Fixing test_iss1601

pypdf/generic/_data_structures.py

@@ -988,6 +988,7 @@ class ContentStream(DecodedStreamObject):
     * when .set_data() is called, ._operations is set to None
     * when .operations is set, ._data is set to None
     """
+
     def __init__(
         self,
         stream: Any,
@@ -1232,7 +1233,7 @@ def write_to_stream(
         self, stream: StreamType, encryption_key: Union[None, str, bytes] = None
     ) -> None:
         if not self._data and self._operations:
-            self.get_data()  # this ensures ._data is rebuilt for ContentStream
+            self.get_data()  # this ensures ._data is rebuilt
         super().write_to_stream(stream, encryption_key)
 
 

tests/__init__.py

@@ -87,3 +87,20 @@ def get_object(self) -> "DummyObj":
 
     def get_reference(self, obj):
         return IndirectObject(idnum=1, generation=1, pdf=self)
+
+
+def is_sublist(child_list, parent_list):
+    """
+    Check if child_list is a sublist of parent_list, with respect to
+    * elements order
+    * elements repetition
+
+    Elements are compared using `==`
+    """
+    if len(child_list) == 0:
+        return True
+    if len(parent_list) == 0:
+        return False
+    if parent_list[0] == child_list[0]:
+        return is_sublist(child_list[1:], parent_list[1:])
+    return is_sublist(child_list, parent_list[1:])

tests/test_utils.py

@@ -25,6 +25,8 @@
 )
 from pypdf.errors import DeprecationError, PdfReadError, PdfStreamError
 
+from . import is_sublist
+
 TESTS_ROOT = Path(__file__).parent.resolve()
 PROJECT_ROOT = TESTS_ROOT.parent
 RESOURCE_ROOT = PROJECT_ROOT / "resources"
@@ -351,3 +353,21 @@ def test_parse_datetime_err():
         parse_iso8824_date("D:20210408T054711Z")
     assert ex.value.args[0] == "Can not convert date: D:20210408T054711Z"
     assert parse_iso8824_date("D:20210408054711").tzinfo is None
+
+
+def test_is_sublist():
+    # Basic checks:
+    assert is_sublist([0, 1], [0, 1, 2]) is True
+    assert is_sublist([0, 2], [0, 1, 2]) is True
+    assert is_sublist([1, 2], [0, 1, 2]) is True
+    assert is_sublist([0, 3], [0, 1, 2]) is False
+    # Ensure order is checked:
+    assert is_sublist([1, 0], [0, 1, 2]) is False
+    # Ensure duplicates are handled:
+    assert is_sublist([0, 1, 1], [0, 1, 1, 2]) is True
+    assert is_sublist([0, 1, 1], [0, 1, 2]) is False
+    # Edge cases with empty lists:
+    assert is_sublist([], [0, 1, 2]) is True
+    assert is_sublist([0, 1], []) is False
+    # Self-sublist edge case:
+    assert is_sublist([0, 1, 2], [0, 1, 2]) is True

tests/test_writer.py

@@ -28,7 +28,7 @@
     TextStringObject,
 )
 
-from . import get_data_from_url
+from . import get_data_from_url, is_sublist
 
 TESTS_ROOT = Path(__file__).parent.resolve()
 PROJECT_ROOT = TESTS_ROOT.parent
@@ -1238,23 +1238,20 @@ def test_iss1601():
     url = "https://github.com/py-pdf/pypdf/files/10579503/badges-38.pdf"
     name = "badge-38.pdf"
     reader = PdfReader(BytesIO(get_data_from_url(url, name=name)))
+    original_cs_operations = ContentStream(reader.pages[0].get_contents(), reader).operations
     writer = PdfWriter()
     page_1 = writer.add_blank_page(
         reader.pages[0].mediabox[2], reader.pages[0].mediabox[3]
     )
     page_1.merge_transformed_page(reader.pages[0], Transformation())
-    assert (
-        ContentStream(reader.pages[0].get_contents(), reader).get_data()
-        in page_1.get_contents().get_data()
-    )
+    page_1_cs_operations = page_1.get_contents().operations
+    assert is_sublist(original_cs_operations, page_1_cs_operations)
     page_1 = writer.add_blank_page(
         reader.pages[0].mediabox[2], reader.pages[0].mediabox[3]
     )
     page_1.merge_page(reader.pages[0])
-    assert (
-        ContentStream(reader.pages[0].get_contents(), reader).get_data()
-        in page_1.get_contents().get_data()
-    )
+    page_1_cs_operations = page_1.get_contents().operations
+    assert is_sublist(original_cs_operations, page_1_cs_operations)
 
 
 def test_attachments():
@@ -1539,7 +1536,7 @@ def test_watermark():
 
 
 @pytest.mark.enable_socket()
-@pytest.mark.timeout(2)  # this was a lot slower before PR #2086
+@pytest.mark.timeout(4)  # this was a lot slower before PR #2086
 def test_watermarking_speed():
     url = "https://github.com/py-pdf/pypdf/files/11985889/bg.pdf"
     name = "bgwatermark.pdf"