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 "libavutil/avassert.h"
31 #include "libavutil/dict.h"
32 #include "libavutil/intreadwrite.h"
33 #include "libavutil/log.h"
34 #include "libavutil/mathematics.h"
35 #include "libavutil/opt.h"
36 #include "avformat.h"
37 #include "avio.h"
38 #include "avio_internal.h"
39 #include "id3v2.h"
40 #include "internal.h"
41 #include "metadata.h"
42 #include "pcm.h"
43 #include "riff.h"
44 #include "w64.h"
45 #include "spdif.h"
46 
47 typedef struct WAVDemuxContext {
48  const AVClass *class;
49  int64_t data_end;
50  int w64;
51  int64_t smv_data_ofs;
54  int smv_block;
56  int smv_eof;
57  int audio_eof;
59  int max_size;
60  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 
68 {
69  if (CONFIG_SPDIF_DEMUXER && s->streams[0]->codecpar->codec_tag == 1) {
70  enum AVCodecID codec;
71  int len = 1<<16;
72  int ret = ffio_ensure_seekback(s->pb, len);
73 
74  if (ret >= 0) {
75  uint8_t *buf = av_malloc(len);
76  if (!buf) {
77  ret = AVERROR(ENOMEM);
78  } else {
79  int64_t pos = avio_tell(s->pb);
80  len = ret = avio_read(s->pb, buf, len);
81  if (len >= 0) {
82  ret = ff_spdif_probe(buf, len, &codec);
83  if (ret > AVPROBE_SCORE_EXTENSION) {
84  s->streams[0]->codecpar->codec_id = codec;
85  wav->spdif = 1;
86  }
87  }
88  avio_seek(s->pb, pos, SEEK_SET);
89  av_free(buf);
90  }
91  }
92 
93  if (ret < 0)
94  av_log(s, AV_LOG_WARNING, "Cannot check for SPDIF\n");
95  }
96 }
97 
98 #if CONFIG_WAV_DEMUXER
99 
100 static int64_t next_tag(AVIOContext *pb, uint32_t *tag, int big_endian)
101 {
102  *tag = avio_rl32(pb);
103  if (!big_endian) {
104  return avio_rl32(pb);
105  } else {
106  return avio_rb32(pb);
107  }
108 }
109 
110 /* RIFF chunks are always at even offsets relative to where they start. */
111 static int64_t wav_seek_tag(WAVDemuxContext * wav, AVIOContext *s, int64_t offset, int whence)
112 {
113  offset += offset < INT64_MAX && offset + wav->unaligned & 1;
114 
115  return avio_seek(s, offset, whence);
116 }
117 
118 /* return the size of the found tag */
119 static int64_t find_tag(WAVDemuxContext * wav, AVIOContext *pb, uint32_t tag1)
120 {
121  unsigned int tag;
122  int64_t size;
123 
124  for (;;) {
125  if (avio_feof(pb))
126  return AVERROR_EOF;
127  size = next_tag(pb, &tag, wav->rifx);
128  if (tag == tag1)
129  break;
130  wav_seek_tag(wav, pb, size, SEEK_CUR);
131  }
132  return size;
133 }
134 
135 static int wav_probe(const AVProbeData *p)
136 {
137  /* check file header */
138  if (p->buf_size <= 32)
139  return 0;
140  if (!memcmp(p->buf + 8, "WAVE", 4)) {
141  if (!memcmp(p->buf, "RIFF", 4) || !memcmp(p->buf, "RIFX", 4))
142  /* Since the ACT demuxer has a standard WAV header at the top of
143  * its own, the returned score is decreased to avoid a probe
144  * conflict between ACT and WAV. */
145  return AVPROBE_SCORE_MAX - 1;
146  else if ((!memcmp(p->buf, "RF64", 4) ||
147  !memcmp(p->buf, "BW64", 4)) &&
148  !memcmp(p->buf + 12, "ds64", 4))
149  return AVPROBE_SCORE_MAX;
150  }
151  return 0;
152 }
153 
154 static void handle_stream_probing(AVStream *st)
155 {
158  st->probe_packets = FFMIN(st->probe_packets, 32);
159  }
160 }
161 
162 static int wav_parse_fmt_tag(AVFormatContext *s, int64_t size, AVStream **st)
163 {
164  AVIOContext *pb = s->pb;
165  WAVDemuxContext *wav = s->priv_data;
166  int ret;
167 
168  /* parse fmt header */
169  *st = avformat_new_stream(s, NULL);
170  if (!*st)
171  return AVERROR(ENOMEM);
172 
173  ret = ff_get_wav_header(s, pb, (*st)->codecpar, size, wav->rifx);
174  if (ret < 0)
175  return ret;
176  handle_stream_probing(*st);
177 
178  (*st)->need_parsing = AVSTREAM_PARSE_FULL_RAW;
179 
180  avpriv_set_pts_info(*st, 64, 1, (*st)->codecpar->sample_rate);
181 
182  return 0;
183 }
184 
185 static int wav_parse_xma2_tag(AVFormatContext *s, int64_t size, AVStream **st)
186 {
187  AVIOContext *pb = s->pb;
188  int version, num_streams, i, channels = 0, ret;
189 
190  if (size < 36)
191  return AVERROR_INVALIDDATA;
192 
193  *st = avformat_new_stream(s, NULL);
194  if (!*st)
195  return AVERROR(ENOMEM);
196 
197  (*st)->codecpar->codec_type = AVMEDIA_TYPE_AUDIO;
198  (*st)->codecpar->codec_id = AV_CODEC_ID_XMA2;
199  (*st)->need_parsing = AVSTREAM_PARSE_FULL_RAW;
200 
201  version = avio_r8(pb);
202  if (version != 3 && version != 4)
203  return AVERROR_INVALIDDATA;
204  num_streams = avio_r8(pb);
205  if (size != (32 + ((version==3)?0:8) + 4*num_streams))
206  return AVERROR_INVALIDDATA;
207  avio_skip(pb, 10);
208  (*st)->codecpar->sample_rate = avio_rb32(pb);
209  if (version == 4)
210  avio_skip(pb, 8);
211  avio_skip(pb, 4);
212  (*st)->duration = avio_rb32(pb);
213  avio_skip(pb, 8);
214 
215  for (i = 0; i < num_streams; i++) {
216  channels += avio_r8(pb);
217  avio_skip(pb, 3);
218  }
219  (*st)->codecpar->channels = channels;
220 
221  if ((*st)->codecpar->channels <= 0 || (*st)->codecpar->sample_rate <= 0)
222  return AVERROR_INVALIDDATA;
223 
224  avpriv_set_pts_info(*st, 64, 1, (*st)->codecpar->sample_rate);
225 
226  avio_seek(pb, -size, SEEK_CUR);
227  if ((ret = ff_get_extradata(s, (*st)->codecpar, pb, size)) < 0)
228  return ret;
229 
230  return 0;
231 }
232 
233 static inline int wav_parse_bext_string(AVFormatContext *s, const char *key,
234  int length)
235 {
236  char temp[257];
237  int ret;
238 
239  av_assert0(length < sizeof(temp));
240  if ((ret = avio_read(s->pb, temp, length)) != length)
241  return ret < 0 ? ret : AVERROR_INVALIDDATA;
242 
243  temp[length] = 0;
244 
245  if (strlen(temp))
246  return av_dict_set(&s->metadata, key, temp, 0);
247 
248  return 0;
249 }
250 
251 static int wav_parse_bext_tag(AVFormatContext *s, int64_t size)
252 {
253  char temp[131], *coding_history;
254  int ret, x;
255  uint64_t time_reference;
256  int64_t umid_parts[8], umid_mask = 0;
257 
258  if ((ret = wav_parse_bext_string(s, "description", 256)) < 0 ||
259  (ret = wav_parse_bext_string(s, "originator", 32)) < 0 ||
260  (ret = wav_parse_bext_string(s, "originator_reference", 32)) < 0 ||
261  (ret = wav_parse_bext_string(s, "origination_date", 10)) < 0 ||
262  (ret = wav_parse_bext_string(s, "origination_time", 8)) < 0)
263  return ret;
264 
265  time_reference = avio_rl64(s->pb);
266  snprintf(temp, sizeof(temp), "%"PRIu64, time_reference);
267  if ((ret = av_dict_set(&s->metadata, "time_reference", temp, 0)) < 0)
268  return ret;
269 
270  /* check if version is >= 1, in which case an UMID may be present */
271  if (avio_rl16(s->pb) >= 1) {
272  for (x = 0; x < 8; x++)
273  umid_mask |= umid_parts[x] = avio_rb64(s->pb);
274 
275  if (umid_mask) {
276  /* the string formatting below is per SMPTE 330M-2004 Annex C */
277  if (umid_parts[4] == 0 && umid_parts[5] == 0 &&
278  umid_parts[6] == 0 && umid_parts[7] == 0) {
279  /* basic UMID */
280  snprintf(temp, sizeof(temp),
281  "0x%016"PRIX64"%016"PRIX64"%016"PRIX64"%016"PRIX64,
282  umid_parts[0], umid_parts[1],
283  umid_parts[2], umid_parts[3]);
284  } else {
285  /* extended UMID */
286  snprintf(temp, sizeof(temp),
287  "0x%016"PRIX64"%016"PRIX64"%016"PRIX64"%016"PRIX64
288  "%016"PRIX64"%016"PRIX64"%016"PRIX64"%016"PRIX64,
289  umid_parts[0], umid_parts[1],
290  umid_parts[2], umid_parts[3],
291  umid_parts[4], umid_parts[5],
292  umid_parts[6], umid_parts[7]);
293  }
294 
295  if ((ret = av_dict_set(&s->metadata, "umid", temp, 0)) < 0)
296  return ret;
297  }
298 
299  avio_skip(s->pb, 190);
300  } else
301  avio_skip(s->pb, 254);
302 
303  if (size > 602) {
304  /* CodingHistory present */
305  size -= 602;
306 
307  if (!(coding_history = av_malloc(size + 1)))
308  return AVERROR(ENOMEM);
309 
310  if ((ret = avio_read(s->pb, coding_history, size)) != size) {
311  av_free(coding_history);
312  return ret < 0 ? ret : AVERROR_INVALIDDATA;
313  }
314 
315  coding_history[size] = 0;
316  if ((ret = av_dict_set(&s->metadata, "coding_history", coding_history,
318  return ret;
319  }
320 
321  return 0;
322 }
323 
324 static const AVMetadataConv wav_metadata_conv[] = {
325  { "description", "comment" },
326  { "originator", "encoded_by" },
327  { "origination_date", "date" },
328  { "origination_time", "creation_time" },
329  { 0 },
330 };
331 
332 /* wav input */
333 static int wav_read_header(AVFormatContext *s)
334 {
335  int64_t size, av_uninit(data_size);
336  int64_t sample_count = 0;
337  int rf64 = 0, bw64 = 0;
338  uint32_t tag;
339  AVIOContext *pb = s->pb;
340  AVStream *st = NULL;
341  WAVDemuxContext *wav = s->priv_data;
342  int ret, got_fmt = 0, got_xma2 = 0;
343  int64_t next_tag_ofs, data_ofs = -1;
344 
345  wav->unaligned = avio_tell(s->pb) & 1;
346 
347  wav->smv_data_ofs = -1;
348 
349  /* read chunk ID */
350  tag = avio_rl32(pb);
351  switch (tag) {
352  case MKTAG('R', 'I', 'F', 'F'):
353  break;
354  case MKTAG('R', 'I', 'F', 'X'):
355  wav->rifx = 1;
356  break;
357  case MKTAG('R', 'F', '6', '4'):
358  rf64 = 1;
359  break;
360  case MKTAG('B', 'W', '6', '4'):
361  bw64 = 1;
362  break;
363  default:
364  av_log(s, AV_LOG_ERROR, "invalid start code %s in RIFF header\n",
365  av_fourcc2str(tag));
366  return AVERROR_INVALIDDATA;
367  }
368 
369  /* read chunk size */
370  avio_rl32(pb);
371 
372  /* read format */
373  if (avio_rl32(pb) != MKTAG('W', 'A', 'V', 'E')) {
374  av_log(s, AV_LOG_ERROR, "invalid format in RIFF header\n");
375  return AVERROR_INVALIDDATA;
376  }
377 
378  if (rf64 || bw64) {
379  if (avio_rl32(pb) != MKTAG('d', 's', '6', '4'))
380  return AVERROR_INVALIDDATA;
381  size = avio_rl32(pb);
382  if (size < 24)
383  return AVERROR_INVALIDDATA;
384  avio_rl64(pb); /* RIFF size */
385 
386  data_size = avio_rl64(pb);
387  sample_count = avio_rl64(pb);
388 
389  if (data_size < 0 || sample_count < 0) {
390  av_log(s, AV_LOG_ERROR, "negative data_size and/or sample_count in "
391  "ds64: data_size = %"PRId64", sample_count = %"PRId64"\n",
392  data_size, sample_count);
393  return AVERROR_INVALIDDATA;
394  }
395  avio_skip(pb, size - 24); /* skip rest of ds64 chunk */
396 
397  }
398 
399  for (;;) {
400  AVStream *vst;
401  size = next_tag(pb, &tag, wav->rifx);
402  next_tag_ofs = avio_tell(pb) + size;
403 
404  if (avio_feof(pb))
405  break;
406 
407  switch (tag) {
408  case MKTAG('f', 'm', 't', ' '):
409  /* only parse the first 'fmt ' tag found */
410  if (!got_xma2 && !got_fmt && (ret = wav_parse_fmt_tag(s, size, &st)) < 0) {
411  return ret;
412  } else if (got_fmt)
413  av_log(s, AV_LOG_WARNING, "found more than one 'fmt ' tag\n");
414 
415  got_fmt = 1;
416  break;
417  case MKTAG('X', 'M', 'A', '2'):
418  /* only parse the first 'XMA2' tag found */
419  if (!got_fmt && !got_xma2 && (ret = wav_parse_xma2_tag(s, size, &st)) < 0) {
420  return ret;
421  } else if (got_xma2)
422  av_log(s, AV_LOG_WARNING, "found more than one 'XMA2' tag\n");
423 
424  got_xma2 = 1;
425  break;
426  case MKTAG('d', 'a', 't', 'a'):
427  if (!(pb->seekable & AVIO_SEEKABLE_NORMAL) && !got_fmt && !got_xma2) {
428  av_log(s, AV_LOG_ERROR,
429  "found no 'fmt ' tag before the 'data' tag\n");
430  return AVERROR_INVALIDDATA;
431  }
432 
433  if (rf64 || bw64) {
434  next_tag_ofs = wav->data_end = avio_tell(pb) + data_size;
435  } else if (size != 0xFFFFFFFF) {
436  data_size = size;
437  next_tag_ofs = wav->data_end = size ? next_tag_ofs : INT64_MAX;
438  } else {
439  av_log(s, AV_LOG_WARNING, "Ignoring maximum wav data size, "
440  "file may be invalid\n");
441  data_size = 0;
442  next_tag_ofs = wav->data_end = INT64_MAX;
443  }
444 
445  data_ofs = avio_tell(pb);
446 
447  /* don't look for footer metadata if we can't seek or if we don't
448  * know where the data tag ends
449  */
450  if (!(pb->seekable & AVIO_SEEKABLE_NORMAL) || (!(rf64 && !bw64) && !size))
451  goto break_loop;
452  break;
453  case MKTAG('f', 'a', 'c', 't'):
454  if (!sample_count)
455  sample_count = (!wav->rifx ? avio_rl32(pb) : avio_rb32(pb));
456  break;
457  case MKTAG('b', 'e', 'x', 't'):
458  if ((ret = wav_parse_bext_tag(s, size)) < 0)
459  return ret;
460  break;
461  case MKTAG('S','M','V','0'):
462  if (!got_fmt) {
463  av_log(s, AV_LOG_ERROR, "found no 'fmt ' tag before the 'SMV0' tag\n");
464  return AVERROR_INVALIDDATA;
465  }
466  // SMV file, a wav file with video appended.
467  if (size != MKTAG('0','2','0','0')) {
468  av_log(s, AV_LOG_ERROR, "Unknown SMV version found\n");
469  goto break_loop;
470  }
471  av_log(s, AV_LOG_DEBUG, "Found SMV data\n");
472  wav->smv_given_first = 0;
473  vst = avformat_new_stream(s, NULL);
474  if (!vst)
475  return AVERROR(ENOMEM);
476  avio_r8(pb);
477  vst->id = 1;
480  vst->codecpar->width = avio_rl24(pb);
481  vst->codecpar->height = avio_rl24(pb);
482  if ((ret = ff_alloc_extradata(vst->codecpar, 4)) < 0) {
483  av_log(s, AV_LOG_ERROR, "Could not allocate extradata.\n");
484  return ret;
485  }
486  size = avio_rl24(pb);
487  wav->smv_data_ofs = avio_tell(pb) + (size - 5) * 3;
488  avio_rl24(pb);
489  wav->smv_block_size = avio_rl24(pb);
490  avpriv_set_pts_info(vst, 32, 1, avio_rl24(pb));
491  vst->duration = avio_rl24(pb);
492  avio_rl24(pb);
493  avio_rl24(pb);
494  wav->smv_frames_per_jpeg = avio_rl24(pb);
495  if (wav->smv_frames_per_jpeg > 65536) {
496  av_log(s, AV_LOG_ERROR, "too many frames per jpeg\n");
497  return AVERROR_INVALIDDATA;
498  }
500  wav->smv_cur_pt = 0;
501  goto break_loop;
502  case MKTAG('L', 'I', 'S', 'T'):
503  case MKTAG('l', 'i', 's', 't'):
504  if (size < 4) {
505  av_log(s, AV_LOG_ERROR, "too short LIST tag\n");
506  return AVERROR_INVALIDDATA;
507  }
508  switch (avio_rl32(pb)) {
509  case MKTAG('I', 'N', 'F', 'O'):
510  ff_read_riff_info(s, size - 4);
511  break;
512  case MKTAG('a', 'd', 't', 'l'):
513  if (s->nb_chapters > 0) {
514  while (avio_tell(pb) < next_tag_ofs &&
515  !avio_feof(pb)) {
516  char cue_label[512];
517  unsigned id, sub_size;
518 
519  if (avio_rl32(pb) != MKTAG('l', 'a', 'b', 'l'))
520  break;
521 
522  sub_size = avio_rl32(pb);
523  if (sub_size < 5)
524  break;
525  id = avio_rl32(pb);
526  avio_get_str(pb, sub_size - 4, cue_label, sizeof(cue_label));
527  avio_skip(pb, avio_tell(pb) & 1);
528 
529  for (int i = 0; i < s->nb_chapters; i++) {
530  if (s->chapters[i]->id == id) {
531  av_dict_set(&s->chapters[i]->metadata, "title", cue_label, 0);
532  break;
533  }
534  }
535  }
536  }
537  break;
538  }
539  break;
540  case MKTAG('I', 'D', '3', ' '):
541  case MKTAG('i', 'd', '3', ' '): {
542  ID3v2ExtraMeta *id3v2_extra_meta = NULL;
543  ff_id3v2_read_dict(pb, &s->internal->id3v2_meta, ID3v2_DEFAULT_MAGIC, &id3v2_extra_meta);
544  if (id3v2_extra_meta) {
545  ff_id3v2_parse_apic(s, id3v2_extra_meta);
546  ff_id3v2_parse_chapters(s, id3v2_extra_meta);
547  ff_id3v2_parse_priv(s, id3v2_extra_meta);
548  }
549  ff_id3v2_free_extra_meta(&id3v2_extra_meta);
550  }
551  break;
552  case MKTAG('c', 'u', 'e', ' '):
553  if (size >= 4 && got_fmt && st->codecpar->sample_rate > 0) {
554  AVRational tb = {1, st->codecpar->sample_rate};
555  unsigned nb_cues = avio_rl32(pb);
556 
557  if (size >= nb_cues * 24LL + 4LL) {
558  for (int i = 0; i < nb_cues; i++) {
559  unsigned offset, id = avio_rl32(pb);
560 
561  avio_skip(pb, 16);
562  offset = avio_rl32(pb);
563 
564  if (!avpriv_new_chapter(s, id, tb, offset, AV_NOPTS_VALUE, NULL))
565  return AVERROR(ENOMEM);
566  }
567  }
568  }
569  break;
570  }
571 
572  /* seek to next tag unless we know that we'll run into EOF */
573  if ((avio_size(pb) > 0 && next_tag_ofs >= avio_size(pb)) ||
574  wav_seek_tag(wav, pb, next_tag_ofs, SEEK_SET) < 0) {
575  break;
576  }
577  }
578 
579 break_loop:
580  if (!got_fmt && !got_xma2) {
581  av_log(s, AV_LOG_ERROR, "no 'fmt ' or 'XMA2' tag found\n");
582  return AVERROR_INVALIDDATA;
583  }
584 
585  if (data_ofs < 0) {
586  av_log(s, AV_LOG_ERROR, "no 'data' tag found\n");
587  return AVERROR_INVALIDDATA;
588  }
589 
590  avio_seek(pb, data_ofs, SEEK_SET);
591 
592  if (data_size > (INT64_MAX>>3)) {
593  av_log(s, AV_LOG_WARNING, "Data size %"PRId64" is too large\n", data_size);
594  data_size = 0;
595  }
596 
597  if ( st->codecpar->bit_rate > 0 && data_size > 0
598  && st->codecpar->sample_rate > 0
599  && sample_count > 0 && st->codecpar->channels > 1
600  && sample_count % st->codecpar->channels == 0) {
601  if (fabs(8.0 * data_size * st->codecpar->channels * st->codecpar->sample_rate /
602  sample_count /st->codecpar->bit_rate - 1.0) < 0.3)
603  sample_count /= st->codecpar->channels;
604  }
605 
606  if ( data_size > 0 && sample_count && st->codecpar->channels
607  && (data_size << 3) / sample_count / st->codecpar->channels > st->codecpar->bits_per_coded_sample + 1) {
608  av_log(s, AV_LOG_WARNING, "ignoring wrong sample_count %"PRId64"\n", sample_count);
609  sample_count = 0;
610  }
611 
612  /* G.729 hack (for Ticket4577)
613  * FIXME: Come up with cleaner, more general solution */
614  if (st->codecpar->codec_id == AV_CODEC_ID_G729 && sample_count && (data_size << 3) > sample_count) {
615  av_log(s, AV_LOG_WARNING, "ignoring wrong sample_count %"PRId64"\n", sample_count);
616  sample_count = 0;
617  }
618 
619  if (!sample_count || av_get_exact_bits_per_sample(st->codecpar->codec_id) > 0)
620  if ( st->codecpar->channels
621  && data_size
623  && wav->data_end <= avio_size(pb))
624  sample_count = (data_size << 3)
625  /
626  (st->codecpar->channels * (uint64_t)av_get_bits_per_sample(st->codecpar->codec_id));
627 
628  if (sample_count)
629  st->duration = sample_count;
630 
632  st->codecpar->block_align == st->codecpar->channels * 4 &&
633  st->codecpar->bits_per_coded_sample == 32 &&
634  st->codecpar->extradata_size == 2 &&
635  AV_RL16(st->codecpar->extradata) == 1) {
638  } else if (st->codecpar->codec_id == AV_CODEC_ID_PCM_S24LE &&
639  st->codecpar->block_align == st->codecpar->channels * 4 &&
640  st->codecpar->bits_per_coded_sample == 24) {
642  } else if (st->codecpar->codec_id == AV_CODEC_ID_XMA1 ||
644  st->codecpar->block_align = 2048;
645  } else if (st->codecpar->codec_id == AV_CODEC_ID_ADPCM_MS && st->codecpar->channels > 2) {
646  st->codecpar->block_align *= st->codecpar->channels;
647  }
648 
649  ff_metadata_conv_ctx(s, NULL, wav_metadata_conv);
651 
652  set_spdif(s, wav);
653 
654  return 0;
655 }
656 
657 /**
658  * Find chunk with w64 GUID by skipping over other chunks.
659  * @return the size of the found chunk
660  */
661 static int64_t find_guid(AVIOContext *pb, const uint8_t guid1[16])
662 {
663  uint8_t guid[16];
664  int64_t size;
665 
666  while (!avio_feof(pb)) {
667  avio_read(pb, guid, 16);
668  size = avio_rl64(pb);
669  if (size <= 24)
670  return AVERROR_INVALIDDATA;
671  if (!memcmp(guid, guid1, 16))
672  return size;
673  avio_skip(pb, FFALIGN(size, INT64_C(8)) - 24);
674  }
675  return AVERROR_EOF;
676 }
677 
678 static int wav_read_packet(AVFormatContext *s, AVPacket *pkt)
679 {
680  int ret, size;
681  int64_t left;
682  AVStream *st;
683  WAVDemuxContext *wav = s->priv_data;
684 
685  if (CONFIG_SPDIF_DEMUXER && wav->spdif == 1)
686  return ff_spdif_read_packet(s, pkt);
687 
688  if (wav->smv_data_ofs > 0) {
689  int64_t audio_dts, video_dts;
690 smv_retry:
691  audio_dts = (int32_t)s->streams[0]->cur_dts;
692  video_dts = (int32_t)s->streams[1]->cur_dts;
693 
694  if (audio_dts != AV_NOPTS_VALUE && video_dts != AV_NOPTS_VALUE) {
695  /*We always return a video frame first to get the pixel format first*/
696  wav->smv_last_stream = wav->smv_given_first ?
697  av_compare_ts(video_dts, s->streams[1]->time_base,
698  audio_dts, s->streams[0]->time_base) > 0 : 0;
699  wav->smv_given_first = 1;
700  }
701  wav->smv_last_stream = !wav->smv_last_stream;
702  wav->smv_last_stream |= wav->audio_eof;
703  wav->smv_last_stream &= !wav->smv_eof;
704  if (wav->smv_last_stream) {
705  uint64_t old_pos = avio_tell(s->pb);
706  uint64_t new_pos = wav->smv_data_ofs +
707  wav->smv_block * wav->smv_block_size;
708  if (avio_seek(s->pb, new_pos, SEEK_SET) < 0) {
709  ret = AVERROR_EOF;
710  goto smv_out;
711  }
712  size = avio_rl24(s->pb);
713  ret = av_get_packet(s->pb, pkt, size);
714  if (ret < 0)
715  goto smv_out;
716  pkt->pos -= 3;
717  pkt->pts = wav->smv_block * wav->smv_frames_per_jpeg + wav->smv_cur_pt;
718  wav->smv_cur_pt++;
719  if (wav->smv_frames_per_jpeg > 0)
720  wav->smv_cur_pt %= wav->smv_frames_per_jpeg;
721  if (!wav->smv_cur_pt)
722  wav->smv_block++;
723 
724  pkt->stream_index = 1;
725 smv_out:
726  avio_seek(s->pb, old_pos, SEEK_SET);
727  if (ret == AVERROR_EOF) {
728  wav->smv_eof = 1;
729  goto smv_retry;
730  }
731  return ret;
732  }
733  }
734 
735  st = s->streams[0];
736 
737  left = wav->data_end - avio_tell(s->pb);
738  if (wav->ignore_length)
739  left = INT_MAX;
740  if (left <= 0) {
741  if (CONFIG_W64_DEMUXER && wav->w64)
742  left = find_guid(s->pb, ff_w64_guid_data) - 24;
743  else
744  left = find_tag(wav, s->pb, MKTAG('d', 'a', 't', 'a'));
745  if (left < 0) {
746  wav->audio_eof = 1;
747  if (wav->smv_data_ofs > 0 && !wav->smv_eof)
748  goto smv_retry;
749  return AVERROR_EOF;
750  }
751  wav->data_end = avio_tell(s->pb) + left;
752  }
753 
754  size = wav->max_size;
755  if (st->codecpar->block_align > 1) {
756  if (size < st->codecpar->block_align)
757  size = st->codecpar->block_align;
758  size = (size / st->codecpar->block_align) * st->codecpar->block_align;
759  }
760  size = FFMIN(size, left);
761  ret = av_get_packet(s->pb, pkt, size);
762  if (ret < 0)
763  return ret;
764  pkt->stream_index = 0;
765 
766  return ret;
767 }
768 
769 static int wav_read_seek(AVFormatContext *s,
770  int stream_index, int64_t timestamp, int flags)
771 {
772  WAVDemuxContext *wav = s->priv_data;
773  AVStream *st;
774  wav->smv_eof = 0;
775  wav->audio_eof = 0;
776  if (wav->smv_data_ofs > 0) {
777  int64_t smv_timestamp = timestamp;
778  if (stream_index == 0)
779  smv_timestamp = av_rescale_q(timestamp, s->streams[0]->time_base, s->streams[1]->time_base);
780  else
781  timestamp = av_rescale_q(smv_timestamp, s->streams[1]->time_base, s->streams[0]->time_base);
782  if (wav->smv_frames_per_jpeg > 0) {
783  wav->smv_block = smv_timestamp / wav->smv_frames_per_jpeg;
784  wav->smv_cur_pt = smv_timestamp % wav->smv_frames_per_jpeg;
785  }
786  }
787 
788  st = s->streams[0];
789  switch (st->codecpar->codec_id) {
790  case AV_CODEC_ID_MP2:
791  case AV_CODEC_ID_MP3:
792  case AV_CODEC_ID_AC3:
793  case AV_CODEC_ID_DTS:
794  case AV_CODEC_ID_XMA2:
795  /* use generic seeking with dynamically generated indexes */
796  return -1;
797  default:
798  break;
799  }
800  return ff_pcm_read_seek(s, stream_index, timestamp, flags);
801 }
802 
803 #define OFFSET(x) offsetof(WAVDemuxContext, x)
804 #define DEC AV_OPT_FLAG_DECODING_PARAM
805 static const AVOption demux_options[] = {
806  { "ignore_length", "Ignore length", OFFSET(ignore_length), AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, DEC },
807  { "max_size", "max size of single packet", OFFSET(max_size), AV_OPT_TYPE_INT, { .i64 = 4096 }, 1024, 1 << 22, DEC },
808  { NULL },
809 };
810 
811 static const AVClass wav_demuxer_class = {
812  .class_name = "WAV demuxer",
813  .item_name = av_default_item_name,
814  .option = demux_options,
815  .version = LIBAVUTIL_VERSION_INT,
816 };
818  .name = "wav",
819  .long_name = NULL_IF_CONFIG_SMALL("WAV / WAVE (Waveform Audio)"),
820  .priv_data_size = sizeof(WAVDemuxContext),
821  .read_probe = wav_probe,
822  .read_header = wav_read_header,
823  .read_packet = wav_read_packet,
824  .read_seek = wav_read_seek,
826  .codec_tag = (const AVCodecTag * const []) { ff_codec_wav_tags, 0 },
827  .priv_class = &wav_demuxer_class,
828 };
829 #endif /* CONFIG_WAV_DEMUXER */
830 
831 #if CONFIG_W64_DEMUXER
832 static int w64_probe(const AVProbeData *p)
833 {
834  if (p->buf_size <= 40)
835  return 0;
836  if (!memcmp(p->buf, ff_w64_guid_riff, 16) &&
837  !memcmp(p->buf + 24, ff_w64_guid_wave, 16))
838  return AVPROBE_SCORE_MAX;
839  else
840  return 0;
841 }
842 
843 static int w64_read_header(AVFormatContext *s)
844 {
845  int64_t size, data_ofs = 0;
846  AVIOContext *pb = s->pb;
847  WAVDemuxContext *wav = s->priv_data;
848  AVStream *st;
849  uint8_t guid[16];
850  int ret;
851 
852  avio_read(pb, guid, 16);
853  if (memcmp(guid, ff_w64_guid_riff, 16))
854  return AVERROR_INVALIDDATA;
855 
856  /* riff + wave + fmt + sizes */
857  if (avio_rl64(pb) < 16 + 8 + 16 + 8 + 16 + 8)
858  return AVERROR_INVALIDDATA;
859 
860  avio_read(pb, guid, 16);
861  if (memcmp(guid, ff_w64_guid_wave, 16)) {
862  av_log(s, AV_LOG_ERROR, "could not find wave guid\n");
863  return AVERROR_INVALIDDATA;
864  }
865 
866  wav->w64 = 1;
867 
868  st = avformat_new_stream(s, NULL);
869  if (!st)
870  return AVERROR(ENOMEM);
871 
872  while (!avio_feof(pb)) {
873  if (avio_read(pb, guid, 16) != 16)
874  break;
875  size = avio_rl64(pb);
876  if (size <= 24 || INT64_MAX - size < avio_tell(pb))
877  return AVERROR_INVALIDDATA;
878 
879  if (!memcmp(guid, ff_w64_guid_fmt, 16)) {
880  /* subtract chunk header size - normal wav file doesn't count it */
881  ret = ff_get_wav_header(s, pb, st->codecpar, size - 24, 0);
882  if (ret < 0)
883  return ret;
884  avio_skip(pb, FFALIGN(size, INT64_C(8)) - size);
885 
886  avpriv_set_pts_info(st, 64, 1, st->codecpar->sample_rate);
887  } else if (!memcmp(guid, ff_w64_guid_fact, 16)) {
888  int64_t samples;
889 
890  samples = avio_rl64(pb);
891  if (samples > 0)
892  st->duration = samples;
893  avio_skip(pb, FFALIGN(size, INT64_C(8)) - 32);
894  } else if (!memcmp(guid, ff_w64_guid_data, 16)) {
895  wav->data_end = avio_tell(pb) + size - 24;
896 
897  data_ofs = avio_tell(pb);
898  if (!(pb->seekable & AVIO_SEEKABLE_NORMAL))
899  break;
900 
901  avio_skip(pb, size - 24);
902  } else if (!memcmp(guid, ff_w64_guid_summarylist, 16)) {
903  int64_t start, end, cur;
904  uint32_t count, chunk_size, i;
905 
906  start = avio_tell(pb);
907  end = start + FFALIGN(size, INT64_C(8)) - 24;
908  count = avio_rl32(pb);
909 
910  for (i = 0; i < count; i++) {
911  char chunk_key[5], *value;
912 
913  if (avio_feof(pb) || (cur = avio_tell(pb)) < 0 || cur > end - 8 /* = tag + size */)
914  break;
915 
916  chunk_key[4] = 0;
917  avio_read(pb, chunk_key, 4);
918  chunk_size = avio_rl32(pb);
919  if (chunk_size == UINT32_MAX)
920  return AVERROR_INVALIDDATA;
921 
922  value = av_mallocz(chunk_size + 1);
923  if (!value)
924  return AVERROR(ENOMEM);
925 
926  ret = avio_get_str16le(pb, chunk_size, value, chunk_size);
927  avio_skip(pb, chunk_size - ret);
928 
929  av_dict_set(&s->metadata, chunk_key, value, AV_DICT_DONT_STRDUP_VAL);
930  }
931 
932  avio_skip(pb, end - avio_tell(pb));
933  } else {
934  av_log(s, AV_LOG_DEBUG, "unknown guid: "FF_PRI_GUID"\n", FF_ARG_GUID(guid));
935  avio_skip(pb, FFALIGN(size, INT64_C(8)) - 24);
936  }
937  }
938 
939  if (!data_ofs)
940  return AVERROR_EOF;
941 
942  ff_metadata_conv_ctx(s, NULL, wav_metadata_conv);
944 
945  handle_stream_probing(st);
947 
948  avio_seek(pb, data_ofs, SEEK_SET);
949 
950  set_spdif(s, wav);
951 
952  return 0;
953 }
954 
955 #define OFFSET(x) offsetof(WAVDemuxContext, x)
956 #define DEC AV_OPT_FLAG_DECODING_PARAM
957 static const AVOption w64_demux_options[] = {
958  { "max_size", "max size of single packet", OFFSET(max_size), AV_OPT_TYPE_INT, { .i64 = 4096 }, 1024, 1 << 22, DEC },
959  { NULL }
960 };
961 
962 static const AVClass w64_demuxer_class = {
963  .class_name = "W64 demuxer",
964  .item_name = av_default_item_name,
965  .option = w64_demux_options,
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 = (const AVCodecTag * const []) { ff_codec_wav_tags, 0 },
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:1580
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:1261
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:317
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:105
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:4869
int probe_packets
Number of packets to buffer for codec probing.
Definition: avformat.h:1086
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:1804
int smv_frames_per_jpeg
Definition: wavdec.c:53
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:67
const char * key
void * av_mallocz(size_t size)
Allocate a memory block with alignment suitable for all memory accesses (including vectors if availab...
Definition: mem.c:237
int max_size
Definition: wavdec.c:59
static AVPacket pkt
AVDictionary * metadata
Definition: avformat.h:1312
uint64_t_TMPL AV_WL64 unsigned int_TMPL AV_WL32 unsigned int_TMPL AV_WL24 unsigned int_TMPL AV_RL16
Definition: bytestream.h:87
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:58
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:4578
Format I/O context.
Definition: avformat.h:1351
static const GUIDParseTable * find_guid(ff_asf_guid guid)
Definition: asfdec_o.c:1658
int64_t audio_dts
Definition: movenc.c:60
int64_t cur_dts
Definition: avformat.h:1079
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:790
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
static av_cold int end(AVCodecContext *avctx)
Definition: avrndec.c:92
int id
unique ID to identify the chapter
Definition: avformat.h:1309
enum AVStreamParseType need_parsing
Definition: avformat.h:1094
int id
Format-specific stream ID.
Definition: avformat.h:883
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:4450
AVStream ** streams
A list of all streams in the file.
Definition: avformat.h:1419
int ff_id3v2_parse_chapters(AVFormatContext *s, ID3v2ExtraMeta *extra_meta)
Create chapters for all CHAP tags found in the ID3v2 header.
Definition: id3v2.c:1170
uint32_t tag
Definition: movenc.c:1532
#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:305
ptrdiff_t size
Definition: opengl_enc.c:100
uint64_t avio_rb64(AVIOContext *s)
Definition: aviobuf.c:911
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:799
#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:1566
AVDictionary * metadata
Metadata that applies to the whole file.
Definition: avformat.h:1591
unsigned int avio_rl32(AVIOContext *s)
Definition: aviobuf.c:759
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
Definition: internal.h:153
unsigned int pos
Definition: spdifenc.c:410
#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:415
#define av_fourcc2str(fourcc)
Definition: avutil.h:348
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 local see the OFFSET() macro
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:1114
AVChapter ** chapters
Definition: avformat.h:1581
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:52
int smv_cur_pt
Definition: wavdec.c:61
GLsizei count
Definition: opengl_enc.c:108
const AVCodecTag ff_codec_wav_tags[]
Definition: riff.c:506
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
#define DEC
Definition: librsvgdec.c:105
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:3259
#define FFMIN(a, b)
Definition: common.h:96
int smv_last_stream
Definition: wavdec.c:55
int av_get_exact_bits_per_sample(enum AVCodecID codec_id)
Return codec bits per sample.
Definition: utils.c:1480
#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:57
int ff_pcm_read_seek(AVFormatContext *s, int stream_index, int64_t timestamp, int flags)
Definition: pcm.c:52
static int read_header(FFV1Context *f)
Definition: ffv1dec.c:527
Stream structure.
Definition: avformat.h:876
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
AVIOContext * pb
I/O context.
Definition: avformat.h:1393
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:1130
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
int64_t smv_data_ofs
Definition: wavdec.c:51
#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:591
#define flags(name, subs,...)
Definition: cbs_av1.c:560
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:994
int64_t duration
Decoding: duration of the stream, in stream time base.
Definition: avformat.h:925
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:743
Main libavformat public API header.
#define FF_PRI_GUID
Definition: riff.h:101
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:3277
int64_t data_end
Definition: wavdec.c:49
#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:1102
int smv_block
Definition: wavdec.c:54
int len
void * priv_data
Format private data.
Definition: avformat.h:1379
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:650
AVCodecParameters * codecpar
Codec parameters associated with this stream.
Definition: avformat.h:1023
int avio_get_str(AVIOContext *pb, int maxlen, char *buf, int buflen)
Read a string from pb into buf.
Definition: aviobuf.c:869
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:751
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:905
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:406
int request_probe
stream probing state -1 -> probing finished 0 -> no probing requested rest -> perform probing with re...
Definition: avformat.h:1133
enum AVCodecID id
This structure stores compressed data.
Definition: packet.h:340
uint64_t avio_rl64(AVIOContext *s)
Definition: aviobuf.c:767
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