[DASH] Fix TTML text input passthrough (regression)

Note that TTML in ISO-BMFF is not supported yet.

Also updated packager_test.py:
- Added a test using TTML passthrough.
- Computed output extension from input extension unless output_format
  is specified.

Fixes #478.

Change-Id: Ia917fc4ed3c326782791ed67601fba02ea28b11d

packager/app/test/packager_test.py

@@ -215,14 +215,12 @@ def _UpdateMpdTimes(mpd_filepath):
     f.write(content)
 
 
-def GetExtension(stream_descriptor, output_format):
+def GetExtension(input_file_path, output_format):
   if output_format:
     return output_format
-  # TODO(rkuroiwa): Support ttml.
+  # Otherwise use the same extension as the input.
-  if stream_descriptor == 'text':
+  ext = os.path.splitext(input_file_path)[1]
-    return 'vtt'
+  return ext[1:]  # Remove the leading '.'.
-  # Default to mp4.
-  return 'mp4'
 
 
 def GetSegmentedExtension(base_extension):
@@ -336,7 +334,7 @@ class PackagerAppTest(unittest.TestCase):
     if skip_encryption:
       stream.Append('skip_encryption', 1)
 
-    base_ext = GetExtension(descriptor, output_format)
+    base_ext = GetExtension(input_file_path, output_format)
     output_file_name_base = stream.GetOutputFileNameBase(output_file_prefix)
 
     if hls:
@@ -516,13 +514,10 @@ class PackagerAppTest(unittest.TestCase):
     self.assertIn('Found 1 stream(s).', stream_info)
     self.assertIn(info, stream_info)
 
-  def _Decrypt(self, file_path, output_format):
+  def _Decrypt(self, file_path):
     streams = [
         self._GetStream(
-            '0',
+            '0', output_file_prefix='decrypted', test_file=file_path)
-            output_file_prefix='decrypted',
-            output_format=output_format,
-            test_file=file_path)
     ]
     self.assertPackageSuccess(streams, self._GetFlags(decryption=True))
 
@@ -560,7 +555,7 @@ class PackagerAppTest(unittest.TestCase):
           continue
         extension = os.path.splitext(file_name)[1][1:]
         if extension not in ['mpd', 'm3u8', 'media_info']:
-          self._Decrypt(os.path.join(self.tmp_dir, file_name), extension)
+          self._Decrypt(os.path.join(self.tmp_dir, file_name))
 
     out_dir = self.tmp_dir
     gold_dir = os.path.join(self.golden_file_dir, test_dir)
@@ -628,12 +623,6 @@ class PackagerFunctionalTest(PackagerAppTest):
         self._GetStreams(['0']), self._GetFlags(output_dash=True))
     self._CheckTestResults('first-stream')
 
-  def testText(self):
-    self.assertPackageSuccess(
-        self._GetStreams(['text'], test_files=['subtitle-english.vtt']),
-        self._GetFlags(output_dash=True))
-    self._CheckTestResults('text')
-
   # Probably one of the most common scenarios is to package audio and video.
   def testAudioVideo(self):
     self.assertPackageSuccess(
@@ -715,14 +704,19 @@ class PackagerFunctionalTest(PackagerAppTest):
         self._GetFlags(output_dash=True))
     self._CheckTestResults('acc-he')
 
-  # Package all video, audio, and text.
+  def testVideoAudioWebVTT(self):
-  def testVideoAudioText(self):
     audio_video_streams = self._GetStreams(['audio', 'video'])
-    text_stream = self._GetStreams(['text'],
+    text_stream = self._GetStreams(['text'], test_files=['bear-english.vtt'])
-                                   test_files=['subtitle-english.vtt'])
     self.assertPackageSuccess(audio_video_streams + text_stream,
                               self._GetFlags(output_dash=True))
-    self._CheckTestResults('video-audio-text')
+    self._CheckTestResults('video-audio-webvtt')
+
+  def testVideoAudioTTML(self):
+    audio_video_streams = self._GetStreams(['audio', 'video'])
+    text_stream = self._GetStreams(['text'], test_files=['bear-english.ttml'])
+    self.assertPackageSuccess(audio_video_streams + text_stream,
+                              self._GetFlags(output_dash=True))
+    self._CheckTestResults('video-audio-ttml')
 
   def testVideoNoEditList(self):
     stream = self._GetStream('video', test_file='bear-640x360-no_edit_list.mp4')
@@ -732,9 +726,7 @@ class PackagerFunctionalTest(PackagerAppTest):
   def testAvcAacTs(self):
     # Currently we only support live packaging for ts.
     self.assertPackageSuccess(
-        self._GetStreams(
+        self._GetStreams(['audio', 'video'],
-            ['audio', 'video'],
-            output_format='ts',
                          segmented=True,
                          hls=True,
                          test_files=['bear-640x360.ts']),
@@ -744,9 +736,7 @@ class PackagerFunctionalTest(PackagerAppTest):
   def testAvcAc3Ts(self):
     # Currently we only support live packaging for ts.
     self.assertPackageSuccess(
-        self._GetStreams(
+        self._GetStreams(['audio', 'video'],
-            ['audio', 'video'],
-            output_format='ts',
                          segmented=True,
                          hls=True,
                          test_files=['bear-640x360-ac3.ts']),
@@ -755,16 +745,16 @@ class PackagerFunctionalTest(PackagerAppTest):
 
   def testAvcAc3TsToMp4(self):
     self.assertPackageSuccess(
-        self._GetStreams(
+        self._GetStreams(['audio', 'video'],
-            ['audio', 'video'], hls=True, test_files=['bear-640x360-ac3.ts']),
+                         output_format='mp4',
+                         hls=True,
+                         test_files=['bear-640x360-ac3.ts']),
         self._GetFlags(output_hls=True))
     self._CheckTestResults('avc-ac3-ts-to-mp4')
 
   def testAvcTsLivePlaylist(self):
     self.assertPackageSuccess(
-        self._GetStreams(
+        self._GetStreams(['audio', 'video'],
-            ['audio', 'video'],
-            output_format='ts',
                          segmented=True,
                          hls=True,
                          test_files=['bear-640x360.ts']),
@@ -776,9 +766,7 @@ class PackagerFunctionalTest(PackagerAppTest):
 
   def testAvcTsLivePlaylistWithKeyRotation(self):
     self.packager.Package(
-        self._GetStreams(
+        self._GetStreams(['audio', 'video'],
-            ['audio', 'video'],
-            output_format='ts',
                          segmented=True,
                          hls=True,
                          test_files=['bear-640x360.ts']),
@@ -792,9 +780,7 @@ class PackagerFunctionalTest(PackagerAppTest):
 
   def testAvcTsEventPlaylist(self):
     self.assertPackageSuccess(
-        self._GetStreams(
+        self._GetStreams(['audio', 'video'],
-            ['audio', 'video'],
-            output_format='ts',
                          segmented=True,
                          hls=True,
                          test_files=['bear-640x360.ts']),
@@ -824,43 +810,34 @@ class PackagerFunctionalTest(PackagerAppTest):
 
   def testVp8Webm(self):
     self.assertPackageSuccess(
-        self._GetStreams(['video'],
+        self._GetStreams(['video'], test_files=['bear-640x360.webm']),
-                         output_format='webm',
-                         test_files=['bear-640x360.webm']),
         self._GetFlags(output_dash=True))
     self._CheckTestResults('vp8-webm')
 
   def testVp9Webm(self):
     self.assertPackageSuccess(
         self._GetStreams(['audio', 'video'],
-                         output_format='webm',
                          test_files=['bear-320x240-vp9-opus.webm']),
         self._GetFlags(output_dash=True))
     self._CheckTestResults('vp9-webm')
 
   def testVp9WebmWithBlockgroup(self):
     self.assertPackageSuccess(
-        self._GetStreams(['video'],
+        self._GetStreams(['video'], test_files=['bear-vp9-blockgroup.webm']),
-                         output_format='webm',
-                         test_files=['bear-vp9-blockgroup.webm']),
         self._GetFlags(output_dash=True))
     self._CheckTestResults('vp9-webm-with-blockgroup')
 
   def testVorbisWebm(self):
     self.assertPackageSuccess(
         self._GetStreams(['audio'],
-                         output_format='webm',
                          test_files=['bear-320x240-audio-only.webm']),
         self._GetFlags(output_dash=True))
     self._CheckTestResults('vorbis-webm')
 
   def testAv1Mp4(self):
     self.assertPackageSuccess(
-        self._GetStreams(['video'],
+        self._GetStreams(['video'], test_files=['bear-av1.mp4']),
-                         output_format='mp4',
+        self._GetFlags(output_dash=True, output_hls=True))
-                         test_files=['bear-av1.mp4']),
-        self._GetFlags(output_dash=True, output_hls=True)
-    )
     self._CheckTestResults('av1-mp4')
 
   def testAv1Mp4ToWebM(self):
@@ -868,8 +845,7 @@ class PackagerFunctionalTest(PackagerAppTest):
         self._GetStreams(['video'],
                          output_format='webm',
                          test_files=['bear-av1.mp4']),
-        self._GetFlags(output_dash=True, output_hls=True)
+        self._GetFlags(output_dash=True, output_hls=True))
-    )
     self._CheckTestResults('av1-mp4-to-webm')
 
   def testAv1WebM(self):
@@ -877,8 +853,7 @@ class PackagerFunctionalTest(PackagerAppTest):
         self._GetStreams(['video'],
                          output_format='mp4',
                          test_files=['bear-av1.webm']),
-        self._GetFlags(output_dash=True, output_hls=True)
+        self._GetFlags(output_dash=True, output_hls=True))
-    )
     self._CheckTestResults('av1-webm')
 
   def testEncryption(self):
@@ -1029,16 +1004,10 @@ class PackagerFunctionalTest(PackagerAppTest):
 
   def testHlsAudioVideoTextWithAdCues(self):
     streams = [
-        self._GetStream('audio',
+        self._GetStream('audio', hls=True, segmented=True),
-                        hls=True,
+        self._GetStream('video', hls=True, segmented=True),
-                        segmented=True),
+        self._GetStream(
-        self._GetStream('video',
+            'text', hls=True, segmented=True, test_file='bear-english.vtt')
-                        hls=True,
-                        segmented=True),
-        self._GetStream('text',
-                        hls=True,
-                        segmented=True,
-                        test_file='bear-subtitle-english.vtt')
     ]
     flags = self._GetFlags(output_hls=True, ad_cues='1.5')
     self.assertPackageSuccess(streams, flags)
@@ -1046,16 +1015,13 @@ class PackagerFunctionalTest(PackagerAppTest):
 
   def testVttTextToMp4WithAdCues(self):
     streams = [
-        self._GetStream('audio',
+        self._GetStream('audio', hls=True, segmented=True),
-                        hls=True,
+        self._GetStream('video', hls=True, segmented=True),
-                        segmented=True),
+        self._GetStream(
-        self._GetStream('video',
+            'text',
-                        hls=True,
-                        segmented=True),
-        self._GetStream('text',
             hls=True,
             segmented=True,
-                        test_file='bear-subtitle-english.vtt',
+            test_file='bear-english.vtt',
             output_format='mp4')
     ]
     flags = self._GetFlags(output_dash=True, output_hls=True,
@@ -1072,9 +1038,7 @@ class PackagerFunctionalTest(PackagerAppTest):
 
   def testWebmSubsampleEncryption(self):
     streams = [
-        self._GetStream('video',
+        self._GetStream('video', test_file='bear-320x180-vp9-altref.webm')
-                        output_format='webm',
-                        test_file='bear-320x180-vp9-altref.webm')
     ]
     self.assertPackageSuccess(streams,
                               self._GetFlags(encryption=True, output_dash=True))
@@ -1082,9 +1046,7 @@ class PackagerFunctionalTest(PackagerAppTest):
 
   def testWebmVp9FullSampleEncryption(self):
     streams = [
-        self._GetStream('video',
+        self._GetStream('video', test_file='bear-320x180-vp9-altref.webm')
-                        output_format='webm',
-                        test_file='bear-320x180-vp9-altref.webm')
     ]
     flags = self._GetFlags(
         encryption=True, vp9_subsample_encryption=False, output_dash=True)
@@ -1096,9 +1058,7 @@ class PackagerFunctionalTest(PackagerAppTest):
   def testAvcTsWithEncryption(self):
     # Currently we only support live packaging for ts.
     self.assertPackageSuccess(
-        self._GetStreams(
+        self._GetStreams(['audio', 'video'],
-            ['audio', 'video'],
-            output_format='ts',
                          segmented=True,
                          hls=True,
                          test_files=['bear-640x360.ts']),
@@ -1115,11 +1075,7 @@ class PackagerFunctionalTest(PackagerAppTest):
             hls=True,
             test_file='bear-640x360.ts'),
         self._GetStream(
-            'video',
+            'video', segmented=True, hls=True, test_file='bear-640x360.ts')
-            output_format='ts',
-            segmented=True,
-            hls=True,
-            test_file='bear-640x360.ts')
     ]
     flags = self._GetFlags(encryption=True, output_hls=True)
 
@@ -1129,9 +1085,7 @@ class PackagerFunctionalTest(PackagerAppTest):
   def testAvcTsWithEncryptionAndFairPlay(self):
     # Currently we only support live packaging for ts.
     self.assertPackageSuccess(
-        self._GetStreams(
+        self._GetStreams(['audio', 'video'],
-            ['audio', 'video'],
-            output_format='ts',
                          segmented=True,
                          hls=True,
                          test_files=['bear-640x360.ts']),
@@ -1141,9 +1095,7 @@ class PackagerFunctionalTest(PackagerAppTest):
   def testAvcAc3TsWithEncryption(self):
     # Currently we only support live packaging for ts.
     self.assertPackageSuccess(
-        self._GetStreams(
+        self._GetStreams(['audio', 'video'],
-            ['audio', 'video'],
-            output_format='ts',
                          segmented=True,
                          hls=True,
                          test_files=['bear-640x360-ac3.ts']),
@@ -1160,11 +1112,7 @@ class PackagerFunctionalTest(PackagerAppTest):
             hls=True,
             test_file='bear-640x360-ac3.ts'),
         self._GetStream(
-            'video',
+            'video', segmented=True, hls=True, test_file='bear-640x360-ac3.ts')
-            output_format='ts',
-            segmented=True,
-            hls=True,
-            test_file='bear-640x360-ac3.ts')
     ]
     flags = self._GetFlags(encryption=True, output_hls=True)
 
@@ -1188,11 +1136,7 @@ class PackagerFunctionalTest(PackagerAppTest):
         'avc-ts-with-encryption-exercise-emulation-prevention')
 
   def testWebmWithEncryption(self):
-    streams = [
+    streams = [self._GetStream('video', test_file='bear-640x360.webm')]
-        self._GetStream('video',
-                        output_format='webm',
-                        test_file='bear-640x360.webm')
-    ]
     flags = self._GetFlags(encryption=True, output_dash=True)
 
     self.assertPackageSuccess(streams, flags)
@@ -1235,8 +1179,7 @@ class PackagerFunctionalTest(PackagerAppTest):
 
   def testFlacWithEncryption(self):
     streams = [
-        self._GetStream(
+        self._GetStream('audio', test_file='bear-flac.mp4'),
-            'audio', output_format='mp4', test_file='bear-flac.mp4'),
     ]
     flags = self._GetFlags(encryption=True, output_dash=True, output_hls=True)
 
@@ -1246,8 +1189,9 @@ class PackagerFunctionalTest(PackagerAppTest):
   def testWvmInput(self):
     self.encryption_key = '9248d245390e0a49d483ba9b43fc69c3'
     self.assertPackageSuccess(
-        self._GetStreams(
+        self._GetStreams(['0', '1', '2', '3'],
-            ['0', '1', '2', '3'], test_files=['bear-multi-configs.wvm']),
+                         output_format='mp4',
+                         test_files=['bear-multi-configs.wvm']),
         self._GetFlags(decryption=True, output_dash=True))
     # Output timescale is 90000.
     self._CheckTestResults('wvm-input')
@@ -1262,8 +1206,9 @@ class PackagerFunctionalTest(PackagerAppTest):
   def testWvmInputWithoutStrippingParameterSetNalus(self):
     self.encryption_key = '9248d245390e0a49d483ba9b43fc69c3'
     self.assertPackageSuccess(
-        self._GetStreams(
+        self._GetStreams(['0', '1', '2', '3'],
-            ['0', '1', '2', '3'], test_files=['bear-multi-configs.wvm']),
+                         output_format='mp4',
+                         test_files=['bear-multi-configs.wvm']),
         self._GetFlags(
             strip_parameter_set_nalus=False, decryption=True, output_dash=True))
     # Output timescale is 90000.
@@ -1389,12 +1334,9 @@ class PackagerFunctionalTest(PackagerAppTest):
     self._CheckTestResults('live-profile')
 
   def testLiveProfileWithWebM(self):
-    streams = self._GetStreams(
+    streams = self._GetStreams(['audio', 'video'],
-        ['audio', 'video'],
                                segmented=True,
-        output_format='webm',
+                               test_file='bear-640x360.webm')
-        test_file='bear-640x360.webm'
-    )
     flags = self._GetFlags(output_dash=True, output_hls=True)
 
     self.assertPackageSuccess(streams, flags)
@@ -1537,7 +1479,7 @@ class PackagerFunctionalTest(PackagerAppTest):
     streams = self._GetStreams(
         ['audio', 'video'], output_format='ts', segmented=True)
     streams += self._GetStreams(
-        ['text'], test_files=['bear-subtitle-english.vtt'], segmented=True)
+        ['text'], test_files=['bear-english.vtt'], segmented=True)
 
     flags = self._GetFlags(output_hls=True)
 

packager/app/test/testdata/hls-audio-video-text-with-ad-cues/bear-english-text.m3u8

@@ -4,16 +4,16 @@
 #EXT-X-TARGETDURATION:2
 #EXT-X-PLAYLIST-TYPE:VOD
 #EXTINF:1.000,
-bear-subtitle-english-text-1.vtt
+bear-english-text-1.vtt
 #EXTINF:1.000,
-bear-subtitle-english-text-2.vtt
+bear-english-text-2.vtt
 #EXTINF:0.001,
-bear-subtitle-english-text-3.vtt
+bear-english-text-3.vtt
 #EXT-X-PLACEMENT-OPPORTUNITY
 #EXTINF:1.000,
-bear-subtitle-english-text-4.vtt
+bear-english-text-4.vtt
 #EXTINF:1.000,
-bear-subtitle-english-text-5.vtt
+bear-english-text-5.vtt
 #EXTINF:1.000,
-bear-subtitle-english-text-6.vtt
+bear-english-text-6.vtt
 #EXT-X-ENDLIST

packager/app/test/testdata/hls-audio-video-text-with-ad-cues/output.m3u8

@@ -3,7 +3,7 @@
 
 #EXT-X-MEDIA:TYPE=AUDIO,URI="bear-640x360-audio.m3u8",GROUP-ID="default-audio-group",NAME="stream_1",AUTOSELECT=YES,CHANNELS="2"
 
-#EXT-X-MEDIA:TYPE=SUBTITLES,URI="bear-subtitle-english-text.m3u8",GROUP-ID="default-text-group",NAME="stream_0",AUTOSELECT=YES
+#EXT-X-MEDIA:TYPE=SUBTITLES,URI="bear-english-text.m3u8",GROUP-ID="default-text-group",NAME="stream_0",AUTOSELECT=YES
 
 #EXT-X-STREAM-INF:BANDWIDTH=1108051,AVERAGE-BANDWIDTH=1005999,CODECS="avc1.64001e,mp4a.40.2",RESOLUTION=640x360,AUDIO="default-audio-group",SUBTITLES="default-text-group"
 bear-640x360-video.m3u8

packager/app/test/testdata/hls-segmented-webvtt/stream_0.m3u8

@@ -4,13 +4,13 @@
 #EXT-X-TARGETDURATION:2
 #EXT-X-PLAYLIST-TYPE:VOD
 #EXTINF:1.000,
-bear-subtitle-english-text-1.vtt
+bear-english-text-1.vtt
 #EXTINF:1.000,
-bear-subtitle-english-text-2.vtt
+bear-english-text-2.vtt
 #EXTINF:1.000,
-bear-subtitle-english-text-3.vtt
+bear-english-text-3.vtt
 #EXTINF:1.000,
-bear-subtitle-english-text-4.vtt
+bear-english-text-4.vtt
 #EXTINF:1.000,
-bear-subtitle-english-text-5.vtt
+bear-english-text-5.vtt
 #EXT-X-ENDLIST

packager/app/test/testdata/video-audio-ttml/bear-english-text.ttml

@@ -0,0 +1,9 @@
+<?xml version="1.0" encoding="utf-8"?>
+<tt xmlns="http://www.w3.org/ns/ttml" xml:lang="en">
+  <body>
+    <div>
+      <p begin="0.1s" end="0.8s">Yup, that's a bear, eh.</p>
+      <p begin="1.0s" end="4.7s">He 's... um... doing bear-like stuff.</p>
+    </div>
+  </body>
+</tt>

packager/app/test/testdata/video-audio-ttml/output.mpd

@@ -0,0 +1,28 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!--Generated with https://github.com/google/shaka-packager version <tag>-<hash>-<test>-->
+<MPD xmlns="urn:mpeg:dash:schema:mpd:2011" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xlink="http://www.w3.org/1999/xlink" xsi:schemaLocation="urn:mpeg:dash:schema:mpd:2011 DASH-MPD.xsd" xmlns:cenc="urn:mpeg:cenc:2013" profiles="urn:mpeg:dash:profile:isoff-on-demand:2011" minBufferTime="PT2S" type="static" mediaPresentationDuration="PT2.7360665798187256S">
+  <Period id="0">
+    <AdaptationSet id="0" contentType="text" subsegmentAlignment="true">
+      <Representation id="0" bandwidth="256" mimeType="application/ttml+xml">
+        <BaseURL>bear-english-text.ttml</BaseURL>
+      </Representation>
+    </AdaptationSet>
+    <AdaptationSet id="1" contentType="video" width="640" height="360" frameRate="30000/1001" subsegmentAlignment="true" par="16:9">
+      <Representation id="1" bandwidth="973483" codecs="avc1.64001e" mimeType="video/mp4" sar="1:1">
+        <BaseURL>bear-640x360-video.mp4</BaseURL>
+        <SegmentBase indexRange="859-926" timescale="30000">
+          <Initialization range="0-858"/>
+        </SegmentBase>
+      </Representation>
+    </AdaptationSet>
+    <AdaptationSet id="2" contentType="audio" subsegmentAlignment="true">
+      <Representation id="2" bandwidth="133334" codecs="mp4a.40.2" mimeType="audio/mp4" audioSamplingRate="44100">
+        <AudioChannelConfiguration schemeIdUri="urn:mpeg:dash:23003:3:audio_channel_configuration:2011" value="2"/>
+        <BaseURL>bear-640x360-audio.mp4</BaseURL>
+        <SegmentBase indexRange="793-860" timescale="44100">
+          <Initialization range="0-792"/>
+        </SegmentBase>
+      </Representation>
+    </AdaptationSet>
+  </Period>
+</MPD>

packager/app/test/testdata/video-audio-webvtt/output.mpd

@@ -0,0 +1,28 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!--Generated with https://github.com/google/shaka-packager version <tag>-<hash>-<test>-->
+<MPD xmlns="urn:mpeg:dash:schema:mpd:2011" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xlink="http://www.w3.org/1999/xlink" xsi:schemaLocation="urn:mpeg:dash:schema:mpd:2011 DASH-MPD.xsd" xmlns:cenc="urn:mpeg:cenc:2013" profiles="urn:mpeg:dash:profile:isoff-on-demand:2011" minBufferTime="PT2S" type="static" mediaPresentationDuration="PT2.7360665798187256S">
+  <Period id="0">
+    <AdaptationSet id="0" contentType="text" subsegmentAlignment="true">
+      <Representation id="0" bandwidth="256" mimeType="text/vtt">
+        <BaseURL>bear-english-text.vtt</BaseURL>
+      </Representation>
+    </AdaptationSet>
+    <AdaptationSet id="1" contentType="video" width="640" height="360" frameRate="30000/1001" subsegmentAlignment="true" par="16:9">
+      <Representation id="1" bandwidth="973483" codecs="avc1.64001e" mimeType="video/mp4" sar="1:1">
+        <BaseURL>bear-640x360-video.mp4</BaseURL>
+        <SegmentBase indexRange="859-926" timescale="30000">
+          <Initialization range="0-858"/>
+        </SegmentBase>
+      </Representation>
+    </AdaptationSet>
+    <AdaptationSet id="2" contentType="audio" subsegmentAlignment="true">
+      <Representation id="2" bandwidth="133334" codecs="mp4a.40.2" mimeType="audio/mp4" audioSamplingRate="44100">
+        <AudioChannelConfiguration schemeIdUri="urn:mpeg:dash:23003:3:audio_channel_configuration:2011" value="2"/>
+        <BaseURL>bear-640x360-audio.mp4</BaseURL>
+        <SegmentBase indexRange="793-860" timescale="44100">
+          <Initialization range="0-792"/>
+        </SegmentBase>
+      </Representation>
+    </AdaptationSet>
+  </Period>
+</MPD>

packager/app/test/testdata/vtt-text-to-mp4-with-ad-cues/bear-english-text.m3u8

@@ -3,18 +3,18 @@
 ## Generated with https://github.com/google/shaka-packager version <tag>-<hash>-<test>
 #EXT-X-TARGETDURATION:2
 #EXT-X-PLAYLIST-TYPE:VOD
-#EXT-X-MAP:URI="bear-subtitle-english-text-init.mp4"
+#EXT-X-MAP:URI="bear-english-text-init.mp4"
 #EXTINF:1.000,
-bear-subtitle-english-text-1.m4s
+bear-english-text-1.m4s
 #EXTINF:1.000,
-bear-subtitle-english-text-2.m4s
+bear-english-text-2.m4s
 #EXTINF:0.001,
-bear-subtitle-english-text-3.m4s
+bear-english-text-3.m4s
 #EXT-X-PLACEMENT-OPPORTUNITY
 #EXTINF:1.000,
-bear-subtitle-english-text-4.m4s
+bear-english-text-4.m4s
 #EXTINF:1.000,
-bear-subtitle-english-text-5.m4s
+bear-english-text-5.m4s
 #EXTINF:1.000,
-bear-subtitle-english-text-6.m4s
+bear-english-text-6.m4s
 #EXT-X-ENDLIST

packager/app/test/testdata/vtt-text-to-mp4-with-ad-cues/output.m3u8

@@ -3,7 +3,7 @@
 
 #EXT-X-MEDIA:TYPE=AUDIO,URI="bear-640x360-audio.m3u8",GROUP-ID="default-audio-group",NAME="stream_1",AUTOSELECT=YES,CHANNELS="2"
 
-#EXT-X-MEDIA:TYPE=SUBTITLES,URI="bear-subtitle-english-text.m3u8",GROUP-ID="default-text-group",NAME="stream_0",AUTOSELECT=YES
+#EXT-X-MEDIA:TYPE=SUBTITLES,URI="bear-english-text.m3u8",GROUP-ID="default-text-group",NAME="stream_0",AUTOSELECT=YES
 
 #EXT-X-STREAM-INF:BANDWIDTH=1108051,AVERAGE-BANDWIDTH=1005999,CODECS="avc1.64001e,mp4a.40.2",RESOLUTION=640x360,AUDIO="default-audio-group",SUBTITLES="default-text-group"
 bear-640x360-video.m3u8

packager/app/test/testdata/vtt-text-to-mp4-with-ad-cues/output.mpd

@@ -5,7 +5,7 @@
     <AdaptationSet id="0" contentType="text" segmentAlignment="true">
       <Role schemeIdUri="urn:mpeg:dash:role:2011" value="subtitle"/>
       <Representation id="0" bandwidth="1896000" codecs="wvtt" mimeType="application/mp4">
-        <SegmentTemplate timescale="1000" initialization="bear-subtitle-english-text-init.mp4" media="bear-subtitle-english-text-$Number$.m4s" startNumber="1">
+        <SegmentTemplate timescale="1000" initialization="bear-english-text-init.mp4" media="bear-english-text-$Number$.m4s" startNumber="1">
           <SegmentTimeline>
             <S t="0" d="1000" r="1"/>
             <S t="2000" d="1"/>
@@ -38,7 +38,7 @@
     <AdaptationSet id="0" contentType="text" segmentAlignment="true">
       <Role schemeIdUri="urn:mpeg:dash:role:2011" value="subtitle"/>
       <Representation id="0" bandwidth="2024" codecs="wvtt" mimeType="application/mp4">
-        <SegmentTemplate timescale="1000" presentationTimeOffset="2001" initialization="bear-subtitle-english-text-init.mp4" media="bear-subtitle-english-text-$Number$.m4s" startNumber="4">
+        <SegmentTemplate timescale="1000" presentationTimeOffset="2001" initialization="bear-english-text-init.mp4" media="bear-english-text-$Number$.m4s" startNumber="4">
           <SegmentTimeline>
             <S t="2001" d="1000" r="2"/>
           </SegmentTimeline>

packager/media/test/data/bear-english.ttml

@@ -0,0 +1,9 @@
+<?xml version="1.0" encoding="utf-8"?>
+<tt xmlns="http://www.w3.org/ns/ttml" xml:lang="en">
+  <body>
+    <div>
+      <p begin="0.1s" end="0.8s">Yup, that's a bear, eh.</p>
+      <p begin="1.0s" end="4.7s">He 's... um... doing bear-like stuff.</p>
+    </div>
+  </body>
+</tt>

packager/media/test/data/bear-english.vtt

@@ -0,0 +1,10 @@
+WEBVTT
+
+STYLE
+::cue { color:lime }
+
+00:00:00.000 --> 00:00:00.800
+Yup, that's a bear, eh.
+
+00:00:01.000 --> 00:00:04.700
+He 's... um... doing bear-like stuff.

packager/media/test/data/subtitle-english.vtt

@@ -1,79 +0,0 @@
-WEBVTT
-
-1
-00:00:03.837 --> 00:00:07.299
-Captain's log, stardate 41636.9.
-
-2
-00:00:07.466 --> 00:00:11.845
-As feared, our examination of the
-overdue Federation freighter Odin,
-
-3
-00:00:12.012 --> 00:00:16.475
-disabled by an asteroid collision,
-revealed no life signs.
-
-4
-00:00:16.642 --> 00:00:19.019
-However three escape pods
-were missing,
-
-5
-00:00:19.186 --> 00:00:21.939
-suggesting
-the possibility of survivors.
-
-6
-00:00:22.606 --> 00:00:27.861
-- Ready to orbit Angel One.
-- What kind of place is this, Data?
-
-7
-00:00:28.028 --> 00:00:31.615
-A Class-M planet supporting
-carbon-based flora and fauna,
-
-8
-00:00:31.782 --> 00:00:34.326
-sparsely populated
-with intelligent life.
-
-9
-00:00:34.493 --> 00:00:38.497
-Similar in technological development
-to mid-20th century Earth.
-
-10
-00:00:38.664 --> 00:00:41.000
-Kinda like being marooned at home.
-
-11
-00:00:41.166 --> 00:00:43.586
-Assuming any survivors
-made it this far.
-
-12
-00:00:43.794 --> 00:00:49.174
-It is the closest planet, but to
-go the distance we did in two days,
-
-13
-00:00:49.341 --> 00:00:52.344
-would've taken the Odin escape pod
-five months.
-
-14
-00:00:52.511 --> 00:00:54.680
-Five months, six days, 11 hours,
-two min...
-
-15
-00:00:54.847 --> 00:00:58.392
-- Thank you, Data.
-- ...and 57 seconds.
-
-16
-00:00:58.559 --> 00:01:01.353
-Receiving an audio signal
-from Angel One.

packager/packager.cc

@@ -536,7 +536,8 @@ Status CreateTextJobs(
   DCHECK(muxer_listener_factory);
   DCHECK(job_manager);
   for (const StreamDescriptor& stream : streams) {
-    // There are currently four options:
+    // There are currently options:
+    //    TEXT TTML --> TEXT TTML [ supported ], for DASH only.
     //    TEXT WEBVTT --> TEXT WEBVTT [ supported ]
     //    TEXT WEBVTT --> MP4 WEBVTT  [ supported ]
     //    MP4 WEBVTT  --> MP4 WEBVTT  [ unsupported ]
@@ -544,12 +545,20 @@ Status CreateTextJobs(
     const auto input_container = DetermineContainerFromFileName(stream.input);
     const auto output_container = GetOutputFormat(stream);
 
-    if (input_container != CONTAINER_WEBVTT) {
+    if (input_container != CONTAINER_WEBVTT &&
+        input_container != CONTAINER_TTML) {
       return Status(error::INVALID_ARGUMENT,
                     "Text output format is not support for " + stream.input);
     }
 
     if (output_container == CONTAINER_MOV) {
+      if (input_container == CONTAINER_TTML) {
+        return Status(error::INVALID_ARGUMENT,
+                      "TTML in MP4 is not supported yet. Please follow "
+                      "https://github.com/google/shaka-packager/issues/87 for "
+                      "the updates.");
+      }
+
       std::unique_ptr<MuxerListener> muxer_listener =
           muxer_listener_factory->CreateListener(ToMuxerListenerData(stream));
 
@@ -566,6 +575,10 @@ Status CreateTextJobs(
 
       // Check input to ensure that output is possible.
       if (hls_listener) {
+        if (input_container == CONTAINER_TTML) {
+          return Status(error::INVALID_ARGUMENT,
+                        "HLS does not support TTML in xml format.");
+        }
         if (stream.segment_template.empty() || !stream.output.empty()) {
           return Status(error::INVALID_ARGUMENT,
                         "segment_template needs to be specified for HLS text "