[Python] How to split book-like pdf pages into half with PyPDF2?

I am using the python package PyPDF2 to split a pdf of a scanned book. Here is my code.

# This is a programme to split the page into half

from PyPDF2 import PdfReader, PdfWriter

reader = PdfReader("mypdf.pdf")
writer = PdfWriter()
number_of_pages = len(reader.pages)
print(f"Original number of page: {number_of_pages}")

# Duplicate the page
for i in range(number_of_pages):
writer.add_page(reader.pages)
writer.add_page(reader.pages)
i += 1

# Save the doubled pdf first
with open("mypdf_splitted.pdf", "wb") as fp:
writer.write(fp)

# Stupidly re-open the file again and split it
reader = PdfReader("mypdf_splitted.pdf")
writer = PdfWriter()
number_of_pages = len(reader.pages)
print(f"Duplicated number of page: {number_of_pages}")

# Get the dimension of every page and crop it
for i in range(number_of_pages):
page = reader.pages
x, y = page.mediabox.getUpperRight_x(), page.mediabox.getUpperRight_y()
if i % 2 == 0:
# print(f'page {i}, width = {x}, height = {y}')
page.mediabox.lowerLeft = (0, 0)
page.mediabox.upperRight = (x / 2, y)
else:
# print(f'page {i}, width = {x}, height = {y}')
page.mediabox.lowerLeft = (x / 2, 0)
page.mediabox.upperRight = (x, y)
writer.add_page(page)

with open("mypdf_splitted.pdf", "wb") as fp:
writer.write(fp)


The code actually works. However, I want to know is there any way to double and crop the pages at once, i.e. without the stupid re-opening.

I tried to combine it. But the output file is of doubled right pages only.

The original scanned books:

page 1 -- 1, 2
page 2 -- 3, 4
page 3 -- 5, 6

What I want is:

page 1 -- 1 page 2 -- 2 page 3 -- 3 (My code works with a dummy write-read)

What I got is: page 1 -- 2 page 2 -- 2 page 3 -- 4 (Only the even (or right) pages are written)

It seems the pages object is not that trivial to handle. After I doubled the pages, the action of cropping the left is overwritten by that of the right.

Continue reading...