Merge pull request #62 from Andrius521/master

Improve *.ttml file format parsing
mantas-done · Jun 22, 2023 · 45809e8 · 45809e8
2 parents 78d0e36 + 2e9d6e2
commit 45809e8
Show file tree

Hide file tree

Showing 4 changed files with 76 additions and 7 deletions.
diff --git a/src/Code/Converters/TtmlConverter.php b/src/Code/Converters/TtmlConverter.php
@@ -12,7 +12,7 @@ public function canParseFileContent($file_content)
     public function fileContentToInternalFormat($file_content)
     {
         $dom = new \DOMDocument();
-        $dom->loadXML($file_content);
+        @$dom->loadXML($file_content);
 
         $array = array();
 
@@ -120,12 +120,14 @@ protected static function ttmlTimeToInternal($ttml_time)
         } elseif (substr($ttml_time, -1) === 's') {
             return rtrim($ttml_time, 's');
         } else {
-            $timeParts = explode(':', $ttml_time);
-            $hours = (int)$timeParts[0];
-            $minutes = (int)$timeParts[1];
-            $seconds = (int)$timeParts[2];
-            $totalSeconds = ($hours * 3600) + ($minutes * 60) + $seconds;
-            return $totalSeconds;
+            $time_parts = explode('.', $ttml_time);
+            $milliseconds = 0;
+            if (isset($time_parts[1])) {
+                $milliseconds = (float) ('0.' . $time_parts[1]);
+            }
+
+            list($hours, $minutes, $seconds) = array_map('intval', explode(':', $time_parts[0]));
+            return ($hours * 3600) + ($minutes * 60) + $seconds + $milliseconds;
         }
     }
 }
diff --git a/tests/files/ttml_with_duplicated_element_ids.ttml b/tests/files/ttml_with_duplicated_element_ids.ttml
@@ -0,0 +1,21 @@
+<tt xml:lang="en"
+xmlns="http://www.w3.org/ns/ttml"
+xmlns:tts="http://www.w3.org/ns/ttml#styling">
+ <head>
+    <metadata xmlns:ttm="http://www.w3.org/ns/ttml#metadata">
+     <ttm:title>TTML Example With Duplicated Element Ids</ttm:title>
+     <ttm:copyright>Thierry Michel 2015</ttm:copyright>
+    </metadata>
+
+    <styling xmlns:tts="http://www.w3.org/ns/ttml#styling">
+     <style xml:id="s1" tts:color="red" tts:textAlign="center"  />
+    </styling>
+</head>
+ <body>
+    <div>
+        <p xml:id="c1" begin="00:00:00" end="00:00:01">First line.</p>
+        <p xml:id="c1" begin="00:00:01" end="00:00:02">Second line.</p>
+        <p xml:id="c3" begin="00:00:02" end="00:00:03">Third line.</p>
+    </div>
+ </body>
+</tt>
diff --git a/tests/files/ttml_with_mixed_timestamps.ttml b/tests/files/ttml_with_mixed_timestamps.ttml
@@ -0,0 +1,22 @@
+<tt xml:lang="en"
+xmlns="http://www.w3.org/ns/ttml"
+xmlns:tts="http://www.w3.org/ns/ttml#styling">
+ <head>
+    <metadata xmlns:ttm="http://www.w3.org/ns/ttml#metadata">
+     <ttm:title>TTML Example Mixed Timestamp formats</ttm:title>
+     <ttm:copyright>Thierry Michel 2015</ttm:copyright>
+    </metadata>
+
+    <styling xmlns:tts="http://www.w3.org/ns/ttml#styling">
+     <style xml:id="s1" tts:color="red" tts:textAlign="center"  />
+    </styling>
+</head>
+
+ <body>
+    <div>
+        <p xml:id="c1" begin="00:00:00" end="00:00:10">First line.</p>
+        <p xml:id="c2" begin="00:00:11.200" end="00:00:13.300">Second line.</p>
+        <p xml:id="c3" begin="15s" end="18s">Third line.</p>
+    </div>
+ </body>
+</tt>
diff --git a/tests/formats/TtmlTest.php b/tests/formats/TtmlTest.php
@@ -63,4 +63,28 @@ public function testParses2()
         $this->assertInternalFormatsEqual($expected, $actual);
     }
 
+    public function testDifferentTimestampParse()
+    {
+        $ttml_path = './tests/files/ttml_with_mixed_timestamps.ttml';
+        $actual = Subtitles::loadFromFile($ttml_path)->getInternalFormat();
+        $expected = (new Subtitles())
+            ->add(0, 10, 'First line.')
+            ->add(11.2, 13.3, ['Second line.'])
+            ->add(15, 18, ['Third line.'])
+            ->getInternalFormat();
+        $this->assertInternalFormatsEqual($expected, $actual);
+    }
+
+    public function testDuplicatedElementIdsParse()
+    {
+        $ttml_path = './tests/files/ttml_with_duplicated_element_ids.ttml';
+        $actual = Subtitles::loadFromFile($ttml_path)->getInternalFormat();
+        $expected = (new Subtitles())
+            ->add(0, 1, 'First line.')
+            ->add(1, 2, ['Second line.'])
+            ->add(2, 3, ['Third line.'])
+            ->getInternalFormat();
+        $this->assertInternalFormatsEqual($expected, $actual);
+    }
+
 }