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;
62  int unaligned; // e.g. if an odd number of bytes ID3 tag was prepended
63  int rifx; // RIFX: integer byte order for parameters is big endian
65 
67 {
68  if (CONFIG_SPDIF_DEMUXER && s->streams[0]->codecpar->codec_tag == 1) {
69  enum AVCodecID codec;
70  int len = 1<<16;
71  int ret = ffio_ensure_seekback(s->pb, len);
72 
73  if (ret >= 0) {
74  uint8_t *buf = av_malloc(len);
75  if (!buf) {
76  ret = AVERROR(ENOMEM);
77  } else {
78  int64_t pos = avio_tell(s->pb);
79  len = ret = avio_read(s->pb, buf, len);
80  if (len >= 0) {
81  ret = ff_spdif_probe(buf, len, &codec);
82  if (ret > AVPROBE_SCORE_EXTENSION) {
83  s->streams[0]->codecpar->codec_id = codec;
84  wav->spdif = 1;
85  }
86  }
87  avio_seek(s->pb, pos, SEEK_SET);
88  av_free(buf);
89  }
90  }
91 
92  if (ret < 0)
93  av_log(s, AV_LOG_WARNING, "Cannot check for SPDIF\n");
94  }
95 }
96 
97 #if CONFIG_WAV_DEMUXER
98 
99 static int64_t next_tag(AVIOContext *pb, uint32_t *tag, int big_endian)
100 {
101  *tag = avio_rl32(pb);
102  if (!big_endian) {
103  return avio_rl32(pb);
104  } else {
105  return avio_rb32(pb);
106  }
107 }
108 
109 /* RIFF chunks are always at even offsets relative to where they start. */
110 static int64_t wav_seek_tag(WAVDemuxContext * wav, AVIOContext *s, int64_t offset, int whence)
111 {
112  offset += offset < INT64_MAX && offset + wav->unaligned & 1;
113 
114  return avio_seek(s, offset, whence);
115 }
116 
117 /* return the size of the found tag */
118 static int64_t find_tag(WAVDemuxContext * wav, AVIOContext *pb, uint32_t tag1)
119 {
120  unsigned int tag;
121  int64_t size;
122 
123  for (;;) {
124  if (avio_feof(pb))
125  return AVERROR_EOF;
126  size = next_tag(pb, &tag, wav->rifx);
127  if (tag == tag1)
128  break;
129  wav_seek_tag(wav, pb, size, SEEK_CUR);
130  }
131  return size;
132 }
133 
134 static int wav_probe(const AVProbeData *p)
135 {
136  /* check file header */
137  if (p->buf_size <= 32)
138  return 0;
139  if (!memcmp(p->buf + 8, "WAVE", 4)) {
140  if (!memcmp(p->buf, "RIFF", 4) || !memcmp(p->buf, "RIFX", 4))
141  /* Since the ACT demuxer has a standard WAV header at the top of
142  * its own, the returned score is decreased to avoid a probe
143  * conflict between ACT and WAV. */
144  return AVPROBE_SCORE_MAX - 1;
145  else if ((!memcmp(p->buf, "RF64", 4) ||
146  !memcmp(p->buf, "BW64", 4)) &&
147  !memcmp(p->buf + 12, "ds64", 4))
148  return AVPROBE_SCORE_MAX;
149  }
150  return 0;
151 }
152 
153 static void handle_stream_probing(AVStream *st)
154 {
157  st->probe_packets = FFMIN(st->probe_packets, 32);
158  }
159 }
160 
161 static int wav_parse_fmt_tag(AVFormatContext *s, int64_t size, AVStream **st)
162 {
163  AVIOContext *pb = s->pb;
164  WAVDemuxContext *wav = s->priv_data;
165  int ret;
166 
167  /* parse fmt header */
168  *st = avformat_new_stream(s, NULL);
169  if (!*st)
170  return AVERROR(ENOMEM);
171 
172  ret = ff_get_wav_header(s, pb, (*st)->codecpar, size, wav->rifx);
173  if (ret < 0)
174  return ret;
175  handle_stream_probing(*st);
176 
177  (*st)->need_parsing = AVSTREAM_PARSE_FULL_RAW;
178 
179  avpriv_set_pts_info(*st, 64, 1, (*st)->codecpar->sample_rate);
180 
181  return 0;
182 }
183 
184 static int wav_parse_xma2_tag(AVFormatContext *s, int64_t size, AVStream **st)
185 {
186  AVIOContext *pb = s->pb;
187  int version, num_streams, i, channels = 0, ret;
188 
189  if (size < 36)
190  return AVERROR_INVALIDDATA;
191 
192  *st = avformat_new_stream(s, NULL);
193  if (!*st)
194  return AVERROR(ENOMEM);
195 
196  (*st)->codecpar->codec_type = AVMEDIA_TYPE_AUDIO;
197  (*st)->codecpar->codec_id = AV_CODEC_ID_XMA2;
198  (*st)->need_parsing = AVSTREAM_PARSE_FULL_RAW;
199 
200  version = avio_r8(pb);
201  if (version != 3 && version != 4)
202  return AVERROR_INVALIDDATA;
203  num_streams = avio_r8(pb);
204  if (size != (32 + ((version==3)?0:8) + 4*num_streams))
205  return AVERROR_INVALIDDATA;
206  avio_skip(pb, 10);
207  (*st)->codecpar->sample_rate = avio_rb32(pb);
208  if (version == 4)
209  avio_skip(pb, 8);
210  avio_skip(pb, 4);
211  (*st)->duration = avio_rb32(pb);
212  avio_skip(pb, 8);
213 
214  for (i = 0; i < num_streams; i++) {
215  channels += avio_r8(pb);
216  avio_skip(pb, 3);
217  }
218  (*st)->codecpar->channels = channels;
219 
220  if ((*st)->codecpar->channels <= 0 || (*st)->codecpar->sample_rate <= 0)
221  return AVERROR_INVALIDDATA;
222 
223  avpriv_set_pts_info(*st, 64, 1, (*st)->codecpar->sample_rate);
224 
225  avio_seek(pb, -size, SEEK_CUR);
226  if ((ret = ff_get_extradata(s, (*st)->codecpar, pb, size)) < 0)
227  return ret;
228 
229  return 0;
230 }
231 
232 static inline int wav_parse_bext_string(AVFormatContext *s, const char *key,
233  int length)
234 {
235  char temp[257];
236  int ret;
237 
238  av_assert0(length < sizeof(temp));
239  if ((ret = avio_read(s->pb, temp, length)) != length)
240  return ret < 0 ? ret : AVERROR_INVALIDDATA;
241 
242  temp[length] = 0;
243 
244  if (strlen(temp))
245  return av_dict_set(&s->metadata, key, temp, 0);
246 
247  return 0;
248 }
249 
250 static int wav_parse_bext_tag(AVFormatContext *s, int64_t size)
251 {
252  char temp[131], *coding_history;
253  int ret, x;
254  uint64_t time_reference;
255  int64_t umid_parts[8], umid_mask = 0;
256 
257  if ((ret = wav_parse_bext_string(s, "description", 256)) < 0 ||
258  (ret = wav_parse_bext_string(s, "originator", 32)) < 0 ||
259  (ret = wav_parse_bext_string(s, "originator_reference", 32)) < 0 ||
260  (ret = wav_parse_bext_string(s, "origination_date", 10)) < 0 ||
261  (ret = wav_parse_bext_string(s, "origination_time", 8)) < 0)
262  return ret;
263 
264  time_reference = avio_rl64(s->pb);
265  snprintf(temp, sizeof(temp), "%"PRIu64, time_reference);
266  if ((ret = av_dict_set(&s->metadata, "time_reference", temp, 0)) < 0)
267  return ret;
268 
269  /* check if version is >= 1, in which case an UMID may be present */
270  if (avio_rl16(s->pb) >= 1) {
271  for (x = 0; x < 8; x++)
272  umid_mask |= umid_parts[x] = avio_rb64(s->pb);
273 
274  if (umid_mask) {
275  /* the string formatting below is per SMPTE 330M-2004 Annex C */
276  if (umid_parts[4] == 0 && umid_parts[5] == 0 &&
277  umid_parts[6] == 0 && umid_parts[7] == 0) {
278  /* basic UMID */
279  snprintf(temp, sizeof(temp),
280  "0x%016"PRIX64"%016"PRIX64"%016"PRIX64"%016"PRIX64,
281  umid_parts[0], umid_parts[1],
282  umid_parts[2], umid_parts[3]);
283  } else {
284  /* extended UMID */
285  snprintf(temp, sizeof(temp),
286  "0x%016"PRIX64"%016"PRIX64"%016"PRIX64"%016"PRIX64
287  "%016"PRIX64"%016"PRIX64"%016"PRIX64"%016"PRIX64,
288  umid_parts[0], umid_parts[1],
289  umid_parts[2], umid_parts[3],
290  umid_parts[4], umid_parts[5],
291  umid_parts[6], umid_parts[7]);
292  }
293 
294  if ((ret = av_dict_set(&s->metadata, "umid", temp, 0)) < 0)
295  return ret;
296  }
297 
298  avio_skip(s->pb, 190);
299  } else
300  avio_skip(s->pb, 254);
301 
302  if (size > 602) {
303  /* CodingHistory present */
304  size -= 602;
305 
306  if (!(coding_history = av_malloc(size + 1)))
307  return AVERROR(ENOMEM);
308 
309  if ((ret = avio_read(s->pb, coding_history, size)) != size) {
310  av_free(coding_history);
311  return ret < 0 ? ret : AVERROR_INVALIDDATA;
312  }
313 
314  coding_history[size] = 0;
315  if ((ret = av_dict_set(&s->metadata, "coding_history", coding_history,
317  return ret;
318  }
319 
320  return 0;
321 }
322 
323 static const AVMetadataConv wav_metadata_conv[] = {
324  { "description", "comment" },
325  { "originator", "encoded_by" },
326  { "origination_date", "date" },
327  { "origination_time", "creation_time" },
328  { 0 },
329 };
330 
331 /* wav input */
332 static int wav_read_header(AVFormatContext *s)
333 {
334  int64_t size, av_uninit(data_size);
335  int64_t sample_count = 0;
336  int rf64 = 0, bw64 = 0;
337  uint32_t tag;
338  AVIOContext *pb = s->pb;
339  AVStream *st = NULL;
340  WAVDemuxContext *wav = s->priv_data;
341  int ret, got_fmt = 0, got_xma2 = 0;
342  int64_t next_tag_ofs, data_ofs = -1;
343 
344  wav->unaligned = avio_tell(s->pb) & 1;
345 
346  wav->smv_data_ofs = -1;
347 
348  /* read chunk ID */
349  tag = avio_rl32(pb);
350  switch (tag) {
351  case MKTAG('R', 'I', 'F', 'F'):
352  break;
353  case MKTAG('R', 'I', 'F', 'X'):
354  wav->rifx = 1;
355  break;
356  case MKTAG('R', 'F', '6', '4'):
357  rf64 = 1;
358  break;
359  case MKTAG('B', 'W', '6', '4'):
360  bw64 = 1;
361  break;
362  default:
363  av_log(s, AV_LOG_ERROR, "invalid start code %s in RIFF header\n",
364  av_fourcc2str(tag));
365  return AVERROR_INVALIDDATA;
366  }
367 
368  /* read chunk size */
369  avio_rl32(pb);
370 
371  /* read format */
372  if (avio_rl32(pb) != MKTAG('W', 'A', 'V', 'E')) {
373  av_log(s, AV_LOG_ERROR, "invalid format in RIFF header\n");
374  return AVERROR_INVALIDDATA;
375  }
376 
377  if (rf64 || bw64) {
378  if (avio_rl32(pb) != MKTAG('d', 's', '6', '4'))
379  return AVERROR_INVALIDDATA;
380  size = avio_rl32(pb);
381  if (size < 24)
382  return AVERROR_INVALIDDATA;
383  avio_rl64(pb); /* RIFF size */
384 
385  data_size = avio_rl64(pb);
386  sample_count = avio_rl64(pb);
387 
388  if (data_size < 0 || sample_count < 0) {
389  av_log(s, AV_LOG_ERROR, "negative data_size and/or sample_count in "
390  "ds64: data_size = %"PRId64", sample_count = %"PRId64"\n",
391  data_size, sample_count);
392  return AVERROR_INVALIDDATA;
393  }
394  avio_skip(pb, size - 24); /* skip rest of ds64 chunk */
395 
396  }
397 
398  for (;;) {
399  AVStream *vst;
400  size = next_tag(pb, &tag, wav->rifx);
401  next_tag_ofs = avio_tell(pb) + size;
402 
403  if (avio_feof(pb))
404  break;
405 
406  switch (tag) {
407  case MKTAG('f', 'm', 't', ' '):
408  /* only parse the first 'fmt ' tag found */
409  if (!got_xma2 && !got_fmt && (ret = wav_parse_fmt_tag(s, size, &st)) < 0) {
410  return ret;
411  } else if (got_fmt)
412  av_log(s, AV_LOG_WARNING, "found more than one 'fmt ' tag\n");
413 
414  got_fmt = 1;
415  break;
416  case MKTAG('X', 'M', 'A', '2'):
417  /* only parse the first 'XMA2' tag found */
418  if (!got_fmt && !got_xma2 && (ret = wav_parse_xma2_tag(s, size, &st)) < 0) {
419  return ret;
420  } else if (got_xma2)
421  av_log(s, AV_LOG_WARNING, "found more than one 'XMA2' tag\n");
422 
423  got_xma2 = 1;
424  break;
425  case MKTAG('d', 'a', 't', 'a'):
426  if (!(pb->seekable & AVIO_SEEKABLE_NORMAL) && !got_fmt && !got_xma2) {
427  av_log(s, AV_LOG_ERROR,
428  "found no 'fmt ' tag before the 'data' tag\n");
429  return AVERROR_INVALIDDATA;
430  }
431 
432  if (rf64 || bw64) {
433  next_tag_ofs = wav->data_end = avio_tell(pb) + data_size;
434  } else if (size != 0xFFFFFFFF) {
435  data_size = size;
436  next_tag_ofs = wav->data_end = size ? next_tag_ofs : INT64_MAX;
437  } else {
438  av_log(s, AV_LOG_WARNING, "Ignoring maximum wav data size, "
439  "file may be invalid\n");
440  data_size = 0;
441  next_tag_ofs = wav->data_end = INT64_MAX;
442  }
443 
444  data_ofs = avio_tell(pb);
445 
446  /* don't look for footer metadata if we can't seek or if we don't
447  * know where the data tag ends
448  */
449  if (!(pb->seekable & AVIO_SEEKABLE_NORMAL) || (!(rf64 && !bw64) && !size))
450  goto break_loop;
451  break;
452  case MKTAG('f', 'a', 'c', 't'):
453  if (!sample_count)
454  sample_count = (!wav->rifx ? avio_rl32(pb) : avio_rb32(pb));
455  break;
456  case MKTAG('b', 'e', 'x', 't'):
457  if ((ret = wav_parse_bext_tag(s, size)) < 0)
458  return ret;
459  break;
460  case MKTAG('S','M','V','0'):
461  if (!got_fmt) {
462  av_log(s, AV_LOG_ERROR, "found no 'fmt ' tag before the 'SMV0' tag\n");
463  return AVERROR_INVALIDDATA;
464  }
465  // SMV file, a wav file with video appended.
466  if (size != MKTAG('0','2','0','0')) {
467  av_log(s, AV_LOG_ERROR, "Unknown SMV version found\n");
468  goto break_loop;
469  }
470  av_log(s, AV_LOG_DEBUG, "Found SMV data\n");
471  wav->smv_given_first = 0;
472  vst = avformat_new_stream(s, NULL);
473  if (!vst)
474  return AVERROR(ENOMEM);
475  avio_r8(pb);
476  vst->id = 1;
479  vst->codecpar->width = avio_rl24(pb);
480  vst->codecpar->height = avio_rl24(pb);
481  if ((ret = ff_alloc_extradata(vst->codecpar, 4)) < 0) {
482  av_log(s, AV_LOG_ERROR, "Could not allocate extradata.\n");
483  return ret;
484  }
485  size = avio_rl24(pb);
486  wav->smv_data_ofs = avio_tell(pb) + (size - 5) * 3;
487  avio_rl24(pb);
488  wav->smv_block_size = avio_rl24(pb);
489  avpriv_set_pts_info(vst, 32, 1, avio_rl24(pb));
490  vst->duration = avio_rl24(pb);
491  avio_rl24(pb);
492  avio_rl24(pb);
493  wav->smv_frames_per_jpeg = avio_rl24(pb);
494  if (wav->smv_frames_per_jpeg > 65536) {
495  av_log(s, AV_LOG_ERROR, "too many frames per jpeg\n");
496  return AVERROR_INVALIDDATA;
497  }
499  goto break_loop;
500  case MKTAG('L', 'I', 'S', 'T'):
501  case MKTAG('l', 'i', 's', 't'):
502  if (size < 4) {
503  av_log(s, AV_LOG_ERROR, "too short LIST tag\n");
504  return AVERROR_INVALIDDATA;
505  }
506  switch (avio_rl32(pb)) {
507  case MKTAG('I', 'N', 'F', 'O'):
508  ff_read_riff_info(s, size - 4);
509  break;
510  case MKTAG('a', 'd', 't', 'l'):
511  if (s->nb_chapters > 0) {
512  while (avio_tell(pb) < next_tag_ofs &&
513  !avio_feof(pb)) {
514  char cue_label[512];
515  unsigned id, sub_size;
516 
517  if (avio_rl32(pb) != MKTAG('l', 'a', 'b', 'l'))
518  break;
519 
520  sub_size = avio_rl32(pb);
521  if (sub_size < 5)
522  break;
523  id = avio_rl32(pb);
524  avio_get_str(pb, sub_size - 4, cue_label, sizeof(cue_label));
525  avio_skip(pb, avio_tell(pb) & 1);
526 
527  for (int i = 0; i < s->nb_chapters; i++) {
528  if (s->chapters[i]->id == id) {
529  av_dict_set(&s->chapters[i]->metadata, "title", cue_label, 0);
530  break;
531  }
532  }
533  }
534  }
535  break;
536  }
537  break;
538  case MKTAG('I', 'D', '3', ' '):
539  case MKTAG('i', 'd', '3', ' '): {
540  ID3v2ExtraMeta *id3v2_extra_meta = NULL;
541  ff_id3v2_read_dict(pb, &s->internal->id3v2_meta, ID3v2_DEFAULT_MAGIC, &id3v2_extra_meta);
542  if (id3v2_extra_meta) {
543  ff_id3v2_parse_apic(s, id3v2_extra_meta);
544  ff_id3v2_parse_chapters(s, id3v2_extra_meta);
545  ff_id3v2_parse_priv(s, id3v2_extra_meta);
546  }
547  ff_id3v2_free_extra_meta(&id3v2_extra_meta);
548  }
549  break;
550  case MKTAG('c', 'u', 'e', ' '):
551  if (size >= 4 && got_fmt && st->codecpar->sample_rate > 0) {
552  AVRational tb = {1, st->codecpar->sample_rate};
553  unsigned nb_cues = avio_rl32(pb);
554 
555  if (size >= nb_cues * 24LL + 4LL) {
556  for (int i = 0; i < nb_cues; i++) {
557  unsigned offset, id = avio_rl32(pb);
558 
559  if (avio_feof(pb))
560  return AVERROR_INVALIDDATA;
561 
562  avio_skip(pb, 16);
563  offset = avio_rl32(pb);
564 
565  if (!avpriv_new_chapter(s, id, tb, offset, AV_NOPTS_VALUE, NULL))
566  return AVERROR(ENOMEM);
567  }
568  }
569  }
570  break;
571  }
572 
573  /* seek to next tag unless we know that we'll run into EOF */
574  if ((avio_size(pb) > 0 && next_tag_ofs >= avio_size(pb)) ||
575  wav_seek_tag(wav, pb, next_tag_ofs, SEEK_SET) < 0) {
576  break;
577  }
578  }
579 
580 break_loop:
581  if (!got_fmt && !got_xma2) {
582  av_log(s, AV_LOG_ERROR, "no 'fmt ' or 'XMA2' tag found\n");
583  return AVERROR_INVALIDDATA;
584  }
585 
586  if (data_ofs < 0) {
587  av_log(s, AV_LOG_ERROR, "no 'data' tag found\n");
588  return AVERROR_INVALIDDATA;
589  }
590 
591  avio_seek(pb, data_ofs, SEEK_SET);
592 
593  if (data_size > (INT64_MAX>>3)) {
594  av_log(s, AV_LOG_WARNING, "Data size %"PRId64" is too large\n", data_size);
595  data_size = 0;
596  }
597 
598  if ( st->codecpar->bit_rate > 0 && data_size > 0
599  && st->codecpar->sample_rate > 0
600  && sample_count > 0 && st->codecpar->channels > 1
601  && sample_count % st->codecpar->channels == 0) {
602  if (fabs(8.0 * data_size * st->codecpar->channels * st->codecpar->sample_rate /
603  sample_count /st->codecpar->bit_rate - 1.0) < 0.3)
604  sample_count /= st->codecpar->channels;
605  }
606 
607  if ( data_size > 0 && sample_count && st->codecpar->channels
608  && (data_size << 3) / sample_count / st->codecpar->channels > st->codecpar->bits_per_coded_sample + 1) {
609  av_log(s, AV_LOG_WARNING, "ignoring wrong sample_count %"PRId64"\n", sample_count);
610  sample_count = 0;
611  }
612 
613  /* G.729 hack (for Ticket4577)
614  * FIXME: Come up with cleaner, more general solution */
615  if (st->codecpar->codec_id == AV_CODEC_ID_G729 && sample_count && (data_size << 3) > sample_count) {
616  av_log(s, AV_LOG_WARNING, "ignoring wrong sample_count %"PRId64"\n", sample_count);
617  sample_count = 0;
618  }
619 
620  if (!sample_count || av_get_exact_bits_per_sample(st->codecpar->codec_id) > 0)
621  if ( st->codecpar->channels
622  && data_size
624  && wav->data_end <= avio_size(pb))
625  sample_count = (data_size << 3)
626  /
627  (st->codecpar->channels * (uint64_t)av_get_bits_per_sample(st->codecpar->codec_id));
628 
629  if (sample_count)
630  st->duration = sample_count;
631 
633  st->codecpar->block_align == st->codecpar->channels * 4 &&
634  st->codecpar->bits_per_coded_sample == 32 &&
635  st->codecpar->extradata_size == 2 &&
636  AV_RL16(st->codecpar->extradata) == 1) {
639  } else if (st->codecpar->codec_id == AV_CODEC_ID_PCM_S24LE &&
640  st->codecpar->block_align == st->codecpar->channels * 4 &&
641  st->codecpar->bits_per_coded_sample == 24) {
643  } else if (st->codecpar->codec_id == AV_CODEC_ID_XMA1 ||
645  st->codecpar->block_align = 2048;
646  } else if (st->codecpar->codec_id == AV_CODEC_ID_ADPCM_MS && st->codecpar->channels > 2) {
647  st->codecpar->block_align *= st->codecpar->channels;
648  }
649 
650  ff_metadata_conv_ctx(s, NULL, wav_metadata_conv);
652 
653  set_spdif(s, wav);
654 
655  return 0;
656 }
657 
658 /**
659  * Find chunk with w64 GUID by skipping over other chunks.
660  * @return the size of the found chunk
661  */
662 static int64_t find_guid(AVIOContext *pb, const uint8_t guid1[16])
663 {
664  uint8_t guid[16];
665  int64_t size;
666 
667  while (!avio_feof(pb)) {
668  avio_read(pb, guid, 16);
669  size = avio_rl64(pb);
670  if (size <= 24 || size > INT64_MAX - 8)
671  return AVERROR_INVALIDDATA;
672  if (!memcmp(guid, guid1, 16))
673  return size;
674  avio_skip(pb, FFALIGN(size, INT64_C(8)) - 24);
675  }
676  return AVERROR_EOF;
677 }
678 
679 static int wav_read_packet(AVFormatContext *s, AVPacket *pkt)
680 {
681  int ret, size;
682  int64_t left;
683  AVStream *st;
684  WAVDemuxContext *wav = s->priv_data;
685 
686  if (CONFIG_SPDIF_DEMUXER && wav->spdif == 1)
687  return ff_spdif_read_packet(s, pkt);
688 
689  if (wav->smv_data_ofs > 0) {
690  int64_t audio_dts, video_dts;
691 smv_retry:
692  audio_dts = (int32_t)s->streams[0]->cur_dts;
693  video_dts = (int32_t)s->streams[1]->cur_dts;
694 
695  if (audio_dts != AV_NOPTS_VALUE && video_dts != AV_NOPTS_VALUE) {
696  /*We always return a video frame first to get the pixel format first*/
697  wav->smv_last_stream = wav->smv_given_first ?
698  av_compare_ts(video_dts, s->streams[1]->time_base,
699  audio_dts, s->streams[0]->time_base) > 0 : 0;
700  wav->smv_given_first = 1;
701  }
702  wav->smv_last_stream = !wav->smv_last_stream;
703  wav->smv_last_stream |= wav->audio_eof;
704  wav->smv_last_stream &= !wav->smv_eof;
705  if (wav->smv_last_stream) {
706  uint64_t old_pos = avio_tell(s->pb);
707  uint64_t new_pos = wav->smv_data_ofs +
708  wav->smv_block * wav->smv_block_size;
709  if (avio_seek(s->pb, new_pos, SEEK_SET) < 0) {
710  ret = AVERROR_EOF;
711  goto smv_out;
712  }
713  size = avio_rl24(s->pb);
714  ret = av_get_packet(s->pb, pkt, size);
715  if (ret < 0)
716  goto smv_out;
717  pkt->pos -= 3;
718  pkt->pts = wav->smv_block * wav->smv_frames_per_jpeg;
719  pkt->duration = wav->smv_frames_per_jpeg;
720  wav->smv_block++;
721 
722  pkt->stream_index = 1;
723 smv_out:
724  avio_seek(s->pb, old_pos, SEEK_SET);
725  if (ret == AVERROR_EOF) {
726  wav->smv_eof = 1;
727  goto smv_retry;
728  }
729  return ret;
730  }
731  }
732 
733  st = s->streams[0];
734 
735  left = wav->data_end - avio_tell(s->pb);
736  if (wav->ignore_length)
737  left = INT_MAX;
738  if (left <= 0) {
739  if (CONFIG_W64_DEMUXER && wav->w64)
740  left = find_guid(s->pb, ff_w64_guid_data) - 24;
741  else
742  left = find_tag(wav, s->pb, MKTAG('d', 'a', 't', 'a'));
743  if (left < 0) {
744  wav->audio_eof = 1;
745  if (wav->smv_data_ofs > 0 && !wav->smv_eof)
746  goto smv_retry;
747  return AVERROR_EOF;
748  }
749  wav->data_end = avio_tell(s->pb) + left;
750  }
751 
752  size = wav->max_size;
753  if (st->codecpar->block_align > 1) {
754  if (size < st->codecpar->block_align)
755  size = st->codecpar->block_align;
756  size = (size / st->codecpar->block_align) * st->codecpar->block_align;
757  }
758  size = FFMIN(size, left);
759  ret = av_get_packet(s->pb, pkt, size);
760  if (ret < 0)
761  return ret;
762  pkt->stream_index = 0;
763 
764  return ret;
765 }
766 
767 static int wav_read_seek(AVFormatContext *s,
768  int stream_index, int64_t timestamp, int flags)
769 {
770  WAVDemuxContext *wav = s->priv_data;
771  AVStream *st;
772  wav->smv_eof = 0;
773  wav->audio_eof = 0;
774  if (wav->smv_data_ofs > 0) {
775  int64_t smv_timestamp = timestamp;
776  if (stream_index == 0)
777  smv_timestamp = av_rescale_q(timestamp, s->streams[0]->time_base, s->streams[1]->time_base);
778  else
779  timestamp = av_rescale_q(smv_timestamp, s->streams[1]->time_base, s->streams[0]->time_base);
780  if (wav->smv_frames_per_jpeg > 0) {
781  wav->smv_block = smv_timestamp / wav->smv_frames_per_jpeg;
782  }
783  }
784 
785  st = s->streams[0];
786  switch (st->codecpar->codec_id) {
787  case AV_CODEC_ID_MP2:
788  case AV_CODEC_ID_MP3:
789  case AV_CODEC_ID_AC3:
790  case AV_CODEC_ID_DTS:
791  case AV_CODEC_ID_XMA2:
792  /* use generic seeking with dynamically generated indexes */
793  return -1;
794  default:
795  break;
796  }
797  return ff_pcm_read_seek(s, stream_index, timestamp, flags);
798 }
799 
800 #define OFFSET(x) offsetof(WAVDemuxContext, x)
801 #define DEC AV_OPT_FLAG_DECODING_PARAM
802 static const AVOption demux_options[] = {
803  { "ignore_length", "Ignore length", OFFSET(ignore_length), AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, DEC },
804  { "max_size", "max size of single packet", OFFSET(max_size), AV_OPT_TYPE_INT, { .i64 = 4096 }, 1024, 1 << 22, DEC },
805  { NULL },
806 };
807 
808 static const AVClass wav_demuxer_class = {
809  .class_name = "WAV demuxer",
810  .item_name = av_default_item_name,
811  .option = demux_options,
812  .version = LIBAVUTIL_VERSION_INT,
813 };
815  .name = "wav",
816  .long_name = NULL_IF_CONFIG_SMALL("WAV / WAVE (Waveform Audio)"),
817  .priv_data_size = sizeof(WAVDemuxContext),
818  .read_probe = wav_probe,
819  .read_header = wav_read_header,
820  .read_packet = wav_read_packet,
821  .read_seek = wav_read_seek,
823  .codec_tag = (const AVCodecTag * const []) { ff_codec_wav_tags, 0 },
824  .priv_class = &wav_demuxer_class,
825 };
826 #endif /* CONFIG_WAV_DEMUXER */
827 
828 #if CONFIG_W64_DEMUXER
829 static int w64_probe(const AVProbeData *p)
830 {
831  if (p->buf_size <= 40)
832  return 0;
833  if (!memcmp(p->buf, ff_w64_guid_riff, 16) &&
834  !memcmp(p->buf + 24, ff_w64_guid_wave, 16))
835  return AVPROBE_SCORE_MAX;
836  else
837  return 0;
838 }
839 
840 static int w64_read_header(AVFormatContext *s)
841 {
842  int64_t size, data_ofs = 0;
843  AVIOContext *pb = s->pb;
844  WAVDemuxContext *wav = s->priv_data;
845  AVStream *st;
846  uint8_t guid[16];
847  int ret;
848 
849  avio_read(pb, guid, 16);
850  if (memcmp(guid, ff_w64_guid_riff, 16))
851  return AVERROR_INVALIDDATA;
852 
853  /* riff + wave + fmt + sizes */
854  if (avio_rl64(pb) < 16 + 8 + 16 + 8 + 16 + 8)
855  return AVERROR_INVALIDDATA;
856 
857  avio_read(pb, guid, 16);
858  if (memcmp(guid, ff_w64_guid_wave, 16)) {
859  av_log(s, AV_LOG_ERROR, "could not find wave guid\n");
860  return AVERROR_INVALIDDATA;
861  }
862 
863  wav->w64 = 1;
864 
865  st = avformat_new_stream(s, NULL);
866  if (!st)
867  return AVERROR(ENOMEM);
868 
869  while (!avio_feof(pb)) {
870  if (avio_read(pb, guid, 16) != 16)
871  break;
872  size = avio_rl64(pb);
873  if (size <= 24 || INT64_MAX - size < avio_tell(pb))
874  return AVERROR_INVALIDDATA;
875 
876  if (!memcmp(guid, ff_w64_guid_fmt, 16)) {
877  /* subtract chunk header size - normal wav file doesn't count it */
878  ret = ff_get_wav_header(s, pb, st->codecpar, size - 24, 0);
879  if (ret < 0)
880  return ret;
881  avio_skip(pb, FFALIGN(size, INT64_C(8)) - size);
882 
883  avpriv_set_pts_info(st, 64, 1, st->codecpar->sample_rate);
884  } else if (!memcmp(guid, ff_w64_guid_fact, 16)) {
885  int64_t samples;
886 
887  samples = avio_rl64(pb);
888  if (samples > 0)
889  st->duration = samples;
890  avio_skip(pb, FFALIGN(size, INT64_C(8)) - 32);
891  } else if (!memcmp(guid, ff_w64_guid_data, 16)) {
892  wav->data_end = avio_tell(pb) + size - 24;
893 
894  data_ofs = avio_tell(pb);
895  if (!(pb->seekable & AVIO_SEEKABLE_NORMAL))
896  break;
897 
898  avio_skip(pb, size - 24);
899  } else if (!memcmp(guid, ff_w64_guid_summarylist, 16)) {
900  int64_t start, end, cur;
901  uint32_t count, chunk_size, i;
902  int64_t filesize = avio_size(s->pb);
903 
904  start = avio_tell(pb);
905  end = start + FFALIGN(size, INT64_C(8)) - 24;
906  count = avio_rl32(pb);
907 
908  for (i = 0; i < count; i++) {
909  char chunk_key[5], *value;
910 
911  if (avio_feof(pb) || (cur = avio_tell(pb)) < 0 || cur > end - 8 /* = tag + size */)
912  break;
913 
914  chunk_key[4] = 0;
915  avio_read(pb, chunk_key, 4);
916  chunk_size = avio_rl32(pb);
917  if (chunk_size == UINT32_MAX || (filesize >= 0 && chunk_size > filesize))
918  return AVERROR_INVALIDDATA;
919 
920  value = av_malloc(chunk_size + 1);
921  if (!value)
922  return AVERROR(ENOMEM);
923 
924  ret = avio_get_str16le(pb, chunk_size, value, chunk_size);
925  avio_skip(pb, chunk_size - ret);
926 
927  av_dict_set(&s->metadata, chunk_key, value, AV_DICT_DONT_STRDUP_VAL);
928  }
929 
930  avio_skip(pb, end - avio_tell(pb));
931  } else {
932  av_log(s, AV_LOG_DEBUG, "unknown guid: "FF_PRI_GUID"\n", FF_ARG_GUID(guid));
933  avio_skip(pb, FFALIGN(size, INT64_C(8)) - 24);
934  }
935  }
936 
937  if (!data_ofs)
938  return AVERROR_EOF;
939 
940  ff_metadata_conv_ctx(s, NULL, wav_metadata_conv);
942 
943  handle_stream_probing(st);
945 
946  avio_seek(pb, data_ofs, SEEK_SET);
947 
948  set_spdif(s, wav);
949 
950  return 0;
951 }
952 
953 #define OFFSET(x) offsetof(WAVDemuxContext, x)
954 #define DEC AV_OPT_FLAG_DECODING_PARAM
955 static const AVOption w64_demux_options[] = {
956  { "max_size", "max size of single packet", OFFSET(max_size), AV_OPT_TYPE_INT, { .i64 = 4096 }, 1024, 1 << 22, DEC },
957  { NULL }
958 };
959 
960 static const AVClass w64_demuxer_class = {
961  .class_name = "W64 demuxer",
962  .item_name = av_default_item_name,
963  .option = w64_demux_options,
964  .version = LIBAVUTIL_VERSION_INT,
965 };
966 
968  .name = "w64",
969  .long_name = NULL_IF_CONFIG_SMALL("Sony Wave64"),
970  .priv_data_size = sizeof(WAVDemuxContext),
971  .read_probe = w64_probe,
972  .read_header = w64_read_header,
973  .read_packet = wav_read_packet,
974  .read_seek = wav_read_seek,
976  .codec_tag = (const AVCodecTag * const []) { ff_codec_wav_tags, 0 },
977  .priv_class = &w64_demuxer_class,
978 };
979 #endif /* CONFIG_W64_DEMUXER */
unsigned int nb_chapters
Number of chapters in AVChapter array.
Definition: avformat.h:1468
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:1266
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: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:4900
int probe_packets
Number of packets to buffer for codec probing.
Definition: avformat.h:1080
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:1704
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:66
const char * key
int max_size
Definition: wavdec.c:59
static AVPacket pkt
AVDictionary * metadata
Definition: avformat.h:1200
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: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:4606
Format I/O context.
Definition: avformat.h:1239
static const GUIDParseTable * find_guid(ff_asf_guid guid)
Definition: asfdec_o.c:1643
int64_t audio_dts
Definition: movenc.c:60
int64_t cur_dts
Definition: avformat.h:1073
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
static av_cold int end(AVCodecContext *avctx)
Definition: avrndec.c:92
int id
unique ID to identify the chapter
Definition: avformat.h:1197
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:1088
int id
Format-specific stream ID.
Definition: avformat.h:887
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:1307
int ff_id3v2_parse_chapters(AVFormatContext *s, ID3v2ExtraMeta *extra_meta)
Create chapters for all CHAP tags found in the ID3v2 header.
Definition: id3v2.c:1175
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:803
#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:1479
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: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:422
#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:1119
int request_probe
stream probing state -1 -> probing finished 0 -> no probing requested rest -> perform probing with re...
Definition: internal.h:242
AVChapter ** chapters
Definition: avformat.h:1469
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
GLsizei count
Definition: opengl_enc.c:108
const AVCodecTag ff_codec_wav_tags[]
Definition: riff.c:507
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:3286
#define FFMIN(a, b)
Definition: common.h:104
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: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:57
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:880
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:1119
AVIOContext * pb
I/O context.
Definition: avformat.h:1281
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:1135
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:592
#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:929
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:61
unsigned int avio_rl16(AVIOContext *s)
Definition: aviobuf.c:738
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:3304
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:1107
int smv_block
Definition: wavdec.c:54
int len
void * priv_data
Format private data.
Definition: avformat.h:1267
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:652
AVCodecParameters * codecpar
Codec parameters associated with this stream.
Definition: avformat.h:1045
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:909
Filter the word “frame” indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel sample they are references to shared objects When the negotiation mechanism computes the intersection of the formats supported at each end of a all references to both lists are replaced with a reference to the intersection And when a single format is eventually chosen for a link amongst the remaining all references to the list are updated That means that if a filter requires that its input and output have the same format amongst a supported all it has to do is use a reference to the same list of formats query_formats can leave some formats unset and return AVERROR(EAGAIN) to cause the negotiation mechanism toagain later.That can be used by filters with complex requirements to use the format negotiated on one link to set the formats supported on another.Frame references ownership and permissions
#define MKTAG(a, b, c, d)
Definition: common.h:413
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:62
#define AV_WL32(p, v)
Definition: intreadwrite.h:426