MAINT: Move tests for images to own module (#2071)

tests/test_images.py

@@ -0,0 +1,109 @@
+"""
+Tests which ensure that image extraction works properly go here.
+
+Typically, tests in here should compare the extracted images count, names,
+and/or the actual image data with the expected value.
+"""
+
+from io import BytesIO
+from pathlib import Path
+
+import pytest
+
+from pypdf import PdfReader
+from pypdf._page import PageObject
+
+from . import get_pdf_from_url
+
+TESTS_ROOT = Path(__file__).parent.resolve()
+PROJECT_ROOT = TESTS_ROOT.parent
+RESOURCE_ROOT = PROJECT_ROOT / "resources"
+SAMPLE_ROOT = PROJECT_ROOT / "sample-files"
+
+
+@pytest.mark.enable_socket()
+def test_image_new_property():
+ url = "https://github.com/py-pdf/pypdf/files/11219022/pdf_font_garbled.pdf"
+ name = "pdf_font_garbled.pdf"
+ reader = PdfReader(BytesIO(get_pdf_from_url(url, name=name)))
+ assert reader.pages[0].images.keys() == [
+ "/I0",
+ "/I1",
+ "/I2",
+ "/I3",
+ "/I4",
+ "/I5",
+ "/I6",
+ "/I7",
+ "/I8",
+ "/I9",
+ ["/TPL1", "/Image5"],
+ ["/TPL2", "/Image53"],
+ ["/TPL2", "/Image37"],
+ ["/TPL2", "/Image49"],
+ ["/TPL2", "/Image51"],
+ ["/TPL2", "/Image39"],
+ ["/TPL2", "/Image57"],
+ ["/TPL2", "/Image55"],
+ ["/TPL2", "/Image43"],
+ ["/TPL2", "/Image30"],
+ ["/TPL2", "/Image22"],
+ ["/TPL2", "/Image41"],
+ ["/TPL2", "/Image47"],
+ ["/TPL2", "/Image45"],
+ ["/TPL3", "/Image65"],
+ ["/TPL3", "/Image30"],
+ ["/TPL3", "/Image61"],
+ ["/TPL4", "/Image30"],
+ ["/TPL5", "/Image30"],
+ ["/TPL6", "/Image30"],
+ ["/TPL7", "/Image30"],
+ ["/TPL8", "/Image30"],
+ ["/TPL9", "/Image30"],
+ ["/TPL10", "/Image30"],
+ ["/TPL11", "/Image30"],
+ ["/TPL12", "/Image30"],
+ ]
+ assert len(reader.pages[0].images.items()) == 36
+ assert reader.pages[0].images[0].name == "I0.png"
+ assert len(reader.pages[0].images[-1].data) == 15168
+ assert reader.pages[0].images["/TPL1", "/Image5"].image.format == "JPEG"
+ assert (
+ reader.pages[0].images["/I0"].indirect_reference.get_object()
+ == reader.pages[0]["/Resources"]["/XObject"]["/I0"]
+ )
+ list(reader.pages[0].images[0:2])
+ with pytest.raises(TypeError):
+ reader.pages[0].images[b"0"]
+ with pytest.raises(IndexError):
+ reader.pages[0].images[9999]
+ # just for test coverage:
+ with pytest.raises(KeyError):
+ reader.pages[0]._get_image(["test"], reader.pages[0])
+ assert list(PageObject(None, None).images) == []
+
+
+@pytest.mark.parametrize(
+ ("src", "page_index", "image_key", "expected"),
+ [
+ (
+ SAMPLE_ROOT / "009-pdflatex-geotopo/GeoTopo.pdf",
+ 23,
+ "/Im2",
+ SAMPLE_ROOT / "009-pdflatex-geotopo/page-23-Im2.png",
+ ),
+ # (SAMPLE_ROOT / "009-pdflatex-geotopo/GeoTopo.pdf", 30, '/Fm22',
+ # SAMPLE_ROOT / "009-pdflatex-geotopo/page-30-Fm22.png"),
+ ],
+)
+@pytest.mark.samples()
+def test_image_extraction(src, page_index, image_key, expected):
+ reader = PdfReader(src)
+ actual_image = reader.pages[page_index].images[image_key]
+ if not expected.exists():
+ # A little helper for test generation
+ with open(f"page-{page_index}-{actual_image.name}", "wb") as fp:
+ fp.write(actual_image.data)
+ with open(expected, "rb") as fp:
+ expected_data = fp.read()
+ assert actual_image.data == expected_data

tests/test_page.py

@@ -1184,31 +1184,6 @@ def test_pdf_pages_missing_type():
  writer.pages[0]
 
 
-@pytest.mark.enable_socket()
-def test_image_new_property():
- url = "https://github.com/py-pdf/pypdf/files/11219022/pdf_font_garbled.pdf"
- name = "pdf_font_garbled.pdf"
- reader = PdfReader(BytesIO(get_pdf_from_url(url, name=name)))
- reader.pages[0].images.keys()
- reader.pages[0].images.items()
- reader.pages[0].images[0].name
- reader.pages[0].images[-1].data
- reader.pages[0].images["/TPL1", "/Image5"].image
- assert (
- reader.pages[0].images["/I0"].indirect_reference.get_object()
- == reader.pages[0]["/Resources"]["/XObject"]["/I0"]
- )
- list(reader.pages[0].images[0:2])
- with pytest.raises(TypeError):
- reader.pages[0].images[b"0"]
- with pytest.raises(IndexError):
- reader.pages[0].images[9999]
- # just for test coverage:
- with pytest.raises(KeyError):
- reader.pages[0]._get_image(["test"], reader.pages[0])
- assert list(PageObject(None, None).images) == []
-
-
 @pytest.mark.samples()
 def test_compression():
  """Test for issue #1897"""

tests/test_pdfa.py

@@ -32,6 +32,7 @@ def document_information_has_analoguos_xml(src: BytesIO) -> bool:
  return document_information_has_analoguos_xml(src)
 
 
+@pytest.mark.samples()
 @pytest.mark.parametrize(
  ("src", "diagnostic_write_name"),
  [