fix: allow to parse null or empty header element (#560)

* fix: allow to parse null or empty header element * fix: change return null by ElementNull * feat: add unit test for issue557 pdf file * fix: make empty array for empty headers * fix: PHP CS files fix * fix: remove empty() to replace it by raw comparison Co-authored-by: Konrad Abicht <[email protected]> * removed obsolete ) Co-authored-by: Enzo <[email protected]> Co-authored-by: Konrad Abicht <[email protected]>
smalot · Dec 5, 2022 · e2e3581 · e2e3581
1 parent 13a8178
commit e2e3581
Show file tree

Hide file tree

Showing 3 changed files with 29 additions and 0 deletions.
diff --git a/samples/bugs/Issue557.pdf b/samples/bugs/Issue557.pdf
diff --git a/src/Smalot/PdfParser/Parser.php b/src/Smalot/PdfParser/Parser.php
@@ -262,6 +262,11 @@ protected function parseHeader(array $structure, ?Document $document): Header
      */
     protected function parseHeaderElement(?string $type, $value, ?Document $document)
     {
+        $valueIsEmpty = null == $value || '' == $value || false == $value;
+        if (('<<' === $type || '>>' === $type) && $valueIsEmpty) {
+            $value = [];
+        }
+
         switch ($type) {
             case '<<':
             case '>>':

diff --git a/tests/Integration/ParserTest.php b/tests/Integration/ParserTest.php
@@ -270,6 +270,30 @@ public function testIssue391(): void
         );
     }
 
+    /**
+     * Tests if a PDF with null or empty string headers trigger an Exception.
+     *
+     * It happened because there was a check missing in Parser.php (parseHeaderElement function).
+     *
+     * @see https://github.com/smalot/pdfparser/issues/557
+     */
+    public function testIssue557(): void
+    {
+        /**
+         * PDF provided by @DogLoc for usage in our test environment.
+         *
+         * @see https://github.com/smalot/pdfparser/pull/560#issue-1461437944
+         */
+        $filename = $this->rootDir.'/samples/bugs/Issue557.pdf';
+
+        $document = $this->fixture->parseFile($filename);
+
+        $this->assertStringContainsString(
+            'Metal Face Inductive Sensor',
+            $document->getText()
+        );
+    }
+
     /**
      * Tests behavior when changing default font space limit (-50).
      *