aboutsummaryrefslogtreecommitdiffstats
path: root/libavcodec/libmp3lame.c
blob: 6da68ad3b799a02312a9fcdb14532f28311641f1 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
/*
 * Interface to libmp3lame for mp3 encoding
 * Copyright (c) 2002 Lennert Buytenhek <buytenh@gnu.org>
 *
 * This file is part of Libav.
 *
 * Libav 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.
 *
 * Libav 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 Libav; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

/**
 * @file
 * Interface to libmp3lame for mp3 encoding.
 */

#include <lame/lame.h>

#include "libavutil/channel_layout.h"
#include "libavutil/common.h"
#include "libavutil/float_dsp.h"
#include "libavutil/intreadwrite.h"
#include "libavutil/log.h"
#include "libavutil/opt.h"
#include "avcodec.h"
#include "audio_frame_queue.h"
#include "internal.h"
#include "mpegaudio.h"
#include "mpegaudiodecheader.h"

#define BUFFER_SIZE (7200 + 2 * MPA_FRAME_SIZE + MPA_FRAME_SIZE / 4)

typedef struct LAMEContext {
    AVClass *class;
    AVCodecContext *avctx;
    lame_global_flags *gfp;
    uint8_t *buffer;
    int buffer_index;
    int buffer_size;
    int reservoir;
    int joint_stereo;
    float *samples_flt[2];
    AudioFrameQueue afq;
    AVFloatDSPContext fdsp;
} LAMEContext;


static int realloc_buffer(LAMEContext *s)
{
    if (!s->buffer || s->buffer_size - s->buffer_index < BUFFER_SIZE) {
        int new_size = s->buffer_index + 2 * BUFFER_SIZE, err;

        av_dlog(s->avctx, "resizing output buffer: %d -> %d\n", s->buffer_size,
                new_size);
        if ((err = av_reallocp(&s->buffer, new_size)) < 0) {
            s->buffer_size = s->buffer_index = 0;
            return err;
        }
        s->buffer_size = new_size;
    }
    return 0;
}

static av_cold int mp3lame_encode_close(AVCodecContext *avctx)
{
    LAMEContext *s = avctx->priv_data;

    av_freep(&s->samples_flt[0]);
    av_freep(&s->samples_flt[1]);
    av_freep(&s->buffer);

    ff_af_queue_close(&s->afq);

    lame_close(s->gfp);
    return 0;
}

static av_cold int mp3lame_encode_init(AVCodecContext *avctx)
{
    LAMEContext *s = avctx->priv_data;
    int ret;

    s->avctx = avctx;

    /* initialize LAME and get defaults */
    if ((s->gfp = lame_init()) == NULL)
        return AVERROR(ENOMEM);

    lame_set_num_channels(s->gfp, avctx->channels);
    lame_set_mode(s->gfp, avctx->channels > 1 ? s->joint_stereo ? JOINT_STEREO : STEREO : MONO);

    /* sample rate */
    lame_set_in_samplerate (s->gfp, avctx->sample_rate);
    lame_set_out_samplerate(s->gfp, avctx->sample_rate);

    /* algorithmic quality */
    if (avctx->compression_level == FF_COMPRESSION_DEFAULT)
        lame_set_quality(s->gfp, 5);
    else
        lame_set_quality(s->gfp, avctx->compression_level);

    /* rate control */
    if (avctx->flags & CODEC_FLAG_QSCALE) {
        lame_set_VBR(s->gfp, vbr_default);
        lame_set_VBR_quality(s->gfp, avctx->global_quality / (float)FF_QP2LAMBDA);
    } else {
        if (avctx->bit_rate)
            lame_set_brate(s->gfp, avctx->bit_rate / 1000);
    }

    /* do not get a Xing VBR header frame from LAME */
    lame_set_bWriteVbrTag(s->gfp,0);

    /* bit reservoir usage */
    lame_set_disable_reservoir(s->gfp, !s->reservoir);

    /* set specified parameters */
    if (lame_init_params(s->gfp) < 0) {
        ret = -1;
        goto error;
    }

    /* get encoder delay */
    avctx->delay = lame_get_encoder_delay(s->gfp) + 528 + 1;
    ff_af_queue_init(avctx, &s->afq);

    avctx->frame_size  = lame_get_framesize(s->gfp);

    /* allocate float sample buffers */
    if (avctx->sample_fmt == AV_SAMPLE_FMT_FLTP) {
        int ch;
        for (ch = 0; ch < avctx->channels; ch++) {
            s->samples_flt[ch] = av_malloc(avctx->frame_size *
                                           sizeof(*s->samples_flt[ch]));
            if (!s->samples_flt[ch]) {
                ret = AVERROR(ENOMEM);
                goto error;
            }
        }
    }

    ret = realloc_buffer(s);
    if (ret < 0)
        goto error;

    avpriv_float_dsp_init(&s->fdsp, avctx->flags & CODEC_FLAG_BITEXACT);

    return 0;
error:
    mp3lame_encode_close(avctx);
    return ret;
}

#define ENCODE_BUFFER(func, buf_type, buf_name) do {                        \
    lame_result = func(s->gfp,                                              \
                       (const buf_type *)buf_name[0],                       \
                       (const buf_type *)buf_name[1], frame->nb_samples,    \
                       s->buffer + s->buffer_index,                         \
                       s->buffer_size - s->buffer_index);                   \
} while (0)

static int mp3lame_encode_frame(AVCodecContext *avctx, AVPacket *avpkt,
                                const AVFrame *frame, int *got_packet_ptr)
{
    LAMEContext *s = avctx->priv_data;
    MPADecodeHeader hdr;
    int len, ret, ch;
    int lame_result;

    if (frame) {
        switch (avctx->sample_fmt) {
        case AV_SAMPLE_FMT_S16P:
            ENCODE_BUFFER(lame_encode_buffer, int16_t, frame->data);
            break;
        case AV_SAMPLE_FMT_S32P:
            ENCODE_BUFFER(lame_encode_buffer_int, int32_t, frame->data);
            break;
        case AV_SAMPLE_FMT_FLTP:
            if (frame->linesize[0] < 4 * FFALIGN(frame->nb_samples, 8)) {
                av_log(avctx, AV_LOG_ERROR, "inadequate AVFrame plane padding\n");
                return AVERROR(EINVAL);
            }
            for (ch = 0; ch < avctx->channels; ch++) {
                s->fdsp.vector_fmul_scalar(s->samples_flt[ch],
                                           (const float *)frame->data[ch],
                                           32768.0f,
                                           FFALIGN(frame->nb_samples, 8));
            }
            ENCODE_BUFFER(lame_encode_buffer_float, float, s->samples_flt);
            break;
        default:
            return AVERROR_BUG;
        }
    } else {
        lame_result = lame_encode_flush(s->gfp, s->buffer + s->buffer_index,
                                        s->buffer_size - s->buffer_index);
    }
    if (lame_result < 0) {
        if (lame_result == -1) {
            av_log(avctx, AV_LOG_ERROR,
                   "lame: output buffer too small (buffer index: %d, free bytes: %d)\n",
                   s->buffer_index, s->buffer_size - s->buffer_index);
        }
        return -1;
    }
    s->buffer_index += lame_result;
    ret = realloc_buffer(s);
    if (ret < 0) {
        av_log(avctx, AV_LOG_ERROR, "error reallocating output buffer\n");
        return ret;
    }

    /* add current frame to the queue */
    if (frame) {
        if ((ret = ff_af_queue_add(&s->afq, frame)) < 0)
            return ret;
    }

    /* Move 1 frame from the LAME buffer to the output packet, if available.
       We have to parse the first frame header in the output buffer to
       determine the frame size. */
    if (s->buffer_index < 4)
        return 0;
    if (avpriv_mpegaudio_decode_header(&hdr, AV_RB32(s->buffer))) {
        av_log(avctx, AV_LOG_ERROR, "free format output not supported\n");
        return -1;
    }
    len = hdr.frame_size;
    av_dlog(avctx, "in:%d packet-len:%d index:%d\n", avctx->frame_size, len,
            s->buffer_index);
    if (len <= s->buffer_index) {
        if ((ret = ff_alloc_packet(avpkt, len))) {
            av_log(avctx, AV_LOG_ERROR, "Error getting output packet\n");
            return ret;
        }
        memcpy(avpkt->data, s->buffer, len);
        s->buffer_index -= len;
        memmove(s->buffer, s->buffer + len, s->buffer_index);

        /* Get the next frame pts/duration */
        ff_af_queue_remove(&s->afq, avctx->frame_size, &avpkt->pts,
                           &avpkt->duration);

        avpkt->size = len;
        *got_packet_ptr = 1;
    }
    return 0;
}

#define OFFSET(x) offsetof(LAMEContext, x)
#define AE AV_OPT_FLAG_AUDIO_PARAM | AV_OPT_FLAG_ENCODING_PARAM
static const AVOption options[] = {
    { "reservoir", "Use bit reservoir.", OFFSET(reservoir), AV_OPT_TYPE_INT, { .i64 = 1 }, 0, 1, AE },
    { "joint_stereo", "Use joint stereo.", OFFSET(joint_stereo), AV_OPT_TYPE_INT, { .i64 = 1 }, 0, 1, AE },
    { NULL },
};

static const AVClass libmp3lame_class = {
    .class_name = "libmp3lame encoder",
    .item_name  = av_default_item_name,
    .option     = options,
    .version    = LIBAVUTIL_VERSION_INT,
};

static const AVCodecDefault libmp3lame_defaults[] = {
    { "b",          "0" },
    { NULL },
};

static const int libmp3lame_sample_rates[] = {
    44100, 48000,  32000, 22050, 24000, 16000, 11025, 12000, 8000, 0
};

AVCodec ff_libmp3lame_encoder = {
    .name                  = "libmp3lame",
    .long_name             = NULL_IF_CONFIG_SMALL("libmp3lame MP3 (MPEG audio layer 3)"),
    .type                  = AVMEDIA_TYPE_AUDIO,
    .id                    = AV_CODEC_ID_MP3,
    .priv_data_size        = sizeof(LAMEContext),
    .init                  = mp3lame_encode_init,
    .encode2               = mp3lame_encode_frame,
    .close                 = mp3lame_encode_close,
    .capabilities          = CODEC_CAP_DELAY | CODEC_CAP_SMALL_LAST_FRAME,
    .sample_fmts           = (const enum AVSampleFormat[]) { AV_SAMPLE_FMT_S32P,
                                                             AV_SAMPLE_FMT_FLTP,
                                                             AV_SAMPLE_FMT_S16P,
                                                             AV_SAMPLE_FMT_NONE },
    .supported_samplerates = libmp3lame_sample_rates,
    .channel_layouts       = (const uint64_t[]) { AV_CH_LAYOUT_MONO,
                                                  AV_CH_LAYOUT_STEREO,
                                                  0 },
    .priv_class            = &libmp3lame_class,
    .defaults              = libmp3lame_defaults,
};