1 |
|
|
/* |
2 |
|
|
* Sierra VMD audio decoder |
3 |
|
|
* Copyright (c) 2004 The FFmpeg Project |
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 |
|
|
* Sierra VMD audio decoder |
25 |
|
|
* by Vladimir "VAG" Gneushev (vagsoft at mail.ru) |
26 |
|
|
* for more information on the Sierra VMD format, visit: |
27 |
|
|
* http://www.pcisys.net/~melanson/codecs/ |
28 |
|
|
* |
29 |
|
|
* The audio decoder, expects each encoded data |
30 |
|
|
* chunk to be prepended with the appropriate 16-byte frame information |
31 |
|
|
* record from the VMD file. It does not require the 0x330-byte VMD file |
32 |
|
|
* header, but it does need the audio setup parameters passed in through |
33 |
|
|
* normal libavcodec API means. |
34 |
|
|
*/ |
35 |
|
|
|
36 |
|
|
#include <string.h> |
37 |
|
|
|
38 |
|
|
#include "libavutil/avassert.h" |
39 |
|
|
#include "libavutil/channel_layout.h" |
40 |
|
|
#include "libavutil/common.h" |
41 |
|
|
#include "libavutil/intreadwrite.h" |
42 |
|
|
|
43 |
|
|
#include "avcodec.h" |
44 |
|
|
#include "internal.h" |
45 |
|
|
|
46 |
|
|
#define BLOCK_TYPE_AUDIO 1 |
47 |
|
|
#define BLOCK_TYPE_INITIAL 2 |
48 |
|
|
#define BLOCK_TYPE_SILENCE 3 |
49 |
|
|
|
50 |
|
|
typedef struct VmdAudioContext { |
51 |
|
|
int out_bps; |
52 |
|
|
int chunk_size; |
53 |
|
|
} VmdAudioContext; |
54 |
|
|
|
55 |
|
|
static const uint16_t vmdaudio_table[128] = { |
56 |
|
|
0x000, 0x008, 0x010, 0x020, 0x030, 0x040, 0x050, 0x060, 0x070, 0x080, |
57 |
|
|
0x090, 0x0A0, 0x0B0, 0x0C0, 0x0D0, 0x0E0, 0x0F0, 0x100, 0x110, 0x120, |
58 |
|
|
0x130, 0x140, 0x150, 0x160, 0x170, 0x180, 0x190, 0x1A0, 0x1B0, 0x1C0, |
59 |
|
|
0x1D0, 0x1E0, 0x1F0, 0x200, 0x208, 0x210, 0x218, 0x220, 0x228, 0x230, |
60 |
|
|
0x238, 0x240, 0x248, 0x250, 0x258, 0x260, 0x268, 0x270, 0x278, 0x280, |
61 |
|
|
0x288, 0x290, 0x298, 0x2A0, 0x2A8, 0x2B0, 0x2B8, 0x2C0, 0x2C8, 0x2D0, |
62 |
|
|
0x2D8, 0x2E0, 0x2E8, 0x2F0, 0x2F8, 0x300, 0x308, 0x310, 0x318, 0x320, |
63 |
|
|
0x328, 0x330, 0x338, 0x340, 0x348, 0x350, 0x358, 0x360, 0x368, 0x370, |
64 |
|
|
0x378, 0x380, 0x388, 0x390, 0x398, 0x3A0, 0x3A8, 0x3B0, 0x3B8, 0x3C0, |
65 |
|
|
0x3C8, 0x3D0, 0x3D8, 0x3E0, 0x3E8, 0x3F0, 0x3F8, 0x400, 0x440, 0x480, |
66 |
|
|
0x4C0, 0x500, 0x540, 0x580, 0x5C0, 0x600, 0x640, 0x680, 0x6C0, 0x700, |
67 |
|
|
0x740, 0x780, 0x7C0, 0x800, 0x900, 0xA00, 0xB00, 0xC00, 0xD00, 0xE00, |
68 |
|
|
0xF00, 0x1000, 0x1400, 0x1800, 0x1C00, 0x2000, 0x3000, 0x4000 |
69 |
|
|
}; |
70 |
|
|
|
71 |
|
4 |
static av_cold int vmdaudio_decode_init(AVCodecContext *avctx) |
72 |
|
|
{ |
73 |
|
4 |
VmdAudioContext *s = avctx->priv_data; |
74 |
|
|
|
75 |
✓✗✗✓
|
4 |
if (avctx->channels < 1 || avctx->channels > 2) { |
76 |
|
|
av_log(avctx, AV_LOG_ERROR, "invalid number of channels\n"); |
77 |
|
|
return AVERROR(EINVAL); |
78 |
|
|
} |
79 |
✓✗✓✗
|
4 |
if (avctx->block_align < 1 || avctx->block_align % avctx->channels || |
80 |
✗✓ |
4 |
avctx->block_align > INT_MAX - avctx->channels |
81 |
|
|
) { |
82 |
|
|
av_log(avctx, AV_LOG_ERROR, "invalid block align\n"); |
83 |
|
|
return AVERROR(EINVAL); |
84 |
|
|
} |
85 |
|
|
|
86 |
✓✗ |
4 |
avctx->channel_layout = avctx->channels == 1 ? AV_CH_LAYOUT_MONO : |
87 |
|
|
AV_CH_LAYOUT_STEREO; |
88 |
|
|
|
89 |
✗✓ |
4 |
if (avctx->bits_per_coded_sample == 16) |
90 |
|
|
avctx->sample_fmt = AV_SAMPLE_FMT_S16; |
91 |
|
|
else |
92 |
|
4 |
avctx->sample_fmt = AV_SAMPLE_FMT_U8; |
93 |
|
4 |
s->out_bps = av_get_bytes_per_sample(avctx->sample_fmt); |
94 |
|
|
|
95 |
|
4 |
s->chunk_size = avctx->block_align + avctx->channels * (s->out_bps == 2); |
96 |
|
|
|
97 |
|
4 |
av_log(avctx, AV_LOG_DEBUG, "%d channels, %d bits/sample, " |
98 |
|
|
"block align = %d, sample rate = %d\n", |
99 |
|
|
avctx->channels, avctx->bits_per_coded_sample, avctx->block_align, |
100 |
|
|
avctx->sample_rate); |
101 |
|
|
|
102 |
|
4 |
return 0; |
103 |
|
|
} |
104 |
|
|
|
105 |
|
|
static void decode_audio_s16(int16_t *out, const uint8_t *buf, int buf_size, |
106 |
|
|
int channels) |
107 |
|
|
{ |
108 |
|
|
int ch; |
109 |
|
|
const uint8_t *buf_end = buf + buf_size; |
110 |
|
|
int predictor[2]; |
111 |
|
|
int st = channels - 1; |
112 |
|
|
|
113 |
|
|
/* decode initial raw sample */ |
114 |
|
|
for (ch = 0; ch < channels; ch++) { |
115 |
|
|
predictor[ch] = (int16_t)AV_RL16(buf); |
116 |
|
|
buf += 2; |
117 |
|
|
*out++ = predictor[ch]; |
118 |
|
|
} |
119 |
|
|
|
120 |
|
|
/* decode DPCM samples */ |
121 |
|
|
ch = 0; |
122 |
|
|
while (buf < buf_end) { |
123 |
|
|
uint8_t b = *buf++; |
124 |
|
|
if (b & 0x80) |
125 |
|
|
predictor[ch] -= vmdaudio_table[b & 0x7F]; |
126 |
|
|
else |
127 |
|
|
predictor[ch] += vmdaudio_table[b]; |
128 |
|
|
predictor[ch] = av_clip_int16(predictor[ch]); |
129 |
|
|
*out++ = predictor[ch]; |
130 |
|
|
ch ^= st; |
131 |
|
|
} |
132 |
|
|
} |
133 |
|
|
|
134 |
|
215 |
static int vmdaudio_decode_frame(AVCodecContext *avctx, void *data, |
135 |
|
|
int *got_frame_ptr, AVPacket *avpkt) |
136 |
|
|
{ |
137 |
|
215 |
AVFrame *frame = data; |
138 |
|
215 |
const uint8_t *buf = avpkt->data; |
139 |
|
|
const uint8_t *buf_end; |
140 |
|
215 |
int buf_size = avpkt->size; |
141 |
|
215 |
VmdAudioContext *s = avctx->priv_data; |
142 |
|
|
int block_type, silent_chunks, audio_chunks; |
143 |
|
|
int ret; |
144 |
|
|
uint8_t *output_samples_u8; |
145 |
|
|
int16_t *output_samples_s16; |
146 |
|
|
|
147 |
✗✓ |
215 |
if (buf_size < 16) { |
148 |
|
|
av_log(avctx, AV_LOG_WARNING, "skipping small junk packet\n"); |
149 |
|
|
*got_frame_ptr = 0; |
150 |
|
|
return buf_size; |
151 |
|
|
} |
152 |
|
|
|
153 |
|
215 |
block_type = buf[6]; |
154 |
✓✗✗✓
|
215 |
if (block_type < BLOCK_TYPE_AUDIO || block_type > BLOCK_TYPE_SILENCE) { |
155 |
|
|
av_log(avctx, AV_LOG_ERROR, "unknown block type: %d\n", block_type); |
156 |
|
|
return AVERROR(EINVAL); |
157 |
|
|
} |
158 |
|
215 |
buf += 16; |
159 |
|
215 |
buf_size -= 16; |
160 |
|
|
|
161 |
|
|
/* get number of silent chunks */ |
162 |
|
215 |
silent_chunks = 0; |
163 |
✓✓ |
215 |
if (block_type == BLOCK_TYPE_INITIAL) { |
164 |
|
|
uint32_t flags; |
165 |
✗✓ |
1 |
if (buf_size < 4) { |
166 |
|
|
av_log(avctx, AV_LOG_ERROR, "packet is too small\n"); |
167 |
|
|
return AVERROR(EINVAL); |
168 |
|
|
} |
169 |
|
1 |
flags = AV_RB32(buf); |
170 |
|
1 |
silent_chunks = av_popcount(flags); |
171 |
|
1 |
buf += 4; |
172 |
|
1 |
buf_size -= 4; |
173 |
✓✓ |
214 |
} else if (block_type == BLOCK_TYPE_SILENCE) { |
174 |
|
29 |
silent_chunks = 1; |
175 |
|
29 |
buf_size = 0; // should already be zero but set it just to be sure |
176 |
|
|
} |
177 |
|
|
|
178 |
|
|
/* ensure output buffer is large enough */ |
179 |
|
215 |
audio_chunks = buf_size / s->chunk_size; |
180 |
|
|
|
181 |
|
|
/* drop incomplete chunks */ |
182 |
|
215 |
buf_size = audio_chunks * s->chunk_size; |
183 |
|
|
|
184 |
✗✓ |
215 |
if (silent_chunks + audio_chunks >= INT_MAX / avctx->block_align) |
185 |
|
|
return AVERROR_INVALIDDATA; |
186 |
|
|
|
187 |
|
|
/* get output buffer */ |
188 |
|
215 |
frame->nb_samples = ((silent_chunks + audio_chunks) * avctx->block_align) / |
189 |
|
215 |
avctx->channels; |
190 |
✗✓ |
215 |
if ((ret = ff_get_buffer(avctx, frame, 0)) < 0) |
191 |
|
|
return ret; |
192 |
|
215 |
output_samples_u8 = frame->data[0]; |
193 |
|
215 |
output_samples_s16 = (int16_t *)frame->data[0]; |
194 |
|
|
|
195 |
|
|
/* decode silent chunks */ |
196 |
✓✓ |
215 |
if (silent_chunks > 0) { |
197 |
|
29 |
int silent_size = avctx->block_align * silent_chunks; |
198 |
✗✓ |
29 |
av_assert0(avctx->block_align * silent_chunks <= frame->nb_samples * avctx->channels); |
199 |
|
|
|
200 |
✗✓ |
29 |
if (s->out_bps == 2) { |
201 |
|
|
memset(output_samples_s16, 0x00, silent_size * 2); |
202 |
|
|
output_samples_s16 += silent_size; |
203 |
|
|
} else { |
204 |
|
29 |
memset(output_samples_u8, 0x80, silent_size); |
205 |
|
29 |
output_samples_u8 += silent_size; |
206 |
|
|
} |
207 |
|
|
} |
208 |
|
|
|
209 |
|
|
/* decode audio chunks */ |
210 |
✓✓ |
215 |
if (audio_chunks > 0) { |
211 |
|
186 |
buf_end = buf + buf_size; |
212 |
✗✓ |
186 |
av_assert0((buf_size & (avctx->channels > 1)) == 0); |
213 |
✓✓ |
399 |
while (buf_end - buf >= s->chunk_size) { |
214 |
✗✓ |
213 |
if (s->out_bps == 2) { |
215 |
|
|
decode_audio_s16(output_samples_s16, buf, s->chunk_size, |
216 |
|
|
avctx->channels); |
217 |
|
|
output_samples_s16 += avctx->block_align; |
218 |
|
|
} else { |
219 |
|
213 |
memcpy(output_samples_u8, buf, s->chunk_size); |
220 |
|
213 |
output_samples_u8 += avctx->block_align; |
221 |
|
|
} |
222 |
|
213 |
buf += s->chunk_size; |
223 |
|
|
} |
224 |
|
|
} |
225 |
|
|
|
226 |
|
215 |
*got_frame_ptr = 1; |
227 |
|
|
|
228 |
|
215 |
return avpkt->size; |
229 |
|
|
} |
230 |
|
|
|
231 |
|
|
AVCodec ff_vmdaudio_decoder = { |
232 |
|
|
.name = "vmdaudio", |
233 |
|
|
.long_name = NULL_IF_CONFIG_SMALL("Sierra VMD audio"), |
234 |
|
|
.type = AVMEDIA_TYPE_AUDIO, |
235 |
|
|
.id = AV_CODEC_ID_VMDAUDIO, |
236 |
|
|
.priv_data_size = sizeof(VmdAudioContext), |
237 |
|
|
.init = vmdaudio_decode_init, |
238 |
|
|
.decode = vmdaudio_decode_frame, |
239 |
|
|
.capabilities = AV_CODEC_CAP_DR1, |
240 |
|
|
}; |