feat: add fill color -> Ink classification helper (#1688)

src/ocrmypdf/pdfinfo/_contentstream.py

@@ -15,7 +15,7 @@ from pikepdf import Matrix, Object, PdfInlineImage, parse_content_stream
 
 from ocrmypdf.exceptions import InputFileError
 from ocrmypdf.helpers import Resolution
-from ocrmypdf.pdfinfo._types import UNIT_SQUARE
+from ocrmypdf.pdfinfo._types import UNIT_SQUARE, Ink
 
 
 class XobjectSettings(NamedTuple):
@@ -67,6 +67,33 @@ def _is_unit_square(shorthand):
     return all(isclose(a, b, rel_tol=1e-3) for a, b in pairwise)
 
 
+_INK_EPSILON = 1e-3
+
+
+def _ink_from_components(space: str, comps: list[float]) -> Ink:
+    """Classify a device-color fill into mono/gray/color.
+
+    ``space`` is one of 'gray', 'rgb', 'cmyk'. Any other value is treated
+    conservatively as color, since we cannot prove it is achromatic.
+    """
+    eps = _INK_EPSILON
+    if space == 'gray' and len(comps) == 1:
+        return Ink.mono if comps[0] <= eps else Ink.gray
+    if space == 'rgb' and len(comps) == 3:
+        r, g, b = comps
+        if max(r, g, b) <= eps:
+            return Ink.mono
+        if abs(r - g) <= eps and abs(g - b) <= eps:
+            return Ink.gray
+        return Ink.color
+    if space == 'cmyk' and len(comps) == 4:
+        c, m, y, k = comps
+        if c <= eps and m <= eps and y <= eps:
+            return Ink.mono if k <= eps else Ink.gray
+        return Ink.color
+    return Ink.color  # conservative-to-color
+
+
 def _normalize_stack(graphobjs):
     """Convert runs of qQ's in the stack into single graphobjs."""
     for operands, operator in graphobjs:

tests/test_pdfinfo.py

@@ -19,7 +19,7 @@ from ocrmypdf import pdfinfo
 from ocrmypdf.exceptions import InputFileError
 from ocrmypdf.helpers import IMG2PDF_KWARGS, Resolution
 from ocrmypdf.pdfinfo import Colorspace, Encoding, Ink
-from ocrmypdf.pdfinfo._contentstream import _interpret_contents
+from ocrmypdf.pdfinfo._contentstream import _ink_from_components, _interpret_contents
 from ocrmypdf.pdfinfo.layout import PDFPage
 
 warnings.filterwarnings(
@@ -296,3 +296,23 @@ def test_ink_enum_is_picklable():
     # ImageInfo crosses the worker-process boundary, so Ink must pickle.
     for member in (Ink.mono, Ink.gray, Ink.color):
         assert pickle.loads(pickle.dumps(member)) is member
+
+
+@pytest.mark.parametrize(
+    "space, comps, expected",
+    [
+        ('gray', [0.0], 'mono'),
+        ('gray', [0.263], 'gray'),
+        ('gray', [1.0], 'gray'),  # white -> gray (harmless)
+        ('rgb', [0.0, 0.0, 0.0], 'mono'),
+        ('rgb', [0.263, 0.263, 0.263], 'gray'),
+        ('rgb', [0.8, 0.2, 0.2], 'color'),
+        ('rgb', [1.0, 1.0, 1.0], 'gray'),
+        ('cmyk', [0.0, 0.0, 0.0, 0.0], 'mono'),  # white
+        ('cmyk', [0.0, 0.0, 0.0, 0.5], 'gray'),
+        ('cmyk', [0.5, 0.1, 0.0, 0.0], 'color'),
+        ('unknown', [0.5], 'color'),  # conservative fallback
+    ],
+)
+def test_ink_from_components(space, comps, expected):
+    assert _ink_from_components(space, comps) is Ink[expected]