From 296dd3bad77bf28bbf464f2903dd84055b683b4c Mon Sep 17 00:00:00 2001
From: Matthias Valvekens <matthias@mvalvekens.be>
Date: Wed, 27 Mar 2024 22:03:35 +0100
Subject: [PATCH] Improve error messages on malformed keys

---
 pyhanko/pdf_utils/generic.py | 10 ++++++++--
 pyhanko_tests/test_utils.py  | 17 +++++++++++++++++
 2 files changed, 25 insertions(+), 2 deletions(-)

diff --git a/pyhanko/pdf_utils/generic.py b/pyhanko/pdf_utils/generic.py
index fcf85af5..8831d89e 100644
--- a/pyhanko/pdf_utils/generic.py
+++ b/pyhanko/pdf_utils/generic.py
@@ -1259,7 +1259,7 @@ def read_from_stream(
         tmp = stream.read(2)
         if tmp != b"<<":
             raise PdfReadError(
-                "Dictionary read error at byte %s: "
+                "Dictionary read error at byte 0x%s: "
                 "stream must begin with '<<'" % hex(stream.tell())
             )
         data = {}
@@ -1270,7 +1270,13 @@ def read_from_stream(
                 stream.read(1)
                 break
             stream.seek(-1, os.SEEK_CUR)
-            key = read_object(stream, container_ref)
+            try:
+                key = NameObject.read_from_stream(stream)
+            except Exception as ex:
+                raise PdfReadError(
+                    "Failed to read dictionary key at byte 0x%s; expected PDF name"
+                    % hex(stream.tell())
+                ) from ex
             read_non_whitespace(stream)
             stream.seek(-1, os.SEEK_CUR)
             value = read_object(stream, container_ref)
diff --git a/pyhanko_tests/test_utils.py b/pyhanko_tests/test_utils.py
index b87d7ecd..c6b165d5 100644
--- a/pyhanko_tests/test_utils.py
+++ b/pyhanko_tests/test_utils.py
@@ -20,6 +20,7 @@
 from pyhanko.pdf_utils.incremental_writer import IncrementalPdfFileWriter
 from pyhanko.pdf_utils.layout import BoxConstraints, BoxSpecificationError
 from pyhanko.pdf_utils.metadata.model import DocumentMetadata
+from pyhanko.pdf_utils.misc import PdfReadError
 from pyhanko.pdf_utils.reader import (
     HistoricalResolver,
     PdfFileReader,
@@ -1521,6 +1522,22 @@ def test_parse_comments(input_str):
     assert result['/C'] == '/D'
 
 
+@pytest.mark.parametrize(
+    'input_str',
+    [
+        '<</2 0 R>>',
+        '<</Name 0 0>>',
+        '<<(blah) 0>>',
+    ],
+)
+def test_parse_malformed_dictionary(input_str):
+    strm = BytesIO(input_str.strip().encode('utf8'))
+    with pytest.raises(PdfReadError, match="Failed to read dictionary key"):
+        generic.DictionaryObject.read_from_stream(
+            strm, container_ref=generic.Reference(1, 0, pdf=None)
+        )
+
+
 NONEXISTENT_XREF_PATH = os.path.join(
     PDF_DATA_DIR, 'minimal-with-nonexistent-refs.pdf'
 )