Add support for extracting ITU-T T.35 (it35) timed metadata tracks

Update the MP4 parsing pipeline in `BoxParser` to natively recognize and
expose ITU-T T.35 timed metadata tracks. The parsed tracks are assigned
the `application/x-itut-t35` MIME type, allowing them to be safely
exposed as `C.TRACK_TYPE_METADATA` standalone tracks.

This extraction logic mirrors the `it35` muxing support and adheres to the accepted T.35 metadata sample entry spec.

PiperOrigin-RevId: 917318574

Author: rohitjoins

RELEASENOTES.md

@@ -56,6 +56,8 @@
     *   Fix an issue where `ExportResult.fileSizeBytes` may be over-reported.
 *   Track selection:
 *   Extractors:
+    *   MP4: Add support for extracting ITU-T T.35 (`it35`) timed metadata
+        tracks.
     *   MP4: Add support for extracting chapter metadata (both Nero and
         QuickTime formats). QuickTime chapters are preferred if both are
         present. The extracted chapter information is exposed as `Chapter`

libraries/container/src/main/java/androidx/media3/container/Mp4Box.java

@@ -461,6 +461,9 @@ public abstract class Mp4Box {
   @SuppressWarnings("ConstantCaseForConstants")
   public static final int TYPE_mett = 0x6d657474;
 
+  @SuppressWarnings({"ConstantCaseForConstants", "IdentifierName"})
+  public static final int TYPE_it35 = 0x69743335;
+
   @SuppressWarnings("ConstantCaseForConstants")
   public static final int TYPE_alac = 0x616c6163;
 

libraries/extractor/src/main/java/androidx/media3/extractor/mp4/BoxParser.java

@@ -1343,7 +1343,7 @@ private static StsdData parseStsd(
           || childAtomType == TYPE_text) {
         parseTextSampleEntry(
             stsd, childAtomType, childStartPosition, childAtomSize, tkhdData, language, out);
-      } else if (childAtomType == Mp4Box.TYPE_mett) {
+      } else if (childAtomType == Mp4Box.TYPE_mett || childAtomType == Mp4Box.TYPE_it35) {
         parseMetaDataSampleEntry(stsd, childAtomType, childStartPosition, tkhdData.id, out);
       } else if (childAtomType == Mp4Box.TYPE_camm) {
         out.format =
@@ -1958,6 +1958,16 @@ private static void parseMetaDataSampleEntry(
       if (mimeType != null) {
         out.format = new Format.Builder().setId(trackId).setSampleMimeType(mimeType).build();
       }
+    } else if (atomType == Mp4Box.TYPE_it35) {
+      int identifierLength = parent.readUnsignedByte();
+      byte[] identifier = new byte[identifierLength];
+      parent.readBytes(identifier, 0, identifierLength);
+      out.format =
+          new Format.Builder()
+              .setId(trackId)
+              .setSampleMimeType(MimeTypes.APPLICATION_ITUT_T35)
+              .setInitializationData(ImmutableList.of(identifier))
+              .build();
     }
   }
 

libraries/extractor/src/test/java/androidx/media3/extractor/mp4/Mp4ExtractorParameterizedTest.java

@@ -348,6 +348,11 @@ public void mp4SampleWithNeroChaptersOnly() throws Exception {
     assertExtractorBehavior("media/mp4/sample_with_nero_chapters_only.mp4", /* peekLimit= */ 2152);
   }
 
+  @Test
+  public void mp4SampleWithIt35Track() throws Exception {
+    assertExtractorBehavior("media/mp4/sample_with_it35_track.mp4", /* peekLimit= */ 1386);
+  }
+
   private void assertExtractorBehavior(String file, int peekLimit) throws IOException {
     ExtractorAsserts.AssertionConfig.Builder assertionConfigBuilder =
         new ExtractorAsserts.AssertionConfig.Builder();

libraries/muxer/src/test/java/androidx/media3/muxer/Mp4MuxerEndToEndTest.java

@@ -1094,7 +1094,6 @@ public void writeMp4File_withT35MetadataTrack_matchesExpected() throws Exception
     FakeExtractorOutput fakeExtractorOutput =
         TestUtil.extractAllSamplesFromFilePath(
             new Mp4Extractor(new DefaultSubtitleParserFactory()), checkNotNull(outputFilePath));
-    // Dump file should have T.35 metadata track once extractor starts extracting.
     DumpFileAsserts.assertOutput(
         context,
         fakeExtractorOutput,

libraries/test_data/src/test/assets/extractordumps/mp4/sample_with_it35_track.mp4.0.dump

@@ -0,0 +1,287 @@
+seekMap:
+  isSeekable = true
+  duration = 1003000
+  getPosition(0) = [[timeUs=0, position=48]]
+  getPosition(1) = [[timeUs=0, position=48]]
+  getPosition(501500) = [[timeUs=0, position=48]]
+  getPosition(1003000) = [[timeUs=0, position=48]]
+numberOfTracks = 2
+track 0:
+  total output bytes = 43683
+  sample count = 30
+  track duration = 1003000
+  format 0:
+    averageBitrate = 348493
+    peakBitrate = 348493
+    id = 1
+    containerMimeType = video/mp4
+    sampleMimeType = video/av01
+    codecs = av01.0.00M.10
+    maxInputSize = 18085
+    width = 480
+    height = 270
+    frameRate = 29.92
+    colorInfo:
+      colorSpace = 6
+      colorRange = 2
+      colorTransfer = 6
+      lumaBitdepth = 10
+      chromaBitdepth = 10
+    language = en
+    metadata = entries=[TSSE: description=null: values=[Lavf62.12.100], Mp4Timestamp: creation time=0, modification time=0, timescale=1000, ThumbnailMetadata: presentationTimeUs=0]
+    initializationData:
+      data = length 20, hash 27B68A02
+  sample 0:
+    time = 0
+    flags = 1
+    data = length 18055, hash 4A4CEBD8
+  sample 1:
+    time = 33426
+    flags = 0
+    data = length 10390, hash 1A97B887
+  sample 2:
+    time = 66852
+    flags = 0
+    data = length 3, hash D5E0
+  sample 3:
+    time = 100278
+    flags = 0
+    data = length 111, hash F52A3D2C
+  sample 4:
+    time = 133704
+    flags = 0
+    data = length 3, hash D5D0
+  sample 5:
+    time = 167130
+    flags = 0
+    data = length 748, hash 7FA30B17
+  sample 6:
+    time = 200557
+    flags = 0
+    data = length 3, hash D610
+  sample 7:
+    time = 233983
+    flags = 0
+    data = length 94, hash D426B3A3
+  sample 8:
+    time = 267409
+    flags = 0
+    data = length 3, hash D5C0
+  sample 9:
+    time = 300835
+    flags = 0
+    data = length 1487, hash 1B89F2D3
+  sample 10:
+    time = 334261
+    flags = 0
+    data = length 3, hash D5D0
+  sample 11:
+    time = 367688
+    flags = 0
+    data = length 243, hash 83AC8EF9
+  sample 12:
+    time = 401114
+    flags = 0
+    data = length 3, hash D600
+  sample 13:
+    time = 434540
+    flags = 0
+    data = length 719, hash 3F9A2D2F
+  sample 14:
+    time = 467966
+    flags = 0
+    data = length 3, hash D5E0
+  sample 15:
+    time = 501392
+    flags = 0
+    data = length 398, hash 29898A8D
+  sample 16:
+    time = 534818
+    flags = 0
+    data = length 200, hash 18678644
+  sample 17:
+    time = 568245
+    flags = 0
+    data = length 3, hash D5B0
+  sample 18:
+    time = 601671
+    flags = 0
+    data = length 8443, hash 90C5F9E
+  sample 19:
+    time = 635097
+    flags = 0
+    data = length 3, hash D5C0
+  sample 20:
+    time = 668523
+    flags = 0
+    data = length 299, hash 3219B04
+  sample 21:
+    time = 701949
+    flags = 0
+    data = length 222, hash FBB32D8
+  sample 22:
+    time = 735376
+    flags = 0
+    data = length 3, hash D600
+  sample 23:
+    time = 768802
+    flags = 0
+    data = length 645, hash 3FF6F2DF
+  sample 24:
+    time = 802228
+    flags = 0
+    data = length 3, hash D5E0
+  sample 25:
+    time = 835654
+    flags = 0
+    data = length 251, hash 83CE17A6
+  sample 26:
+    time = 869080
+    flags = 0
+    data = length 253, hash 34714BEA
+  sample 27:
+    time = 902506
+    flags = 0
+    data = length 3, hash D610
+  sample 28:
+    time = 935933
+    flags = 0
+    data = length 482, hash 4D8E647E
+  sample 29:
+    time = 969359
+    flags = 536870912
+    data = length 607, hash 230AC585
+track 1:
+  total output bytes = 7350
+  sample count = 30
+  track duration = 1003000
+  format 0:
+    averageBitrate = 58637
+    id = 2
+    containerMimeType = video/mp4
+    sampleMimeType = application/x-itut-t35
+    maxInputSize = 275
+    metadata = entries=[TSSE: description=null: values=[Lavf62.12.100], Mp4Timestamp: creation time=0, modification time=0, timescale=1000]
+    initializationData:
+      data = length 5, hash FD925165
+  sample 0:
+    time = 0
+    flags = 1
+    data = length 245, hash 9BFA0B
+  sample 1:
+    time = 33426
+    flags = 1
+    data = length 245, hash EA2065CD
+  sample 2:
+    time = 66852
+    flags = 1
+    data = length 245, hash F3A0DA5C
+  sample 3:
+    time = 100278
+    flags = 1
+    data = length 245, hash CF1BDEB2
+  sample 4:
+    time = 133704
+    flags = 1
+    data = length 245, hash BA53650F
+  sample 5:
+    time = 167130
+    flags = 1
+    data = length 245, hash 88941B39
+  sample 6:
+    time = 200557
+    flags = 1
+    data = length 245, hash 302A69B6
+  sample 7:
+    time = 233983
+    flags = 1
+    data = length 245, hash 86A9B7CF
+  sample 8:
+    time = 267409
+    flags = 1
+    data = length 245, hash 6C8AF8DF
+  sample 9:
+    time = 300835
+    flags = 1
+    data = length 245, hash A8D264F
+  sample 10:
+    time = 334261
+    flags = 1
+    data = length 245, hash BF90E97B
+  sample 11:
+    time = 367688
+    flags = 1
+    data = length 245, hash 3A036FF7
+  sample 12:
+    time = 401114
+    flags = 1
+    data = length 245, hash 4C76613D
+  sample 13:
+    time = 434540
+    flags = 1
+    data = length 245, hash 9C103E0D
+  sample 14:
+    time = 467966
+    flags = 1
+    data = length 245, hash 546611FD
+  sample 15:
+    time = 501392
+    flags = 1
+    data = length 245, hash 2EEAD77C
+  sample 16:
+    time = 534818
+    flags = 1
+    data = length 245, hash 45EE18FD
+  sample 17:
+    time = 568245
+    flags = 1
+    data = length 245, hash 6C4A5B60
+  sample 18:
+    time = 601671
+    flags = 1
+    data = length 245, hash C55E7F4F
+  sample 19:
+    time = 635097
+    flags = 1
+    data = length 245, hash 8C43E61B
+  sample 20:
+    time = 668523
+    flags = 1
+    data = length 245, hash 88CB156F
+  sample 21:
+    time = 701949
+    flags = 1
+    data = length 245, hash 975E8975
+  sample 22:
+    time = 735376
+    flags = 1
+    data = length 245, hash D7DAF392
+  sample 23:
+    time = 768802
+    flags = 1
+    data = length 245, hash 72F8BC8A
+  sample 24:
+    time = 802228
+    flags = 1
+    data = length 245, hash 69F05BD3
+  sample 25:
+    time = 835654
+    flags = 1
+    data = length 245, hash E12DFF0E
+  sample 26:
+    time = 869080
+    flags = 1
+    data = length 245, hash 5DD18D72
+  sample 27:
+    time = 902506
+    flags = 1
+    data = length 245, hash 2B6AB923
+  sample 28:
+    time = 935933
+    flags = 1
+    data = length 245, hash 68205A3F
+  sample 29:
+    time = 969359
+    flags = 536870913
+    data = length 245, hash 7D03F36A
+tracksEnded = true