Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

BUG: Fix indexed/CMYK images #2039

Merged
merged 2 commits into from
Jul 30, 2023
Merged

BUG: Fix indexed/CMYK images #2039

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
ebe94c5
BUG: Fix indexed/CMYK images
pubpub-zz Jul 29, 2023
bc73a42
add ref on TODO
pubpub-zz Jul 29, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
67 changes: 50 additions & 17 deletions pypdf/filters.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -57,6 +57,7 @@
from .errors import PdfReadError, PdfStreamError
from .generic import (
ArrayObject,
DecodedStreamObject,
DictionaryObject,
IndirectObject,
NullObject,
Expand Down Expand Up @@ -837,23 +838,43 @@ def bits2byte(data: bytes, size: Tuple[int, int], bits: int) -> bytes:
if color_space == "/Indexed":
from .generic import TextStringObject

if isinstance(lookup, DecodedStreamObject):
lookup = lookup.get_data()
if isinstance(lookup, TextStringObject):
lookup = lookup.original_bytes
if isinstance(lookup, bytes):
try:
nb, conv, mode = { # type: ignore
"1": (0, "", ""),
"L": (1, "P", "L"),
"P": (0, "", ""),
"RGB": (3, "P", "RGB"),
"CMYK": (4, "P", "CMYK"),
}[_get_imagemode(base, 0, "")[0]]
except KeyError: # pragma: no cover
logger_warning(
f"Base {base} not coded please share the pdf file with pypdf dev team",
__name__,
if isinstance(lookup, str):
lookup = lookup.encode()
try:
nb, conv, mode = { # type: ignore
"1": (0, "", ""),
"L": (1, "P", "L"),
"P": (0, "", ""),
"RGB": (3, "P", "RGB"),
"CMYK": (4, "P", "CMYK"),
}[_get_imagemode(base, 0, "")[0]]
except KeyError: # pragma: no cover
logger_warning(
f"Base {base} not coded please share the pdf file with pypdf dev team",
__name__,
)
lookup = None
else:
if img.mode == "1":
colors_arr = [
lookup[x - nb : x] for x in range(nb, len(lookup), nb)
]
arr = b"".join(
[
b"".join(
[
colors_arr[1 if img.getpixel((x, y)) > 127 else 0]
for x in range(img.size[0])
]
)
for y in range(img.size[1])
]
)
lookup = None
img = Image.frombytes(mode, img.size, arr)
else:
img = img.convert(conv)
if len(lookup) != (hival + 1) * nb:
Expand All @@ -865,11 +886,23 @@ def bits2byte(data: bytes, size: Tuple[int, int], bits: int) -> bytes:
# 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)
else:
img.putpalette(lookup.get_data())
img = img.convert("L" if base == ColorSpaces.DEVICE_GRAY else "RGB")
img = img.convert("L" if base == ColorSpaces.DEVICE_GRAY else "RGB")
elif not isinstance(color_space, NullObject) and color_space[0] == "/ICCBased":
# see Table 66 - Additional Entries Specific to an ICC Profile
# Stream Dictionary
Expand Down
15 changes: 15 additions & 0 deletions tests/test_filters.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -513,6 +513,21 @@ def test_index_lookup():
diff.size[0] * diff.size[1]
)
assert d < 0.001
# indexed CMYK images
# currently with a TODO as we convert to RBG the palette
url = "https://corpora.tika.apache.org/base/docs/govdocs1/972/972174.pdf"
name = "tika-972174.pdf"
reader = PdfReader(BytesIO(get_pdf_from_url(url, name=name)))
url_png = "https://github.com/py-pdf/pypdf/assets/4083478/56c93021-33cd-4387-ae13-5cbe7e673f42"
name_png = "usa.png"
refimg = Image.open(BytesIO(get_pdf_from_url(url_png, name=name_png)))
data = reader.pages[0].images["/Im3"]
# assert data.image.mode == "PA" but currently "RGBA"
diff = ImageChops.difference(data.image, refimg)
d = sqrt(
sum([(a * a + b * b + c * c + d * d) for a, b, c, d in diff.getdata()])
) / (diff.size[0] * diff.size[1])
assert d < 0.001


@pytest.mark.enable_socket()
Expand Down