import argparse
import fitz  # PyMuPDF

def remove_text_from_pdf(input_pdf_path, output_pdf_path, watermark):
    doc = fitz.open(input_pdf_path)
    for page in doc:
        text_instances = page.search_for(watermark)
        for text in text_instances:
            print(f'Rect found: {text}')
            page.add_redact_annot(text)

        page.apply_redactions(fitz.PDF_REDACT_IMAGE_NONE)

    doc.save(output_pdf_path, garbage=4, deflate=True, deflate_images=True, deflate_fonts=True)
    doc.close()
    print(f"Processed PDF saved as '{output_pdf_path}'.")

if __name__ == "__main__":
    parser = argparse.ArgumentParser(description="Remove specified text and links from a PDF.")
    parser.add_argument("input_pdf", help="Path to the source PDF file.")
    parser.add_argument("output_pdf", help="Path to the destination PDF file.")
    parser.add_argument("watermark", help="Text to remove from the output PDF file.")
    args = parser.parse_args()
    remove_text_from_pdf(args.input_pdf, args.output_pdf, args.watermark)