It seems that merge_page corrupts the file, but I can't determine if it actually corrupts the file.

[ssjkamei]

The files were merged using merge_page.
I am not familiar with PDF specifications, so any advice on what is broken in the file would be very helpful.

• Adobe Acrobat shows that the merged file is smaller and rotated relative to the base.
• The base is a file created with WeasyPrint 61.2 (PDF-1.7).
• The file to be merged is the file created by reportlab 4.2.0 (PDF-1.3).

The following is a simple code.

from pypdf import PdfReader, PdfWriter

with open("merge_base.pdf", "rb") as data_f:
    merge_base = PdfReader(data_f)

    with open("add_file.pdf", "rb") as data_f:
        add_file = PdfReader(data_f)

        pdf_file_writer = PdfWriter()
        for i in range(0, 1):
            pdf_file_writer.add_page(merge_base.pages[i])

        for i in range(0, 1):
            pdf_file_writer.pages[i].merge_page(
                add_file.pages[i])

        pdf_file_writer.write("JoinedPDF.pdf")

merge_base.pdf
add_file.pdf
JoinedPDF.pdf

[stefan6419846]

Which version of pypdf are you using? Did you try with the latest release? I just tested it with (nearly) the latest head and rendering works correct with some minified code:

from pypdf import PdfReader, PdfWriter

merge_base = PdfReader("merge_base.pdf")
add_file = PdfReader("add_file.pdf")

pdf_file_writer = PdfWriter()
for page in merge_base.pages:
    pdf_file_writer.add_page(page)

for i in range(len(pdf_file_writer.pages)):
    pdf_file_writer.pages[i].merge_page(add_file.pages[i])

pdf_file_writer.write("JoinedPDF.pdf")

[ssjkamei]

I was running on 4.0.2. It worked fine with 4.2.0.
Thank you. Very helpful, I was struggling for about 2 days...

When upgrading from 4.0.2 to 4.1.0 or higher, I started getting errors in parallel processing using concurrent.futures.ProcessPoolExecutor, and I forgot I was not using the latest version.

I will register this matter for another discussion.
Thanks so much for your help.