How to remove a "text-based" watermark from a PDF

remove_watermark.py

## This code was generated by ChatGPT-4o after several iterations.

from PyPDF2 import PdfReader, PdfWriter
from PyPDF2.generic import StreamObject, ArrayObject, NameObject

def remove_watermark(input_pdf, output_pdf, watermark_text=r"your_watermark"):
    reader = PdfReader(input_pdf)
    writer = PdfWriter()

    for page_num, page in enumerate(reader.pages):
        try:
            print(f"Processing page {page_num + 1}/{len(reader.pages)}")
            # Check if the page has a content stream
            if "/Contents" in page:
                contents = page["/Contents"]
                if isinstance(contents, ArrayObject):
                    # Multiple content streams
                    new_streams = []
                    for content in contents:
                        stream = content.get_object()
                        raw_stream = stream.get_data().decode("utf-8", errors="ignore")

                        # Remove the watermark text
                        updated_stream = raw_stream.replace(watermark_text, "")

                        # Convert back to bytes and update
                        new_stream = StreamObject()
                        new_stream._data = updated_stream.encode("utf-8")
                        new_streams.append(new_stream)
                    # Update the page contents
                    page[NameObject("/Contents")] = ArrayObject(new_streams)
                else:
                    # Single content stream
                    stream = contents.get_object()
                    raw_stream = stream.get_data().decode("utf-8", errors="ignore")

                    # Remove the watermark text
                    updated_stream = raw_stream.replace(watermark_text, "")
                   
                    # Convert back to bytes and update
                    new_stream = StreamObject()
                    new_stream._data = updated_stream.encode("utf-8")
                    page[NameObject("/Contents")] = new_stream
            # Add the modified page to the writer
            writer.add_page(page)
        except Exception as e:
            print(f"Error on page {page_num + 1}: {e}")
            continue

    # Save the modified PDF
    with open(output_pdf, "wb") as f:
        writer.write(f)

    print(f"Watermark '{watermark_text}' removed and saved as {output_pdf}")


# Replace these paths with your actual file paths
input_pdf_path = "/path/to/pdf.pdf"
output_pdf_path = "/path/to/new.pdf"

remove_watermark(input_pdf_path, output_pdf_path)