BUG: Correctly handle image mode 1 with FlateDecode (#2249)

Fixes #2248

Author: stefan6419846

pypdf/_xobj_image_helpers.py

@@ -186,7 +186,9 @@ def bits2byte(data: bytes, size: Tuple[int, int], bits: int) -> bytes:
             lookup = None
         else:
             if img.mode == "1":
-                colors_arr = [lookup[x - nb : x] for x in range(nb, len(lookup), nb)]
+                # Two values ("high" and "low").
+                assert len(lookup) == 2 * nb, len(lookup)
+                colors_arr = [lookup[:nb], lookup[nb:]]
                 arr = b"".join(
                     [
                         b"".join(

tests/test_filters.py

@@ -22,6 +22,7 @@
 from pypdf.generic import ArrayObject, DictionaryObject, NameObject, NumberObject
 
 from . import get_data_from_url
+from .test_encryption import HAS_AES
 from .test_images import image_similarity
 
 filter_inputs = (
@@ -637,3 +638,14 @@ def test_nested_device_n_color_space():
     name = "issue2240.pdf"
     reader = PdfReader(BytesIO(get_data_from_url(url, name=name)))
     reader.pages[0].images[0]
+
+
+@pytest.mark.enable_socket()
+@pytest.mark.skipif(not HAS_AES, reason="No AES implementation")
+def test_flate_decode_with_image_mode_1():
+    """From #2248"""
+    url = "https://github.com/py-pdf/pypdf/files/12847339/Prototype-Declaration-VDE4110-HYD-5000-20000-ZSS-DE.pdf"
+    name = "issue2248.pdf"
+    reader = PdfReader(BytesIO(get_data_from_url(url, name=name)))
+    for image in reader.pages[7].images:
+        _ = image