avformat/mpeg: demux ivtv captions
[ffmpeg.git] / libavcodec / libtwolame.c
blobf3bd4771d713635fdf8888b26c93ccb9acbdf468
1 /*
2 * Interface to libtwolame for mp2 encoding
3 * Copyright (c) 2012 Paul B Mahol
5 * This file is part of FFmpeg.
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.
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.
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
22 /**
23 * @file
24 * Interface to libtwolame for mp2 encoding.
27 #include <twolame.h>
29 #include "libavutil/channel_layout.h"
30 #include "libavutil/common.h"
31 #include "libavutil/opt.h"
33 #include "avcodec.h"
34 #include "codec_internal.h"
35 #include "encode.h"
36 #include "mpegaudio.h"
38 typedef struct TWOLAMEContext {
39 AVClass *class;
40 int mode;
41 int psymodel;
42 int energy;
43 int error_protection;
44 int copyright;
45 int original;
46 int verbosity;
48 twolame_options *glopts;
49 int64_t next_pts;
50 } TWOLAMEContext;
52 static av_cold int twolame_encode_close(AVCodecContext *avctx)
54 TWOLAMEContext *s = avctx->priv_data;
55 twolame_close(&s->glopts);
56 return 0;
59 static av_cold int twolame_encode_init(AVCodecContext *avctx)
61 TWOLAMEContext *s = avctx->priv_data;
62 int ret;
64 avctx->frame_size = TWOLAME_SAMPLES_PER_FRAME;
65 avctx->initial_padding = 512 - 32 + 1;
67 s->glopts = twolame_init();
68 if (!s->glopts)
69 return AVERROR(ENOMEM);
71 twolame_set_verbosity(s->glopts, s->verbosity);
72 twolame_set_mode(s->glopts, s->mode);
73 twolame_set_psymodel(s->glopts, s->psymodel);
74 twolame_set_energy_levels(s->glopts, s->energy);
75 twolame_set_error_protection(s->glopts, s->error_protection);
76 twolame_set_copyright(s->glopts, s->copyright);
77 twolame_set_original(s->glopts, s->original);
79 twolame_set_num_channels(s->glopts, avctx->ch_layout.nb_channels);
80 twolame_set_in_samplerate(s->glopts, avctx->sample_rate);
81 twolame_set_out_samplerate(s->glopts, avctx->sample_rate);
83 if (!avctx->bit_rate) {
84 if ((s->mode == TWOLAME_AUTO_MODE && avctx->ch_layout.nb_channels == 1) || s->mode == TWOLAME_MONO)
85 avctx->bit_rate = avctx->sample_rate < 28000 ? 80000 : 192000;
86 else
87 avctx->bit_rate = avctx->sample_rate < 28000 ? 160000 : 384000;
90 if (avctx->flags & AV_CODEC_FLAG_QSCALE || !avctx->bit_rate) {
91 twolame_set_VBR(s->glopts, TRUE);
92 twolame_set_VBR_level(s->glopts,
93 avctx->global_quality / (float) FF_QP2LAMBDA);
94 av_log(avctx, AV_LOG_WARNING,
95 "VBR in MP2 is a hack, use another codec that supports it.\n");
96 } else {
97 twolame_set_bitrate(s->glopts, avctx->bit_rate / 1000);
100 ret = twolame_init_params(s->glopts);
101 if (ret) {
102 twolame_encode_close(avctx);
103 return AVERROR_UNKNOWN;
106 return 0;
109 static int twolame_encode_frame(AVCodecContext *avctx, AVPacket *avpkt,
110 const AVFrame *frame, int *got_packet_ptr)
112 TWOLAMEContext *s = avctx->priv_data;
113 int ret;
115 if ((ret = ff_alloc_packet(avctx, avpkt, MPA_MAX_CODED_FRAME_SIZE)) < 0)
116 return ret;
118 if (frame) {
119 switch (avctx->sample_fmt) {
120 case AV_SAMPLE_FMT_FLT:
121 ret = twolame_encode_buffer_float32_interleaved(s->glopts,
122 (const float *)frame->data[0],
123 frame->nb_samples,
124 avpkt->data,
125 avpkt->size);
126 break;
127 case AV_SAMPLE_FMT_FLTP:
128 ret = twolame_encode_buffer_float32(s->glopts,
129 (const float *)frame->data[0],
130 (const float *)frame->data[1],
131 frame->nb_samples,
132 avpkt->data, avpkt->size);
133 break;
134 case AV_SAMPLE_FMT_S16:
135 ret = twolame_encode_buffer_interleaved(s->glopts,
136 (const short int *)frame->data[0],
137 frame->nb_samples,
138 avpkt->data, avpkt->size);
139 break;
140 case AV_SAMPLE_FMT_S16P:
141 ret = twolame_encode_buffer(s->glopts,
142 (const short int *)frame->data[0],
143 (const short int *)frame->data[1],
144 frame->nb_samples,
145 avpkt->data, avpkt->size);
146 break;
147 default:
148 av_log(avctx, AV_LOG_ERROR,
149 "Unsupported sample format %d.\n", avctx->sample_fmt);
150 return AVERROR_BUG;
152 } else {
153 ret = twolame_encode_flush(s->glopts, avpkt->data, avpkt->size);
156 if (!ret) // no bytes written
157 return 0;
158 if (ret < 0) // twolame error
159 return AVERROR_UNKNOWN;
161 if (frame) {
162 avpkt->duration = ff_samples_to_time_base(avctx, frame->nb_samples);
163 if (frame->pts != AV_NOPTS_VALUE)
164 avpkt->pts = frame->pts - ff_samples_to_time_base(avctx, avctx->initial_padding);
165 } else {
166 avpkt->pts = s->next_pts;
168 // this is for setting pts for flushed packet(s).
169 if (avpkt->pts != AV_NOPTS_VALUE)
170 s->next_pts = avpkt->pts + avpkt->duration;
172 av_shrink_packet(avpkt, ret);
173 *got_packet_ptr = 1;
174 return 0;
177 #define OFFSET(x) offsetof(TWOLAMEContext, x)
178 #define AE AV_OPT_FLAG_AUDIO_PARAM | AV_OPT_FLAG_ENCODING_PARAM
179 static const AVOption options[] = {
180 { "mode", "Mpeg Mode", OFFSET(mode), AV_OPT_TYPE_INT, { .i64 = TWOLAME_AUTO_MODE }, TWOLAME_AUTO_MODE, TWOLAME_MONO, AE, .unit = "mode"},
181 { "auto", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = TWOLAME_AUTO_MODE }, 0, 0, AE, .unit = "mode" },
182 { "stereo", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = TWOLAME_STEREO }, 0, 0, AE, .unit = "mode" },
183 { "joint_stereo", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = TWOLAME_JOINT_STEREO }, 0, 0, AE, .unit = "mode" },
184 { "dual_channel", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = TWOLAME_DUAL_CHANNEL }, 0, 0, AE, .unit = "mode" },
185 { "mono", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = TWOLAME_MONO }, 0, 0, AE, .unit = "mode" },
186 { "psymodel", "Psychoacoustic Model", OFFSET(psymodel), AV_OPT_TYPE_INT, { .i64 = 3 }, -1, 4, AE},
187 { "energy_levels","enable energy levels", OFFSET(energy), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, AE},
188 { "error_protection","enable CRC error protection", OFFSET(error_protection), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, AE},
189 { "copyright", "set MPEG Audio Copyright flag", OFFSET(copyright), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, AE},
190 { "original", "set MPEG Audio Original flag", OFFSET(original), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, AE},
191 { "verbosity", "set library optput level (0-10)", OFFSET(verbosity), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 10, AE},
192 { NULL },
195 static const AVClass twolame_class = {
196 .class_name = "libtwolame encoder",
197 .item_name = av_default_item_name,
198 .option = options,
199 .version = LIBAVUTIL_VERSION_INT,
202 static const FFCodecDefault twolame_defaults[] = {
203 { "b", "0" },
204 { NULL },
207 static const int twolame_samplerates[] = {
208 16000, 22050, 24000, 32000, 44100, 48000, 0
211 const FFCodec ff_libtwolame_encoder = {
212 .p.name = "libtwolame",
213 CODEC_LONG_NAME("libtwolame MP2 (MPEG audio layer 2)"),
214 .p.type = AVMEDIA_TYPE_AUDIO,
215 .p.id = AV_CODEC_ID_MP2,
216 .p.capabilities = AV_CODEC_CAP_DR1 | AV_CODEC_CAP_DELAY,
217 .caps_internal = FF_CODEC_CAP_NOT_INIT_THREADSAFE,
218 .priv_data_size = sizeof(TWOLAMEContext),
219 .init = twolame_encode_init,
220 FF_CODEC_ENCODE_CB(twolame_encode_frame),
221 .close = twolame_encode_close,
222 .defaults = twolame_defaults,
223 .p.priv_class = &twolame_class,
224 .p.sample_fmts = (const enum AVSampleFormat[]) {
225 AV_SAMPLE_FMT_FLT,
226 AV_SAMPLE_FMT_FLTP,
227 AV_SAMPLE_FMT_S16,
228 AV_SAMPLE_FMT_S16P,
229 AV_SAMPLE_FMT_NONE
231 .p.ch_layouts = (const AVChannelLayout[]) {
232 AV_CHANNEL_LAYOUT_MONO,
233 AV_CHANNEL_LAYOUT_STEREO,
234 { 0 },
236 .p.supported_samplerates = twolame_samplerates,
237 .p.wrapper_name = "libtwolame",