FFmpeg
 All Data Structures Namespaces Files Functions Variables Typedefs Enumerations Enumerator Macros Groups Pages
vmdaudio.c
Go to the documentation of this file.
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"
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;
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 
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 :
86 
87  if (avctx->bits_per_coded_sample == 16)
89  else
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 
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),
234  .capabilities = CODEC_CAP_DR1,
235 };
const char * s
Definition: avisynth_c.h:631
This structure describes decoded (raw) audio or video data.
Definition: frame.h:171
ptrdiff_t const GLvoid * data
Definition: opengl_enc.c:101
#define AV_LOG_WARNING
Something somehow does not look correct.
Definition: log.h:182
static av_cold int init(AVCodecContext *avctx)
Definition: avrndec.c:35
int size
Definition: avcodec.h:1163
const char * b
Definition: vf_curves.c:109
#define AV_CH_LAYOUT_STEREO
AVCodec.
Definition: avcodec.h:3181
uint64_t_TMPL AV_WL64 unsigned int_TMPL AV_WL32 unsigned int_TMPL AV_WL24 unsigned int_TMPL AV_RL16
Definition: bytestream.h:85
int block_align
number of bytes per packet if constant and known or 0 Used by some WAV based audio codecs...
Definition: avcodec.h:2022
#define av_assert0(cond)
assert() equivalent, that is always enabled.
Definition: avassert.h:37
if()
Definition: avfilter.c:975
enum AVSampleFormat sample_fmt
audio sample format
Definition: avcodec.h:1993
uint8_t
#define av_cold
Definition: attributes.h:74
AV_SAMPLE_FMT_U8
uint64_t_TMPL AV_WL64 unsigned int_TMPL AV_WL32 unsigned int_TMPL AV_WL24 unsigned int_TMPL AV_WL16 uint64_t_TMPL AV_WB64 unsigned int_TMPL AV_RB32
Definition: bytestream.h:85
#define CODEC_CAP_DR1
Codec uses get_buffer() for allocating buffers and supports custom allocators.
Definition: avcodec.h:789
static AVFrame * frame
uint8_t * data
Definition: avcodec.h:1162
int bits_per_coded_sample
bits per sample/pixel from the demuxer (needed for huffyuv).
Definition: avcodec.h:2720
#define av_log(a,...)
static av_cold int vmdaudio_decode_init(AVCodecContext *avctx)
Definition: vmdaudio.c:71
static void predictor(uint8_t *src, int size)
Definition: exr.c:220
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
Definition: log.h:176
#define AVERROR(e)
Definition: error.h:43
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
Definition: internal.h:175
#define AV_LOG_DEBUG
Stuff which is only useful for libav* developers.
Definition: log.h:197
simple assert() macros that are a bit more flexible than ISO C assert().
const char * name
Name of the codec implementation.
Definition: avcodec.h:3188
static void decode_audio_s16(int16_t *out, const uint8_t *buf, int buf_size, int channels)
Definition: vmdaudio.c:103
Libavcodec external API header.
uint64_t channel_layout
Audio channel layout.
Definition: avcodec.h:2046
audio channel layout utility functions
ret
Definition: avfilter.c:974
#define BLOCK_TYPE_SILENCE
Definition: vmdaudio.c:48
static const uint16_t vmdaudio_table[128]
Definition: vmdaudio.c:55
int sample_rate
samples per second
Definition: avcodec.h:1985
main external API structure.
Definition: avcodec.h:1241
int ff_get_buffer(AVCodecContext *avctx, AVFrame *frame, int flags)
Get a buffer for a frame.
Definition: utils.c:1035
void * buf
Definition: avisynth_c.h:553
static int flags
Definition: cpu.c:47
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
Definition: frame.h:182
int av_get_bytes_per_sample(enum AVSampleFormat sample_fmt)
Return number of bytes per sample.
Definition: samplefmt.c:104
#define BLOCK_TYPE_INITIAL
Definition: vmdaudio.c:47
static int decode(AVCodecContext *avctx, void *data, int *got_sub, AVPacket *avpkt)
Definition: ccaption_dec.c:522
common internal api header.
common internal and external API header
signed 16 bits
Definition: samplefmt.h:62
AVCodec ff_vmdaudio_decoder
Definition: vmdaudio.c:226
void * priv_data
Definition: avcodec.h:1283
int channels
number of audio channels
Definition: avcodec.h:1986
uint8_t pi<< 24) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_U8, uint8_t,(*(constuint8_t *) pi-0x80)*(1.0f/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_U8, uint8_t,(*(constuint8_t *) pi-0x80)*(1.0/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S16, int16_t,(*(constint16_t *) pi >>8)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S16, int16_t,*(constint16_t *) pi *(1.0f/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S16, int16_t,*(constint16_t *) pi *(1.0/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S32, int32_t,(*(constint32_t *) pi >>24)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S32, int32_t,*(constint32_t *) pi *(1.0f/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S32, int32_t,*(constint32_t *) pi *(1.0/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_FLT, float, av_clip_uint8(lrintf(*(constfloat *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_FLT, float, av_clip_int16(lrintf(*(constfloat *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_FLT, float, av_clipl_int32(llrintf(*(constfloat *) pi *(1U<< 31)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_DBL, double, av_clip_uint8(lrint(*(constdouble *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_DBL, double, av_clip_int16(lrint(*(constdouble *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_DBL, double, av_clipl_int32(llrint(*(constdouble *) pi *(1U<< 31))))#defineSET_CONV_FUNC_GROUP(ofmt, ifmt) staticvoidset_generic_function(AudioConvert *ac){}voidff_audio_convert_free(AudioConvert **ac){if(!*ac) return;ff_dither_free(&(*ac) ->dc);av_freep(ac);}AudioConvert *ff_audio_convert_alloc(AVAudioResampleContext *avr, enumAVSampleFormatout_fmt, enumAVSampleFormatin_fmt, intchannels, intsample_rate, intapply_map){AudioConvert *ac;intin_planar, out_planar;ac=av_mallocz(sizeof(*ac));if(!ac) returnNULL;ac->avr=avr;ac->out_fmt=out_fmt;ac->in_fmt=in_fmt;ac->channels=channels;ac->apply_map=apply_map;if(avr->dither_method!=AV_RESAMPLE_DITHER_NONE &&av_get_packed_sample_fmt(out_fmt)==AV_SAMPLE_FMT_S16 &&av_get_bytes_per_sample(in_fmt)>2){ac->dc=ff_dither_alloc(avr, out_fmt, in_fmt, channels, sample_rate, apply_map);if(!ac->dc){av_free(ac);returnNULL;}returnac;}in_planar=ff_sample_fmt_is_planar(in_fmt, channels);out_planar=ff_sample_fmt_is_planar(out_fmt, channels);if(in_planar==out_planar){ac->func_type=CONV_FUNC_TYPE_FLAT;ac->planes=in_planar?ac->channels:1;}elseif(in_planar) ac->func_type=CONV_FUNC_TYPE_INTERLEAVE;elseac->func_type=CONV_FUNC_TYPE_DEINTERLEAVE;set_generic_function(ac);if(ARCH_AARCH64) ff_audio_convert_init_aarch64(ac);if(ARCH_ARM) ff_audio_convert_init_arm(ac);if(ARCH_X86) ff_audio_convert_init_x86(ac);returnac;}intff_audio_convert(AudioConvert *ac, AudioData *out, AudioData *in){intuse_generic=1;intlen=in->nb_samples;intp;if(ac->dc){av_log(ac->avr, AV_LOG_TRACE,"%dsamples-audio_convert:%sto%s(dithered)\n", len, av_get_sample_fmt_name(ac->in_fmt), av_get_sample_fmt_name(ac->out_fmt));returnff_convert_dither(ac-> out
#define AV_CH_LAYOUT_MONO
This structure stores compressed data.
Definition: avcodec.h:1139
int nb_samples
number of audio samples (per channel) described by this frame
Definition: frame.h:225
static int vmdaudio_decode_frame(AVCodecContext *avctx, void *data, int *got_frame_ptr, AVPacket *avpkt)
Definition: vmdaudio.c:132