Baseencoding fallback (#669)

* Use 'StandardEncoding' as default for BaseEncoding

When a document doesn't include a BaseEncoding, 'StandardEncoding' should be assumed as the default instead of an empty string.

* PHP-CS-Fixer edits

* Strict check for preg_replace output

src/Smalot/PdfParser/Encoding.php

@@ -145,6 +145,12 @@ protected function getEncodingClass(): string
     {
         // Load reference table charset.
         $baseEncoding = preg_replace('/[^A-Z0-9]/is', '', $this->get('BaseEncoding')->getContent());
+
+        // Check for empty BaseEncoding field value
+        if (!\is_string($baseEncoding) || 0 == \strlen($baseEncoding)) {
+            $baseEncoding = 'StandardEncoding';
+        }
+
         $className = '\\Smalot\\PdfParser\\Encoding\\'.$baseEncoding;
 
         if (!class_exists($className)) {

tests/PHPUnit/Integration/EncodingTest.php

@@ -42,9 +42,17 @@
 use Smalot\PdfParser\Encoding\StandardEncoding;
 use Smalot\PdfParser\Exception\EncodingNotFoundException;
 use Smalot\PdfParser\Header;
+use Smalot\PdfParser\Parser;
 
 class EncodingTest extends TestCase
 {
+    protected function setUp(): void
+    {
+        parent::setUp();
+
+        $this->fixture = new Parser();
+    }
+
     public function testGetEncodingClass(): void
     {
         $header = new Header(['BaseEncoding' => new Element('StandardEncoding')]);
@@ -103,4 +111,20 @@ public function testToStringGetEncodingClassMissingClassException(): void
             $encoding->__toString();
         }
     }
+
+    /**
+     * Fall back to 'StandardEncoding' when the document has none
+     *
+     * @see https://github.com/smalot/pdfparser/issues/665
+     */
+    public function testEmptyBaseEncodingFallback(): void
+    {
+        $filename = $this->rootDir.'/samples/bugs/Issue665.pdf';
+
+        $document = $this->fixture->parseFile($filename);
+        $objects = $document->getObjects();
+
+        $this->assertEquals(25, \count($objects));
+        $this->assertArrayHasKey('3_0', $objects);
+    }
 }