blob: e8c8a064c7dcd3edc70a2e4fa8bf826b16a7f702
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 | static av_cold int vmdaudio_decode_init(AVCodecContext *avctx) |
72 | { |
73 | VmdAudioContext *s = avctx->priv_data; |
74 | |
75 | 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 | if (avctx->block_align < 1 || avctx->block_align % avctx->channels) { |
80 | av_log(avctx, AV_LOG_ERROR, "invalid block align\n"); |
81 | return AVERROR(EINVAL); |
82 | } |
83 | |
84 | avctx->channel_layout = avctx->channels == 1 ? AV_CH_LAYOUT_MONO : |
85 | AV_CH_LAYOUT_STEREO; |
86 | |
87 | if (avctx->bits_per_coded_sample == 16) |
88 | avctx->sample_fmt = AV_SAMPLE_FMT_S16; |
89 | else |
90 | avctx->sample_fmt = AV_SAMPLE_FMT_U8; |
91 | s->out_bps = av_get_bytes_per_sample(avctx->sample_fmt); |
92 | |
93 | s->chunk_size = avctx->block_align + avctx->channels * (s->out_bps == 2); |
94 | |
95 | av_log(avctx, AV_LOG_DEBUG, "%d channels, %d bits/sample, " |
96 | "block align = %d, sample rate = %d\n", |
97 | avctx->channels, avctx->bits_per_coded_sample, avctx->block_align, |
98 | avctx->sample_rate); |
99 | |
100 | return 0; |
101 | } |
102 | |
103 | static void decode_audio_s16(int16_t *out, const uint8_t *buf, int buf_size, |
104 | int channels) |
105 | { |
106 | int ch; |
107 | const uint8_t *buf_end = buf + buf_size; |
108 | int predictor[2]; |
109 | int st = channels - 1; |
110 | |
111 | /* decode initial raw sample */ |
112 | for (ch = 0; ch < channels; ch++) { |
113 | predictor[ch] = (int16_t)AV_RL16(buf); |
114 | buf += 2; |
115 | *out++ = predictor[ch]; |
116 | } |
117 | |
118 | /* decode DPCM samples */ |
119 | ch = 0; |
120 | while (buf < buf_end) { |
121 | uint8_t b = *buf++; |
122 | if (b & 0x80) |
123 | predictor[ch] -= vmdaudio_table[b & 0x7F]; |
124 | else |
125 | predictor[ch] += vmdaudio_table[b]; |
126 | predictor[ch] = av_clip_int16(predictor[ch]); |
127 | *out++ = predictor[ch]; |
128 | ch ^= st; |
129 | } |
130 | } |
131 | |
132 | static int vmdaudio_decode_frame(AVCodecContext *avctx, void *data, |
133 | int *got_frame_ptr, AVPacket *avpkt) |
134 | { |
135 | AVFrame *frame = data; |
136 | const uint8_t *buf = avpkt->data; |
137 | const uint8_t *buf_end; |
138 | int buf_size = avpkt->size; |
139 | VmdAudioContext *s = avctx->priv_data; |
140 | int block_type, silent_chunks, audio_chunks; |
141 | int ret; |
142 | uint8_t *output_samples_u8; |
143 | int16_t *output_samples_s16; |
144 | |
145 | if (buf_size < 16) { |
146 | av_log(avctx, AV_LOG_WARNING, "skipping small junk packet\n"); |
147 | *got_frame_ptr = 0; |
148 | return buf_size; |
149 | } |
150 | |
151 | block_type = buf[6]; |
152 | if (block_type < BLOCK_TYPE_AUDIO || block_type > BLOCK_TYPE_SILENCE) { |
153 | av_log(avctx, AV_LOG_ERROR, "unknown block type: %d\n", block_type); |
154 | return AVERROR(EINVAL); |
155 | } |
156 | buf += 16; |
157 | buf_size -= 16; |
158 | |
159 | /* get number of silent chunks */ |
160 | silent_chunks = 0; |
161 | if (block_type == BLOCK_TYPE_INITIAL) { |
162 | uint32_t flags; |
163 | if (buf_size < 4) { |
164 | av_log(avctx, AV_LOG_ERROR, "packet is too small\n"); |
165 | return AVERROR(EINVAL); |
166 | } |
167 | flags = AV_RB32(buf); |
168 | silent_chunks = av_popcount(flags); |
169 | buf += 4; |
170 | buf_size -= 4; |
171 | } else if (block_type == BLOCK_TYPE_SILENCE) { |
172 | silent_chunks = 1; |
173 | buf_size = 0; // should already be zero but set it just to be sure |
174 | } |
175 | |
176 | /* ensure output buffer is large enough */ |
177 | audio_chunks = buf_size / s->chunk_size; |
178 | |
179 | /* drop incomplete chunks */ |
180 | buf_size = audio_chunks * s->chunk_size; |
181 | |
182 | /* get output buffer */ |
183 | frame->nb_samples = ((silent_chunks + audio_chunks) * avctx->block_align) / |
184 | avctx->channels; |
185 | if ((ret = ff_get_buffer(avctx, frame, 0)) < 0) |
186 | return ret; |
187 | output_samples_u8 = frame->data[0]; |
188 | output_samples_s16 = (int16_t *)frame->data[0]; |
189 | |
190 | /* decode silent chunks */ |
191 | if (silent_chunks > 0) { |
192 | int silent_size = avctx->block_align * silent_chunks; |
193 | av_assert0(avctx->block_align * silent_chunks <= frame->nb_samples * avctx->channels); |
194 | |
195 | if (s->out_bps == 2) { |
196 | memset(output_samples_s16, 0x00, silent_size * 2); |
197 | output_samples_s16 += silent_size; |
198 | } else { |
199 | memset(output_samples_u8, 0x80, silent_size); |
200 | output_samples_u8 += silent_size; |
201 | } |
202 | } |
203 | |
204 | /* decode audio chunks */ |
205 | if (audio_chunks > 0) { |
206 | buf_end = buf + buf_size; |
207 | av_assert0((buf_size & (avctx->channels > 1)) == 0); |
208 | while (buf_end - buf >= s->chunk_size) { |
209 | if (s->out_bps == 2) { |
210 | decode_audio_s16(output_samples_s16, buf, s->chunk_size, |
211 | avctx->channels); |
212 | output_samples_s16 += avctx->block_align; |
213 | } else { |
214 | memcpy(output_samples_u8, buf, s->chunk_size); |
215 | output_samples_u8 += avctx->block_align; |
216 | } |
217 | buf += s->chunk_size; |
218 | } |
219 | } |
220 | |
221 | *got_frame_ptr = 1; |
222 | |
223 | return avpkt->size; |
224 | } |
225 | |
226 | AVCodec ff_vmdaudio_decoder = { |
227 | .name = "vmdaudio", |
228 | .long_name = NULL_IF_CONFIG_SMALL("Sierra VMD audio"), |
229 | .type = AVMEDIA_TYPE_AUDIO, |
230 | .id = AV_CODEC_ID_VMDAUDIO, |
231 | .priv_data_size = sizeof(VmdAudioContext), |
232 | .init = vmdaudio_decode_init, |
233 | .decode = vmdaudio_decode_frame, |
234 | .capabilities = AV_CODEC_CAP_DR1, |
235 | }; |
236 |