FFmpeg
riffenc.c
Go to the documentation of this file.
1 /*
2  * RIFF muxing functions
3  * Copyright (c) 2000 Fabrice Bellard
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 #include "libavutil/dict.h"
23 #include "libavutil/log.h"
24 #include "libavutil/mathematics.h"
25 #include "libavcodec/avcodec.h"
26 #include "libavcodec/bytestream.h"
27 #include "avformat.h"
28 #include "avio_internal.h"
29 #include "riff.h"
30 
31 int64_t ff_start_tag(AVIOContext *pb, const char *tag)
32 {
33  ffio_wfourcc(pb, tag);
34  avio_wl32(pb, -1);
35  return avio_tell(pb);
36 }
37 
38 void ff_end_tag(AVIOContext *pb, int64_t start)
39 {
40  int64_t pos;
41 
42  av_assert0((start&1) == 0);
43 
44  pos = avio_tell(pb);
45  if (pos & 1)
46  avio_w8(pb, 0);
47  avio_seek(pb, start - 4, SEEK_SET);
48  avio_wl32(pb, (uint32_t)(pos - start));
49  avio_seek(pb, FFALIGN(pos, 2), SEEK_SET);
50 }
51 
52 /* WAVEFORMATEX header */
53 /* returns the size or -1 on error */
55  AVCodecParameters *par, int flags)
56 {
57  int bps, blkalign, bytespersec, frame_size;
58  int hdrsize;
59  int64_t hdrstart = avio_tell(pb);
60  int waveformatextensible;
61  uint8_t temp[256];
62  uint8_t *riff_extradata = temp;
63  uint8_t *riff_extradata_start = temp;
64 
65  if (!par->codec_tag || par->codec_tag > 0xffff)
66  return -1;
67 
68  if (par->codec_id == AV_CODEC_ID_ADPCM_SWF && par->block_align == 0) {
69  av_log(s, AV_LOG_ERROR, "%s can only be written to WAVE with a constant frame size\n",
71  return AVERROR(EINVAL);
72  }
73 
74  /* We use the known constant frame size for the codec if known, otherwise
75  * fall back on using AVCodecContext.frame_size, which is not as reliable
76  * for indicating packet duration. */
77  frame_size = av_get_audio_frame_duration2(par, par->block_align);
78 
79  waveformatextensible = (par->channels > 2 && par->channel_layout) ||
80  par->channels == 1 && par->channel_layout && par->channel_layout != AV_CH_LAYOUT_MONO ||
81  par->channels == 2 && par->channel_layout && par->channel_layout != AV_CH_LAYOUT_STEREO ||
82  par->sample_rate > 48000 ||
83  par->codec_id == AV_CODEC_ID_EAC3 ||
85 
86  if (waveformatextensible)
87  avio_wl16(pb, 0xfffe);
88  else
89  avio_wl16(pb, par->codec_tag);
90 
91  avio_wl16(pb, par->channels);
92  avio_wl32(pb, par->sample_rate);
93  if (par->codec_id == AV_CODEC_ID_ATRAC3 ||
94  par->codec_id == AV_CODEC_ID_G723_1 ||
95  par->codec_id == AV_CODEC_ID_MP2 ||
96  par->codec_id == AV_CODEC_ID_MP3 ||
97  par->codec_id == AV_CODEC_ID_GSM_MS) {
98  bps = 0;
99  } else {
100  if (!(bps = av_get_bits_per_sample(par->codec_id))) {
101  if (par->bits_per_coded_sample)
102  bps = par->bits_per_coded_sample;
103  else
104  bps = 16; // default to 16
105  }
106  }
107  if (bps != par->bits_per_coded_sample && par->bits_per_coded_sample) {
109  "requested bits_per_coded_sample (%d) "
110  "and actually stored (%d) differ\n",
111  par->bits_per_coded_sample, bps);
112  }
113 
114  if (par->codec_id == AV_CODEC_ID_MP2) {
115  blkalign = (144 * par->bit_rate - 1)/par->sample_rate + 1;
116  } else if (par->codec_id == AV_CODEC_ID_MP3) {
117  blkalign = 576 * (par->sample_rate <= (24000 + 32000)/2 ? 1 : 2);
118  } else if (par->codec_id == AV_CODEC_ID_AC3) {
119  blkalign = 3840; /* maximum bytes per frame */
120  } else if (par->codec_id == AV_CODEC_ID_AAC) {
121  blkalign = 768 * par->channels; /* maximum bytes per frame */
122  } else if (par->codec_id == AV_CODEC_ID_G723_1) {
123  blkalign = 24;
124  } else if (par->block_align != 0) { /* specified by the codec */
125  blkalign = par->block_align;
126  } else
127  blkalign = bps * par->channels / av_gcd(8, bps);
128  if (par->codec_id == AV_CODEC_ID_PCM_U8 ||
134  bytespersec = par->sample_rate * blkalign;
135  } else if (par->codec_id == AV_CODEC_ID_G723_1) {
136  bytespersec = 800;
137  } else {
138  bytespersec = par->bit_rate / 8;
139  }
140  avio_wl32(pb, bytespersec); /* bytes per second */
141  avio_wl16(pb, blkalign); /* block align */
142  avio_wl16(pb, bps); /* bits per sample */
143  if (par->codec_id == AV_CODEC_ID_MP3) {
144  bytestream_put_le16(&riff_extradata, 1); /* wID */
145  bytestream_put_le32(&riff_extradata, 2); /* fdwFlags */
146  bytestream_put_le16(&riff_extradata, 1152); /* nBlockSize */
147  bytestream_put_le16(&riff_extradata, 1); /* nFramesPerBlock */
148  bytestream_put_le16(&riff_extradata, 1393); /* nCodecDelay */
149  } else if (par->codec_id == AV_CODEC_ID_MP2) {
150  /* fwHeadLayer */
151  bytestream_put_le16(&riff_extradata, 2);
152  /* dwHeadBitrate */
153  bytestream_put_le32(&riff_extradata, par->bit_rate);
154  /* fwHeadMode */
155  bytestream_put_le16(&riff_extradata, par->channels == 2 ? 1 : 8);
156  /* fwHeadModeExt */
157  bytestream_put_le16(&riff_extradata, 0);
158  /* wHeadEmphasis */
159  bytestream_put_le16(&riff_extradata, 1);
160  /* fwHeadFlags */
161  bytestream_put_le16(&riff_extradata, 16);
162  /* dwPTSLow */
163  bytestream_put_le32(&riff_extradata, 0);
164  /* dwPTSHigh */
165  bytestream_put_le32(&riff_extradata, 0);
166  } else if (par->codec_id == AV_CODEC_ID_G723_1) {
167  bytestream_put_le32(&riff_extradata, 0x9ace0002); /* extradata needed for msacm g723.1 codec */
168  bytestream_put_le32(&riff_extradata, 0xaea2f732);
169  bytestream_put_le16(&riff_extradata, 0xacde);
170  } else if (par->codec_id == AV_CODEC_ID_GSM_MS ||
172  /* wSamplesPerBlock */
173  bytestream_put_le16(&riff_extradata, frame_size);
174  } else if (par->extradata_size) {
175  riff_extradata_start = par->extradata;
176  riff_extradata = par->extradata + par->extradata_size;
177  }
178  /* write WAVEFORMATEXTENSIBLE extensions */
179  if (waveformatextensible) {
180  int write_channel_mask = !(flags & FF_PUT_WAV_HEADER_SKIP_CHANNELMASK) &&
182  par->channel_layout < 0x40000);
183  /* 22 is WAVEFORMATEXTENSIBLE size */
184  avio_wl16(pb, riff_extradata - riff_extradata_start + 22);
185  /* ValidBitsPerSample || SamplesPerBlock || Reserved */
186  avio_wl16(pb, bps);
187  /* dwChannelMask */
188  avio_wl32(pb, write_channel_mask ? par->channel_layout : 0);
189  /* GUID + next 3 */
190  if (par->codec_id == AV_CODEC_ID_EAC3) {
192  } else {
193  avio_wl32(pb, par->codec_tag);
194  avio_wl32(pb, 0x00100000);
195  avio_wl32(pb, 0xAA000080);
196  avio_wl32(pb, 0x719B3800);
197  }
198  } else if ((flags & FF_PUT_WAV_HEADER_FORCE_WAVEFORMATEX) ||
199  par->codec_tag != 0x0001 /* PCM */ ||
200  riff_extradata - riff_extradata_start) {
201  /* WAVEFORMATEX */
202  avio_wl16(pb, riff_extradata - riff_extradata_start); /* cbSize */
203  } /* else PCMWAVEFORMAT */
204  avio_write(pb, riff_extradata_start, riff_extradata - riff_extradata_start);
205  hdrsize = avio_tell(pb) - hdrstart;
206  if (hdrsize & 1) {
207  hdrsize++;
208  avio_w8(pb, 0);
209  }
210 
211  return hdrsize;
212 }
213 
214 /* BITMAPINFOHEADER header */
216  int for_asf, int ignore_extradata, int rgb_frame_is_flipped)
217 {
218  int flipped_extradata = (par->extradata_size >= 9 &&
219  !memcmp(par->extradata + par->extradata_size - 9, "BottomUp", 9));
220  int keep_height = flipped_extradata || rgb_frame_is_flipped;
221  int extradata_size = par->extradata_size - 9*flipped_extradata;
222  enum AVPixelFormat pix_fmt = par->format;
223  int pal_avi;
224 
225  if (pix_fmt == AV_PIX_FMT_NONE && par->bits_per_coded_sample == 1)
226  pix_fmt = AV_PIX_FMT_MONOWHITE;
227  pal_avi = !for_asf &&
228  (pix_fmt == AV_PIX_FMT_PAL8 ||
229  pix_fmt == AV_PIX_FMT_MONOWHITE ||
230  pix_fmt == AV_PIX_FMT_MONOBLACK);
231 
232  /* Size (not including the size of the color table or color masks) */
233  avio_wl32(pb, 40 + (ignore_extradata || pal_avi ? 0 : extradata_size));
234  avio_wl32(pb, par->width);
235  //We always store RGB TopDown
236  avio_wl32(pb, par->codec_tag || keep_height ? par->height : -par->height);
237  /* planes */
238  avio_wl16(pb, 1);
239  /* depth */
241  /* compression type */
242  avio_wl32(pb, par->codec_tag);
243  avio_wl32(pb, (par->width * par->height * (par->bits_per_coded_sample ? par->bits_per_coded_sample : 24)+7) / 8);
244  avio_wl32(pb, 0);
245  avio_wl32(pb, 0);
246  /* Number of color indices in the color table that are used.
247  * A value of 0 means 2^biBitCount indices, but this doesn't work
248  * with Windows Media Player and files containing xxpc chunks. */
249  avio_wl32(pb, pal_avi ? 1 << par->bits_per_coded_sample : 0);
250  avio_wl32(pb, 0);
251 
252  if (!ignore_extradata) {
253  if (par->extradata_size) {
254  avio_write(pb, par->extradata, extradata_size);
255  if (!for_asf && extradata_size & 1)
256  avio_w8(pb, 0);
257  } else if (pal_avi) {
258  int i;
259  for (i = 0; i < 1 << par->bits_per_coded_sample; i++) {
260  /* Initialize 1 bpp palette to black & white */
261  if (i == 0 && pix_fmt == AV_PIX_FMT_MONOWHITE)
262  avio_wl32(pb, 0xffffff);
263  else if (i == 1 && pix_fmt == AV_PIX_FMT_MONOBLACK)
264  avio_wl32(pb, 0xffffff);
265  else
266  avio_wl32(pb, 0);
267  }
268  }
269  }
270 }
271 
272 void ff_parse_specific_params(AVStream *st, int *au_rate,
273  int *au_ssize, int *au_scale)
274 {
275  AVCodecParameters *par = st->codecpar;
276  int gcd;
277  int audio_frame_size;
278 
279  audio_frame_size = av_get_audio_frame_duration2(par, 0);
280  if (!audio_frame_size)
281  audio_frame_size = par->frame_size;
282 
283  *au_ssize = par->block_align;
284  if (audio_frame_size && par->sample_rate) {
285  *au_scale = audio_frame_size;
286  *au_rate = par->sample_rate;
287  } else if (par->codec_type == AVMEDIA_TYPE_VIDEO ||
288  par->codec_type == AVMEDIA_TYPE_DATA ||
290  *au_scale = st->time_base.num;
291  *au_rate = st->time_base.den;
292  } else {
293  *au_scale = par->block_align ? par->block_align * 8 : 8;
294  *au_rate = par->bit_rate ? par->bit_rate :
295  8 * par->sample_rate;
296  }
297  gcd = av_gcd(*au_scale, *au_rate);
298  *au_scale /= gcd;
299  *au_rate /= gcd;
300 }
301 
302 void ff_riff_write_info_tag(AVIOContext *pb, const char *tag, const char *str)
303 {
304  size_t len = strlen(str);
305  if (len > 0 && len < UINT32_MAX) {
306  len++;
307  ffio_wfourcc(pb, tag);
308  avio_wl32(pb, len);
309  avio_put_str(pb, str);
310  if (len & 1)
311  avio_w8(pb, 0);
312  }
313 }
314 
315 static const char riff_tags[][5] = {
316  "IARL", "IART", "IAS1", "IAS2", "IAS3", "IAS4", "IAS5", "IAS6", "IAS7",
317  "IAS8", "IAS9", "ICMS", "ICMT", "ICOP", "ICRD", "ICRP", "IDIM", "IDPI",
318  "IENG", "IGNR", "IKEY", "ILGT", "ILNG", "IMED", "INAM", "IPLT", "IPRD",
319  "IPRT", "ITRK", "ISBJ", "ISFT", "ISHP", "ISMP", "ISRC", "ISRF", "ITCH",
320  { 0 }
321 };
322 
324 {
325  int i;
326 
327  for (i = 0; *riff_tags[i]; i++)
329  return 1;
330 
331  return 0;
332 }
333 
335 {
336  AVIOContext *pb = s->pb;
337  int i;
338  int64_t list_pos;
339  AVDictionaryEntry *t = NULL;
340 
342 
343  /* writing empty LIST is not nice and may cause problems */
344  if (!riff_has_valid_tags(s))
345  return;
346 
347  list_pos = ff_start_tag(pb, "LIST");
348  ffio_wfourcc(pb, "INFO");
349  for (i = 0; *riff_tags[i]; i++)
350  if ((t = av_dict_get(s->metadata, riff_tags[i],
352  ff_riff_write_info_tag(s->pb, t->key, t->value);
353  ff_end_tag(pb, list_pos);
354 }
355 
357 {
358  av_assert0(sizeof(*g) == 16);
359  avio_write(s, *g, sizeof(*g));
360 }
361 
362 const ff_asf_guid *ff_get_codec_guid(enum AVCodecID id, const AVCodecGuid *av_guid)
363 {
364  int i;
365  for (i = 0; av_guid[i].id != AV_CODEC_ID_NONE; i++) {
366  if (id == av_guid[i].id)
367  return &(av_guid[i].guid);
368  }
369  return NULL;
370 }
#define NULL
Definition: coverity.c:32
void avio_wl16(AVIOContext *s, unsigned int val)
Definition: aviobuf.c:459
Bytestream IO Context.
Definition: avio.h:161
static enum AVPixelFormat pix_fmt
#define AV_LOG_WARNING
Something somehow does not look correct.
Definition: log.h:200
else temp
Definition: vf_mcdeint.c:256
const char * g
Definition: vf_curves.c:115
enum AVCodecID id
Definition: riff.h:95
enum AVCodecID codec_id
Specific type of the encoded data (the codec used).
Definition: codec_par.h:60
ff_asf_guid guid
Definition: riff.h:96
int num
Numerator.
Definition: rational.h:59
int64_t avio_seek(AVIOContext *s, int64_t offset, int whence)
fseek() equivalent for AVIOContext.
Definition: aviobuf.c:253
#define AV_CH_LAYOUT_STEREO
int strict_std_compliance
Allow non-standard and experimental extension.
Definition: avformat.h:1547
This struct describes the properties of an encoded stream.
Definition: codec_par.h:52
void ff_parse_specific_params(AVStream *st, int *au_rate, int *au_ssize, int *au_scale)
Definition: riffenc.c:272
int frame_size
Audio only.
Definition: codec_par.h:181
const AVCodecGuid ff_codec_wav_guids[]
Definition: riff.c:619
Format I/O context.
Definition: avformat.h:1239
#define av_assert0(cond)
assert() equivalent, that is always enabled.
Definition: avassert.h:37
Public dictionary API.
void ff_put_guid(AVIOContext *s, const ff_asf_guid *g)
Definition: riffenc.c:356
void avio_wl32(AVIOContext *s, unsigned int val)
Definition: aviobuf.c:379
uint8_t
Opaque data information usually continuous.
Definition: avutil.h:203
int width
Video only.
Definition: codec_par.h:126
8 bits with AV_PIX_FMT_RGB32 palette
Definition: pixfmt.h:77
AVDictionaryEntry * av_dict_get(const AVDictionary *m, const char *key, const AVDictionaryEntry *prev, int flags)
Get a dictionary entry with matching key.
Definition: dict.c:40
uint32_t tag
Definition: movenc.c:1597
static av_always_inline int64_t avio_tell(AVIOContext *s)
ftell() equivalent for AVIOContext.
Definition: avio.h:557
void avio_write(AVIOContext *s, const unsigned char *buf, int size)
Definition: aviobuf.c:225
static av_always_inline void ffio_wfourcc(AVIOContext *pb, const uint8_t *s)
Definition: avio_internal.h:58
#define FFALIGN(x, a)
Definition: macros.h:48
uint64_t channel_layout
Audio only.
Definition: codec_par.h:162
#define av_log(a,...)
void ff_riff_write_info(AVFormatContext *s)
Write all recognized RIFF tags from s->metadata.
Definition: riffenc.c:334
int64_t bit_rate
The average bitrate of the encoded data (in bits per second).
Definition: codec_par.h:89
const ff_asf_guid * ff_get_codec_guid(enum AVCodecID id, const AVCodecGuid *av_guid)
Definition: riffenc.c:362
AVCodecID
Identify the syntax and semantics of the bitstream.
Definition: codec_id.h:46
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
Definition: log.h:194
#define AV_DICT_MATCH_CASE
Only get an entry with exact-case key match.
Definition: dict.h:69
AVDictionary * metadata
Metadata that applies to the whole file.
Definition: avformat.h:1479
int av_get_bits_per_sample(enum AVCodecID codec_id)
Return codec bits per sample.
Definition: utils.c:1600
void ff_put_bmp_header(AVIOContext *pb, AVCodecParameters *par, int for_asf, int ignore_extradata, int rgb_frame_is_flipped)
Definition: riffenc.c:215
unsigned int pos
Definition: spdifenc.c:410
preferred ID for decoding MPEG audio layer 1, 2 or 3
Definition: codec_id.h:419
enum AVMediaType codec_type
General type of the encoded data.
Definition: codec_par.h:56
int64_t av_gcd(int64_t a, int64_t b)
Compute the greatest common divisor of two integer operands.
Definition: mathematics.c:37
int av_get_audio_frame_duration2(AVCodecParameters *par, int frame_bytes)
This function is the same as av_get_audio_frame_duration(), except it works with AVCodecParameters in...
Definition: utils.c:1823
int extradata_size
Size of the extradata content in bytes.
Definition: codec_par.h:78
int block_align
Audio only.
Definition: codec_par.h:177
void ff_end_tag(AVIOContext *pb, int64_t start)
Definition: riffenc.c:38
internal header for RIFF based (de)muxers do NOT include this in end user applications ...
void ff_riff_write_info_tag(AVIOContext *pb, const char *tag, const char *str)
Write a single RIFF info tag.
Definition: riffenc.c:302
#define s(width, name)
Definition: cbs_vp9.c:257
int avio_put_str(AVIOContext *s, const char *str)
Write a NULL-terminated string.
Definition: aviobuf.c:395
uint8_t ff_asf_guid[16]
Definition: riff.h:92
Stream structure.
Definition: avformat.h:880
int frame_size
Definition: mxfenc.c:2166
Libavcodec external API header.
int64_t ff_start_tag(AVIOContext *pb, const char *tag)
Definition: riffenc.c:31
AVIOContext * pb
I/O context.
Definition: avformat.h:1281
static const char riff_tags[][5]
Definition: riffenc.c:315
void avio_w8(AVIOContext *s, int b)
Definition: aviobuf.c:203
static int riff_has_valid_tags(AVFormatContext *s)
Definition: riffenc.c:323
#define FF_COMPLIANCE_NORMAL
Definition: avcodec.h:1597
const AVMetadataConv ff_riff_info_conv[]
Definition: riff.c:592
#define flags(name, subs,...)
Definition: cbs_av1.c:560
int ff_put_wav_header(AVFormatContext *s, AVIOContext *pb, AVCodecParameters *par, int flags)
Write WAVEFORMAT header structure.
Definition: riffenc.c:54
int sample_rate
Audio only.
Definition: codec_par.h:170
Y , 1bpp, 0 is black, 1 is white, in each byte pixels are ordered from the msb to the lsb...
Definition: pixfmt.h:76
Main libavformat public API header.
Y , 1bpp, 0 is white, 1 is black, in each byte pixels are ordered from the msb to the lsb...
Definition: pixfmt.h:75
const char * avcodec_get_name(enum AVCodecID id)
Get the name of a codec.
Definition: utils.c:1226
char * key
Definition: dict.h:86
int den
Denominator.
Definition: rational.h:60
unsigned bps
Definition: movenc.c:1598
void ff_metadata_conv(AVDictionary **pm, const AVMetadataConv *d_conv, const AVMetadataConv *s_conv)
Definition: metadata.c:26
char * value
Definition: dict.h:87
int len
#define FF_PUT_WAV_HEADER_SKIP_CHANNELMASK
Tell ff_put_wav_header() to write an empty channel mask.
Definition: riff.h:59
int bits_per_coded_sample
The number of bits per sample in the codedwords.
Definition: codec_par.h:102
uint8_t * extradata
Extra binary data needed for initializing the decoder, codec-dependent.
Definition: codec_par.h:74
int channels
Audio only.
Definition: codec_par.h:166
AVCodecParameters * codecpar
Codec parameters associated with this stream.
Definition: avformat.h:1045
uint32_t codec_tag
Additional information about the codec (corresponds to the AVI FOURCC).
Definition: codec_par.h:64
AVRational time_base
This is the fundamental unit of time (in seconds) in terms of which frame timestamps are represented...
Definition: avformat.h:909
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
#define AV_CH_LAYOUT_MONO
AVPixelFormat
Pixel format.
Definition: pixfmt.h:64
int i
Definition: input.c:407
#define FF_PUT_WAV_HEADER_FORCE_WAVEFORMATEX
Tell ff_put_wav_header() to use WAVEFORMATEX even for PCM codecs.
Definition: riff.h:54