Fix SMI decoding

mantas-done · Jan 2, 2024 · 3b83a22 · 3b83a22
1 parent 5a34c5a
commit 3b83a22
Show file tree

Hide file tree

Showing 2 changed files with 17 additions and 0 deletions.
diff --git a/src/Code/Converters/SmiConverter.php b/src/Code/Converters/SmiConverter.php
@@ -19,6 +19,7 @@ public function fileContentToInternalFormat($file_content, $original_file_conten
     {
         $internal_format = []; // array - where file content will be stored
 
+        $file_content = mb_convert_encoding($file_content, 'HTML-ENTITIES', "UTF-8");
         if (strpos($file_content, '</SYNC>') === false) {
             $file_content = str_replace('<SYNC ', '</SYNC><SYNC ', $file_content);
         }

diff --git a/tests/formats/SmiTest.php b/tests/formats/SmiTest.php
@@ -134,6 +134,22 @@ public function testClientFile2()
         $this->assertInternalFormatsEqual($expected, $actual);
     }
 
+    public function testNonEnglishWords()
+    {
+        $actual = Subtitles::loadFromString('
+<SAMI>
+<BODY>
+<SYNC Start=0><P Class=ENUSCC>늘</P></SYNC>
+<SYNC Start=140400><P Class=ENUSCC>&nbsp;</P></SYNC>
+</BODY>
+</SAMI>
+        ')->getInternalFormat();
+        $expected = (new Subtitles())
+            ->add(0, 140.4, '늘')
+            ->getInternalFormat();
+        $this->assertInternalFormatsEqual($expected, $actual);
+    }
+
     // ---------------------------------- private ----------------------------------------------------------------------
 
     private static function fileContent()