203 204 av_vorbis_parse_free(&s->vp); 205 206 return 0; 207} 208 209static av_cold int libvorbis_encode_init(AVCodecContext *avctx) 210{ 211 LibvorbisEncContext *s = avctx->priv_data; 212 ogg_packet header, header_comm, header_code; 213 uint8_t *p; 214 unsigned int offset; 215 int ret; 216 217 vorbis_info_init(&s->vi); 218 if ((ret = libvorbis_setup(&s->vi, avctx))) { 219 av_log(avctx, AV_LOG_ERROR, "encoder setup failed\n"); 220 goto error; 221 } 222 if ((ret = vorbis_analysis_init(&s->vd, &s->vi))) { 223 av_log(avctx, AV_LOG_ERROR, "analysis init failed\n"); 224 ret = vorbis_error_to_averror(ret); 225 goto error; 226 } 227 s->dsp_initialized = 1; 228 if ((ret = vorbis_block_init(&s->vd, &s->vb))) { 229 av_log(avctx, AV_LOG_ERROR, "dsp init failed\n"); 230 ret = vorbis_error_to_averror(ret); 231 goto error; 232 } 233 234 vorbis_comment_init(&s->vc); 235 if (!(avctx->flags & AV_CODEC_FLAG_BITEXACT)) 236 vorbis_comment_add_tag(&s->vc, "encoder", LIBAVCODEC_IDENT); 237 238 if ((ret = vorbis_analysis_headerout(&s->vd, &s->vc, &header, &header_comm, 239 &header_code))) { 240 ret = vorbis_error_to_averror(ret); 241 goto error; 242 } 243 244 avctx->extradata_size = 1 + xiph_len(header.bytes) + 245 xiph_len(header_comm.bytes) + 246 header_code.bytes; 247 p = avctx->extradata = av_malloc(avctx->extradata_size + 248 AV_INPUT_BUFFER_PADDING_SIZE); 249 if (!p) { 250 ret = AVERROR(ENOMEM); 251 goto error; 252 } 253 p[0] = 2; 254 offset = 1; 255 offset += av_xiphlacing(&p[offset], header.bytes); 256 offset += av_xiphlacing(&p[offset], header_comm.bytes); 257 memcpy(&p[offset], header.packet, header.bytes); 258 offset += header.bytes; 259 memcpy(&p[offset], header_comm.packet, header_comm.bytes); 260 offset += header_comm.bytes; 261 memcpy(&p[offset], header_code.packet, header_code.bytes); 262 offset += header_code.bytes; 263 av_assert0(offset == avctx->extradata_size); 264 265 s->vp = av_vorbis_parse_init(avctx->extradata, avctx->extradata_size); 266 if (!s->vp) { 267 av_log(avctx, AV_LOG_ERROR, "invalid extradata\n"); 268 return ret; 269 } 270 271 vorbis_comment_clear(&s->vc); 272 273 avctx->frame_size = LIBVORBIS_FRAME_SIZE; 274 ff_af_queue_init(avctx, &s->afq); 275 276 s->pkt_fifo = av_fifo_alloc2(BUFFER_SIZE, 1, 0); 277 if (!s->pkt_fifo) { 278 ret = AVERROR(ENOMEM); 279 goto error; 280 } 281 282 return 0; 283error: 284 libvorbis_encode_close(avctx); 285 return ret; 286} 287 288static int libvorbis_encode_frame(AVCodecContext *avctx, AVPacket *avpkt, 289 const AVFrame *frame, int *got_packet_ptr) 290{ 291 LibvorbisEncContext *s = avctx->priv_data; 292 ogg_packet op; 293 int ret, duration; 294 295 /* send samples to libvorbis */ 296 if (frame) { 297 const int samples = frame->nb_samples; 298 float **buffer; 299 int c, channels = s->vi.channels; 300 301 buffer = vorbis_analysis_buffer(&s->vd, samples); 302 for (c = 0; c < channels; c++) { 303 int co = (channels > 8) ? c : 304 vorbis_encoding_channel_layout_offsets[channels - 1][c]; 305 memcpy(buffer[c], frame->extended_data[co], 306 samples * sizeof(*buffer[c])); 307 } 308 if ((ret = vorbis_analysis_wrote(&s->vd, samples)) < 0) { 309 av_log(avctx, AV_LOG_ERROR, "error in vorbis_analysis_wrote()\n"); 310 return vorbis_error_to_averror(ret); 311 } 312 if ((ret = ff_af_queue_add(&s->afq, frame)) < 0) 313 return ret; 314 } else { 315 if (!s->eof && s->afq.frame_alloc) 316 if ((ret = vorbis_analysis_wrote(&s->vd, 0)) < 0) { 317 av_log(avctx, AV_LOG_ERROR, "error in vorbis_analysis_wrote()\n"); 318 return vorbis_error_to_averror(ret); 319 } 320 s->eof = 1; 321 } 322 323 /* retrieve available packets from libvorbis */ 324 while ((ret = vorbis_analysis_blockout(&s->vd, &s->vb)) == 1) { 325 if ((ret = vorbis_analysis(&s->vb, NULL)) < 0) 326 break; 327 if ((ret = vorbis_bitrate_addblock(&s->vb)) < 0) 328 break; 329 330 /* add any available packets to the output packet buffer */ 331 while ((ret = vorbis_bitrate_flushpacket(&s->vd, &op)) == 1) { 332 if (av_fifo_can_write(s->pkt_fifo) < sizeof(ogg_packet) + op.bytes) { 333 av_log(avctx, AV_LOG_ERROR, "packet buffer is too small\n"); 334 return AVERROR_BUG; 335 } 336 av_fifo_write(s->pkt_fifo, &op, sizeof(ogg_packet)); 337 av_fifo_write(s->pkt_fifo, op.packet, op.bytes); 338 } 339 if (ret < 0) { 340 av_log(avctx, AV_LOG_ERROR, "error getting available packets\n"); 341 break; 342 } 343 } 344 if (ret < 0) { 345 av_log(avctx, AV_LOG_ERROR, "error getting available packets\n"); 346 return vorbis_error_to_averror(ret); 347 } 348 349 /* Read an available packet if possible */ 350 if (av_fifo_read(s->pkt_fifo, &op, sizeof(ogg_packet)) < 0) 351 return 0; 352 353 if ((ret = ff_get_encode_buffer(avctx, avpkt, op.bytes, 0)) < 0) 354 return ret; 355 av_fifo_read(s->pkt_fifo, avpkt->data, op.bytes); 356 357 avpkt->pts = ff_samples_to_time_base(avctx, op.granulepos); 358 359 duration = av_vorbis_parse_frame(s->vp, avpkt->data, avpkt->size); 360 if (duration > 0) { 361 /* we do not know encoder delay until we get the first packet from 362 * libvorbis, so we have to update the AudioFrameQueue counts */ 363 if (!avctx->initial_padding && s->afq.frames) { 364 avctx->initial_padding = duration; 365 av_assert0(!s->afq.remaining_delay); 366 s->afq.frames->duration += duration; 367 if (s->afq.frames->pts != AV_NOPTS_VALUE) 368 s->afq.frames->pts -= duration; 369 s->afq.remaining_samples += duration; 370 } 371 ff_af_queue_remove(&s->afq, duration, &avpkt->pts, &avpkt->duration); 372 } 373 374 *got_packet_ptr = 1; 375 return 0; 376} 377 378const FFCodec ff_libvorbis_encoder = { 379 .p.name = "libvorbis", 380 .p.long_name = NULL_IF_CONFIG_SMALL("libvorbis"), 381 .p.type = AVMEDIA_TYPE_AUDIO, 382 .p.id = AV_CODEC_ID_VORBIS, 383 .p.capabilities = AV_CODEC_CAP_DR1 | AV_CODEC_CAP_DELAY | 384 AV_CODEC_CAP_SMALL_LAST_FRAME, 385 .priv_data_size = sizeof(LibvorbisEncContext), 386 .init = libvorbis_encode_init, 387 FF_CODEC_ENCODE_CB(libvorbis_encode_frame), 388 .close = libvorbis_encode_close, 389 .p.sample_fmts = (const enum AVSampleFormat[]) { AV_SAMPLE_FMT_FLTP, 390 AV_SAMPLE_FMT_NONE }, 391 .p.priv_class = &vorbis_class, 392 .defaults = defaults, 393 .p.wrapper_name = "libvorbis", 394}; 395