Watermark a PDF file (all pages)

watermarker.py

import os
from PyPDF2 import PdfReader, PdfWriter


def main():
    mark_it_up("watermark.pdf", "/source/", "/output/")


def mark_it_up(watermark_file, pdf_folder, output_folder):
    """
    watermark_file: str - the file that is the watermark
    pdf_folder: str - the folder containing the PDF files to process
    output_folder: str - the folder to save the watermarked PDFs
    """
    os.makedirs(output_folder, exist_ok=True)

    # Iterate through all PDF files in the folder
    for filename in os.listdir(pdf_folder):
        if filename.endswith(".pdf"):
            # Full path to the input PDF
            input_pdf_path = os.path.join(pdf_folder, filename)

            # Read the input PDF
            input_pdf = PdfReader(input_pdf_path)

            # Create a PdfWriter for the output PDF
            output_pdf = PdfWriter()

            # Open the watermark PDF
            with open(watermark_file, "rb") as watermark_file_obj:
                watermark_pdf = PdfReader(watermark_file_obj)
                watermark_page = watermark_pdf.pages[0]

                # Merge the watermark with each page in the input PDF
                for page in input_pdf.pages:
                    page.merge_page(watermark_page)
                    output_pdf.add_page(page)

            # Define the output file name based on the original file
            output_pdf_path = os.path.join(output_folder, f"output_{filename}")
            try:
                # Write the watermarked PDF to the output file
                with open(output_pdf_path, "wb") as output_pdf_file:
                    output_pdf.write(output_pdf_file)
            except Exception as e:
                print(f"Error processing {filename}: {e}")
                break  # Stop processing this folder
    print("Watermarking and merging complete.")


if __name__ == "__main__":
    main()