FFmpeg
wavdec.c
Go to the documentation of this file.
1 /*
2  * WAV demuxer
3  * Copyright (c) 2001, 2002 Fabrice Bellard
4  *
5  * Sony Wave64 demuxer
6  * RF64 demuxer
7  * Copyright (c) 2009 Daniel Verkamp
8  *
9  * BW64 demuxer
10  *
11  * This file is part of FFmpeg.
12  *
13  * FFmpeg is free software; you can redistribute it and/or
14  * modify it under the terms of the GNU Lesser General Public
15  * License as published by the Free Software Foundation; either
16  * version 2.1 of the License, or (at your option) any later version.
17  *
18  * FFmpeg is distributed in the hope that it will be useful,
19  * but WITHOUT ANY WARRANTY; without even the implied warranty of
20  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
21  * Lesser General Public License for more details.
22  *
23  * You should have received a copy of the GNU Lesser General Public
24  * License along with FFmpeg; if not, write to the Free Software
25  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
26  */
27 
28 #include <stdint.h>
29 
30 #include "config.h"
31 #include "libavutil/avassert.h"
32 #include "libavutil/dict.h"
33 #include "libavutil/intreadwrite.h"
34 #include "libavutil/log.h"
35 #include "libavutil/mathematics.h"
36 #include "libavutil/opt.h"
37 #include "avformat.h"
38 #include "avio.h"
39 #include "avio_internal.h"
40 #include "id3v2.h"
41 #include "internal.h"
42 #include "metadata.h"
43 #include "pcm.h"
44 #include "riff.h"
45 #include "w64.h"
46 #include "spdif.h"
47 
48 typedef struct WAVDemuxContext {
49  const AVClass *class;
50  int64_t data_end;
51  int w64;
52  int64_t smv_data_ofs;
55  int smv_block;
57  int smv_eof;
58  int audio_eof;
60  int max_size;
61  int spdif;
63  int unaligned; // e.g. if an odd number of bytes ID3 tag was prepended
64  int rifx; // RIFX: integer byte order for parameters is big endian
66 
67 #define OFFSET(x) offsetof(WAVDemuxContext, x)
68 #define DEC AV_OPT_FLAG_DECODING_PARAM
69 static const AVOption demux_options[] = {
70 #define W64_DEMUXER_OPTIONS_OFFSET (1 * CONFIG_WAV_DEMUXER)
71 #if CONFIG_WAV_DEMUXER
72  { "ignore_length", "Ignore length", OFFSET(ignore_length), AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, DEC },
73 #endif
74  { "max_size", "max size of single packet", OFFSET(max_size), AV_OPT_TYPE_INT, { .i64 = 4096 }, 1024, 1 << 22, DEC },
75  { NULL },
76 };
77 
79 {
80  if (CONFIG_SPDIF_DEMUXER && s->streams[0]->codecpar->codec_tag == 1) {
81  enum AVCodecID codec;
82  int len = 1<<16;
83  int ret = ffio_ensure_seekback(s->pb, len);
84 
85  if (ret >= 0) {
87  if (!buf) {
88  ret = AVERROR(ENOMEM);
89  } else {
90  int64_t pos = avio_tell(s->pb);
91  len = ret = avio_read(s->pb, buf, len);
92  if (len >= 0) {
93  ret = ff_spdif_probe(buf, len, &codec);
94  if (ret > AVPROBE_SCORE_EXTENSION) {
95  s->streams[0]->codecpar->codec_id = codec;
96  wav->spdif = 1;
97  }
98  }
99  avio_seek(s->pb, pos, SEEK_SET);
100  av_free(buf);
101  }
102  }
103 
104  if (ret < 0)
105  av_log(s, AV_LOG_WARNING, "Cannot check for SPDIF\n");
106  }
107 }
108 
109 #if CONFIG_WAV_DEMUXER
110 
111 static int64_t next_tag(AVIOContext *pb, uint32_t *tag, int big_endian)
112 {
113  *tag = avio_rl32(pb);
114  if (!big_endian) {
115  return avio_rl32(pb);
116  } else {
117  return avio_rb32(pb);
118  }
119 }
120 
121 /* RIFF chunks are always at even offsets relative to where they start. */
122 static int64_t wav_seek_tag(WAVDemuxContext * wav, AVIOContext *s, int64_t offset, int whence)
123 {
124  offset += offset < INT64_MAX && offset + wav->unaligned & 1;
125 
126  return avio_seek(s, offset, whence);
127 }
128 
129 /* return the size of the found tag */
130 static int64_t find_tag(WAVDemuxContext * wav, AVIOContext *pb, uint32_t tag1)
131 {
132  unsigned int tag;
133  int64_t size;
134 
135  for (;;) {
136  if (avio_feof(pb))
137  return AVERROR_EOF;
138  size = next_tag(pb, &tag, wav->rifx);
139  if (tag == tag1)
140  break;
141  wav_seek_tag(wav, pb, size, SEEK_CUR);
142  }
143  return size;
144 }
145 
146 static int wav_probe(const AVProbeData *p)
147 {
148  /* check file header */
149  if (p->buf_size <= 32)
150  return 0;
151  if (!memcmp(p->buf + 8, "WAVE", 4)) {
152  if (!memcmp(p->buf, "RIFF", 4) || !memcmp(p->buf, "RIFX", 4))
153  /* Since the ACT demuxer has a standard WAV header at the top of
154  * its own, the returned score is decreased to avoid a probe
155  * conflict between ACT and WAV. */
156  return AVPROBE_SCORE_MAX - 1;
157  else if ((!memcmp(p->buf, "RF64", 4) ||
158  !memcmp(p->buf, "BW64", 4)) &&
159  !memcmp(p->buf + 12, "ds64", 4))
160  return AVPROBE_SCORE_MAX;
161  }
162  return 0;
163 }
164 
165 static void handle_stream_probing(AVStream *st)
166 {
169  st->probe_packets = FFMIN(st->probe_packets, 32);
170  }
171 }
172 
173 static int wav_parse_fmt_tag(AVFormatContext *s, int64_t size, AVStream **st)
174 {
175  AVIOContext *pb = s->pb;
176  WAVDemuxContext *wav = s->priv_data;
177  int ret;
178 
179  /* parse fmt header */
180  *st = avformat_new_stream(s, NULL);
181  if (!*st)
182  return AVERROR(ENOMEM);
183 
184  ret = ff_get_wav_header(s, pb, (*st)->codecpar, size, wav->rifx);
185  if (ret < 0)
186  return ret;
187  handle_stream_probing(*st);
188 
189  (*st)->need_parsing = AVSTREAM_PARSE_FULL_RAW;
190 
191  avpriv_set_pts_info(*st, 64, 1, (*st)->codecpar->sample_rate);
192 
193  return 0;
194 }
195 
196 static int wav_parse_xma2_tag(AVFormatContext *s, int64_t size, AVStream **st)
197 {
198  AVIOContext *pb = s->pb;
199  int version, num_streams, i, channels = 0, ret;
200 
201  if (size < 36)
202  return AVERROR_INVALIDDATA;
203 
204  *st = avformat_new_stream(s, NULL);
205  if (!*st)
206  return AVERROR(ENOMEM);
207 
208  (*st)->codecpar->codec_type = AVMEDIA_TYPE_AUDIO;
209  (*st)->codecpar->codec_id = AV_CODEC_ID_XMA2;
210  (*st)->need_parsing = AVSTREAM_PARSE_FULL_RAW;
211 
212  version = avio_r8(pb);
213  if (version != 3 && version != 4)
214  return AVERROR_INVALIDDATA;
215  num_streams = avio_r8(pb);
216  if (size != (32 + ((version==3)?0:8) + 4*num_streams))
217  return AVERROR_INVALIDDATA;
218  avio_skip(pb, 10);
219  (*st)->codecpar->sample_rate = avio_rb32(pb);
220  if (version == 4)
221  avio_skip(pb, 8);
222  avio_skip(pb, 4);
223  (*st)->duration = avio_rb32(pb);
224  avio_skip(pb, 8);
225 
226  for (i = 0; i < num_streams; i++) {
227  channels += avio_r8(pb);
228  avio_skip(pb, 3);
229  }
230  (*st)->codecpar->channels = channels;
231 
232  if ((*st)->codecpar->channels <= 0 || (*st)->codecpar->sample_rate <= 0)
233  return AVERROR_INVALIDDATA;
234 
235  avpriv_set_pts_info(*st, 64, 1, (*st)->codecpar->sample_rate);
236 
237  avio_seek(pb, -size, SEEK_CUR);
238  if ((ret = ff_get_extradata(s, (*st)->codecpar, pb, size)) < 0)
239  return ret;
240 
241  return 0;
242 }
243 
244 static inline int wav_parse_bext_string(AVFormatContext *s, const char *key,
245  int length)
246 {
247  char temp[257];
248  int ret;
249 
250  av_assert0(length < sizeof(temp));
251  if ((ret = avio_read(s->pb, temp, length)) != length)
252  return ret < 0 ? ret : AVERROR_INVALIDDATA;
253 
254  temp[length] = 0;
255 
256  if (strlen(temp))
257  return av_dict_set(&s->metadata, key, temp, 0);
258 
259  return 0;
260 }
261 
262 static int wav_parse_bext_tag(AVFormatContext *s, int64_t size)
263 {
264  char temp[131], *coding_history;
265  int ret, x;
266  uint64_t time_reference;
267  int64_t umid_parts[8], umid_mask = 0;
268 
269  if ((ret = wav_parse_bext_string(s, "description", 256)) < 0 ||
270  (ret = wav_parse_bext_string(s, "originator", 32)) < 0 ||
271  (ret = wav_parse_bext_string(s, "originator_reference", 32)) < 0 ||
272  (ret = wav_parse_bext_string(s, "origination_date", 10)) < 0 ||
273  (ret = wav_parse_bext_string(s, "origination_time", 8)) < 0)
274  return ret;
275 
276  time_reference = avio_rl64(s->pb);
277  snprintf(temp, sizeof(temp), "%"PRIu64, time_reference);
278  if ((ret = av_dict_set(&s->metadata, "time_reference", temp, 0)) < 0)
279  return ret;
280 
281  /* check if version is >= 1, in which case an UMID may be present */
282  if (avio_rl16(s->pb) >= 1) {
283  for (x = 0; x < 8; x++)
284  umid_mask |= umid_parts[x] = avio_rb64(s->pb);
285 
286  if (umid_mask) {
287  /* the string formatting below is per SMPTE 330M-2004 Annex C */
288  if (umid_parts[4] == 0 && umid_parts[5] == 0 &&
289  umid_parts[6] == 0 && umid_parts[7] == 0) {
290  /* basic UMID */
291  snprintf(temp, sizeof(temp),
292  "0x%016"PRIX64"%016"PRIX64"%016"PRIX64"%016"PRIX64,
293  umid_parts[0], umid_parts[1],
294  umid_parts[2], umid_parts[3]);
295  } else {
296  /* extended UMID */
297  snprintf(temp, sizeof(temp),
298  "0x%016"PRIX64"%016"PRIX64"%016"PRIX64"%016"PRIX64
299  "%016"PRIX64"%016"PRIX64"%016"PRIX64"%016"PRIX64,
300  umid_parts[0], umid_parts[1],
301  umid_parts[2], umid_parts[3],
302  umid_parts[4], umid_parts[5],
303  umid_parts[6], umid_parts[7]);
304  }
305 
306  if ((ret = av_dict_set(&s->metadata, "umid", temp, 0)) < 0)
307  return ret;
308  }
309 
310  avio_skip(s->pb, 190);
311  } else
312  avio_skip(s->pb, 254);
313 
314  if (size > 602) {
315  /* CodingHistory present */
316  size -= 602;
317 
318  if (!(coding_history = av_malloc(size + 1)))
319  return AVERROR(ENOMEM);
320 
321  if ((ret = avio_read(s->pb, coding_history, size)) != size) {
322  av_free(coding_history);
323  return ret < 0 ? ret : AVERROR_INVALIDDATA;
324  }
325 
326  coding_history[size] = 0;
327  if ((ret = av_dict_set(&s->metadata, "coding_history", coding_history,
329  return ret;
330  }
331 
332  return 0;
333 }
334 
335 static const AVMetadataConv wav_metadata_conv[] = {
336  { "description", "comment" },
337  { "originator", "encoded_by" },
338  { "origination_date", "date" },
339  { "origination_time", "creation_time" },
340  { 0 },
341 };
342 
343 /* wav input */
344 static int wav_read_header(AVFormatContext *s)
345 {
346  int64_t size, av_uninit(data_size);
347  int64_t sample_count = 0;
348  int rf64 = 0, bw64 = 0;
349  uint32_t tag;
350  AVIOContext *pb = s->pb;
351  AVStream *st = NULL;
352  WAVDemuxContext *wav = s->priv_data;
353  int ret, got_fmt = 0, got_xma2 = 0;
354  int64_t next_tag_ofs, data_ofs = -1;
355 
356  wav->unaligned = avio_tell(s->pb) & 1;
357 
358  wav->smv_data_ofs = -1;
359 
360  /* read chunk ID */
361  tag = avio_rl32(pb);
362  switch (tag) {
363  case MKTAG('R', 'I', 'F', 'F'):
364  break;
365  case MKTAG('R', 'I', 'F', 'X'):
366  wav->rifx = 1;
367  break;
368  case MKTAG('R', 'F', '6', '4'):
369  rf64 = 1;
370  break;
371  case MKTAG('B', 'W', '6', '4'):
372  bw64 = 1;
373  break;
374  default:
375  av_log(s, AV_LOG_ERROR, "invalid start code %s in RIFF header\n",
376  av_fourcc2str(tag));
377  return AVERROR_INVALIDDATA;
378  }
379 
380  /* read chunk size */
381  avio_rl32(pb);
382 
383  /* read format */
384  if (avio_rl32(pb) != MKTAG('W', 'A', 'V', 'E')) {
385  av_log(s, AV_LOG_ERROR, "invalid format in RIFF header\n");
386  return AVERROR_INVALIDDATA;
387  }
388 
389  if (rf64 || bw64) {
390  if (avio_rl32(pb) != MKTAG('d', 's', '6', '4'))
391  return AVERROR_INVALIDDATA;
392  size = avio_rl32(pb);
393  if (size < 24)
394  return AVERROR_INVALIDDATA;
395  avio_rl64(pb); /* RIFF size */
396 
397  data_size = avio_rl64(pb);
398  sample_count = avio_rl64(pb);
399 
400  if (data_size < 0 || sample_count < 0) {
401  av_log(s, AV_LOG_ERROR, "negative data_size and/or sample_count in "
402  "ds64: data_size = %"PRId64", sample_count = %"PRId64"\n",
403  data_size, sample_count);
404  return AVERROR_INVALIDDATA;
405  }
406  avio_skip(pb, size - 24); /* skip rest of ds64 chunk */
407 
408  }
409 
410  for (;;) {
411  AVStream *vst;
412  size = next_tag(pb, &tag, wav->rifx);
413  next_tag_ofs = avio_tell(pb) + size;
414 
415  if (avio_feof(pb))
416  break;
417 
418  switch (tag) {
419  case MKTAG('f', 'm', 't', ' '):
420  /* only parse the first 'fmt ' tag found */
421  if (!got_xma2 && !got_fmt && (ret = wav_parse_fmt_tag(s, size, &st)) < 0) {
422  return ret;
423  } else if (got_fmt)
424  av_log(s, AV_LOG_WARNING, "found more than one 'fmt ' tag\n");
425 
426  got_fmt = 1;
427  break;
428  case MKTAG('X', 'M', 'A', '2'):
429  /* only parse the first 'XMA2' tag found */
430  if (!got_fmt && !got_xma2 && (ret = wav_parse_xma2_tag(s, size, &st)) < 0) {
431  return ret;
432  } else if (got_xma2)
433  av_log(s, AV_LOG_WARNING, "found more than one 'XMA2' tag\n");
434 
435  got_xma2 = 1;
436  break;
437  case MKTAG('d', 'a', 't', 'a'):
438  if (!(pb->seekable & AVIO_SEEKABLE_NORMAL) && !got_fmt && !got_xma2) {
439  av_log(s, AV_LOG_ERROR,
440  "found no 'fmt ' tag before the 'data' tag\n");
441  return AVERROR_INVALIDDATA;
442  }
443 
444  if (rf64 || bw64) {
445  next_tag_ofs = wav->data_end = avio_tell(pb) + data_size;
446  } else if (size != 0xFFFFFFFF) {
447  data_size = size;
448  next_tag_ofs = wav->data_end = size ? next_tag_ofs : INT64_MAX;
449  } else {
450  av_log(s, AV_LOG_WARNING, "Ignoring maximum wav data size, "
451  "file may be invalid\n");
452  data_size = 0;
453  next_tag_ofs = wav->data_end = INT64_MAX;
454  }
455 
456  data_ofs = avio_tell(pb);
457 
458  /* don't look for footer metadata if we can't seek or if we don't
459  * know where the data tag ends
460  */
461  if (!(pb->seekable & AVIO_SEEKABLE_NORMAL) || (!(rf64 && !bw64) && !size))
462  goto break_loop;
463  break;
464  case MKTAG('f', 'a', 'c', 't'):
465  if (!sample_count)
466  sample_count = (!wav->rifx ? avio_rl32(pb) : avio_rb32(pb));
467  break;
468  case MKTAG('b', 'e', 'x', 't'):
469  if ((ret = wav_parse_bext_tag(s, size)) < 0)
470  return ret;
471  break;
472  case MKTAG('S','M','V','0'):
473  if (!got_fmt) {
474  av_log(s, AV_LOG_ERROR, "found no 'fmt ' tag before the 'SMV0' tag\n");
475  return AVERROR_INVALIDDATA;
476  }
477  // SMV file, a wav file with video appended.
478  if (size != MKTAG('0','2','0','0')) {
479  av_log(s, AV_LOG_ERROR, "Unknown SMV version found\n");
480  goto break_loop;
481  }
482  av_log(s, AV_LOG_DEBUG, "Found SMV data\n");
483  wav->smv_given_first = 0;
484  vst = avformat_new_stream(s, NULL);
485  if (!vst)
486  return AVERROR(ENOMEM);
487  avio_r8(pb);
488  vst->id = 1;
491  vst->codecpar->width = avio_rl24(pb);
492  vst->codecpar->height = avio_rl24(pb);
493  if ((ret = ff_alloc_extradata(vst->codecpar, 4)) < 0) {
494  av_log(s, AV_LOG_ERROR, "Could not allocate extradata.\n");
495  return ret;
496  }
497  size = avio_rl24(pb);
498  wav->smv_data_ofs = avio_tell(pb) + (size - 5) * 3;
499  avio_rl24(pb);
500  wav->smv_block_size = avio_rl24(pb);
501  avpriv_set_pts_info(vst, 32, 1, avio_rl24(pb));
502  vst->duration = avio_rl24(pb);
503  avio_rl24(pb);
504  avio_rl24(pb);
505  wav->smv_frames_per_jpeg = avio_rl24(pb);
506  if (wav->smv_frames_per_jpeg > 65536) {
507  av_log(s, AV_LOG_ERROR, "too many frames per jpeg\n");
508  return AVERROR_INVALIDDATA;
509  }
511  goto break_loop;
512  case MKTAG('L', 'I', 'S', 'T'):
513  case MKTAG('l', 'i', 's', 't'):
514  if (size < 4) {
515  av_log(s, AV_LOG_ERROR, "too short LIST tag\n");
516  return AVERROR_INVALIDDATA;
517  }
518  switch (avio_rl32(pb)) {
519  case MKTAG('I', 'N', 'F', 'O'):
520  ff_read_riff_info(s, size - 4);
521  break;
522  case MKTAG('a', 'd', 't', 'l'):
523  if (s->nb_chapters > 0) {
524  while (avio_tell(pb) < next_tag_ofs &&
525  !avio_feof(pb)) {
526  char cue_label[512];
527  unsigned id, sub_size;
528 
529  if (avio_rl32(pb) != MKTAG('l', 'a', 'b', 'l'))
530  break;
531 
532  sub_size = avio_rl32(pb);
533  if (sub_size < 5)
534  break;
535  id = avio_rl32(pb);
536  avio_get_str(pb, sub_size - 4, cue_label, sizeof(cue_label));
537  avio_skip(pb, avio_tell(pb) & 1);
538 
539  for (int i = 0; i < s->nb_chapters; i++) {
540  if (s->chapters[i]->id == id) {
541  av_dict_set(&s->chapters[i]->metadata, "title", cue_label, 0);
542  break;
543  }
544  }
545  }
546  }
547  break;
548  }
549  break;
550  case MKTAG('I', 'D', '3', ' '):
551  case MKTAG('i', 'd', '3', ' '): {
552  ID3v2ExtraMeta *id3v2_extra_meta = NULL;
553  ff_id3v2_read_dict(pb, &s->internal->id3v2_meta, ID3v2_DEFAULT_MAGIC, &id3v2_extra_meta);
554  if (id3v2_extra_meta) {
555  ff_id3v2_parse_apic(s, id3v2_extra_meta);
556  ff_id3v2_parse_chapters(s, id3v2_extra_meta);
557  ff_id3v2_parse_priv(s, id3v2_extra_meta);
558  }
559  ff_id3v2_free_extra_meta(&id3v2_extra_meta);
560  }
561  break;
562  case MKTAG('c', 'u', 'e', ' '):
563  if (size >= 4 && got_fmt && st->codecpar->sample_rate > 0) {
564  AVRational tb = {1, st->codecpar->sample_rate};
565  unsigned nb_cues = avio_rl32(pb);
566 
567  if (size >= nb_cues * 24LL + 4LL) {
568  for (int i = 0; i < nb_cues; i++) {
569  unsigned offset, id = avio_rl32(pb);
570 
571  if (avio_feof(pb))
572  return AVERROR_INVALIDDATA;
573 
574  avio_skip(pb, 16);
575  offset = avio_rl32(pb);
576 
577  if (!avpriv_new_chapter(s, id, tb, offset, AV_NOPTS_VALUE, NULL))
578  return AVERROR(ENOMEM);
579  }
580  }
581  }
582  break;
583  }
584 
585  /* seek to next tag unless we know that we'll run into EOF */
586  if ((avio_size(pb) > 0 && next_tag_ofs >= avio_size(pb)) ||
587  wav_seek_tag(wav, pb, next_tag_ofs, SEEK_SET) < 0) {
588  break;
589  }
590  }
591 
592 break_loop:
593  if (!got_fmt && !got_xma2) {
594  av_log(s, AV_LOG_ERROR, "no 'fmt ' or 'XMA2' tag found\n");
595  return AVERROR_INVALIDDATA;
596  }
597 
598  if (data_ofs < 0) {
599  av_log(s, AV_LOG_ERROR, "no 'data' tag found\n");
600  return AVERROR_INVALIDDATA;
601  }
602 
603  avio_seek(pb, data_ofs, SEEK_SET);
604 
605  if (data_size > (INT64_MAX>>3)) {
606  av_log(s, AV_LOG_WARNING, "Data size %"PRId64" is too large\n", data_size);
607  data_size = 0;
608  }
609 
610  if ( st->codecpar->bit_rate > 0 && data_size > 0
611  && st->codecpar->sample_rate > 0
612  && sample_count > 0 && st->codecpar->channels > 1
613  && sample_count % st->codecpar->channels == 0) {
614  if (fabs(8.0 * data_size * st->codecpar->channels * st->codecpar->sample_rate /
615  sample_count /st->codecpar->bit_rate - 1.0) < 0.3)
616  sample_count /= st->codecpar->channels;
617  }
618 
619  if ( data_size > 0 && sample_count && st->codecpar->channels
620  && (data_size << 3) / sample_count / st->codecpar->channels > st->codecpar->bits_per_coded_sample + 1) {
621  av_log(s, AV_LOG_WARNING, "ignoring wrong sample_count %"PRId64"\n", sample_count);
622  sample_count = 0;
623  }
624 
625  /* G.729 hack (for Ticket4577)
626  * FIXME: Come up with cleaner, more general solution */
627  if (st->codecpar->codec_id == AV_CODEC_ID_G729 && sample_count && (data_size << 3) > sample_count) {
628  av_log(s, AV_LOG_WARNING, "ignoring wrong sample_count %"PRId64"\n", sample_count);
629  sample_count = 0;
630  }
631 
632  if (!sample_count || av_get_exact_bits_per_sample(st->codecpar->codec_id) > 0)
633  if ( st->codecpar->channels
634  && data_size
636  && wav->data_end <= avio_size(pb))
637  sample_count = (data_size << 3)
638  /
639  (st->codecpar->channels * (uint64_t)av_get_bits_per_sample(st->codecpar->codec_id));
640 
641  if (sample_count)
642  st->duration = sample_count;
643 
645  st->codecpar->block_align == st->codecpar->channels * 4 &&
646  st->codecpar->bits_per_coded_sample == 32 &&
647  st->codecpar->extradata_size == 2 &&
648  AV_RL16(st->codecpar->extradata) == 1) {
651  } else if (st->codecpar->codec_id == AV_CODEC_ID_PCM_S24LE &&
652  st->codecpar->block_align == st->codecpar->channels * 4 &&
653  st->codecpar->bits_per_coded_sample == 24) {
655  } else if (st->codecpar->codec_id == AV_CODEC_ID_XMA1 ||
657  st->codecpar->block_align = 2048;
658  } else if (st->codecpar->codec_id == AV_CODEC_ID_ADPCM_MS && st->codecpar->channels > 2 &&
659  st->codecpar->block_align < INT_MAX / st->codecpar->channels) {
660  st->codecpar->block_align *= st->codecpar->channels;
661  }
662 
663  ff_metadata_conv_ctx(s, NULL, wav_metadata_conv);
665 
666  set_spdif(s, wav);
667 
668  return 0;
669 }
670 
671 /**
672  * Find chunk with w64 GUID by skipping over other chunks.
673  * @return the size of the found chunk
674  */
675 static int64_t find_guid(AVIOContext *pb, const uint8_t guid1[16])
676 {
677  uint8_t guid[16];
678  int64_t size;
679 
680  while (!avio_feof(pb)) {
681  avio_read(pb, guid, 16);
682  size = avio_rl64(pb);
683  if (size <= 24 || size > INT64_MAX - 8)
684  return AVERROR_INVALIDDATA;
685  if (!memcmp(guid, guid1, 16))
686  return size;
687  avio_skip(pb, FFALIGN(size, INT64_C(8)) - 24);
688  }
689  return AVERROR_EOF;
690 }
691 
692 static int wav_read_packet(AVFormatContext *s, AVPacket *pkt)
693 {
694  int ret, size;
695  int64_t left;
696  AVStream *st;
697  WAVDemuxContext *wav = s->priv_data;
698 
699  if (CONFIG_SPDIF_DEMUXER && wav->spdif == 1)
700  return ff_spdif_read_packet(s, pkt);
701 
702  if (wav->smv_data_ofs > 0) {
703  int64_t audio_dts, video_dts;
704 smv_retry:
705  audio_dts = (int32_t)s->streams[0]->cur_dts;
706  video_dts = (int32_t)s->streams[1]->cur_dts;
707 
708  if (audio_dts != AV_NOPTS_VALUE && video_dts != AV_NOPTS_VALUE) {
709  /*We always return a video frame first to get the pixel format first*/
710  wav->smv_last_stream = wav->smv_given_first ?
711  av_compare_ts(video_dts, s->streams[1]->time_base,
712  audio_dts, s->streams[0]->time_base) > 0 : 0;
713  wav->smv_given_first = 1;
714  }
715  wav->smv_last_stream = !wav->smv_last_stream;
716  wav->smv_last_stream |= wav->audio_eof;
717  wav->smv_last_stream &= !wav->smv_eof;
718  if (wav->smv_last_stream) {
719  uint64_t old_pos = avio_tell(s->pb);
720  uint64_t new_pos = wav->smv_data_ofs +
721  wav->smv_block * wav->smv_block_size;
722  if (avio_seek(s->pb, new_pos, SEEK_SET) < 0) {
723  ret = AVERROR_EOF;
724  goto smv_out;
725  }
726  size = avio_rl24(s->pb);
727  ret = av_get_packet(s->pb, pkt, size);
728  if (ret < 0)
729  goto smv_out;
730  pkt->pos -= 3;
731  pkt->pts = wav->smv_block * wav->smv_frames_per_jpeg;
732  pkt->duration = wav->smv_frames_per_jpeg;
733  wav->smv_block++;
734 
735  pkt->stream_index = 1;
736 smv_out:
737  avio_seek(s->pb, old_pos, SEEK_SET);
738  if (ret == AVERROR_EOF) {
739  wav->smv_eof = 1;
740  goto smv_retry;
741  }
742  return ret;
743  }
744  }
745 
746  st = s->streams[0];
747 
748  left = wav->data_end - avio_tell(s->pb);
749  if (wav->ignore_length)
750  left = INT_MAX;
751  if (left <= 0) {
752  if (CONFIG_W64_DEMUXER && wav->w64)
753  left = find_guid(s->pb, ff_w64_guid_data) - 24;
754  else
755  left = find_tag(wav, s->pb, MKTAG('d', 'a', 't', 'a'));
756  if (left < 0) {
757  wav->audio_eof = 1;
758  if (wav->smv_data_ofs > 0 && !wav->smv_eof)
759  goto smv_retry;
760  return AVERROR_EOF;
761  }
762  wav->data_end = avio_tell(s->pb) + left;
763  }
764 
765  size = wav->max_size;
766  if (st->codecpar->block_align > 1) {
767  if (size < st->codecpar->block_align)
768  size = st->codecpar->block_align;
769  size = (size / st->codecpar->block_align) * st->codecpar->block_align;
770  }
771  size = FFMIN(size, left);
772  ret = av_get_packet(s->pb, pkt, size);
773  if (ret < 0)
774  return ret;
775  pkt->stream_index = 0;
776 
777  return ret;
778 }
779 
780 static int wav_read_seek(AVFormatContext *s,
781  int stream_index, int64_t timestamp, int flags)
782 {
783  WAVDemuxContext *wav = s->priv_data;
784  AVStream *st;
785  wav->smv_eof = 0;
786  wav->audio_eof = 0;
787  if (wav->smv_data_ofs > 0) {
788  int64_t smv_timestamp = timestamp;
789  if (stream_index == 0)
790  smv_timestamp = av_rescale_q(timestamp, s->streams[0]->time_base, s->streams[1]->time_base);
791  else
792  timestamp = av_rescale_q(smv_timestamp, s->streams[1]->time_base, s->streams[0]->time_base);
793  if (wav->smv_frames_per_jpeg > 0) {
794  wav->smv_block = smv_timestamp / wav->smv_frames_per_jpeg;
795  }
796  }
797 
798  st = s->streams[0];
799  switch (st->codecpar->codec_id) {
800  case AV_CODEC_ID_MP2:
801  case AV_CODEC_ID_MP3:
802  case AV_CODEC_ID_AC3:
803  case AV_CODEC_ID_DTS:
804  case AV_CODEC_ID_XMA2:
805  /* use generic seeking with dynamically generated indexes */
806  return -1;
807  default:
808  break;
809  }
810  return ff_pcm_read_seek(s, stream_index, timestamp, flags);
811 }
812 
813 static const AVClass wav_demuxer_class = {
814  .class_name = "WAV demuxer",
815  .item_name = av_default_item_name,
816  .option = demux_options,
817  .version = LIBAVUTIL_VERSION_INT,
818 };
820  .name = "wav",
821  .long_name = NULL_IF_CONFIG_SMALL("WAV / WAVE (Waveform Audio)"),
822  .priv_data_size = sizeof(WAVDemuxContext),
823  .read_probe = wav_probe,
824  .read_header = wav_read_header,
825  .read_packet = wav_read_packet,
826  .read_seek = wav_read_seek,
828  .codec_tag = ff_wav_codec_tags_list,
829  .priv_class = &wav_demuxer_class,
830 };
831 #endif /* CONFIG_WAV_DEMUXER */
832 
833 #if CONFIG_W64_DEMUXER
834 static int w64_probe(const AVProbeData *p)
835 {
836  if (p->buf_size <= 40)
837  return 0;
838  if (!memcmp(p->buf, ff_w64_guid_riff, 16) &&
839  !memcmp(p->buf + 24, ff_w64_guid_wave, 16))
840  return AVPROBE_SCORE_MAX;
841  else
842  return 0;
843 }
844 
845 static int w64_read_header(AVFormatContext *s)
846 {
847  int64_t size, data_ofs = 0;
848  AVIOContext *pb = s->pb;
849  WAVDemuxContext *wav = s->priv_data;
850  AVStream *st;
851  uint8_t guid[16];
852  int ret;
853 
854  avio_read(pb, guid, 16);
855  if (memcmp(guid, ff_w64_guid_riff, 16))
856  return AVERROR_INVALIDDATA;
857 
858  /* riff + wave + fmt + sizes */
859  if (avio_rl64(pb) < 16 + 8 + 16 + 8 + 16 + 8)
860  return AVERROR_INVALIDDATA;
861 
862  avio_read(pb, guid, 16);
863  if (memcmp(guid, ff_w64_guid_wave, 16)) {
864  av_log(s, AV_LOG_ERROR, "could not find wave guid\n");
865  return AVERROR_INVALIDDATA;
866  }
867 
868  wav->w64 = 1;
869 
870  st = avformat_new_stream(s, NULL);
871  if (!st)
872  return AVERROR(ENOMEM);
873 
874  while (!avio_feof(pb)) {
875  if (avio_read(pb, guid, 16) != 16)
876  break;
877  size = avio_rl64(pb);
878  if (size <= 24 || INT64_MAX - size < avio_tell(pb))
879  return AVERROR_INVALIDDATA;
880 
881  if (!memcmp(guid, ff_w64_guid_fmt, 16)) {
882  /* subtract chunk header size - normal wav file doesn't count it */
883  ret = ff_get_wav_header(s, pb, st->codecpar, size - 24, 0);
884  if (ret < 0)
885  return ret;
886  avio_skip(pb, FFALIGN(size, INT64_C(8)) - size);
887 
888  avpriv_set_pts_info(st, 64, 1, st->codecpar->sample_rate);
889  } else if (!memcmp(guid, ff_w64_guid_fact, 16)) {
890  int64_t samples;
891 
892  samples = avio_rl64(pb);
893  if (samples > 0)
894  st->duration = samples;
895  avio_skip(pb, FFALIGN(size, INT64_C(8)) - 32);
896  } else if (!memcmp(guid, ff_w64_guid_data, 16)) {
897  wav->data_end = avio_tell(pb) + size - 24;
898 
899  data_ofs = avio_tell(pb);
900  if (!(pb->seekable & AVIO_SEEKABLE_NORMAL))
901  break;
902 
903  avio_skip(pb, size - 24);
904  } else if (!memcmp(guid, ff_w64_guid_summarylist, 16)) {
905  int64_t start, end, cur;
906  uint32_t count, chunk_size, i;
907  int64_t filesize = avio_size(s->pb);
908 
909  start = avio_tell(pb);
910  end = start + FFALIGN(size, INT64_C(8)) - 24;
911  count = avio_rl32(pb);
912 
913  for (i = 0; i < count; i++) {
914  char chunk_key[5], *value;
915 
916  if (avio_feof(pb) || (cur = avio_tell(pb)) < 0 || cur > end - 8 /* = tag + size */)
917  break;
918 
919  chunk_key[4] = 0;
920  avio_read(pb, chunk_key, 4);
921  chunk_size = avio_rl32(pb);
922  if (chunk_size == UINT32_MAX || (filesize >= 0 && chunk_size > filesize))
923  return AVERROR_INVALIDDATA;
924 
925  value = av_malloc(chunk_size + 1);
926  if (!value)
927  return AVERROR(ENOMEM);
928 
929  ret = avio_get_str16le(pb, chunk_size, value, chunk_size);
930  if (ret < 0) {
931  av_free(value);
932  return ret;
933  }
934  avio_skip(pb, chunk_size - ret);
935 
936  av_dict_set(&s->metadata, chunk_key, value, AV_DICT_DONT_STRDUP_VAL);
937  }
938 
939  avio_skip(pb, end - avio_tell(pb));
940  } else {
941  av_log(s, AV_LOG_DEBUG, "unknown guid: "FF_PRI_GUID"\n", FF_ARG_GUID(guid));
942  avio_skip(pb, FFALIGN(size, INT64_C(8)) - 24);
943  }
944  }
945 
946  if (!data_ofs)
947  return AVERROR_EOF;
948 
949  ff_metadata_conv_ctx(s, NULL, wav_metadata_conv);
951 
952  handle_stream_probing(st);
954 
955  avio_seek(pb, data_ofs, SEEK_SET);
956 
957  set_spdif(s, wav);
958 
959  return 0;
960 }
961 
962 static const AVClass w64_demuxer_class = {
963  .class_name = "W64 demuxer",
964  .item_name = av_default_item_name,
965  .option = &demux_options[W64_DEMUXER_OPTIONS_OFFSET],
966  .version = LIBAVUTIL_VERSION_INT,
967 };
968 
970  .name = "w64",
971  .long_name = NULL_IF_CONFIG_SMALL("Sony Wave64"),
972  .priv_data_size = sizeof(WAVDemuxContext),
973  .read_probe = w64_probe,
974  .read_header = w64_read_header,
975  .read_packet = wav_read_packet,
976  .read_seek = wav_read_seek,
978  .codec_tag = ff_wav_codec_tags_list,
979  .priv_class = &w64_demuxer_class,
980 };
981 #endif /* CONFIG_W64_DEMUXER */
unsigned int nb_chapters
Number of chapters in AVChapter array.
Definition: avformat.h:1472
int ff_read_riff_info(AVFormatContext *s, int64_t size)
Definition: riffdec.c:228
#define NULL
Definition: coverity.c:32
int ff_id3v2_parse_priv(AVFormatContext *s, ID3v2ExtraMeta *extra_meta)
Add metadata for all PRIV tags in the ID3v2 header.
Definition: id3v2.c:1269
AVDictionary * id3v2_meta
ID3v2 tag useful for MP3 demuxing.
Definition: internal.h:139
Bytestream IO Context.
Definition: avio.h:161
#define AVERROR_INVALIDDATA
Invalid data found when processing input.
Definition: error.h:59
int64_t video_dts
Definition: movenc.c:60
void ff_metadata_conv_ctx(AVFormatContext *ctx, const AVMetadataConv *d_conv, const AVMetadataConv *s_conv)
Definition: metadata.c:59
int64_t avio_size(AVIOContext *s)
Get the filesize.
Definition: aviobuf.c:346
const uint8_t ff_w64_guid_wave[16]
Definition: w64.c:28
Buffered I/O operations.
version
Definition: libkvazaar.c:320
int ff_spdif_probe(const uint8_t *p_buf, int buf_size, enum AVCodecID *codec)
Definition: spdifdec.c:116
AVOption.
Definition: opt.h:248
const uint8_t ff_w64_guid_fact[16]
Definition: w64.c:38
#define AV_LOG_WARNING
Something somehow does not look correct.
Definition: log.h:200
#define LIBAVUTIL_VERSION_INT
Definition: version.h:85
int64_t pos
byte position in stream, -1 if unknown
Definition: packet.h:383
else temp
Definition: vf_mcdeint.c:256
#define FF_ARG_GUID(g)
Definition: riff.h:109
void avpriv_set_pts_info(AVStream *s, int pts_wrap_bits, unsigned int pts_num, unsigned int pts_den)
Set the time base and wrapping info for a given stream.
Definition: utils.c:4905
int probe_packets
Number of packets to buffer for codec probing.
Definition: avformat.h:1084
static int read_seek(AVFormatContext *ctx, int stream_index, int64_t timestamp, int flags)
Definition: libcdio.c:153
enum AVCodecID codec_id
Specific type of the encoded data (the codec used).
Definition: codec_par.h:60
#define ID3v2_DEFAULT_MAGIC
Default magic bytes for ID3v2 header: "ID3".
Definition: id3v2.h:35
int64_t avio_seek(AVIOContext *s, int64_t offset, int whence)
fseek() equivalent for AVIOContext.
Definition: aviobuf.c:253
const char * av_default_item_name(void *ptr)
Return the context name.
Definition: log.c:235
AVFormatInternal * internal
An opaque field for libavformat internal usage.
Definition: avformat.h:1708
int smv_frames_per_jpeg
Definition: wavdec.c:54
const AVCodecTag *const ff_wav_codec_tags_list[]
int64_t avio_skip(AVIOContext *s, int64_t offset)
Skip given number of bytes forward.
Definition: aviobuf.c:341
static void set_spdif(AVFormatContext *s, WAVDemuxContext *wav)
Definition: wavdec.c:78
const char * key
int max_size
Definition: wavdec.c:60
static AVPacket pkt
AVDictionary * metadata
Definition: avformat.h:1204
uint64_t_TMPL AV_WL64 unsigned int_TMPL AV_WL32 unsigned int_TMPL AV_WL24 unsigned int_TMPL AV_RL16
Definition: bytestream.h:91
int avio_get_str16le(AVIOContext *pb, int maxlen, char *buf, int buflen)
Read a UTF-16 string from pb and convert it to UTF-8.
int ignore_length
Definition: wavdec.c:59
AVChapter * avpriv_new_chapter(AVFormatContext *s, int id, AVRational time_base, int64_t start, int64_t end, const char *title)
Add a new chapter.
Definition: utils.c:4606
Format I/O context.
Definition: avformat.h:1243
static const GUIDParseTable * find_guid(ff_asf_guid guid)
Definition: asfdec_o.c:1645
int64_t audio_dts
Definition: movenc.c:60
int64_t cur_dts
Definition: avformat.h:1077
const char * class_name
The name of the class; usually it is the same name as the context structure type to which the AVClass...
Definition: log.h:72
#define av_assert0(cond)
assert() equivalent, that is always enabled.
Definition: avassert.h:37
internal metadata API header see avformat.h or the public API!
Public dictionary API.
uint8_t
#define av_malloc(s)
int width
Video only.
Definition: codec_par.h:126
AVOptions.
unsigned int avio_rb32(AVIOContext *s)
Definition: aviobuf.c:785
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 offset
int id
unique ID to identify the chapter
Definition: avformat.h:1201
int64_t duration
Duration of this packet in AVStream->time_base units, 0 if unknown.
Definition: packet.h:381
enum AVStreamParseType need_parsing
Definition: avformat.h:1092
int id
Format-specific stream ID.
Definition: avformat.h:891
GLsizei GLboolean const GLfloat * value
Definition: opengl_enc.c:108
AVStream * avformat_new_stream(AVFormatContext *s, const AVCodec *c)
Add a new stream to a media file.
Definition: utils.c:4481
AVStream ** streams
A list of all streams in the file.
Definition: avformat.h:1311
int ff_id3v2_parse_chapters(AVFormatContext *s, ID3v2ExtraMeta *extra_meta)
Create chapters for all CHAP tags found in the ID3v2 header.
Definition: id3v2.c:1178
uint32_t tag
Definition: movenc.c:1597
#define AVERROR_EOF
End of file.
Definition: error.h:55
int av_get_packet(AVIOContext *s, AVPacket *pkt, int size)
Allocate and read the payload of a packet and initialize its fields with default values.
Definition: utils.c:310
ptrdiff_t size
Definition: opengl_enc.c:100
uint64_t avio_rb64(AVIOContext *s)
Definition: aviobuf.c:906
static av_always_inline int64_t avio_tell(AVIOContext *s)
ftell() equivalent for AVIOContext.
Definition: avio.h:557
channels
Definition: aptx.h:33
#define FFALIGN(x, a)
Definition: macros.h:48
full parsing and repack with timestamp and position generation by parser for raw this assumes that ea...
Definition: avformat.h:807
#define av_log(a,...)
int avio_read(AVIOContext *s, unsigned char *buf, int size)
Read size bytes from AVIOContext into buf.
Definition: aviobuf.c:637
int ff_get_wav_header(AVFormatContext *s, AVIOContext *pb, AVCodecParameters *par, int size, int big_endian)
Definition: riffdec.c:91
int64_t bit_rate
The average bitrate of the encoded data (in bits per second).
Definition: codec_par.h:89
int64_t av_rescale_q(int64_t a, AVRational bq, AVRational cq)
Rescale a 64-bit integer by 2 rational numbers.
Definition: mathematics.c:142
int ff_spdif_read_packet(AVFormatContext *s, AVPacket *pkt)
Definition: spdifdec.c:173
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
int av_get_bits_per_sample(enum AVCodecID codec_id)
Return codec bits per sample.
Definition: utils.c:1610
AVDictionary * metadata
Metadata that applies to the whole file.
Definition: avformat.h:1483
#define OFFSET(x)
Definition: wavdec.c:67
static __device__ float fabs(float a)
Definition: cuda_runtime.h:182
unsigned int avio_rl32(AVIOContext *s)
Definition: aviobuf.c:754
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
Definition: internal.h:115
unsigned int pos
Definition: spdifenc.c:412
#define AV_LOG_DEBUG
Stuff which is only useful for libav* developers.
Definition: log.h:215
preferred ID for decoding MPEG audio layer 1, 2 or 3
Definition: codec_id.h:423
#define av_fourcc2str(fourcc)
Definition: avutil.h:348
enum AVMediaType codec_type
General type of the encoded data.
Definition: codec_par.h:56
void ff_id3v2_free_extra_meta(ID3v2ExtraMeta **extra_meta)
Free memory allocated parsing special (non-text) metadata.
Definition: id3v2.c:1122
int request_probe
stream probing state -1 -> probing finished 0 -> no probing requested rest -> perform probing with re...
Definition: internal.h:247
AVChapter ** chapters
Definition: avformat.h:1473
simple assert() macros that are a bit more flexible than ISO C assert().
GLsizei GLsizei * length
Definition: opengl_enc.c:114
int smv_block_size
Definition: wavdec.c:53
static const AVOption demux_options[]
Definition: wavdec.c:69
GLsizei count
Definition: opengl_enc.c:108
int av_compare_ts(int64_t ts_a, AVRational tb_a, int64_t ts_b, AVRational tb_b)
Compare two timestamps each in its own time base.
Definition: mathematics.c:147
int extradata_size
Size of the extradata content in bytes.
Definition: codec_par.h:78
int avio_r8(AVIOContext *s)
Definition: aviobuf.c:628
AVInputFormat ff_wav_demuxer
int buf_size
Size of buf except extra allocated bytes.
Definition: avformat.h:444
unsigned char * buf
Buffer must have AVPROBE_PADDING_SIZE of extra allocated bytes filled with zero.
Definition: avformat.h:443
int block_align
Audio only.
Definition: codec_par.h:177
int seekable
A combination of AVIO_SEEKABLE_ flags or 0 when the stream is not seekable.
Definition: avio.h:260
int ff_alloc_extradata(AVCodecParameters *par, int size)
Allocate extradata with additional AV_INPUT_BUFFER_PADDING_SIZE at end which is always set to 0...
Definition: utils.c:3291
#define FFMIN(a, b)
Definition: common.h:105
int smv_last_stream
Definition: wavdec.c:56
int av_get_exact_bits_per_sample(enum AVCodecID codec_id)
Return codec bits per sample.
Definition: utils.c:1522
#define AV_DICT_DONT_STRDUP_VAL
Take ownership of a value that&#39;s been allocated with av_malloc() or another memory allocation functio...
Definition: dict.h:76
const uint8_t ff_w64_guid_data[16]
Definition: w64.c:42
internal header for RIFF based (de)muxers do NOT include this in end user applications ...
int32_t
const uint8_t ff_w64_guid_riff[16]
Definition: w64.c:23
#define s(width, name)
Definition: cbs_vp9.c:257
int audio_eof
Definition: wavdec.c:58
int ff_pcm_read_seek(AVFormatContext *s, int stream_index, int64_t timestamp, int flags)
Definition: pcm.c:56
static int read_header(FFV1Context *f)
Definition: ffv1dec.c:527
Stream structure.
Definition: avformat.h:884
static int read_packet(void *opaque, uint8_t *buf, int buf_size)
Definition: avio_reading.c:42
#define AVIO_SEEKABLE_NORMAL
Seeking works like for a local file.
Definition: avio.h:40
AVStreamInternal * internal
An opaque field for libavformat internal usage.
Definition: avformat.h:1123
AVIOContext * pb
I/O context.
Definition: avformat.h:1285
Tag MUST be and< 10hcoeff half pel interpolation filter coefficients, hcoeff[0] are the 2 middle coefficients[1] are the next outer ones and so on, resulting in a filter like:...eff[2], hcoeff[1], hcoeff[0], hcoeff[0], hcoeff[1], hcoeff[2]...the sign of the coefficients is not explicitly stored but alternates after each coeff and coeff[0] is positive, so...,+,-,+,-,+,+,-,+,-,+,...hcoeff[0] is not explicitly stored but found by subtracting the sum of all stored coefficients with signs from 32 hcoeff[0]=32-hcoeff[1]-hcoeff[2]-...a good choice for hcoeff and htaps is htaps=6 hcoeff={40,-10, 2}an alternative which requires more computations at both encoder and decoder side and may or may not be better is htaps=8 hcoeff={42,-14, 6,-2}ref_frames minimum of the number of available reference frames and max_ref_frames for example the first frame after a key frame always has ref_frames=1spatial_decomposition_type wavelet type 0 is a 9/7 symmetric compact integer wavelet 1 is a 5/3 symmetric compact integer wavelet others are reserved stored as delta from last, last is reset to 0 if always_reset||keyframeqlog quality(logarithmic quantizer scale) stored as delta from last, last is reset to 0 if always_reset||keyframemv_scale stored as delta from last, last is reset to 0 if always_reset||keyframe FIXME check that everything works fine if this changes between framesqbias dequantization bias stored as delta from last, last is reset to 0 if always_reset||keyframeblock_max_depth maximum depth of the block tree stored as delta from last, last is reset to 0 if always_reset||keyframequant_table quantization tableHighlevel bitstream structure:==============================--------------------------------------------|Header|--------------------------------------------|------------------------------------|||Block0||||split?||||yes no||||.........intra?||||:Block01:yes no||||:Block02:.................||||:Block03::y DC::ref index:||||:Block04::cb DC::motion x:||||.........:cr DC::motion y:||||.................|||------------------------------------||------------------------------------|||Block1|||...|--------------------------------------------|------------------------------------|||Y subbands||Cb subbands||Cr subbands||||------||------||------|||||LL0||HL0||||LL0||HL0||||LL0||HL0|||||------||------||------||||------||------||------|||||LH0||HH0||||LH0||HH0||||LH0||HH0|||||------||------||------||||------||------||------|||||HL1||LH1||||HL1||LH1||||HL1||LH1|||||------||------||------||||------||------||------|||||HH1||HL2||||HH1||HL2||||HH1||HL2|||||...||...||...|||------------------------------------|--------------------------------------------Decoding process:=================------------|||Subbands|------------||||------------|Intra DC||||LL0 subband prediction------------|\Dequantization-------------------\||Reference frames|\IDWT|--------------|Motion\|||Frame 0||Frame 1||Compensation.OBMC v-------|--------------|--------------.\------> Frame n output Frame Frame<----------------------------------/|...|-------------------Range Coder:============Binary Range Coder:-------------------The implemented range coder is an adapted version based upon"Range encoding: an algorithm for removing redundancy from a digitised message."by G.N.N.Martin.The symbols encoded by the Snow range coder are bits(0|1).The associated probabilities are not fix but change depending on the symbol mix seen so far.bit seen|new state---------+-----------------------------------------------0|256-state_transition_table[256-old_state];1|state_transition_table[old_state];state_transition_table={0, 0, 0, 0, 0, 0, 0, 0, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 190, 191, 192, 194, 194, 195, 196, 197, 198, 199, 200, 201, 202, 202, 204, 205, 206, 207, 208, 209, 209, 210, 211, 212, 213, 215, 215, 216, 217, 218, 219, 220, 220, 222, 223, 224, 225, 226, 227, 227, 229, 229, 230, 231, 232, 234, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 248, 0, 0, 0, 0, 0, 0, 0};FIXME Range Coding of integers:-------------------------FIXME Neighboring Blocks:===================left and top are set to the respective blocks unless they are outside of the image in which case they are set to the Null block top-left is set to the top left block unless it is outside of the image in which case it is set to the left block if this block has no larger parent block or it is at the left side of its parent block and the top right block is not outside of the image then the top right block is used for top-right else the top-left block is used Null block y, cb, cr are 128 level, ref, mx and my are 0 Motion Vector Prediction:=========================1.the motion vectors of all the neighboring blocks are scaled to compensate for the difference of reference frames scaled_mv=(mv *(256 *(current_reference+1)/(mv.reference+1))+128)> the median of the scaled left
Definition: snow.txt:206
int ff_id3v2_parse_apic(AVFormatContext *s, ID3v2ExtraMeta *extra_meta)
Create a stream for each APIC (attached picture) extracted from the ID3v2 header. ...
Definition: id3v2.c:1138
int av_dict_set(AVDictionary **pm, const char *key, const char *value, int flags)
Set the given entry in *pm, overwriting an existing entry.
Definition: dict.c:70
Describe the class of an AVClass context structure.
Definition: log.h:67
#define AVFMT_GENERIC_INDEX
Use generic index building code.
Definition: avformat.h:463
Rational number (pair of numerator and denominator).
Definition: rational.h:58
#define W64_DEMUXER_OPTIONS_OFFSET
int64_t smv_data_ofs
Definition: wavdec.c:52
#define DEC
Definition: wavdec.c:68
#define snprintf
Definition: snprintf.h:34
#define AVPROBE_SCORE_EXTENSION
score for file extension
Definition: avformat.h:451
This structure contains the data a format has to probe a file.
Definition: avformat.h:441
const AVMetadataConv ff_riff_info_conv[]
Definition: riff.c:603
#define flags(name, subs,...)
Definition: cbs_av1.c:561
const uint8_t ff_w64_guid_fmt[16]
Definition: w64.c:33
static int read_probe(const AVProbeData *pd)
Definition: jvdec.c:55
int ffio_ensure_seekback(AVIOContext *s, int64_t buf_size)
Ensures that the requested seekback buffer size will be available.
Definition: aviobuf.c:1002
int64_t duration
Decoding: duration of the stream, in stream time base.
Definition: avformat.h:933
int sample_rate
Audio only.
Definition: codec_par.h:170
#define AVPROBE_SCORE_MAX
maximum score
Definition: avformat.h:453
int smv_given_first
Definition: wavdec.c:62
unsigned int avio_rl16(AVIOContext *s)
Definition: aviobuf.c:738
Main libavformat public API header.
#define FF_PRI_GUID
Definition: riff.h:105
int ff_get_extradata(AVFormatContext *s, AVCodecParameters *par, AVIOContext *pb, int size)
Allocate extradata with additional AV_INPUT_BUFFER_PADDING_SIZE at end which is always set to 0 and f...
Definition: utils.c:3309
int64_t data_end
Definition: wavdec.c:50
#define AV_INPUT_BUFFER_PADDING_SIZE
Required number of additionally allocated bytes at the end of the input bitstream for decoding...
Definition: avcodec.h:215
#define av_free(p)
void ff_id3v2_read_dict(AVIOContext *pb, AVDictionary **metadata, const char *magic, ID3v2ExtraMeta **extra_meta)
Read an ID3v2 tag into specified dictionary and retrieve supported extra metadata.
Definition: id3v2.c:1110
int smv_block
Definition: wavdec.c:55
int len
void * priv_data
Format private data.
Definition: avformat.h:1271
const uint8_t ff_w64_guid_summarylist[16]
Definition: w64.c:47
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
#define av_uninit(x)
Definition: attributes.h:154
AVInputFormat ff_w64_demuxer
Filter the word “frame” indicates either a video frame or a group of audio samples
const char * name
A comma separated list of short names for the format.
Definition: avformat.h:654
AVCodecParameters * codecpar
Codec parameters associated with this stream.
Definition: avformat.h:1049
int avio_get_str(AVIOContext *pb, int maxlen, char *buf, int buflen)
Read a string from pb into buf.
Definition: aviobuf.c:864
int avio_feof(AVIOContext *s)
Similar to feof() but also returns nonzero on read errors.
Definition: aviobuf.c:368
uint32_t codec_tag
Additional information about the codec (corresponds to the AVI FOURCC).
Definition: codec_par.h:64
unsigned int avio_rl24(AVIOContext *s)
Definition: aviobuf.c:746
int stream_index
Definition: packet.h:365
AVRational time_base
This is the fundamental unit of time (in seconds) in terms of which frame timestamps are represented...
Definition: avformat.h:913
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 MKTAG(a, b, c, d)
Definition: common.h:478
enum AVCodecID id
This structure stores compressed data.
Definition: packet.h:340
uint64_t avio_rl64(AVIOContext *s)
Definition: aviobuf.c:762
int64_t pts
Presentation timestamp in AVStream->time_base units; the time at which the decompressed packet will b...
Definition: packet.h:356
int i
Definition: input.c:407
#define AV_NOPTS_VALUE
Undefined timestamp value.
Definition: avutil.h:248
#define tb
Definition: regdef.h:68
int unaligned
Definition: wavdec.c:63
#define AV_WL32(p, v)
Definition: intreadwrite.h:426