TST: Increase test coverage for flate handling of image mode 1 (#2339)

As mentioned in #2331, this will improve the test coverage for the edge cases.

Further refactoring was necessary as iterating over bytes will yield integers instead of single bytes and thus the whitespace check has been broken.

Additionally, the whitespace check has previously always been performed on the shortened bytes data.

Author: stefan6419846

pypdf/_utils.py

@@ -190,6 +190,23 @@ def skip_over_whitespace(stream: StreamType) -> bool:
     return cnt > 1
 
 
+def check_if_whitespace_only(value: bytes) -> bool:
+    """
+    Check if the given value consists of whitespace characters only.
+
+    Args:
+        value: The bytes to check.
+
+    Returns:
+        True if the value only has whitespace characters, otherwise return False.
+    """
+    for index in range(len(value)):
+        current = value[index:index + 1]
+        if current not in WHITESPACES:
+            return False
+    return True
+
+
 def skip_over_comment(stream: StreamType) -> None:
     tok = stream.read(1)
     stream.seek(-1, 1)

pypdf/_xobj_image_helpers.py

@@ -4,7 +4,7 @@
 from io import BytesIO
 from typing import Any, List, Tuple, Union, cast
 
-from ._utils import WHITESPACES, logger_warning
+from ._utils import check_if_whitespace_only, logger_warning
 from .constants import ColorSpaces
 from .errors import PdfReadError
 from .generic import (
@@ -199,9 +199,9 @@ def bits2byte(data: bytes, size: Tuple[int, int], bits: int) -> bytes:
                 if len(lookup) != expected_count:
                     if len(lookup) < expected_count:
                         raise PdfReadError(f"Not enough lookup values: Expected {expected_count}, got {len(lookup)}.")
-                    lookup = lookup[:expected_count]
-                    if not all(_value in WHITESPACES for _value in lookup[expected_count:]):
+                    if not check_if_whitespace_only(lookup[expected_count:]):
                         raise PdfReadError(f"Too many lookup values: Expected {expected_count}, got {len(lookup)}.")
+                    lookup = lookup[:expected_count]
                 colors_arr = [lookup[:nb], lookup[nb:]]
                 arr = b"".join(
                     [

tests/test_utils.py

@@ -10,6 +10,7 @@
     Version,
     _get_max_pdf_version_header,
     _human_readable_bytes,
+    check_if_whitespace_only,
     deprecate_with_replacement,
     deprecation_bookmark,
     deprecation_no_replacement,
@@ -48,6 +49,23 @@ def test_skip_over_whitespace(stream, expected):
     assert skip_over_whitespace(stream) == expected
 
 
+@pytest.mark.parametrize(
+    ("value", "expected"),
+    [
+        (b"foo", False),
+        (b" a", False),
+        (b" a\n b", False),
+        (b"", True),
+        (b" ", True),
+        (b"  ", True),
+        (b"  \n", True),
+        (b"    \n", True),
+    ],
+)
+def test_check_if_whitespace_only(value, expected):
+    assert check_if_whitespace_only(value) is expected
+
+
 def test_read_until_whitespace():
     assert read_until_whitespace(io.BytesIO(b"foo"), maxchars=1) == b"f"
 

tests/test_xobject_image_helpers.py

@@ -4,7 +4,9 @@
 import pytest
 
 from pypdf import PdfReader
+from pypdf._xobj_image_helpers import _handle_flate
 from pypdf.errors import PdfReadError
+from pypdf.generic import ArrayObject, DecodedStreamObject, NameObject, NumberObject
 
 from . import get_data_from_url
 
@@ -25,3 +27,61 @@ def test_get_imagemode_recursion_depth():
         match="Color spaces nested too deep. If required, consider increasing MAX_IMAGE_MODE_NESTING_DEPTH.",
     ):
         reader.pages[0].images[0]
+
+
+def test_handle_flate__image_mode_1():
+    data = b"\x00\xe0\x00"
+    lookup = DecodedStreamObject()
+    expected_data = [
+        (66, 66, 66), (66, 66, 66), (66, 66, 66),
+        (0, 19, 55), (0, 19, 55), (0, 19, 55),
+        (66, 66, 66), (66, 66, 66), (66, 66, 66)
+    ]
+
+    # No trailing data.
+    lookup.set_data(b"\x42\x42\x42\x00\x13\x37")
+    result = _handle_flate(
+        size=(3, 3),
+        data=data,
+        mode="1",
+        color_space=ArrayObject([NameObject("/Indexed"), NameObject("/DeviceRGB"), NumberObject(1), lookup]),
+        colors=2,
+        obj_as_text="dummy"
+    )
+    assert expected_data == list(result[0].getdata())
+
+    # Trailing whitespace.
+    lookup.set_data(b"\x42\x42\x42\x00\x13\x37  \x0a")
+    result = _handle_flate(
+        size=(3, 3),
+        data=data,
+        mode="1",
+        color_space=ArrayObject([NameObject("/Indexed"), NameObject("/DeviceRGB"), NumberObject(1), lookup]),
+        colors=2,
+        obj_as_text="dummy"
+    )
+    assert expected_data == list(result[0].getdata())
+
+    # Trailing non-whitespace character.
+    lookup.set_data(b"\x42\x42\x42\x00\x13\x37\x12")
+    with pytest.raises(PdfReadError, match=r"^Too many lookup values: Expected 6, got 7\.$"):
+        _handle_flate(
+            size=(3, 3),
+            data=data,
+            mode="1",
+            color_space=ArrayObject([NameObject("/Indexed"), NameObject("/DeviceRGB"), NumberObject(1), lookup]),
+            colors=2,
+            obj_as_text="dummy"
+        )
+
+    # Not enough lookup data.
+    lookup.set_data(b"\x42\x42\x42\x00\x13")
+    with pytest.raises(PdfReadError, match=r"^Not enough lookup values: Expected 6, got 5\.$"):
+        _handle_flate(
+            size=(3, 3),
+            data=data,
+            mode="1",
+            color_space=ArrayObject([NameObject("/Indexed"), NameObject("/DeviceRGB"), NumberObject(1), lookup]),
+            colors=2,
+            obj_as_text="dummy"
+        )