| 1 | /* |
| 2 | * Interface to libtwolame for mp2 encoding |
| 3 | * Copyright (c) 2012 Paul B Mahol |
| 4 | * |
| 5 | * This file is part of FFmpeg. |
| 6 | * |
| 7 | * FFmpeg is free software; you can redistribute it and/or |
| 8 | * modify it under the terms of the GNU Lesser General Public |
| 9 | * License as published by the Free Software Foundation; either |
| 10 | * version 2.1 of the License, or (at your option) any later version. |
| 11 | * |
| 12 | * FFmpeg is distributed in the hope that it will be useful, |
| 13 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 14 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 15 | * Lesser General Public License for more details. |
| 16 | * |
| 17 | * You should have received a copy of the GNU Lesser General Public |
| 18 | * License along with FFmpeg; if not, write to the Free Software |
| 19 | * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA |
| 20 | */ |
| 21 | |
| 22 | /** |
| 23 | * @file |
| 24 | * Interface to libtwolame for mp2 encoding. |
| 25 | */ |
| 26 | |
| 27 | #include <twolame.h> |
| 28 | |
| 29 | #include "libavutil/common.h" |
| 30 | #include "libavutil/opt.h" |
| 31 | |
| 32 | #include "avcodec.h" |
| 33 | #include "internal.h" |
| 34 | #include "mpegaudio.h" |
| 35 | |
| 36 | typedef struct TWOLAMEContext { |
| 37 | AVClass *class; |
| 38 | int mode; |
| 39 | int psymodel; |
| 40 | int energy; |
| 41 | int error_protection; |
| 42 | int copyright; |
| 43 | int original; |
| 44 | int verbosity; |
| 45 | |
| 46 | twolame_options *glopts; |
| 47 | int64_t next_pts; |
| 48 | } TWOLAMEContext; |
| 49 | |
| 50 | static av_cold int twolame_encode_close(AVCodecContext *avctx) |
| 51 | { |
| 52 | TWOLAMEContext *s = avctx->priv_data; |
| 53 | twolame_close(&s->glopts); |
| 54 | return 0; |
| 55 | } |
| 56 | |
| 57 | static av_cold int twolame_encode_init(AVCodecContext *avctx) |
| 58 | { |
| 59 | TWOLAMEContext *s = avctx->priv_data; |
| 60 | int ret; |
| 61 | |
| 62 | avctx->frame_size = TWOLAME_SAMPLES_PER_FRAME; |
| 63 | avctx->initial_padding = 512 - 32 + 1; |
| 64 | |
| 65 | s->glopts = twolame_init(); |
| 66 | if (!s->glopts) |
| 67 | return AVERROR(ENOMEM); |
| 68 | |
| 69 | twolame_set_verbosity(s->glopts, s->verbosity); |
| 70 | twolame_set_mode(s->glopts, s->mode); |
| 71 | twolame_set_psymodel(s->glopts, s->psymodel); |
| 72 | twolame_set_energy_levels(s->glopts, s->energy); |
| 73 | twolame_set_error_protection(s->glopts, s->error_protection); |
| 74 | twolame_set_copyright(s->glopts, s->copyright); |
| 75 | twolame_set_original(s->glopts, s->original); |
| 76 | |
| 77 | twolame_set_num_channels(s->glopts, avctx->channels); |
| 78 | twolame_set_in_samplerate(s->glopts, avctx->sample_rate); |
| 79 | twolame_set_out_samplerate(s->glopts, avctx->sample_rate); |
| 80 | |
| 81 | if (!avctx->bit_rate) |
| 82 | avctx->bit_rate = avctx->sample_rate < 28000 ? 160000 : 384000; |
| 83 | |
| 84 | if (avctx->flags & CODEC_FLAG_QSCALE || !avctx->bit_rate) { |
| 85 | twolame_set_VBR(s->glopts, TRUE); |
| 86 | twolame_set_VBR_level(s->glopts, |
| 87 | avctx->global_quality / (float) FF_QP2LAMBDA); |
| 88 | av_log(avctx, AV_LOG_WARNING, |
| 89 | "VBR in MP2 is a hack, use another codec that supports it.\n"); |
| 90 | } else { |
| 91 | twolame_set_bitrate(s->glopts, avctx->bit_rate / 1000); |
| 92 | } |
| 93 | |
| 94 | ret = twolame_init_params(s->glopts); |
| 95 | if (ret) { |
| 96 | twolame_encode_close(avctx); |
| 97 | return AVERROR_UNKNOWN; |
| 98 | } |
| 99 | |
| 100 | return 0; |
| 101 | } |
| 102 | |
| 103 | static int twolame_encode_frame(AVCodecContext *avctx, AVPacket *avpkt, |
| 104 | const AVFrame *frame, int *got_packet_ptr) |
| 105 | { |
| 106 | TWOLAMEContext *s = avctx->priv_data; |
| 107 | int ret; |
| 108 | |
| 109 | if ((ret = ff_alloc_packet2(avctx, avpkt, MPA_MAX_CODED_FRAME_SIZE)) < 0) |
| 110 | return ret; |
| 111 | |
| 112 | if (frame) { |
| 113 | switch (avctx->sample_fmt) { |
| 114 | case AV_SAMPLE_FMT_FLT: |
| 115 | ret = twolame_encode_buffer_float32_interleaved(s->glopts, |
| 116 | (const float *)frame->data[0], |
| 117 | frame->nb_samples, |
| 118 | avpkt->data, |
| 119 | avpkt->size); |
| 120 | break; |
| 121 | case AV_SAMPLE_FMT_FLTP: |
| 122 | ret = twolame_encode_buffer_float32(s->glopts, |
| 123 | (const float *)frame->data[0], |
| 124 | (const float *)frame->data[1], |
| 125 | frame->nb_samples, |
| 126 | avpkt->data, avpkt->size); |
| 127 | break; |
| 128 | case AV_SAMPLE_FMT_S16: |
| 129 | ret = twolame_encode_buffer_interleaved(s->glopts, |
| 130 | (const short int *)frame->data[0], |
| 131 | frame->nb_samples, |
| 132 | avpkt->data, avpkt->size); |
| 133 | break; |
| 134 | case AV_SAMPLE_FMT_S16P: |
| 135 | ret = twolame_encode_buffer(s->glopts, |
| 136 | (const short int *)frame->data[0], |
| 137 | (const short int *)frame->data[1], |
| 138 | frame->nb_samples, |
| 139 | avpkt->data, avpkt->size); |
| 140 | break; |
| 141 | default: |
| 142 | av_log(avctx, AV_LOG_ERROR, |
| 143 | "Unsupported sample format %d.\n", avctx->sample_fmt); |
| 144 | return AVERROR_BUG; |
| 145 | } |
| 146 | } else { |
| 147 | ret = twolame_encode_flush(s->glopts, avpkt->data, avpkt->size); |
| 148 | } |
| 149 | |
| 150 | if (!ret) // no bytes written |
| 151 | return 0; |
| 152 | if (ret < 0) // twolame error |
| 153 | return AVERROR_UNKNOWN; |
| 154 | |
| 155 | if (frame) { |
| 156 | avpkt->duration = ff_samples_to_time_base(avctx, frame->nb_samples); |
| 157 | if (frame->pts != AV_NOPTS_VALUE) |
| 158 | avpkt->pts = frame->pts - ff_samples_to_time_base(avctx, avctx->initial_padding); |
| 159 | } else { |
| 160 | avpkt->pts = s->next_pts; |
| 161 | } |
| 162 | // this is for setting pts for flushed packet(s). |
| 163 | if (avpkt->pts != AV_NOPTS_VALUE) |
| 164 | s->next_pts = avpkt->pts + avpkt->duration; |
| 165 | |
| 166 | av_shrink_packet(avpkt, ret); |
| 167 | *got_packet_ptr = 1; |
| 168 | return 0; |
| 169 | } |
| 170 | |
| 171 | #define OFFSET(x) offsetof(TWOLAMEContext, x) |
| 172 | #define AE AV_OPT_FLAG_AUDIO_PARAM | AV_OPT_FLAG_ENCODING_PARAM |
| 173 | static const AVOption options[] = { |
| 174 | { "mode", "Mpeg Mode", OFFSET(mode), AV_OPT_TYPE_INT, { .i64 = TWOLAME_AUTO_MODE }, TWOLAME_AUTO_MODE, TWOLAME_MONO, AE, "mode"}, |
| 175 | { "auto", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = TWOLAME_AUTO_MODE }, 0, 0, AE, "mode" }, |
| 176 | { "stereo", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = TWOLAME_STEREO }, 0, 0, AE, "mode" }, |
| 177 | { "joint_stereo", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = TWOLAME_JOINT_STEREO }, 0, 0, AE, "mode" }, |
| 178 | { "dual_channel", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = TWOLAME_DUAL_CHANNEL }, 0, 0, AE, "mode" }, |
| 179 | { "mono", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = TWOLAME_MONO }, 0, 0, AE, "mode" }, |
| 180 | { "psymodel", "Psychoacoustic Model", OFFSET(psymodel), AV_OPT_TYPE_INT, { .i64 = 3 }, -1, 4, AE}, |
| 181 | { "energy_levels","enable energy levels", OFFSET(energy), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, AE}, |
| 182 | { "error_protection","enable CRC error protection", OFFSET(error_protection), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, AE}, |
| 183 | { "copyright", "set MPEG Audio Copyright flag", OFFSET(copyright), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, AE}, |
| 184 | { "original", "set MPEG Audio Original flag", OFFSET(original), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, AE}, |
| 185 | { "verbosity", "set library optput level (0-10)", OFFSET(verbosity), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 10, AE}, |
| 186 | { NULL }, |
| 187 | }; |
| 188 | |
| 189 | static const AVClass twolame_class = { |
| 190 | .class_name = "libtwolame encoder", |
| 191 | .item_name = av_default_item_name, |
| 192 | .option = options, |
| 193 | .version = LIBAVUTIL_VERSION_INT, |
| 194 | }; |
| 195 | |
| 196 | static const AVCodecDefault twolame_defaults[] = { |
| 197 | { "b", "0" }, |
| 198 | { NULL }, |
| 199 | }; |
| 200 | |
| 201 | static const int twolame_samplerates[] = { |
| 202 | 16000, 22050, 24000, 32000, 44100, 48000, 0 |
| 203 | }; |
| 204 | |
| 205 | AVCodec ff_libtwolame_encoder = { |
| 206 | .name = "libtwolame", |
| 207 | .long_name = NULL_IF_CONFIG_SMALL("libtwolame MP2 (MPEG audio layer 2)"), |
| 208 | .type = AVMEDIA_TYPE_AUDIO, |
| 209 | .id = AV_CODEC_ID_MP2, |
| 210 | .priv_data_size = sizeof(TWOLAMEContext), |
| 211 | .init = twolame_encode_init, |
| 212 | .encode2 = twolame_encode_frame, |
| 213 | .close = twolame_encode_close, |
| 214 | .capabilities = CODEC_CAP_DELAY, |
| 215 | .defaults = twolame_defaults, |
| 216 | .priv_class = &twolame_class, |
| 217 | .sample_fmts = (const enum AVSampleFormat[]) { |
| 218 | AV_SAMPLE_FMT_FLT, |
| 219 | AV_SAMPLE_FMT_FLTP, |
| 220 | AV_SAMPLE_FMT_S16, |
| 221 | AV_SAMPLE_FMT_S16P, |
| 222 | AV_SAMPLE_FMT_NONE |
| 223 | }, |
| 224 | .channel_layouts = (const uint64_t[]) { |
| 225 | AV_CH_LAYOUT_MONO, |
| 226 | AV_CH_LAYOUT_STEREO, |
| 227 | 0 }, |
| 228 | .supported_samplerates = twolame_samplerates, |
| 229 | }; |