Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
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
24 changes: 15 additions & 9 deletions pypdf/_xobj_image_helpers.py
Original file line number Diff line number Diff line change
Expand Up @@ -367,16 +367,19 @@ def _apply_decode(
def _get_mode_and_invert_color(
x_object_obj: dict[str, Any], colors: int, color_space: Union[str, list[Any], Any]
) -> tuple[mode_str_type, bool]:
if (
IA.COLOR_SPACE in x_object_obj
and x_object_obj[IA.COLOR_SPACE] == ColorSpaces.DEVICE_RGB
):
# https://pillow.readthedocs.io/en/stable/handbook/concepts.html#modes
mode: mode_str_type = "RGB"

preferred_mode: mode_str_type = ""
if IA.COLOR_SPACE in x_object_obj:
preferred_mode = {
ColorSpaces.DEVICE_RGB: "RGB",
ColorSpaces.DEVICE_GRAY: "L",
ColorSpaces.DEVICE_CMYK: "CMYK",
}.get(cast(str, x_object_obj[IA.COLOR_SPACE]), "")

if x_object_obj.get("/BitsPerComponent", 8) < 8:
mode, invert_color = _get_imagemode(
f"{x_object_obj.get('/BitsPerComponent', 8)}bit", 0, ""
)
f"{x_object_obj.get('/BitsPerComponent', 8)}bit", 0, preferred_mode
)
else:
mode, invert_color = _get_imagemode(
color_space,
Expand All @@ -389,6 +392,9 @@ def _get_mode_and_invert_color(
)
)
else colors,
"",
preferred_mode,
)
if mode == "" and preferred_mode:
mode = preferred_mode
invert_color = preferred_mode == "CMYK"
return mode, invert_color
38 changes: 35 additions & 3 deletions tests/test_xobject_image_helpers.py
Original file line number Diff line number Diff line change
Expand Up @@ -5,10 +5,14 @@
import pytest

from pypdf import PdfReader
from pypdf._xobj_image_helpers import _extended_image_frombytes, _handle_flate
from pypdf._xobj_image_helpers import (
_extended_image_frombytes,
_get_mode_and_invert_color,
_handle_flate,
)
from pypdf.errors import EmptyImageDataError, PdfReadError
from pypdf.generic import ArrayObject, DecodedStreamObject, NameObject, NumberObject

from pypdf.generic import ArrayObject, DecodedStreamObject, NameObject, NullObject, NumberObject
from pypdf.constants import ColorSpaces, ImageAttributes as IA
from . import get_data_from_url

TESTS_ROOT = Path(__file__).parent.resolve()
Expand Down Expand Up @@ -160,3 +164,31 @@ def test_get_mode_and_invert_color():
page = reader.pages[12]
for _name, image in page.images.items(): # noqa: PERF102
image.image.load()
def test_get_mode_and_invert_color_prefers_color_space_hint():
x_object = {IA.COLOR_SPACE: ColorSpaces.DEVICE_RGB, "/BitsPerComponent": 8}
mode, invert_color = _get_mode_and_invert_color(
x_object, colors=3, color_space=NullObject()
)

assert mode == "RGB"
assert invert_color is False


def test_get_mode_and_invert_color_prefers_gray():
x_object = {IA.COLOR_SPACE: ColorSpaces.DEVICE_GRAY, "/BitsPerComponent": 8}
mode, invert_color = _get_mode_and_invert_color(
x_object, colors=1, color_space=NullObject()
)

assert mode == "L"
assert invert_color is False


def test_get_mode_and_invert_color_prefers_cmyk_sets_invert_flag():
x_object = {IA.COLOR_SPACE: ColorSpaces.DEVICE_CMYK, "/BitsPerComponent": 8}
mode, invert_color = _get_mode_and_invert_color(
x_object, colors=4, color_space=NullObject()
)

assert mode == "CMYK"
assert invert_color is True
Loading