libtwolame MP2 encoding support
[libav.git] / libavcodec / libtwolame.c
1 /*
2 * Interface to libtwolame for mp2 encoding
3 * Copyright (c) 2012 Paul B Mahol
4 *
5 * This file is part of Libav.
6 *
7 * Libav 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 * Libav 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 Libav; 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->delay = 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 if (avctx->flags & CODEC_FLAG_QSCALE || !avctx->bit_rate) {
81 twolame_set_VBR(s->glopts, TRUE);
82 twolame_set_VBR_level(s->glopts,
83 avctx->global_quality / (float) FF_QP2LAMBDA);
84 av_log(avctx, AV_LOG_WARNING,
85 "VBR in MP2 is a hack, use another codec that supports it.\n");
86 } else {
87 twolame_set_bitrate(s->glopts, avctx->bit_rate / 1000);
88 }
89
90 ret = twolame_init_params(s->glopts);
91 if (ret) {
92 twolame_encode_close(avctx);
93 return AVERROR_UNKNOWN;
94 }
95
96 return 0;
97 }
98
99 static int twolame_encode_frame(AVCodecContext *avctx, AVPacket *avpkt,
100 const AVFrame *frame, int *got_packet_ptr)
101 {
102 TWOLAMEContext *s = avctx->priv_data;
103 int ret;
104
105 if ((ret = ff_alloc_packet(avpkt, MPA_MAX_CODED_FRAME_SIZE)) < 0)
106 return ret;
107
108 if (frame) {
109 switch (avctx->sample_fmt) {
110 case AV_SAMPLE_FMT_FLT:
111 ret = twolame_encode_buffer_float32_interleaved(s->glopts,
112 (const float *)frame->data[0],
113 frame->nb_samples,
114 avpkt->data,
115 avpkt->size);
116 break;
117 case AV_SAMPLE_FMT_FLTP:
118 ret = twolame_encode_buffer_float32(s->glopts,
119 (const float *)frame->data[0],
120 (const float *)frame->data[1],
121 frame->nb_samples,
122 avpkt->data, avpkt->size);
123 break;
124 case AV_SAMPLE_FMT_S16:
125 ret = twolame_encode_buffer_interleaved(s->glopts,
126 (const short int *)frame->data[0],
127 frame->nb_samples,
128 avpkt->data, avpkt->size);
129 break;
130 case AV_SAMPLE_FMT_S16P:
131 ret = twolame_encode_buffer(s->glopts,
132 (const short int *)frame->data[0],
133 (const short int *)frame->data[1],
134 frame->nb_samples,
135 avpkt->data, avpkt->size);
136 break;
137 default:
138 av_log(avctx, AV_LOG_ERROR,
139 "Unsupported sample format %d.\n", avctx->sample_fmt);
140 return AVERROR_BUG;
141 }
142 } else {
143 ret = twolame_encode_flush(s->glopts, avpkt->data, avpkt->size);
144 }
145
146 if (!ret) // no bytes written
147 return 0;
148 if (ret < 0) // twolame error
149 return AVERROR_UNKNOWN;
150
151 avpkt->duration = ff_samples_to_time_base(avctx, frame->nb_samples);
152 if (frame) {
153 if (frame->pts != AV_NOPTS_VALUE)
154 avpkt->pts = frame->pts - ff_samples_to_time_base(avctx, avctx->delay);
155 } else {
156 avpkt->pts = s->next_pts;
157 }
158 // this is for setting pts for flushed packet(s).
159 if (avpkt->pts != AV_NOPTS_VALUE)
160 s->next_pts = avpkt->pts + avpkt->duration;
161
162 av_shrink_packet(avpkt, ret);
163 *got_packet_ptr = 1;
164 return 0;
165 }
166
167 #define OFFSET(x) offsetof(TWOLAMEContext, x)
168 #define AE AV_OPT_FLAG_AUDIO_PARAM | AV_OPT_FLAG_ENCODING_PARAM
169 static const AVOption options[] = {
170 { "mode", "Mpeg Mode", OFFSET(mode), AV_OPT_TYPE_INT, { .i64 = TWOLAME_AUTO_MODE }, TWOLAME_AUTO_MODE, TWOLAME_MONO, AE, "mode"},
171 { "auto", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = TWOLAME_AUTO_MODE }, 0, 0, AE, "mode" },
172 { "stereo", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = TWOLAME_STEREO }, 0, 0, AE, "mode" },
173 { "joint_stereo", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = TWOLAME_JOINT_STEREO }, 0, 0, AE, "mode" },
174 { "dual_channel", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = TWOLAME_DUAL_CHANNEL }, 0, 0, AE, "mode" },
175 { "mono", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = TWOLAME_MONO }, 0, 0, AE, "mode" },
176 { "psymodel", "Psychoacoustic Model", OFFSET(psymodel), AV_OPT_TYPE_INT, { .i64 = 3 }, -1, 4, AE},
177 { "energy_levels","enable energy levels", OFFSET(energy), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, AE},
178 { "error_protection","enable CRC error protection", OFFSET(error_protection), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, AE},
179 { "copyright", "set MPEG Audio Copyright flag", OFFSET(copyright), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, AE},
180 { "original", "set MPEG Audio Original flag", OFFSET(original), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, AE},
181 { "verbosity", "set library optput level (0-10)", OFFSET(verbosity), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 10, AE},
182 { NULL },
183 };
184
185 static const AVClass twolame_class = {
186 .class_name = "libtwolame encoder",
187 .item_name = av_default_item_name,
188 .option = options,
189 .version = LIBAVUTIL_VERSION_INT,
190 };
191
192 static const AVCodecDefault twolame_defaults[] = {
193 { "b", "384000" },
194 { "ar", "48000" },
195 { NULL },
196 };
197
198 static const int twolame_samplerates[] = {
199 16000, 22050, 24000, 32000, 44100, 48000, 0
200 };
201
202 AVCodec ff_libtwolame_encoder = {
203 .name = "libtwolame",
204 .long_name = NULL_IF_CONFIG_SMALL("libtwolame MP2 (MPEG audio layer 2)"),
205 .type = AVMEDIA_TYPE_AUDIO,
206 .id = AV_CODEC_ID_MP2,
207 .priv_data_size = sizeof(TWOLAMEContext),
208 .init = twolame_encode_init,
209 .encode2 = twolame_encode_frame,
210 .close = twolame_encode_close,
211 .capabilities = CODEC_CAP_DELAY,
212 .defaults = twolame_defaults,
213 .priv_class = &twolame_class,
214 .sample_fmts = (const enum AVSampleFormat[]) {
215 AV_SAMPLE_FMT_FLT,
216 AV_SAMPLE_FMT_FLTP,
217 AV_SAMPLE_FMT_S16,
218 AV_SAMPLE_FMT_S16P,
219 AV_SAMPLE_FMT_NONE
220 },
221 .channel_layouts = (const uint64_t[]) {
222 AV_CH_LAYOUT_MONO,
223 AV_CH_LAYOUT_STEREO,
224 0 },
225 .supported_samplerates = twolame_samplerates,
226 };