Properly fixing test_remove_images[side-by-side-subfig.pdf]

pypdf/_writer.py

@@ -2194,6 +2194,7 @@ def clean(content: ContentStream) -> None:
                     i += 1
                 else:
                     i += 1
+            content.get_data()  # this ensures ._data is rebuilt
 
         try:
             d = cast(dict, cast(DictionaryObject, page["/Resources"])["/XObject"])

pypdf/generic/_data_structures.py

@@ -854,7 +854,7 @@ def write_to_stream(
             deprecate_no_replacement(
                 "the encryption_key parameter of write_to_stream", "5.0.0"
             )
-        self[NameObject(SA.LENGTH)] = NumberObject(len(self.get_data()))
+        self[NameObject(SA.LENGTH)] = NumberObject(len(self._data))
         DictionaryObject.write_to_stream(self, stream)
         del self[SA.LENGTH]
         stream.write(b"\nstream\n")

tests/test_writer.py

@@ -312,6 +312,8 @@ def test_remove_images(pdf_file_path, input_path):
     page = reader.pages[0]
     writer.insert_page(page, 0)
     writer.remove_images()
+    page_contents_stream = writer.pages[0]["/Contents"]._data
+    assert len(page_contents_stream.strip())
 
     # finally, write "output" to pypdf-output.pdf
     with open(pdf_file_path, "wb") as output_stream: