Ignore Form as well as Image XObjects when assembling the text array for a PDFObject. (#783)

* Ignore Form as well as Image XObjects when assembling the text array for a PDFObject.

* Add test coverage for the change.

Author: rupertj

src/Smalot/PdfParser/PDFObject.php

@@ -788,8 +788,9 @@ public function getTextArray(?Page $page = null): array
                             break;
                         }
 
-                        // If the PDFObject is an image, do nothing, as images aren't text.
-                        if ($xobject instanceof Image) {
+                        // If the PDFObject is an Image or a Form, do nothing as
+                        // neither of these XObject types are text.
+                        if ($xobject instanceof Image || $xobject instanceof Form) {
                             break;
                         }
 

tests/PHPUnit/Unit/PDFObjectTest.php

@@ -11,6 +11,7 @@
 use Smalot\PdfParser\Header;
 use Smalot\PdfParser\Page;
 use Smalot\PdfParser\PDFObject;
+use Smalot\PdfParser\XObject\Form;
 use Smalot\PdfParser\XObject\Image;
 
 class PDFObjectTest extends TestCase
@@ -33,6 +34,7 @@ public function testTextArrayObjects(): void
         $document->init();
 
         $image = new Image($document);
+        $form = new Form($document);
         $xObject = new PDFObject($document);
 
         $header1 = new Header([
@@ -41,24 +43,37 @@ public function testTextArrayObjects(): void
                     'Im0' => $image,
                 ])
             ]),
-            'Contents' => new ElementArray([new Element('/Imo Do', $document)], $document),
+            'Contents' => new ElementArray([new Element('/Im0 Do', $document)], $document),
         ]);
         $page1 = new Page($document, $header1);
 
         $header2 = new Header([
+            'Resources' => new Header([
+                'XObject' => new Header([
+                    'Fr0' => $form,
+                ])
+            ]),
+            'Contents' => new ElementArray([new Element('/Fr0 Do', $document)], $document),
+        ]);
+        $page2 = new Page($document, $header2);
+
+        $header3 = new Header([
             'Resources' => new Header([
                 'XObject' => new Header([
                     'Ps0' => $xObject,
                 ])
             ]),
             'Contents' => new ElementArray([new Element('/Ps0 Do', $document)], $document),
         ]);
-        $page2 = new Page($document, $header2);
+        $page3 = new Page($document, $header3);
 
         // Page 1 contains an image, which should not appear in the text array.
         self::assertSame([], $page1->getTextArray());
 
-        // Page 2 contains a non-image object, which should appear in the text array.
-        self::assertSame([' '], $page2->getTextArray());
+        // Page 2 contains a form, which should not appear in the text array.
+        self::assertSame([], $page2->getTextArray());
+
+        // Page 3 contains a non-image object, which should appear in the text array.
+        self::assertSame([' '], $page3->getTextArray());
     }
 }