summaryrefslogtreecommitdiff
path: root/chromium/media/webm/webm_cluster_parser.cc
blob: 87cccae4da10683d8c11fdeac2aaec178719f222 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
// Copyright (c) 2012 The Chromium Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

#include "media/webm/webm_cluster_parser.h"

#include <vector>

#include "base/logging.h"
#include "base/sys_byteorder.h"
#include "media/base/buffers.h"
#include "media/base/decrypt_config.h"
#include "media/webm/webm_constants.h"
#include "media/webm/webm_crypto_helpers.h"

namespace media {

WebMClusterParser::TextTrackIterator::TextTrackIterator(
    const TextTrackMap& text_track_map) :
    iterator_(text_track_map.begin()),
    iterator_end_(text_track_map.end()) {
}

WebMClusterParser::TextTrackIterator::TextTrackIterator(
    const TextTrackIterator& rhs) :
    iterator_(rhs.iterator_),
    iterator_end_(rhs.iterator_end_) {
}

WebMClusterParser::TextTrackIterator::~TextTrackIterator() {
}

bool WebMClusterParser::TextTrackIterator::operator()(
  int* track_num,
  const BufferQueue** buffers) {
  if (iterator_ == iterator_end_) {
    *track_num = 0;
    *buffers = NULL;

    return false;
  }

  *track_num = iterator_->first;
  *buffers = &iterator_->second.buffers();

  ++iterator_;
  return true;
}

WebMClusterParser::WebMClusterParser(
    int64 timecode_scale, int audio_track_num, int video_track_num,
    const WebMTracksParser::TextTracks& text_tracks,
    const std::set<int64>& ignored_tracks,
    const std::string& audio_encryption_key_id,
    const std::string& video_encryption_key_id,
    const LogCB& log_cb)
    : timecode_multiplier_(timecode_scale / 1000.0),
      ignored_tracks_(ignored_tracks),
      audio_encryption_key_id_(audio_encryption_key_id),
      video_encryption_key_id_(video_encryption_key_id),
      parser_(kWebMIdCluster, this),
      last_block_timecode_(-1),
      block_data_size_(-1),
      block_duration_(-1),
      block_add_id_(-1),
      block_additional_data_size_(-1),
      discard_padding_(-1),
      cluster_timecode_(-1),
      cluster_start_time_(kNoTimestamp()),
      cluster_ended_(false),
      audio_(audio_track_num, false),
      video_(video_track_num, true),
      log_cb_(log_cb) {
  for (WebMTracksParser::TextTracks::const_iterator it = text_tracks.begin();
       it != text_tracks.end();
       ++it) {
    text_track_map_.insert(std::make_pair(it->first, Track(it->first, false)));
  }
}

WebMClusterParser::~WebMClusterParser() {}

void WebMClusterParser::Reset() {
  last_block_timecode_ = -1;
  cluster_timecode_ = -1;
  cluster_start_time_ = kNoTimestamp();
  cluster_ended_ = false;
  parser_.Reset();
  audio_.Reset();
  video_.Reset();
  ResetTextTracks();
}

int WebMClusterParser::Parse(const uint8* buf, int size) {
  audio_.Reset();
  video_.Reset();
  ResetTextTracks();

  int result = parser_.Parse(buf, size);

  if (result < 0) {
    cluster_ended_ = false;
    return result;
  }

  cluster_ended_ = parser_.IsParsingComplete();
  if (cluster_ended_) {
    // If there were no buffers in this cluster, set the cluster start time to
    // be the |cluster_timecode_|.
    if (cluster_start_time_ == kNoTimestamp()) {
      DCHECK_GT(cluster_timecode_, -1);
      cluster_start_time_ = base::TimeDelta::FromMicroseconds(
          cluster_timecode_ * timecode_multiplier_);
    }

    // Reset the parser if we're done parsing so that
    // it is ready to accept another cluster on the next
    // call.
    parser_.Reset();

    last_block_timecode_ = -1;
    cluster_timecode_ = -1;
  }

  return result;
}

WebMClusterParser::TextTrackIterator
WebMClusterParser::CreateTextTrackIterator() const {
  return TextTrackIterator(text_track_map_);
}

WebMParserClient* WebMClusterParser::OnListStart(int id) {
  if (id == kWebMIdCluster) {
    cluster_timecode_ = -1;
    cluster_start_time_ = kNoTimestamp();
  } else if (id == kWebMIdBlockGroup) {
    block_data_.reset();
    block_data_size_ = -1;
    block_duration_ = -1;
    discard_padding_ = -1;
    discard_padding_set_ = false;
  } else if (id == kWebMIdBlockAdditions) {
    block_add_id_ = -1;
    block_additional_data_.reset();
    block_additional_data_size_ = -1;
  }

  return this;
}

bool WebMClusterParser::OnListEnd(int id) {
  if (id != kWebMIdBlockGroup)
    return true;

  // Make sure the BlockGroup actually had a Block.
  if (block_data_size_ == -1) {
    MEDIA_LOG(log_cb_) << "Block missing from BlockGroup.";
    return false;
  }

  bool result = ParseBlock(false, block_data_.get(), block_data_size_,
                           block_additional_data_.get(),
                           block_additional_data_size_, block_duration_,
                           discard_padding_set_ ? discard_padding_ : 0);
  block_data_.reset();
  block_data_size_ = -1;
  block_duration_ = -1;
  block_add_id_ = -1;
  block_additional_data_.reset();
  block_additional_data_size_ = -1;
  discard_padding_ = -1;
  discard_padding_set_ = false;
  return result;
}

bool WebMClusterParser::OnUInt(int id, int64 val) {
  int64* dst;
  switch (id) {
    case kWebMIdTimecode:
      dst = &cluster_timecode_;
      break;
    case kWebMIdBlockDuration:
      dst = &block_duration_;
      break;
    case kWebMIdBlockAddID:
      dst = &block_add_id_;
      break;
    case kWebMIdDiscardPadding:
      if (discard_padding_set_)
        return false;
      discard_padding_set_ = true;
      discard_padding_ = val;
      return true;
    default:
      return true;
  }
  if (*dst != -1)
    return false;
  *dst = val;
  return true;
}

bool WebMClusterParser::ParseBlock(bool is_simple_block, const uint8* buf,
                                   int size, const uint8* additional,
                                   int additional_size, int duration,
                                   int64 discard_padding) {
  if (size < 4)
    return false;

  // Return an error if the trackNum > 127. We just aren't
  // going to support large track numbers right now.
  if (!(buf[0] & 0x80)) {
    MEDIA_LOG(log_cb_) << "TrackNumber over 127 not supported";
    return false;
  }

  int track_num = buf[0] & 0x7f;
  int timecode = buf[1] << 8 | buf[2];
  int flags = buf[3] & 0xff;
  int lacing = (flags >> 1) & 0x3;

  if (lacing) {
    MEDIA_LOG(log_cb_) << "Lacing " << lacing << " is not supported yet.";
    return false;
  }

  // Sign extend negative timecode offsets.
  if (timecode & 0x8000)
    timecode |= ~0xffff;

  const uint8* frame_data = buf + 4;
  int frame_size = size - (frame_data - buf);
  return OnBlock(is_simple_block, track_num, timecode, duration, flags,
                 frame_data, frame_size, additional, additional_size,
                 discard_padding);
}

bool WebMClusterParser::OnBinary(int id, const uint8* data, int size) {
  switch (id) {
    case kWebMIdSimpleBlock:
      return ParseBlock(true, data, size, NULL, -1, -1, 0);

    case kWebMIdBlock:
      if (block_data_) {
        MEDIA_LOG(log_cb_) << "More than 1 Block in a BlockGroup is not "
                              "supported.";
        return false;
      }
      block_data_.reset(new uint8[size]);
      memcpy(block_data_.get(), data, size);
      block_data_size_ = size;
      return true;

    case kWebMIdBlockAdditional: {
      uint64 block_add_id = base::HostToNet64(block_add_id_);
      if (block_additional_data_) {
        // TODO(vigneshv): Technically, more than 1 BlockAdditional is allowed
        // as per matroska spec. But for now we don't have a use case to
        // support parsing of such files. Take a look at this again when such a
        // case arises.
        MEDIA_LOG(log_cb_) << "More than 1 BlockAdditional in a BlockGroup is "
                              "not supported.";
        return false;
      }
      // First 8 bytes of side_data in DecoderBuffer is the BlockAddID
      // element's value in Big Endian format. This is done to mimic ffmpeg
      // demuxer's behavior.
      block_additional_data_size_ = size + sizeof(block_add_id);
      block_additional_data_.reset(new uint8[block_additional_data_size_]);
      memcpy(block_additional_data_.get(), &block_add_id,
             sizeof(block_add_id));
      memcpy(block_additional_data_.get() + 8, data, size);
      return true;
    }

    default:
      return true;
  }
}

bool WebMClusterParser::OnBlock(bool is_simple_block, int track_num,
                                int timecode,
                                int  block_duration,
                                int flags,
                                const uint8* data, int size,
                                const uint8* additional, int additional_size,
                                int64 discard_padding) {
  DCHECK_GE(size, 0);
  if (cluster_timecode_ == -1) {
    MEDIA_LOG(log_cb_) << "Got a block before cluster timecode.";
    return false;
  }

  // TODO(acolwell): Should relative negative timecode offsets be rejected?  Or
  // only when the absolute timecode is negative?  See http://crbug.com/271794
  if (timecode < 0) {
    MEDIA_LOG(log_cb_) << "Got a block with negative timecode offset "
                       << timecode;
    return false;
  }

  if (last_block_timecode_ != -1 && timecode < last_block_timecode_) {
    MEDIA_LOG(log_cb_)
        << "Got a block with a timecode before the previous block.";
    return false;
  }

  Track* track = NULL;
  std::string encryption_key_id;
  if (track_num == audio_.track_num()) {
    track = &audio_;
    encryption_key_id = audio_encryption_key_id_;
  } else if (track_num == video_.track_num()) {
    track = &video_;
    encryption_key_id = video_encryption_key_id_;
  } else if (ignored_tracks_.find(track_num) != ignored_tracks_.end()) {
    return true;
  } else if (Track* const text_track = FindTextTrack(track_num)) {
    if (is_simple_block)  // BlockGroup is required for WebVTT cues
      return false;
    if (block_duration < 0)  // not specified
      return false;
    track = text_track;
  } else {
    MEDIA_LOG(log_cb_) << "Unexpected track number " << track_num;
    return false;
  }

  last_block_timecode_ = timecode;

  base::TimeDelta timestamp = base::TimeDelta::FromMicroseconds(
      (cluster_timecode_ + timecode) * timecode_multiplier_);

  // The first bit of the flags is set when a SimpleBlock contains only
  // keyframes. If this is a Block, then inspection of the payload is
  // necessary to determine whether it contains a keyframe or not.
  // http://www.matroska.org/technical/specs/index.html
  bool is_keyframe =
      is_simple_block ? (flags & 0x80) != 0 : track->IsKeyframe(data, size);

  scoped_refptr<StreamParserBuffer> buffer =
      StreamParserBuffer::CopyFrom(data, size, additional, additional_size,
                                   is_keyframe);

  // Every encrypted Block has a signal byte and IV prepended to it. Current
  // encrypted WebM request for comments specification is here
  // http://wiki.webmproject.org/encryption/webm-encryption-rfc
  if (!encryption_key_id.empty()) {
    scoped_ptr<DecryptConfig> config(WebMCreateDecryptConfig(
        data, size,
        reinterpret_cast<const uint8*>(encryption_key_id.data()),
        encryption_key_id.size()));
    if (!config)
      return false;
    buffer->set_decrypt_config(config.Pass());
  }

  buffer->set_timestamp(timestamp);
  if (cluster_start_time_ == kNoTimestamp())
    cluster_start_time_ = timestamp;

  if (block_duration >= 0) {
    buffer->set_duration(base::TimeDelta::FromMicroseconds(
        block_duration * timecode_multiplier_));
  }

  if (discard_padding != 0) {
    buffer->set_discard_padding(base::TimeDelta::FromMicroseconds(
                                    discard_padding / 1000));
  }

  return track->AddBuffer(buffer);
}

WebMClusterParser::Track::Track(int track_num, bool is_video)
    : track_num_(track_num),
      is_video_(is_video) {
}

WebMClusterParser::Track::~Track() {}

bool WebMClusterParser::Track::AddBuffer(
    const scoped_refptr<StreamParserBuffer>& buffer) {
  DVLOG(2) << "AddBuffer() : " << track_num_
           << " ts " << buffer->timestamp().InSecondsF()
           << " dur " << buffer->duration().InSecondsF()
           << " kf " << buffer->IsKeyframe()
           << " size " << buffer->data_size();

  buffers_.push_back(buffer);
  return true;
}

void WebMClusterParser::Track::Reset() {
  buffers_.clear();
}

bool WebMClusterParser::Track::IsKeyframe(const uint8* data, int size) const {
  // For now, assume that all blocks are keyframes for datatypes other than
  // video. This is a valid assumption for Vorbis, WebVTT, & Opus.
  if (!is_video_)
    return true;

  // Make sure the block is big enough for the minimal keyframe header size.
  if (size < 7)
    return false;

  // The LSb of the first byte must be a 0 for a keyframe.
  // http://tools.ietf.org/html/rfc6386 Section 19.1
  if ((data[0] & 0x01) != 0)
    return false;

  // Verify VP8 keyframe startcode.
  // http://tools.ietf.org/html/rfc6386 Section 19.1
  if (data[3] != 0x9d || data[4] != 0x01 || data[5] != 0x2a)
    return false;

  return true;
}

void WebMClusterParser::ResetTextTracks() {
  for (TextTrackMap::iterator it = text_track_map_.begin();
       it != text_track_map_.end();
       ++it) {
    it->second.Reset();
  }
}

WebMClusterParser::Track*
WebMClusterParser::FindTextTrack(int track_num) {
  const TextTrackMap::iterator it = text_track_map_.find(track_num);

  if (it == text_track_map_.end())
    return NULL;

  return &it->second;
}

}  // namespace media