Fix decode octal regex (#421)

smalot · May 20, 2021 · 0aa0180 · 0aa0180
1 parent e7e4503
commit 0aa0180
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 2 deletions.
diff --git a/src/Smalot/PdfParser/Font.php b/src/Smalot/PdfParser/Font.php
@@ -292,11 +292,11 @@ public static function decodeHexadecimal($hexa, $add_braces = false)
      */
     public static function decodeOctal($text)
     {
-        $parts = preg_split('/(\\\\\d{3})/s', $text, -1, \PREG_SPLIT_NO_EMPTY | \PREG_SPLIT_DELIM_CAPTURE);
+        $parts = preg_split('/(\\\\[0-7]{3})/s', $text, -1, \PREG_SPLIT_NO_EMPTY | \PREG_SPLIT_DELIM_CAPTURE);
         $text = '';
 
         foreach ($parts as $part) {
-            if (preg_match('/^\\\\\d{3}$/', $part)) {
+            if (preg_match('/^\\\\[0-7]{3}$/', $part)) {
                 $text .= \chr(octdec(trim($part, '\\')));
             } else {
                 $text .= $part;

diff --git a/tests/Integration/FontTest.php b/tests/Integration/FontTest.php
@@ -277,6 +277,7 @@ public function testDecodeOctal()
     {
         $this->assertEquals('AB C', Font::decodeOctal('\\101\\102\\040\\103'));
         $this->assertEquals('AB CD', Font::decodeOctal('\\101\\102\\040\\103D'));
+        $this->assertEquals('AB \199', Font::decodeOctal('\\101\\102\\040\\199'));
     }
 
     public function testDecodeEntities()