Blank pages in output file using PyPDF2 in Python script

[tore_nielsen]

Hi 

I'm trying to split pdf pages vertically in 2, and merge all pages to one file, using script below.

I get an output file where there is double pages of original, but they are all blank?

My script is:

import PyPDF2

def split_pdf(input_pdf, output_pdf):
    pdf_reader = PyPDF2.PdfReader(input_pdf)
    pdf_writer = PyPDF2.PdfWriter()

    for page_num in range(len(pdf_reader.pages)):
        page = pdf_reader.pages[page_num]

        width = page.mediabox.upper_right[0]
        height = page.mediabox.upper_right[1] - page.mediabox.lower_left[1]

        half_width = width / 2

        page_left = pdf_writer.add_blank_page(width=half_width, height=height)
        page_left.merge_page(page)

        page_right = pdf_writer.add_blank_page(width=half_width, height=height)
        page_right.merge_page(page)
        page_right.mediabox.lower_left = (half_width, page_right.mediabox.lower_left[1])
        page_right.trimbox.lower_left = (half_width, page_right.trimbox.lower_left[1])

    with open(output_pdf, 'wb') as output_file:
        pdf_writer.write(output_file)

# Eksempel på brug
input_pdf_path = r"C:\Users\xxxxx\Desktop\Pdf splitter - test\didkdk_annualreport_2022_di.pdf"
output_pdf_path = r"C:\Users\xxxxx\Desktop\Pdf splitter - test\didkdk_annualreport_2022_di_split.pdf"
split_pdf(input_pdf_path, output_pdf_path)

Can anyone help me figuring out what is wrong in the code?

Thanks in advance

[Hammad_Rashid]

Here are some modifications you can make to address these issues:

1. Initialize PdfWriter correctly: Change pdf_writer = PyPDF2.PdfWriter() to pdf_writer = PyPDF2.PdfFileWriter().

2. Adjusting page_right coordinates: After merging the page into page_right, you need to adjust the coordinates of page_right to start from the right half of the page. You can do this by updating the page_right.mediabox.lower_left and page_right.trimbox.lower_left.

Here's the modified script:

import PyPDF2

def split_pdf(input_pdf, output_pdf):
pdf_reader = PyPDF2.PdfReader(input_pdf)
pdf_writer = PyPDF2.PdfFileWriter()

for page_num in range(len(pdf_reader.pages)):
page = pdf_reader.pages[page_num]

width = page.mediabox.upper_right[0]
height = page.mediabox.upper_right[1] - page.mediabox.lower_left[1]

half_width = width / 2

page_left = pdf_writer.addBlankPage(width=half_width, height=height)
page_left.merge_page(page)

page_right = pdf_writer.addBlankPage(width=half_width, height=height)
page_right.merge_page(page)
page_right.mediabox.lower_left = (half_width, page_right.mediabox.lower_left[1])
page_right.trimBox.lower_left = (half_width, page_right.trimBox.lower_left[1])

with open(output_pdf, 'wb') as output_file:
pdf_writer.write(output_file)

# Example of usage
input_pdf_path = r"C:\Users\xxxxx\Desktop\Pdf splitter - test\didkdk_annualreport_2022_di.pdf"
output_pdf_path = r"C:\Users\xxxxx\Desktop\Pdf splitter - test\didkdk_annualreport_2022_di_split.pdf"
split_pdf(input_pdf_path, output_pdf_path)

These changes should help fix the issue you're facing.

[caltang]

When you're merging the left and right pages, you're merging the original page into both the left and right pages.

Try this

import PyPDF2

def split_pdf(input_pdf, output_pdf):
    pdf_reader = PyPDF2.PdfReader(input_pdf)
    pdf_writer = PyPDF2.PdfWriter()

    for page_num in range(len(pdf_reader.pages)):
        page = pdf_reader.pages[page_num]

        width = page.mediabox.upper_right[0]
        height = page.mediabox.upper_right[1] - page.mediabox.lower_left[1]

        half_width = width / 2

        page_left = pdf_writer.add_blank_page(width=half_width, height=height)
        page_left.merge_page(page)

        page_right = pdf_writer.add_blank_page(width=half_width, height=height)
        page_right.merge_page(page.copy())  # Create a copy of the original page
        page_right.mediabox.lower_left = (half_width, page_right.mediabox.lower_left[1])
        page_right.trimbox.lower_left = (half_width, page_right.trimbox.lower_left[1])

    with open(output_pdf, 'wb') as output_file:
        pdf_writer.write(output_file)

# Example usage
input_pdf_path = r"C:\Users\xxxxx\Desktop\Pdf splitter - test\didkdk_annualreport_2022_di.pdf"
output_pdf_path = r"C:\Users\xxxxx\Desktop\Pdf splitter - test\didkdk_annualreport_2022_di_split.pdf"
split_pdf(input_pdf_path, output_pdf_path)

[tore_nielsen]

Hi Caltang

I get this error when run the script?

AttributeError                            Traceback (most recent call last)
<ipython-input-12-fa8bf0cb6d10> in <module>
     27 input_pdf_path = r"C:\Users\DKTINN\Desktop\Pdf splitter - test\didkdk_annualreport_2022_di.pdf"
     28 output_pdf_path = r"C:\Users\DKTINN\Desktop\Pdf splitter - test\didkdk_annualreport_2022_di_split.pdf"
---> 29 split_pdf(input_pdf_path, output_pdf_path)

<ipython-input-12-fa8bf0cb6d10> in split_pdf(input_pdf, output_pdf)
     17 
     18         page_right = pdf_writer.add_blank_page(width=half_width, height=height)
---> 19         page_right.merge_page(page.copy())  # Create a copy of the original page
     20         page_right.mediabox.lower_left = (half_width, page_right.mediabox.lower_left[1])
     21         page_right.trimbox.lower_left = (half_width, page_right.trimbox.lower_left[1])

~\AppData\Roaming\Python\Python38\site-packages\PyPDF2\_page.py in merge_page(self, page2, expand)
    683                 expanded to accommodate the dimensions of the page to be merged.
    684         """
--> 685         self._merge_page(page2, expand=expand)
    686 
    687     def mergePage(self, page2: "PageObject") -> None:  # pragma: no cover

~\AppData\Roaming\Python\Python38\site-packages\PyPDF2\_page.py in _merge_page(self, page2, page2transformation, ctm, expand)
    757             )
    758 
--> 759         page2content = page2.get_contents()
    760         if page2content is not None:
    761             page2content = ContentStream(page2content, self.pdf)

AttributeError: 'dict' object has no attribute 'get_contents'

[caltang]

Hmm... I think there's a version compatibility issue with the PyPDF2 library.

Can you use this:

pip install PyMuPDF

Then use this:

import fitz  # PyMuPDF

def split_pdf(input_pdf, output_pdf):
    pdf_document = fitz.open(input_pdf)
    pdf_writer = fitz.open()

    for page_num in range(pdf_document.page_count):
        page = pdf_document[page_num]

        width = page.rect.width
        height = page.rect.height

        half_width = width / 2

        page_left = pdf_writer.new_page(width=half_width, height=height)
        page_left.show_pdf_page(rect=page.rect, pdf=pdf_document, page_number=page_num)

        page_right = pdf_writer.new_page(width=half_width, height=height)
        page_right.show_pdf_page(rect=fitz.Rect(half_width, 0, width, height), pdf=pdf_document, page_number=page_num)

    pdf_writer.save(output_pdf)
    pdf_writer.close()

I'm not sure if you're really strict on PyPDF2 but let me know if this works?

[tore_nielsen]

Thanks for your help

I'm still getting some errors?

My pdf document is a aprox 1500 pages, where the format is Landscape, where I want to split it vertically, just to make it clear :-)

import fitz  # PyMuPDF

def split_pdf(input_pdf, output_pdf):
    pdf_document = fitz.open(input_pdf)
    pdf_writer = fitz.open()

    for page_num in range(pdf_document.page_count):
        page = pdf_document[page_num]

        width = page.rect.width
        height = page.rect.height

        half_width = width / 2

        page_left = pdf_writer.new_page(width=half_width, height=height)
        page_left.show_pdf_page(rect=page.rect, pdf=pdf_document, page_number=page_num)

        page_right = pdf_writer.new_page(width=half_width, height=height)
        page_right.show_pdf_page(rect=fitz.Rect(half_width, 0, width, height), pdf=pdf_document, page_number=page_num)

    pdf_writer.save(output_pdf)
    pdf_writer.close()
    
# Examble of use
input_pdf_path = r"C:\Users\xxxxxx\Desktop\Pdf splitter - test\didkdk_annualreport_2022_di.pdf"
output_pdf_path = r"C:\Users\xxxxxx\Desktop\Pdf splitter - test\didkdk_annualreport_2022_di_split.pdf"
split_pdf(input_pdf_path, output_pdf_path)


----------------------------------------------------------------------------------------------

ValueError                                Traceback (most recent call last)
<ipython-input-35-709476f53b96> in <module>
     25 input_pdf_path = r"C:\Users\xxxxxx\Desktop\Pdf splitter - test\didkdk_annualreport_2022_di.pdf"
     26 output_pdf_path = r"C:\Users\xxxxxx\Desktop\Pdf splitter - test\didkdk_annualreport_2022_di_split.pdf"
---> 27 split_pdf(input_pdf_path, output_pdf_path)

<ipython-input-35-709476f53b96> in split_pdf(input_pdf, output_pdf)
     14 
     15         page_left = pdf_writer.new_page(width=half_width, height=height)
---> 16         page_left.show_pdf_page(rect=page.rect, pdf=pdf_document, page_number=page_num)
     17 
     18         page_right = pdf_writer.new_page(width=half_width, height=height)

~\AppData\Roaming\Python\Python38\site-packages\fitz\utils.py in show_pdf_page(*args, **kwargs)
    122     """
    123     if len(args) not in (3, 4):
--> 124         raise ValueError("bad number of positional parameters")
    125     pno = None
    126     if len(args) == 3:

ValueError: bad number of positional parameters

[caltang]

Hmm, I'm stuck on this myself. Maybe Stackoverflow is the better place for this?

[tore_nielsen]

I found a solution after all.

It wasn't as advanced as I thought thought it would be.

import PyPDF2
import copy

def split_pdf(input_pdf, output_pdf):
        pdf_reader = PyPDF2.PdfReader(input_pdf)
        pdf_writer = PyPDF2.PdfWriter()

        for page_num in range(len(pdf_reader.pages)):
                    page = pdf_reader.pages[page_num]
                    
                    page_left = copy.copy(page)
                    page_right = copy.copy(page)
                    
                    if page.rotation == 90:                    
                        page_left.Cropbox.top = page.Cropbox.top / 2
                        page_right.Cropbox.bottom = page.Cropbox.top / 2
                    else:
                        page_left.Cropbox.right = page.Cropbox.right / 2
                        page_right.Cropbox.left = page.Cropbox.left / 2
                        
                    pdf_writer.add_page(page_left)
                    pdf_writer.add_page(page_right)

        with open(output_pdf, 'wb') as output_file:
            pdf_writer.write(output_file)