From 9e8b781fc97d8fadfe1b02c86339c5d9d8ec976f Mon Sep 17 00:00:00 2001
From: ClaudioWayne <35531629+ClaudioWayne@users.noreply.github.com>
Date: Tue, 29 Sep 2026 06:38:37 +0000
Subject: [PATCH] Fix peepdf 5.0.0 indirect object resolution (#3260)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

PDF processing logged the following exception when peepdf encountered an indirect reference:

`AttributeError: 'PDFReference' object has no attribute 'getElementByName'`

CAPE attempted to identify references using the nonexistent `type` and `id` attributes. peepdf 5.0.0 exposes these values through `getType()` and `getId()`.

Because the exception was caught, PDF processing continued and most report fields remained unchanged. However, base URI and annotation URL resolution could fail.

- Use `getType()` and `getId()` to resolve peepdf reference objects.

Changed files
- lib/cuckoo/common/integrations/peepdf.py
  - Lines 40–41 now use obj.getType() and obj.getId() to resolve indirect PDF objects.
---
 lib/cuckoo/common/integrations/peepdf.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
 mode change 100644 => 100755 lib/cuckoo/common/integrations/peepdf.py

diff --git a/lib/cuckoo/common/integrations/peepdf.py b/lib/cuckoo/common/integrations/peepdf.py
old mode 100644
new mode 100755
index 3d4d082e23a..560bdc5dcb2
--- a/lib/cuckoo/common/integrations/peepdf.py
+++ b/lib/cuckoo/common/integrations/peepdf.py
@@ -37,8 +37,8 @@ def _load_peepdf():
 
 def _get_obj_val(pdf, version: int, obj):
     with contextlib.suppress(Exception):
-        if obj.type == "reference":
-            return pdf.body[version].getObject(obj.id)
+        if obj.getType() == "reference":
+            return pdf.body[version].getObject(obj.getId())
     return obj