Endless Loop When Processing Certain Large PDF with PdfFileWriter

[suokunlong]

STEPS TO REPRODUCE:

1. Download the test pdf file:
   https://suokunlong.cn/owncloud/index.php/s/bWyTHYfoMii3Yh9
   The file is named 2017-Textbook-EconomicLaw.pdf, which is 65.1MB of 511 pages.

2. Run the following code:

from PyPDF2 import PdfFileWriter, PdfFileReader

pdf_in_filename = r"/path/to/2017-Textbook-EconomicLaw.pdf"
pdf_out_filename = r"/path/to/2017-Textbook-EconomicLaw-new.pdf"

pdf_out = PdfFileWriter()
pdf_in = PdfFileReader(open(pdf_in_filename, 'rb'))

numpages = pdf_in.getNumPages()
for i in range(numpages):
    pdf_out.addPage(pdf_in.getPage(i))

with open(pdf_out_filename, 'wb') as outputStream:
    pdf_out.write(outputStream)

3. The code is running forever at the last row.

OTHER USEFUL INFORMATION:

1. I noticed that if I revise the line:

for i in range(numpages):

to:

for i in range(3):

then I will get the output very quickly.

2. I also noticed that if I open the test pdf file using evince in my Linux desktop, and print it to a new pdf file, then the above code finishes within 5s.

PyPDF2.version
'1.25.1'

[suokunlong]

Just for your information, I noticed the above bug when I was trying to add bookmarks to the pdf file using:
https://github.com/RussellLuo/pdfbookmarker.

[suokunlong]

When I run pdfinfo to the test pdf file, I see:
Encrypted: yes (print:yes copy:no change:no addNotes:no algorithm:RC4)

The re-printed pdf is not encrypted.

The issue may be related to the encryption staff.

[vstoykov]

If there is some broken image or the file is not correctly decripted then probably changes in PR #331 will raise error instead of taking forever. Can you check?

[sekrause]

Issue #329 could be related. As @vstoykov said you should check out my PR #331.

[hecko]

I have just hit this issue with embedded picture in template PDF file on Mac with PyPDF2 - 1.26.0 on Python3

[MartinThoma]

I'm closing this issue as I'm pretty certain this is solved (e.g. via #740) in recent versions of PyPDF2.

If you still encounter this issue with a recent PyPDF2 version, please let me know