Estimate inline samples for named CIE color spaces

README.md

@@ -46,10 +46,14 @@ helpers may discard or retain some delimiter whitespace. The second field is
 **not measured decoded pixels**, not an extracted PNG/JPEG file size, and not
 allocated image memory. Device and Indexed color spaces can be resolved directly
 or by name in the current page/Form `/Resources` dictionary. Indexed images use
-one component per sample, regardless of the palette's base space. Other custom
-spaces (including ICCBased, Separation and DeviceN), missing names, and cyclic
+one component per sample, regardless of the palette's base space. Version 0.3.0
+also resolves named CalGray (one component), CalRGB and Lab (three components)
+resources, including indirect parameter dictionaries. Only the family and
+dictionary shape are inspected; colour conversion parameters are not validated.
+Other custom spaces (including ICCBased, Separation and DeviceN), missing names, and cyclic
 aliases remain explicitly unknown. Lookup is bounded to 32 steps and never
-reads image pixels or decodes profiles. See PDF 32000-1, sections 7.8.3 and 8.9.7
+reads image pixels or decodes profiles. See PDF 32000-1, sections 7.8.3,
+8.6.5.2–8.6.5.4 and 8.9.7
 ([Adobe reference](https://opensource.adobe.com/dc-acrobat-sdk-docs/standards/pdfstandards/pdf/PDF32000_2008.pdf)).
 DCT payloads are never treated as decoded rasters. PDF 32000-1 sections 7.4.9
 and 8.9.7 prohibit JPXDecode and JBIG2Decode for inline images; JPX image XObjects

pdf_inline_audit.py

@@ -17,7 +17,7 @@
 from pypdf import PdfReader
 from pypdf.generic import ContentStream, DictionaryObject, NullObject
 
-VERSION = "0.2.1"
+VERSION = "0.3.0"
 
 
 @dataclass(frozen=True)
@@ -34,14 +34,23 @@ class ScanLimit(Exception):
 
 
 def color_components(color: Any, resources: dict | None = None) -> int | None:
-    """Resolve known device/Indexed spaces, without decoding profiles or pixels."""
+    """Resolve known component counts without decoding profiles or pixels."""
     spaces = None
     seen = set()
     for _ in range(32):
         if hasattr(color, "get_object"):
             color = color.get_object()
         if isinstance(color, list):
-            return 1 if color and str(color[0]) in ("/I", "/Indexed") else None
+            if color and str(color[0]) in ("/I", "/Indexed"):
+                return 1
+            if len(color) != 2:
+                return None
+            parameters = color[1]
+            if hasattr(parameters, "get_object"):
+                parameters = parameters.get_object()
+            if not isinstance(parameters, dict):
+                return None
+            return {"/CalGray": 1, "/CalRGB": 3, "/Lab": 3}.get(str(color[0]))
         if not isinstance(color, str):
             return None
         channels = {

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "pdf-inline-audit"
-version = "0.2.1"
+version = "0.3.0"
 description = "Reproducible corpus statistics for PDF inline image payloads"
 readme = "README.md"
 requires-python = ">=3.10"

tests/test_audit.py

@@ -171,6 +171,40 @@ def test_form_colorspace_overrides_page_resource_name(self):
         assert result['status'] == 'ok'
         assert result['images'][0]['estimated_packed_sample_bytes'] == 12
 
+    def test_named_cie_spaces_resolve_indirect_resources_and_parameters(self):
+        for family, expected in [('/CalGray', 6), ('/CalRGB', 18), ('/Lab', 18)]:
+            with self.subTest(family=family):
+                params = DictionaryObject({NameObject('/WhitePoint'): ArrayObject([
+                    NumberObject(1), NumberObject(1), NumberObject(1)])})
+                space = ArrayObject([NameObject(family), params])
+                payload = b'00' * expected + b'>'
+                path = make_pdf(self.tmp_path / f'{family[1:]}.pdf',
+                                [inline(payload, b'/W 2 /H 3 /BPC 8 /CS /Alias /F /AHx')],
+                                color_spaces=[{'/Alias': NameObject('/CIE'), '/CIE': space}])
+                writer = PdfWriter(clone_from=str(path))
+                color = writer.pages[0]['/Resources']['/ColorSpace']['/CIE'].get_object()
+                color[1] = writer._add_object(color[1])
+                writer.write(path)
+                writer.close()
+                result = scan_pdf(path)
+                assert result['status'] == 'ok'
+                image = result['images'][0]
+                assert image['estimated_packed_sample_bytes'] == expected
+                assert image['estimate_unknown_reason'] is None
+                assert image['parsed_encoded_payload_bytes'] == len(payload)
+
+    def test_cie_row_padding_and_unsupported_descriptions(self):
+        for family, expected in [('/CalGray', 4), ('/CalRGB', 10), ('/Lab', 10)]:
+            with self.subTest(family=family):
+                assert sample_estimate({'/W': 3, '/H': 2, '/BPC': 4,
+                                        '/CS': '/CIE'},
+                                       {'/ColorSpace': {'/CIE': [family, {}]}}) == (expected, None)
+        for color in [[], ['/CalGray'], ['/CalRGB', None], ['/Lab', {}, {}],
+                      ['/ICCBased', {}], ['/Separation', {}], '/CalRGB']:
+            with self.subTest(color=color):
+                assert sample_estimate({'/W': 3, '/H': 2, '/BPC': 4, '/CS': color}) == (
+                    None, 'unresolved_or_unsupported_color_space')
+
     def test_null_form_resources_match_absence_but_empty_dictionary_does_not(self):
         source = make_pdf(self.tmp_path / 'source.pdf', [b'/Fm Do /Fm Do'],
                           inline(b'00>', b'/W 1 /H 1 /BPC 8 /CS /Custom /F /AHx'),