Invalid image lookup tables not handled correctly

[stefan6419846]

Invalid image lookup tables do not seem to be handled correctly and might end up trying to iterate over None:

pypdf/pypdf/filters.py

Lines 900 to 926 in 89eb626

	img = img.convert(conv)
	if len(lookup) != (hival + 1) * nb:
	logger_warning(
	f"Invalid Lookup Table in {obj_as_text}", __name__
	)
	lookup = None
	if mode == "L":
	# gray lookup does not work : it is converted to a similar RGB lookup
	lookup = b"".join([bytes([b, b, b]) for b in lookup])
	mode = "RGB"
	# TODO : cf https://github.com/py-pdf/pypdf/pull/2039
	# this is a work around until PIL is able to process CMYK images
	elif mode == "CMYK":
	_rgb = []
	for _c, _m, _y, _k in (
	lookup[n : n + 4]
	for n in range(0, 4 * (len(lookup) // 4), 4)
	):
	_r = int(255 * (1 - _c / 255) * (1 - _k / 255))
	_g = int(255 * (1 - _m / 255) * (1 - _k / 255))
	_b = int(255 * (1 - _y / 255) * (1 - _k / 255))
	_rgb.append(bytes((_r, _g, _b)))
	lookup = b"".join(_rgb)
	mode = "RGB"
	if lookup is not None:
	img.putpalette(lookup, rawmode=mode)
	img = img.convert("L" if base == ColorSpaces.DEVICE_GRAY else "RGB")

Here you can see that in line 905 the lookup table will be set to None, but both line 908 and lines 915-916 try to iterate over a possibly None value. The condition in line 924 is too late to prevent issues.

Environment

Which environment were you using when you encountered the problem?

$ python -m platform
Linux-5.14.21-150400.24.81-default-x86_64-with-glibc2.31

$ python -c "import pypdf;print(pypdf.__version__)"
3.15.2

Code + PDF

This is a minimal, complete example that shows the issue:

from pypdf import PdfReader


for page in PdfReader('file.pdf').pages:
    for key in page.images.keys():
        print(key)
        page.images[key].image.convert('RGB').save(key[1:] + '.png')

Traceback

This is the complete traceback I see:

Invalid Lookup Table in {'/BitsPerComponent': 8, '/ColorSpace': IndirectObject(37, 0, 140090665353664), '/Filter': '/FlateDecode', '/Height': 77, '/Subtype': '/Image', '/Type': '/XObject', '/Width': 106}
Traceback (most recent call last):
  File "/home/stefan/temp/run.py", line 9, in <module>
    print(page.images[key].indirect_reference)
  File "/home/stefan/temp/venv/lib/python3.9/site-packages/pypdf/_page.py", line 2636, in __getitem__
    return self.get_function(index)
  File "/home/stefan/temp/venv/lib/python3.9/site-packages/pypdf/_page.py", line 544, in _get_image
    imgd = _xobj_to_image(cast(DictionaryObject, xobjs[id]))
  File "/home/stefan/temp/venv/lib/python3.9/site-packages/pypdf/filters.py", line 1026, in _xobj_to_image
    img, image_format, extension, invert_color = _handle_flate(
  File "/home/stefan/temp/venv/lib/python3.9/site-packages/pypdf/filters.py", line 916, in _handle_flate
    for n in range(0, 4 * (len(lookup) // 4), 4)
TypeError: object of type 'NoneType' has no len()

[pubpub-zz]

@stefan6419846 can you provide the pdf file?

[stefan6419846]

A reproducing file has been sent to Martin directly for privacy reasons.