summaryrefslogtreecommitdiffstats
path: root/dom/media/gtest/TestAudioSegment.cpp
blob: ea0e43f0cd6c87a41d92746b6930961b23425365 (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
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
/* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* This Source Code Form is subject to the terms of the Mozilla Public
 * License, v. 2.0. If a copy of the MPL was not distributed with this file,
 * You can obtain one at http://mozilla.org/MPL/2.0/. */

#include "AudioSegment.h"
#include <iostream>
#include "gtest/gtest.h"

#include "AudioGenerator.h"

using namespace mozilla;

namespace audio_segment {

/* Helper function to give us the maximum and minimum value that don't clip,
 * for a given sample format (integer or floating-point). */
template <typename T>
T GetLowValue();

template <typename T>
T GetHighValue();

template <typename T>
T GetSilentValue();

template <>
float GetLowValue<float>() {
  return -1.0;
}

template <>
int16_t GetLowValue<short>() {
  return INT16_MIN;
}

template <>
float GetHighValue<float>() {
  return 1.0;
}

template <>
int16_t GetHighValue<short>() {
  return INT16_MAX;
}

template <>
float GetSilentValue() {
  return 0.0;
}

template <>
int16_t GetSilentValue() {
  return 0;
}

// Get an array of planar audio buffers that has the inverse of the index of the
// channel (1-indexed) as samples.
template <typename T>
const T* const* GetPlanarChannelArray(size_t aChannels, size_t aSize) {
  T** channels = new T*[aChannels];
  for (size_t c = 0; c < aChannels; c++) {
    channels[c] = new T[aSize];
    for (size_t i = 0; i < aSize; i++) {
      channels[c][i] = ConvertAudioSample<T>(1.f / static_cast<float>(c + 1));
    }
  }
  return channels;
}

template <typename T>
void DeletePlanarChannelsArray(const T* const* aArrays, size_t aChannels) {
  for (size_t channel = 0; channel < aChannels; channel++) {
    delete[] aArrays[channel];
  }
  delete[] aArrays;
}

template <typename T>
T** GetPlanarArray(size_t aChannels, size_t aSize) {
  T** channels = new T*[aChannels];
  for (size_t c = 0; c < aChannels; c++) {
    channels[c] = new T[aSize];
    for (size_t i = 0; i < aSize; i++) {
      channels[c][i] = 0.0f;
    }
  }
  return channels;
}

template <typename T>
void DeletePlanarArray(T** aArrays, size_t aChannels) {
  for (size_t channel = 0; channel < aChannels; channel++) {
    delete[] aArrays[channel];
  }
  delete[] aArrays;
}

// Get an array of audio samples that have the inverse of the index of the
// channel (1-indexed) as samples.
template <typename T>
const T* GetInterleavedChannelArray(size_t aChannels, size_t aSize) {
  size_t sampleCount = aChannels * aSize;
  T* samples = new T[sampleCount];
  for (size_t i = 0; i < sampleCount; i++) {
    uint32_t channel = (i % aChannels) + 1;
    samples[i] = ConvertAudioSample<T>(1.f / static_cast<float>(channel));
  }
  return samples;
}

template <typename T>
void DeleteInterleavedChannelArray(const T* aArray) {
  delete[] aArray;
}

bool FuzzyEqual(float aLhs, float aRhs) { return std::abs(aLhs - aRhs) < 0.01; }

template <typename SrcT, typename DstT>
void TestInterleaveAndConvert() {
  size_t arraySize = 1024;
  size_t maxChannels = 8;  // 7.1
  for (uint32_t channels = 1; channels < maxChannels; channels++) {
    const SrcT* const* src = GetPlanarChannelArray<SrcT>(channels, arraySize);
    DstT* dst = new DstT[channels * arraySize];

    InterleaveAndConvertBuffer(src, arraySize, 1.0, channels, dst);

    uint32_t channelIndex = 0;
    for (size_t i = 0; i < arraySize * channels; i++) {
      ASSERT_TRUE(
          FuzzyEqual(dst[i], ConvertAudioSample<DstT>(
                                 1.f / static_cast<float>(channelIndex + 1))));
      channelIndex++;
      channelIndex %= channels;
    }

    DeletePlanarChannelsArray(src, channels);
    delete[] dst;
  }
}

template <typename SrcT, typename DstT>
void TestDeinterleaveAndConvert() {
  size_t arraySize = 1024;
  size_t maxChannels = 8;  // 7.1
  for (uint32_t channels = 1; channels < maxChannels; channels++) {
    const SrcT* src = GetInterleavedChannelArray<SrcT>(channels, arraySize);
    DstT** dst = GetPlanarArray<DstT>(channels, arraySize);

    DeinterleaveAndConvertBuffer(src, arraySize, channels, dst);

    for (size_t channel = 0; channel < channels; channel++) {
      for (size_t i = 0; i < arraySize; i++) {
        ASSERT_TRUE(FuzzyEqual(
            dst[channel][i],
            ConvertAudioSample<DstT>(1.f / static_cast<float>(channel + 1))));
      }
    }

    DeleteInterleavedChannelArray(src);
    DeletePlanarArray(dst, channels);
  }
}

uint8_t gSilence[4096] = {0};

template <typename T>
T* SilentChannel() {
  return reinterpret_cast<T*>(gSilence);
}

template <typename T>
void TestUpmixStereo() {
  size_t arraySize = 1024;
  nsTArray<T*> channels;
  nsTArray<const T*> channelsptr;

  channels.SetLength(1);
  channelsptr.SetLength(1);

  channels[0] = new T[arraySize];

  for (size_t i = 0; i < arraySize; i++) {
    channels[0][i] = GetHighValue<T>();
  }
  channelsptr[0] = channels[0];

  AudioChannelsUpMix(&channelsptr, 2, SilentChannel<T>());

  for (size_t channel = 0; channel < 2; channel++) {
    for (size_t i = 0; i < arraySize; i++) {
      ASSERT_TRUE(channelsptr[channel][i] == GetHighValue<T>());
    }
  }
  delete[] channels[0];
}

template <typename T>
void TestDownmixStereo() {
  const size_t arraySize = 1024;
  nsTArray<const T*> inputptr;
  nsTArray<T*> input;
  T** output;

  output = new T*[1];
  output[0] = new T[arraySize];

  input.SetLength(2);
  inputptr.SetLength(2);

  for (size_t channel = 0; channel < input.Length(); channel++) {
    input[channel] = new T[arraySize];
    for (size_t i = 0; i < arraySize; i++) {
      input[channel][i] = channel == 0 ? GetLowValue<T>() : GetHighValue<T>();
    }
    inputptr[channel] = input[channel];
  }

  AudioChannelsDownMix<T, T>(inputptr, Span(output, 1), arraySize);

  for (size_t i = 0; i < arraySize; i++) {
    ASSERT_TRUE(output[0][i] == GetSilentValue<T>());
    ASSERT_TRUE(output[0][i] == GetSilentValue<T>());
  }

  delete[] output[0];
  delete[] output;
}

TEST(AudioSegment, Test)
{
  TestInterleaveAndConvert<float, float>();
  TestInterleaveAndConvert<float, int16_t>();
  TestInterleaveAndConvert<int16_t, float>();
  TestInterleaveAndConvert<int16_t, int16_t>();
  TestDeinterleaveAndConvert<float, float>();
  TestDeinterleaveAndConvert<float, int16_t>();
  TestDeinterleaveAndConvert<int16_t, float>();
  TestDeinterleaveAndConvert<int16_t, int16_t>();
  TestUpmixStereo<float>();
  TestUpmixStereo<int16_t>();
  TestDownmixStereo<float>();
  TestDownmixStereo<int16_t>();
}

template <class T, uint32_t Channels>
void fillChunk(AudioChunk* aChunk, int aDuration) {
  static_assert(Channels != 0, "Filling 0 channels is a no-op");

  aChunk->mDuration = aDuration;

  AutoTArray<nsTArray<T>, Channels> buffer;
  buffer.SetLength(Channels);
  aChunk->mChannelData.ClearAndRetainStorage();
  aChunk->mChannelData.SetCapacity(Channels);
  for (nsTArray<T>& channel : buffer) {
    T* ch = channel.AppendElements(aDuration);
    for (int i = 0; i < aDuration; ++i) {
      ch[i] = GetHighValue<T>();
    }
    aChunk->mChannelData.AppendElement(ch);
  }

  aChunk->mBuffer = new mozilla::SharedChannelArrayBuffer<T>(std::move(buffer));
  aChunk->mBufferFormat = AudioSampleTypeToFormat<T>::Format;
}

TEST(AudioSegment, FlushAfter_ZeroDuration)
{
  AudioChunk c;
  fillChunk<float, 2>(&c, 10);

  AudioSegment s;
  s.AppendAndConsumeChunk(std::move(c));
  s.FlushAfter(0);
  EXPECT_EQ(s.GetDuration(), 0);
}

TEST(AudioSegment, FlushAfter_SmallerDuration)
{
  // It was crashing when the first chunk was silence (null) and FlushAfter
  // was called for a duration, smaller or equal to the duration of the
  // first chunk.
  TrackTime duration = 10;
  TrackTime smaller_duration = 8;
  AudioChunk c1;
  c1.SetNull(duration);
  AudioChunk c2;
  fillChunk<float, 2>(&c2, duration);

  AudioSegment s;
  s.AppendAndConsumeChunk(std::move(c1));
  s.AppendAndConsumeChunk(std::move(c2));
  s.FlushAfter(smaller_duration);
  EXPECT_EQ(s.GetDuration(), smaller_duration) << "Check new duration";

  TrackTime chunkByChunkDuration = 0;
  for (AudioSegment::ChunkIterator iter(s); !iter.IsEnded(); iter.Next()) {
    chunkByChunkDuration += iter->GetDuration();
  }
  EXPECT_EQ(s.GetDuration(), chunkByChunkDuration)
      << "Confirm duration chunk by chunk";
}

TEST(AudioSegment, MemoizedOutputChannelCount)
{
  AudioSegment s;
  EXPECT_EQ(s.MaxChannelCount(), 0U) << "0 channels on init";

  s.AppendNullData(1);
  EXPECT_EQ(s.MaxChannelCount(), 0U) << "Null data has 0 channels";

  s.Clear();
  EXPECT_EQ(s.MaxChannelCount(), 0U) << "Still 0 after clearing";

  AudioChunk c1;
  fillChunk<float, 1>(&c1, 1);
  s.AppendAndConsumeChunk(std::move(c1));
  EXPECT_EQ(s.MaxChannelCount(), 1U) << "A single chunk's channel count";

  AudioChunk c2;
  fillChunk<float, 2>(&c2, 1);
  s.AppendAndConsumeChunk(std::move(c2));
  EXPECT_EQ(s.MaxChannelCount(), 2U) << "The max of two chunks' channel count";

  s.ForgetUpTo(2);
  EXPECT_EQ(s.MaxChannelCount(), 2U) << "Memoized value with null chunks";

  s.Clear();
  EXPECT_EQ(s.MaxChannelCount(), 2U) << "Still memoized after clearing";

  AudioChunk c3;
  fillChunk<float, 1>(&c3, 1);
  s.AppendAndConsumeChunk(std::move(c3));
  EXPECT_EQ(s.MaxChannelCount(), 1U) << "Real chunk trumps memoized value";

  s.Clear();
  EXPECT_EQ(s.MaxChannelCount(), 1U) << "Memoized value was updated";
}

TEST(AudioSegment, AppendAndConsumeChunk)
{
  AudioChunk c;
  fillChunk<float, 2>(&c, 10);
  AudioChunk temp(c);
  EXPECT_TRUE(c.mBuffer->IsShared());

  AudioSegment s;
  s.AppendAndConsumeChunk(std::move(temp));
  EXPECT_FALSE(s.IsEmpty());
  EXPECT_TRUE(c.mBuffer->IsShared());

  s.Clear();
  EXPECT_FALSE(c.mBuffer->IsShared());
}

TEST(AudioSegment, AppendAndConsumeEmptyChunk)
{
  AudioChunk c;
  AudioSegment s;
  s.AppendAndConsumeChunk(std::move(c));
  EXPECT_TRUE(s.IsEmpty());
}

TEST(AudioSegment, AppendAndConsumeNonEmptyZeroDurationChunk)
{
  AudioChunk c;
  fillChunk<float, 2>(&c, 0);
  AudioChunk temp(c);
  EXPECT_TRUE(c.mBuffer->IsShared());

  AudioSegment s;
  s.AppendAndConsumeChunk(std::move(temp));
  EXPECT_TRUE(s.IsEmpty());
  EXPECT_FALSE(c.mBuffer->IsShared());
}

TEST(AudioSegment, CombineChunksInAppendAndConsumeChunk)
{
  AudioChunk source;
  fillChunk<float, 2>(&source, 10);

  auto checkChunks = [&](const AudioSegment& aSegement,
                         const nsTArray<TrackTime>& aDurations) {
    size_t i = 0;
    for (AudioSegment::ConstChunkIterator iter(aSegement); !iter.IsEnded();
         iter.Next()) {
      EXPECT_EQ(iter->GetDuration(), aDurations[i++]);
    }
    EXPECT_EQ(i, aDurations.Length());
  };

  // The chunks can be merged if their duration are adjacent.
  {
    AudioChunk c1(source);
    c1.SliceTo(2, 5);

    AudioChunk c2(source);
    c2.SliceTo(5, 9);

    AudioSegment s;
    s.AppendAndConsumeChunk(std::move(c1));
    EXPECT_EQ(s.GetDuration(), 3);

    s.AppendAndConsumeChunk(std::move(c2));
    EXPECT_EQ(s.GetDuration(), 7);

    checkChunks(s, {7});
  }
  // Otherwise, they cannot be merged.
  {
    // If durations of chunks are overlapped, they cannot be merged.
    AudioChunk c1(source);
    c1.SliceTo(2, 5);

    AudioChunk c2(source);
    c2.SliceTo(4, 9);

    AudioSegment s;
    s.AppendAndConsumeChunk(std::move(c1));
    EXPECT_EQ(s.GetDuration(), 3);

    s.AppendAndConsumeChunk(std::move(c2));
    EXPECT_EQ(s.GetDuration(), 8);

    checkChunks(s, {3, 5});
  }
  {
    // If durations of chunks are discontinuous, they cannot be merged.
    AudioChunk c1(source);
    c1.SliceTo(2, 4);

    AudioChunk c2(source);
    c2.SliceTo(5, 9);

    AudioSegment s;
    s.AppendAndConsumeChunk(std::move(c1));
    EXPECT_EQ(s.GetDuration(), 2);

    s.AppendAndConsumeChunk(std::move(c2));
    EXPECT_EQ(s.GetDuration(), 6);

    checkChunks(s, {2, 4});
  }
}

TEST(AudioSegment, ConvertFromAndToInterleaved)
{
  const uint32_t channels = 2;
  const uint32_t rate = 44100;
  AudioGenerator<AudioDataValue> generator(channels, rate);

  const size_t frames = 10;
  const size_t bufferSize = frames * channels;
  nsTArray<AudioDataValue> buffer(bufferSize);
  buffer.AppendElements(bufferSize);

  generator.GenerateInterleaved(buffer.Elements(), frames);

  AudioSegment data;
  data.AppendFromInterleavedBuffer(buffer.Elements(), frames, channels,
                                   PRINCIPAL_HANDLE_NONE);

  nsTArray<AudioDataValue> interleaved;
  size_t sampleCount = data.WriteToInterleavedBuffer(interleaved, channels);

  EXPECT_EQ(sampleCount, bufferSize);
  EXPECT_EQ(interleaved, buffer);
}

}  // namespace audio_segment