Add gapless playback support for AAC playback.

media/filters/chunk_demuxer_unittest.cc

Index: media/filters/chunk_demuxer_unittest.cc
diff --git a/media/filters/chunk_demuxer_unittest.cc b/media/filters/chunk_demuxer_unittest.cc
index 55815c169026fcf5923fcef52bb495bed85fa221..59b615367ab7b85c40196b233d3cd61bae958ead 100644
--- a/media/filters/chunk_demuxer_unittest.cc
+++ b/media/filters/chunk_demuxer_unittest.cc
@@ -867,7 +867,10 @@ class ChunkDemuxerTest : public ::testing::TestWithParam<bool> {
     base::SplitString(expected, ' ', &timestamps);
     std::stringstream ss;
     for (size_t i = 0; i < timestamps.size(); ++i) {
-      DemuxerStream::Status status;
+      // Initialize status to kAborted since it's possible for Read() to return
+      // without calling StoreStatusAndBuffer() if it doesn't have any buffers
+      // left to return.
+      DemuxerStream::Status status = DemuxerStream::kAborted;
       scoped_refptr<DecoderBuffer> buffer;
       stream->Read(base::Bind(&ChunkDemuxerTest::StoreStatusAndBuffer,
                               base::Unretained(this), &status, &buffer));
@@ -878,6 +881,13 @@ class ChunkDemuxerTest : public ::testing::TestWithParam<bool> {
       if (i > 0)
         ss << " ";
       ss << buffer->timestamp().InMilliseconds();
+
+      // Handle preroll buffers.
+      if (EndsWith(timestamps[i], "P", true)) {
+        ASSERT_EQ(kInfiniteDuration(), buffer->discard_padding().first);
+        ASSERT_EQ(base::TimeDelta(), buffer->discard_padding().second);
+        ss << "P";
+      }
     }
     EXPECT_EQ(expected, ss.str());
   }
@@ -3035,8 +3045,8 @@ TEST_P(ChunkDemuxerTest, AppendWindow_Video) {
   ASSERT_TRUE(InitDemuxer(HAS_VIDEO));
   DemuxerStream* stream = demuxer_->GetStream(DemuxerStream::VIDEO);
 
-  // Set the append window to [20,280).
-  append_window_start_for_next_append_ = base::TimeDelta::FromMilliseconds(20);
+  // Set the append window to [50,280).
+  append_window_start_for_next_append_ = base::TimeDelta::FromMilliseconds(50);
   append_window_end_for_next_append_ = base::TimeDelta::FromMilliseconds(280);
 
   // Append a cluster that starts before and ends after the append window.
@@ -3049,7 +3059,7 @@ TEST_P(ChunkDemuxerTest, AppendWindow_Video) {
   CheckExpectedRanges(kSourceId, "{ [120,270) }");
   CheckExpectedBuffers(stream, "120 150 180 210 240");
 
-  // Extend the append window to [20,650).
+  // Extend the append window to [50,650).
   append_window_end_for_next_append_ = base::TimeDelta::FromMilliseconds(650);
 
   // Append more data and verify that adding buffers start at the next
@@ -3063,8 +3073,8 @@ TEST_P(ChunkDemuxerTest, AppendWindow_Audio) {
   ASSERT_TRUE(InitDemuxer(HAS_AUDIO));
   DemuxerStream* stream = demuxer_->GetStream(DemuxerStream::AUDIO);
 
-  // Set the append window to [20,280).
-  append_window_start_for_next_append_ = base::TimeDelta::FromMilliseconds(20);
+  // Set the append window to [50,280).
+  append_window_start_for_next_append_ = base::TimeDelta::FromMilliseconds(50);
   append_window_end_for_next_append_ = base::TimeDelta::FromMilliseconds(280);
 
   // Append a cluster that starts before and ends after the append window.
@@ -3076,19 +3086,19 @@ TEST_P(ChunkDemuxerTest, AppendWindow_Audio) {
   // in the buffer. Also verify that buffers that start inside the
   // window and extend beyond the end of the window are not included.
   //
-  // The first 20ms of the first buffer should be trimmed off since it
+  // The first 50ms of the first buffer should be trimmed off since it

[wolenetz, 2014/05/30 19:58:45]

nit: This comment seems confusing to me. The test looks correct, but perhaps
restate comment because we are not actually trimming 50ms off of any buffer.
Instead, 50P should be the entirety of the 0K buffer, re-timestamped and marked
for full discard. And 50 should be the the 30K buffer with start discard. The
net is, yes, 50ms preroll, but across two output buffers from two input buffers.
wdyt?

[DaleCurtis, 2014/05/30 20:54:30]

acolwell@ suggested my previous comment to that effect be removed.  I agree it's
confusing though, so I've put it back with some edits.

   // overlaps the start of the append window.
-  CheckExpectedRanges(kSourceId, "{ [20,270) }");
-  CheckExpectedBuffers(stream, "20 30 60 90 120 150 180 210 240");
+  CheckExpectedRanges(kSourceId, "{ [50,270) }");
+  CheckExpectedBuffers(stream, "50P 50 60 90 120 150 180 210 240");
 
-  // Extend the append window to [20,650).
+  // Extend the append window to [50,650).
   append_window_end_for_next_append_ = base::TimeDelta::FromMilliseconds(650);
 
   // Append more data and verify that a new range is created.
   AppendSingleStreamCluster(
       kSourceId, kAudioTrackNum,
       "360K 390K 420K 450K 480K 510K 540K 570K 600K 630K");
-  CheckExpectedRanges(kSourceId, "{ [20,270) [360,630) }");
+  CheckExpectedRanges(kSourceId, "{ [50,270) [360,630) }");
 }
 
 TEST_P(ChunkDemuxerTest, AppendWindow_AudioOverlapStartAndEnd) {
@@ -3106,6 +3116,65 @@ TEST_P(ChunkDemuxerTest, AppendWindow_AudioOverlapStartAndEnd) {
   CheckExpectedRanges(kSourceId, "{ }");
 }
 
+TEST_P(ChunkDemuxerTest, AppendWindow_WebMFile_AudioOnly) {
+  EXPECT_CALL(*this, DemuxerOpened());
+  demuxer_->Initialize(
+      &host_,
+      CreateInitDoneCB(base::TimeDelta::FromMilliseconds(2744), PIPELINE_OK),
+      true);
+  ASSERT_EQ(ChunkDemuxer::kOk, AddId(kSourceId, HAS_AUDIO));
+
+  // Set the append window to [50,150).
+  append_window_start_for_next_append_ = base::TimeDelta::FromMilliseconds(50);
+  append_window_end_for_next_append_ = base::TimeDelta::FromMilliseconds(150);
+
+  // Read a WebM file into memory and send the data to the demuxer.  The chunk
+  // size has been chosen carefully to ensure the preroll buffer used by the
+  // partial append window trim must come from a previous Append() call.
+  scoped_refptr<DecoderBuffer> buffer =
+      ReadTestDataFile("bear-320x240-audio-only.webm");
+  AppendDataInPieces(buffer->data(), buffer->data_size(), 128);
+
+  DemuxerStream* stream = demuxer_->GetStream(DemuxerStream::AUDIO);
+  CheckExpectedBuffers(stream, "50P 50 62 86 109 122 125 128");
+}
+
+TEST_P(ChunkDemuxerTest, AppendWindow_AudioConfigUpdateRemovesPreroll) {
+  EXPECT_CALL(*this, DemuxerOpened());
+  demuxer_->Initialize(
+      &host_,
+      CreateInitDoneCB(base::TimeDelta::FromMilliseconds(2744), PIPELINE_OK),
+      true);
+  ASSERT_EQ(ChunkDemuxer::kOk, AddId(kSourceId, HAS_AUDIO));
+
+  // Set the append window such that the first file is completely before the
+  // append window.
+  const base::TimeDelta duration_1 = base::TimeDelta::FromMilliseconds(2746);

[wolenetz, 2014/05/30 19:58:45]

nit: Please add similar TODO(wolenetz): Update this to match the
CreateInitDoneCB() init segment's duration expectation once the files are
updated to have consistent duration in init segment. See
http://crbug.com/354284.

[DaleCurtis, 2014/05/30 20:54:30]

Done.

+  append_window_start_for_next_append_ = duration_1;
+
+  // Read a WebM file into memory and append the data.
+  scoped_refptr<DecoderBuffer> buffer =
+      ReadTestDataFile("bear-320x240-audio-only.webm");
+  AppendDataInPieces(buffer->data(), buffer->data_size(), 512);
+  CheckExpectedRanges(kSourceId, "{ }");
+
+  DemuxerStream* stream = demuxer_->GetStream(DemuxerStream::AUDIO);
+  AudioDecoderConfig config_1 = stream->audio_decoder_config();
+
+  // Read a second WebM with a different config in and append the data.
+  scoped_refptr<DecoderBuffer> buffer2 =
+      ReadTestDataFile("bear-320x240-audio-only-48khz.webm");
+  EXPECT_CALL(host_, SetDuration(_)).Times(AnyNumber());
+  ASSERT_TRUE(SetTimestampOffset(kSourceId, duration_1));
+  AppendDataInPieces(buffer2->data(), buffer2->data_size(), 512);
+  CheckExpectedRanges(kSourceId, "{ [2746,5519) }");
+
+  Seek(duration_1);
+  ExpectConfigChanged(DemuxerStream::AUDIO);
+  ASSERT_FALSE(config_1.Matches(stream->audio_decoder_config()));
+  CheckExpectedBuffers(stream, "2746 2767 2789 2810");
+}
+
 TEST_P(ChunkDemuxerTest, AppendWindow_Text) {
   DemuxerStream* text_stream = NULL;
   EXPECT_CALL(host_, AddTextStream(_, _))