fix: handle some invalid PDF cases

dhdaines · Sep 19, 2024 · 2c468dc · 2c468dc
1 parent c0fc003
commit 2c468dc
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 2 deletions.
diff --git a/playa/pdfparser.py b/playa/pdfparser.py
@@ -142,7 +142,12 @@ def flush(self) -> None:
     def do_keyword(self, pos: int, token: PSKeyword) -> None:
         if token is KEYWORD_R:
             # reference to indirect object
-            (_, _object_id), _ = self.pop(2)
+            try:
+                (_, _object_id), _ = self.pop(2)
+            except ValueError:
+                raise PDFSyntaxError(
+                    "Expected generation and object id in indirect object reference"
+                )
             object_id = safe_int(_object_id)
             if object_id is not None:
                 obj = PDFObjRef(self.doc, object_id)

diff --git a/playa/utils.py b/playa/utils.py
@@ -23,7 +23,7 @@
     cast,
 )
 
-from playa.exceptions import PDFTypeError, PDFValueError
+from playa.exceptions import PDFSyntaxError, PDFTypeError, PDFValueError
 
 if TYPE_CHECKING:
     from playa.layout import LTComponent
@@ -246,6 +246,8 @@ def parse_rect(o: Any) -> Rect:
         return float(x0), float(y0), float(x1), float(y1)
     except ValueError:
         raise PDFValueError("Could not parse rectangle")
+    except TypeError:
+        raise PDFSyntaxError("Rectangle contains non-numeric values")
 
 
 def mult_matrix(m1: Matrix, m0: Matrix) -> Matrix: