Reset maxTrackDurationUs when a new fragment is written

When there is both audio and video track, then fragment
creation is driven by video track (a combination of duration
so far + next key frame). But if there is no video track, then
the duration so far drives the fragment creation.

Due to bug, when there is only audio track, only first
fragment was created as expected and then a new fragment is
created for every audio sample.

#cherrypick

PiperOrigin-RevId: 731257696

RELEASENOTES.md

@@ -34,6 +34,8 @@
 *   Muxers:
     *   `writeSampleData()` API now uses muxer specific `BufferInfo` class
         instead of `MediaCodec.BufferInfo`.
+    *   Fix a bug in `FragmentedMp4Muxer` that creates a lot of fragments when
+        only audio track is written.
 *   IMA extension:
 *   Session:
     *   Make `MediaSession.setSessionActivity(PendingIntent)` accept null

libraries/muxer/src/main/java/androidx/media3/muxer/FragmentedMp4Writer.java

@@ -291,6 +291,7 @@ import org.checkerframework.checker.nullness.qual.MonotonicNonNull;
     writeMdatBox(trackInfos);
 
     currentFragmentSequenceNumber++;
+    maxTrackDurationUs = 0;
   }
 
   private void writeMdatBox(List<ProcessedTrackInfo> trackInfos) throws IOException {

libraries/muxer/src/test/java/androidx/media3/muxer/FragmentedMp4MuxerEndToEndTest.java

@@ -55,6 +55,7 @@ public class FragmentedMp4MuxerEndToEndTest {
   private static final String H264_WITH_PYRAMID_B_FRAMES_MP4 =
       "bbb_800x640_768kbps_30fps_avc_pyramid_3b.mp4";
   private static final String H265_HDR10_MP4 = "hdr10-720p.mp4";
+  private static final String AUDIO_ONLY_MP4 = "sample_audio_only_15s.mp4";
 
   @Parameters(name = "{0}")
   public static ImmutableList<String> mediaSamples() {
@@ -133,6 +134,34 @@ public class FragmentedMp4MuxerEndToEndTest {
         MuxerTestUtil.getExpectedDumpFilePath(H265_HDR10_MP4 + "_fragmented_box_structure"));
   }
 
+  @Test
+  public void createFragmentedMp4File_fromAudioOnlyInputFile_writesExpectedFragments()
+      throws Exception {
+    @Nullable FragmentedMp4Muxer fragmentedMp4Muxer = null;
+
+    try {
+      fragmentedMp4Muxer = new FragmentedMp4Muxer.Builder(checkNotNull(outputStream)).build();
+      fragmentedMp4Muxer.addMetadataEntry(
+          new Mp4TimestampData(
+              /* creationTimestampSeconds= */ 100_000_000L,
+              /* modificationTimestampSeconds= */ 500_000_000L));
+      feedInputDataToMuxer(context, fragmentedMp4Muxer, AUDIO_ONLY_MP4);
+    } finally {
+      if (fragmentedMp4Muxer != null) {
+        fragmentedMp4Muxer.close();
+      }
+    }
+
+    DumpableMp4Box dumpableMp4Box =
+        new DumpableMp4Box(
+            ByteBuffer.wrap(TestUtil.getByteArrayFromFilePath(checkNotNull(outputPath))));
+    // For a 15 sec audio, there should be 8 fragments (2 sec fragment duration).
+    DumpFileAsserts.assertOutput(
+        context,
+        dumpableMp4Box,
+        MuxerTestUtil.getExpectedDumpFilePath(AUDIO_ONLY_MP4 + "_fragmented_box_structure"));
+  }
+
   private static void feedInputDataToMuxer(
       Context context, FragmentedMp4Muxer muxer, String inputFileName)
       throws IOException, MuxerException {

libraries/test_data/src/test/assets/muxerdumps/sample_audio_only_15s.mp4_fragmented_box_structure.dump

@@ -0,0 +1,112 @@
+ftyp (28 bytes):
+  Data = length 20, hash EF896440
+moov (570 bytes):
+  mvhd (108 bytes):
+    Data = length 100, hash 2FE65288
+  trak (414 bytes):
+    tkhd (92 bytes):
+      Data = length 84, hash B0AC1B04
+    mdia (314 bytes):
+      mdhd (32 bytes):
+        Data = length 24, hash 2FD02F4F
+      hdlr (44 bytes):
+        Data = length 36, hash 49FC755F
+      minf (230 bytes):
+        smhd (16 bytes):
+          Data = length 8, hash 94446F01
+        dinf (36 bytes):
+          Data = length 28, hash D535436B
+        stbl (170 bytes):
+          stsd (94 bytes):
+            Data = length 86, hash 8D20FFFF
+          stts (16 bytes):
+            Data = length 8, hash 94446F01
+          stsz (20 bytes):
+            Data = length 12, hash EE830681
+          stsc (16 bytes):
+            Data = length 8, hash 94446F01
+          stco (16 bytes):
+            Data = length 8, hash 94446F01
+  mvex (40 bytes):
+    trex (32 bytes):
+      Data = length 24, hash C35D3183
+moof (1216 bytes):
+  mfhd (16 bytes):
+    Data = length 8, hash 94446F02
+  traf (1192 bytes):
+    tfhd (24 bytes):
+      Data = length 16, hash D3715417
+    trun (1160 bytes):
+      Data = length 1152, hash 1DBC5FAE
+mdat (595 bytes):
+  Data = length 587, hash 5606AB27
+moof (1216 bytes):
+  mfhd (16 bytes):
+    Data = length 8, hash 94446F03
+  traf (1192 bytes):
+    tfhd (24 bytes):
+      Data = length 16, hash D3715503
+    trun (1160 bytes):
+      Data = length 1152, hash D7BCFD1D
+mdat (578 bytes):
+  Data = length 570, hash 82739626
+moof (1216 bytes):
+  mfhd (16 bytes):
+    Data = length 8, hash 94446F04
+  traf (1192 bytes):
+    tfhd (24 bytes):
+      Data = length 16, hash D37155DE
+    trun (1160 bytes):
+      Data = length 1152, hash 4B1C22AE
+mdat (595 bytes):
+  Data = length 587, hash 5925807B
+moof (1216 bytes):
+  mfhd (16 bytes):
+    Data = length 8, hash 94446F05
+  traf (1192 bytes):
+    tfhd (24 bytes):
+      Data = length 16, hash D37156CA
+    trun (1160 bytes):
+      Data = length 1152, hash D7BCFD1D
+mdat (578 bytes):
+  Data = length 570, hash 82739626
+moof (1216 bytes):
+  mfhd (16 bytes):
+    Data = length 8, hash 94446F06
+  traf (1192 bytes):
+    tfhd (24 bytes):
+      Data = length 16, hash D37156A5
+    trun (1160 bytes):
+      Data = length 1152, hash D7BCFD1D
+mdat (578 bytes):
+  Data = length 570, hash 82739626
+moof (1216 bytes):
+  mfhd (16 bytes):
+    Data = length 8, hash 94446F07
+  traf (1192 bytes):
+    tfhd (24 bytes):
+      Data = length 16, hash D3715780
+    trun (1160 bytes):
+      Data = length 1152, hash B425E72E
+mdat (595 bytes):
+  Data = length 587, hash 310213B9
+moof (1216 bytes):
+  mfhd (16 bytes):
+    Data = length 8, hash 94446F08
+  traf (1192 bytes):
+    tfhd (24 bytes):
+      Data = length 16, hash D371586C
+    trun (1160 bytes):
+      Data = length 1152, hash D7BCFD1D
+mdat (578 bytes):
+  Data = length 570, hash 82739626
+moof (556 bytes):
+  mfhd (16 bytes):
+    Data = length 8, hash 94446F09
+  traf (532 bytes):
+    tfhd (24 bytes):
+      Data = length 16, hash D3715947
+    trun (500 bytes):
+      Data = length 492, hash EC9CC954
+mdat (248 bytes):
+  Data = length 240, hash BD3FF239