From 7023df52531fe9dc69064f5f21ac1877e59652d5 Mon Sep 17 00:00:00 2001
From: Pieter Marsman <pietermarsman@gmail.com>
Date: Tue, 9 Jul 2024 07:12:19 +0200
Subject: [PATCH] Explicitly parse rectangles to prevent errors downstream

---
 pdfminer/pdfpage.py | 15 +++++++++------
 pdfminer/utils.py   |  8 ++++++++
 2 files changed, 17 insertions(+), 6 deletions(-)

diff --git a/pdfminer/pdfpage.py b/pdfminer/pdfpage.py
index 7df4cf6b..cbc33aeb 100644
--- a/pdfminer/pdfpage.py
+++ b/pdfminer/pdfpage.py
@@ -2,12 +2,12 @@
 import logging
 from typing import BinaryIO, Container, Dict, Iterator, List, Optional, Tuple, Any
 
-from pdfminer.utils import Rect
+from pdfminer.utils import parse_rect
 from . import settings
 from .pdfdocument import PDFDocument, PDFTextExtractionNotAllowed, PDFNoPageLabels
 from .pdfparser import PDFParser
 from .pdftypes import dict_value
-from .pdfexceptions import PDFObjectNotFound
+from .pdfexceptions import PDFObjectNotFound, PDFValueError
 from .pdftypes import int_value
 from .pdftypes import list_value
 from .pdftypes import resolve1
@@ -63,11 +63,14 @@ def __init__(
         mediabox_params: List[Any] = [
             resolve1(mediabox_param) for mediabox_param in self.attrs["MediaBox"]
         ]
-        self.mediabox: Rect = resolve1(mediabox_params)
+        self.mediabox = parse_rect(resolve1(mediabox_params))
+        self.cropbox = self.mediabox
         if "CropBox" in self.attrs:
-            self.cropbox: Rect = resolve1(self.attrs["CropBox"])
-        else:
-            self.cropbox = self.mediabox
+            try:
+                self.cropbox = parse_rect(resolve1(self.attrs["CropBox"]))
+            except PDFValueError:
+                pass
+
         self.rotate = (int_value(self.attrs.get("Rotate", 0)) + 360) % 360
         self.annots = self.attrs.get("Annots")
         self.beads = self.attrs.get("B")
diff --git a/pdfminer/utils.py b/pdfminer/utils.py
index fae1f643..439fb8df 100644
--- a/pdfminer/utils.py
+++ b/pdfminer/utils.py
@@ -238,6 +238,14 @@ def apply_png_predictor(
 MATRIX_IDENTITY: Matrix = (1, 0, 0, 1, 0, 0)
 
 
+def parse_rect(o: Any) -> Rect:
+    try:
+        (x0, y0, x1, y1) = o
+        return float(x0), float(y0), float(x1), float(y1)
+    except ValueError:
+        raise PDFValueError("Could not parse rectangle")
+
+
 def mult_matrix(m1: Matrix, m0: Matrix) -> Matrix:
     (a1, b1, c1, d1, e1, f1) = m1
     (a0, b0, c0, d0, e0, f0) = m0