FFmpeg
roqaudioenc.c
Go to the documentation of this file.
1 /*
2  * RoQ audio encoder
3  *
4  * Copyright (c) 2005 Eric Lasota
5  * Based on RoQ specs (c)2001 Tim Ferguson
6  *
7  * This file is part of FFmpeg.
8  *
9  * FFmpeg is free software; you can redistribute it and/or
10  * modify it under the terms of the GNU Lesser General Public
11  * License as published by the Free Software Foundation; either
12  * version 2.1 of the License, or (at your option) any later version.
13  *
14  * FFmpeg is distributed in the hope that it will be useful,
15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
17  * Lesser General Public License for more details.
18  *
19  * You should have received a copy of the GNU Lesser General Public
20  * License along with FFmpeg; if not, write to the Free Software
21  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
22  */
23 
24 #include "libavutil/mem.h"
25 #include "avcodec.h"
26 #include "bytestream.h"
27 #include "codec_internal.h"
28 #include "encode.h"
29 #include "mathops.h"
30 
31 #define ROQ_FRAME_SIZE 735
32 #define ROQ_HEADER_SIZE 8
33 
34 #define MAX_DPCM (127*127)
35 
36 
37 typedef struct ROQDPCMContext {
38  short lastSample[2];
41  int16_t *frame_buffer;
42  int64_t first_pts;
44 
45 
47 {
49 
50  av_freep(&context->frame_buffer);
51 
52  return 0;
53 }
54 
56 {
58  int channels = avctx->ch_layout.nb_channels;
59 
60  if (channels > 2) {
61  av_log(avctx, AV_LOG_ERROR, "Audio must be mono or stereo\n");
62  return AVERROR(EINVAL);
63  }
64  if (avctx->sample_rate != 22050) {
65  av_log(avctx, AV_LOG_ERROR, "Audio must be 22050 Hz\n");
66  return AVERROR(EINVAL);
67  }
68 
69  avctx->frame_size = ROQ_FRAME_SIZE;
71  (22050 / ROQ_FRAME_SIZE) * 8;
72 
73  context->frame_buffer = av_malloc(8 * ROQ_FRAME_SIZE * channels *
74  sizeof(*context->frame_buffer));
75  if (!context->frame_buffer)
76  return AVERROR(ENOMEM);
77 
78  context->lastSample[0] = context->lastSample[1] = 0;
79 
80  return 0;
81 }
82 
83 static unsigned char dpcm_predict(short *previous, short current)
84 {
85  int diff;
86  int negative;
87  int result;
88  int predicted;
89 
90  diff = current - *previous;
91 
92  negative = diff<0;
93  diff = FFABS(diff);
94 
95  if (diff >= MAX_DPCM)
96  result = 127;
97  else {
98  result = ff_sqrt(diff);
100  }
101 
102  /* See if this overflows */
103  retry:
104  diff = result*result;
105  if (negative)
106  diff = -diff;
107  predicted = *previous + diff;
108 
109  /* If it overflows, back off a step */
110  if (predicted > 32767 || predicted < -32768) {
111  result--;
112  goto retry;
113  }
114 
115  /* Add the sign bit */
116  result |= negative << 7; //if (negative) result |= 128;
117 
118  *previous = predicted;
119 
120  return result;
121 }
122 
124  const AVFrame *frame, int *got_packet_ptr)
125 {
126  int i, stereo, data_size, ret;
127  const int16_t *in = frame ? (const int16_t *)frame->data[0] : NULL;
128  int channels = avctx->ch_layout.nb_channels;
129  uint8_t *out;
130  ROQDPCMContext *context = avctx->priv_data;
131 
132  stereo = (channels == 2);
133 
134  if (!in && context->input_frames >= 8)
135  return 0;
136 
137  if (in && context->input_frames < 8) {
138  memcpy(&context->frame_buffer[context->buffered_samples * channels],
139  in, avctx->frame_size * channels * sizeof(*in));
140  context->buffered_samples += avctx->frame_size;
141  if (context->input_frames == 0)
142  context->first_pts = frame->pts;
143  if (context->input_frames < 7) {
144  context->input_frames++;
145  return 0;
146  }
147  }
148  if (context->input_frames < 8)
149  in = context->frame_buffer;
150 
151  if (stereo) {
152  context->lastSample[0] &= 0xFF00;
153  context->lastSample[1] &= 0xFF00;
154  }
155 
156  if (context->input_frames == 7)
157  data_size = channels * context->buffered_samples;
158  else
159  data_size = channels * avctx->frame_size;
160 
161  ret = ff_get_encode_buffer(avctx, avpkt, ROQ_HEADER_SIZE + data_size, 0);
162  if (ret < 0)
163  return ret;
164  out = avpkt->data;
165 
166  bytestream_put_byte(&out, stereo ? 0x21 : 0x20);
167  bytestream_put_byte(&out, 0x10);
168  bytestream_put_le32(&out, data_size);
169 
170  if (stereo) {
171  bytestream_put_byte(&out, (context->lastSample[1])>>8);
172  bytestream_put_byte(&out, (context->lastSample[0])>>8);
173  } else
174  bytestream_put_le16(&out, context->lastSample[0]);
175 
176  /* Write the actual samples */
177  for (i = 0; i < data_size; i++)
178  *out++ = dpcm_predict(&context->lastSample[(i & 1) & stereo], *in++);
179 
180  avpkt->pts = context->input_frames <= 7 ? context->first_pts : frame->pts;
181  avpkt->duration = data_size / channels;
182 
183  context->input_frames++;
184  if (!in)
185  context->input_frames = FFMAX(context->input_frames, 8);
186 
187  *got_packet_ptr = 1;
188  return 0;
189 }
190 
192  .p.name = "roq_dpcm",
193  CODEC_LONG_NAME("id RoQ DPCM"),
194  .p.type = AVMEDIA_TYPE_AUDIO,
195  .p.id = AV_CODEC_ID_ROQ_DPCM,
196  .p.capabilities = AV_CODEC_CAP_DR1 | AV_CODEC_CAP_DELAY,
197  .priv_data_size = sizeof(ROQDPCMContext),
200  .close = roq_dpcm_encode_close,
201  .p.sample_fmts = (const enum AVSampleFormat[]){ AV_SAMPLE_FMT_S16,
203 };
AVCodecContext::frame_size
int frame_size
Number of samples per channel in an audio frame.
Definition: avcodec.h:1077
AVERROR
Filter the word “frame” indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel sample they are references to shared objects When the negotiation mechanism computes the intersection of the formats supported at each end of a all references to both lists are replaced with a reference to the intersection And when a single format is eventually chosen for a link amongst the remaining all references to the list are updated That means that if a filter requires that its input and output have the same format amongst a supported all it has to do is use a reference to the same list of formats query_formats can leave some formats unset and return AVERROR(EAGAIN) to cause the negotiation mechanism toagain later. That can be used by filters with complex requirements to use the format negotiated on one link to set the formats supported on another. Frame references ownership and permissions
out
FILE * out
Definition: movenc.c:55
AVCodecContext::sample_rate
int sample_rate
samples per second
Definition: avcodec.h:1050
AVFrame
This structure describes decoded (raw) audio or video data.
Definition: frame.h:374
AVPacket::data
uint8_t * data
Definition: packet.h:520
encode.h
FFCodec
Definition: codec_internal.h:126
AVPacket::duration
int64_t duration
Duration of this packet in AVStream->time_base units, 0 if unknown.
Definition: packet.h:538
ROQDPCMContext
Definition: roqaudioenc.c:37
FFMAX
#define FFMAX(a, b)
Definition: macros.h:47
AVChannelLayout::nb_channels
int nb_channels
Number of channels in this layout.
Definition: channel_layout.h:321
av_malloc
#define av_malloc(s)
Definition: tableprint_vlc.h:30
FFCodec::p
AVCodec p
The public AVCodec.
Definition: codec_internal.h:130
AVCodecContext::ch_layout
AVChannelLayout ch_layout
Audio channel layout.
Definition: avcodec.h:1065
FF_CODEC_ENCODE_CB
#define FF_CODEC_ENCODE_CB(func)
Definition: codec_internal.h:295
ff_sqrt
#define ff_sqrt
Definition: mathops.h:216
ff_roq_dpcm_encoder
const FFCodec ff_roq_dpcm_encoder
Definition: roqaudioenc.c:191
ROQDPCMContext::frame_buffer
int16_t * frame_buffer
Definition: roqaudioenc.c:41
AV_LOG_ERROR
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
Definition: log.h:180
av_cold
#define av_cold
Definition: attributes.h:90
ROQ_HEADER_SIZE
#define ROQ_HEADER_SIZE
Definition: roqaudioenc.c:32
ROQDPCMContext::input_frames
int input_frames
Definition: roqaudioenc.c:39
AVMEDIA_TYPE_AUDIO
@ AVMEDIA_TYPE_AUDIO
Definition: avutil.h:202
channels
channels
Definition: aptx.h:31
CODEC_LONG_NAME
#define CODEC_LONG_NAME(str)
Definition: codec_internal.h:271
FFABS
#define FFABS(a)
Absolute value, Note, INT_MIN / INT64_MIN result in undefined behavior as they are not representable ...
Definition: common.h:74
context
it s the only field you need to keep assuming you have a context There is some magic you don t need to care about around this just let it vf default minimum maximum flags name is the option keep it simple and lowercase description are in without and describe what they for example set the foo of the bar offset is the offset of the field in your context
Definition: writing_filters.txt:91
result
and forward the result(frame or status change) to the corresponding input. If nothing is possible
NULL
#define NULL
Definition: coverity.c:32
AVCodecContext::bit_rate
int64_t bit_rate
the average bitrate
Definition: avcodec.h:495
mathops.h
ROQDPCMContext::first_pts
int64_t first_pts
Definition: roqaudioenc.c:42
AV_CODEC_ID_ROQ_DPCM
@ AV_CODEC_ID_ROQ_DPCM
Definition: codec_id.h:429
init
int(* init)(AVBSFContext *ctx)
Definition: dts2pts.c:366
AV_CODEC_CAP_DR1
#define AV_CODEC_CAP_DR1
Codec uses get_buffer() or get_encode_buffer() for allocating buffers and supports custom allocators.
Definition: codec.h:52
codec_internal.h
roq_dpcm_encode_init
static av_cold int roq_dpcm_encode_init(AVCodecContext *avctx)
Definition: roqaudioenc.c:55
AV_SAMPLE_FMT_NONE
@ AV_SAMPLE_FMT_NONE
Definition: samplefmt.h:56
roq_dpcm_encode_frame
static int roq_dpcm_encode_frame(AVCodecContext *avctx, AVPacket *avpkt, const AVFrame *frame, int *got_packet_ptr)
Definition: roqaudioenc.c:123
diff
static av_always_inline int diff(const struct color_info *a, const struct color_info *b, const int trans_thresh)
Definition: vf_paletteuse.c:165
dpcm_predict
static unsigned char dpcm_predict(short *previous, short current)
Definition: roqaudioenc.c:83
i
#define i(width, name, range_min, range_max)
Definition: cbs_h2645.c:256
AVPacket::pts
int64_t pts
Presentation timestamp in AVStream->time_base units; the time at which the decompressed packet will b...
Definition: packet.h:513
AVSampleFormat
AVSampleFormat
Audio sample formats.
Definition: samplefmt.h:55
AV_SAMPLE_FMT_S16
@ AV_SAMPLE_FMT_S16
signed 16 bits
Definition: samplefmt.h:58
AVCodec::name
const char * name
Name of the codec implementation.
Definition: codec.h:194
avcodec.h
ret
ret
Definition: filter_design.txt:187
frame
these buffered frames must be flushed immediately if a new input produces new the filter must not call request_frame to get more It must just process the frame or queue it The task of requesting more frames is left to the filter s request_frame method or the application If a filter has several the filter must be ready for frames arriving randomly on any input any filter with several inputs will most likely require some kind of queuing mechanism It is perfectly acceptable to have a limited queue and to drop frames when the inputs are too unbalanced request_frame For filters that do not use the this method is called when a frame is wanted on an output For a it should directly call filter_frame on the corresponding output For a if there are queued frames already one of these frames should be pushed If the filter should request a frame on one of its repeatedly until at least one frame has been pushed Return or at least make progress towards producing a frame
Definition: filter_design.txt:264
MAX_DPCM
#define MAX_DPCM
Definition: roqaudioenc.c:34
AVCodecContext
main external API structure.
Definition: avcodec.h:445
ROQDPCMContext::buffered_samples
int buffered_samples
Definition: roqaudioenc.c:40
ff_get_encode_buffer
int ff_get_encode_buffer(AVCodecContext *avctx, AVPacket *avpkt, int64_t size, int flags)
Get a buffer for a packet.
Definition: encode.c:106
AV_CODEC_CAP_DELAY
#define AV_CODEC_CAP_DELAY
Encoder or decoder requires flushing with NULL input at the end in order to give the complete and cor...
Definition: codec.h:76
roq_dpcm_encode_close
static av_cold int roq_dpcm_encode_close(AVCodecContext *avctx)
Definition: roqaudioenc.c:46
mem.h
AVPacket
This structure stores compressed data.
Definition: packet.h:497
AVCodecContext::priv_data
void * priv_data
Definition: avcodec.h:472
av_freep
#define av_freep(p)
Definition: tableprint_vlc.h:34
bytestream.h
av_log
#define av_log(a,...)
Definition: tableprint_vlc.h:27
ROQ_FRAME_SIZE
#define ROQ_FRAME_SIZE
Definition: roqaudioenc.c:31
ROQDPCMContext::lastSample
short lastSample[2]
Definition: roqaudioenc.c:38