Change in behavior from 3.10.0 to 3.11.0 #1953
Description
The minimal example below works fine with pypdf 3.10.0, but fails beginning with 3.11.0. The expectation is that the code would work in 3.11.0, or else a deprecation message would be displayed.
Environment
Python 3.11.4
Code + PDF
No PDF is generated due to the error.
from io import BytesIO
from pathlib import Path
from pypdf import PdfReader, PdfWriter, Transformation
LETTER_W = 612
LETTER_H = 792
def create_song_sheet(input_file: Path, output_file: Path) -> None:
pdf_pages = []
reader = PdfReader(input_file)
pdf_page_count = len(reader.pages)
for page in reader.pages:
new_letter_page_writer = PdfWriter()
## scale the PDF to fit a 8.5 x 11 inch US Letter page
## but preserve the aspect ratio by setting the height
## to 11, scaling the width accordingly, and centering
## the result on the page
new_letter_page = new_letter_page_writer.add_blank_page(
width=LETTER_W, height=LETTER_H
)
## set the mediabox width and height so that things don't
## get chopped off when we center the scaled PDF
source_width = page.mediabox.width
source_height = page.mediabox.height
source_aspect = source_width / source_height
new_width = LETTER_H * source_aspect
width_offset = (LETTER_W - new_width) / 2
new_letter_page.merge_transformed_page(
page, Transformation((1, 0, 0, 1, width_offset, 0))
)
tmp_pdf = BytesIO()
new_letter_page_writer.write(tmp_pdf)
new_letter_page_writer.close()
pdf_pages.append(
{"file": tmp_pdf, "name": input_file, "pagecount": pdf_page_count}
)
merger = PdfWriter()
for pdf_page in pdf_pages:
## ERROR OCCURS HERE ON PROCESSING THE FIRST PAGE ##
merger.append(pdf_page["file"])
merger.write(output_file)
merger.close()
create_song_sheet(
Path("/home/moose/Github/py-pdf/pypdf/resources/crazyones.pdf"),
Path("out-1953.pdf"),
)Traceback
Here is the traceback from the file above
Traceback (most recent call last):
File "/home/moose/Github/py-pdf/pypdf/issue-1953.py", line 55, in <module>
create_song_sheet(
File "/home/moose/Github/py-pdf/pypdf/issue-1953.py", line 49, in create_song_sheet
merger.append(pdf_page["file"])
File "/home/moose/Github/py-pdf/pypdf/pypdf/_writer.py", line 2768, in append
self.merge(
File "/home/moose/Github/py-pdf/pypdf/pypdf/_utils.py", line 432, in wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/home/moose/Github/py-pdf/pypdf/pypdf/_writer.py", line 2849, in merge
srcpages[pg.indirect_reference.idnum] = self.add_page(
^^^^^^^^^^^^^^
File "/home/moose/Github/py-pdf/pypdf/pypdf/_writer.py", line 389, in add_page
return self._add_page(page, list.append, excluded_keys)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/moose/Github/py-pdf/pypdf/pypdf/_writer.py", line 322, in _add_page
page = cast("PageObject", page_org.clone(self, False, excluded_keys))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/moose/Github/py-pdf/pypdf/pypdf/generic/_data_structures.py", line 186, in clone
d__._clone(self, pdf_dest, force_duplicate, ignore_fields)
File "/home/moose/Github/py-pdf/pypdf/pypdf/generic/_data_structures.py", line 286, in _clone
v.clone(pdf_dest, force_duplicate, ignore_fields)
File "/home/moose/Github/py-pdf/pypdf/pypdf/generic/_base.py", line 294, in clone
obj.clone(pdf_dest, force_duplicate, ignore_fields)
File "/home/moose/Github/py-pdf/pypdf/pypdf/generic/_data_structures.py", line 102, in clone
arr.append(dup.indirect_reference)
^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'DecodedStreamObject' object has no attribute 'indirect_reference'