Python实现PDF添加水印无效问题排查与解决请求

Let's break down why your watermark isn't appearing and fix it step by step:

Key Issues in Your Code #

• You're adding the original page to the writer before applying the watermark: Once you call pdf_writer.add_page(pdf_page), that unmodified page is locked into the output. Merging the watermark afterward doesn't update the page already in the writer.
• Hardcoded page size (letter) might not match your source PDF: If your input PDF uses a different page size (like A4), the watermark could be drawn outside the visible area.
• Merge order might hide the watermark: By default, merging can place the watermark behind the original content, making it invisible if the original page has a solid background.

Corrected Code #

import os
import io
from PyPDF2 import PdfReader, PdfWriter
from reportlab.lib.pagesizes import letter, A4
from reportlab.pdfgen import canvas

folder_path = 'D:/pdf/input/'
output_folder = 'D:/pdf/output/'
image_path = 'D:/pdf/watermark.jpg'
x, y = 50, 50
page_number = 1  # 1-based index

if not os.path.exists(output_folder):
    os.makedirs(output_folder)

for pdf_file in os.listdir(folder_path):
    if pdf_file.endswith(".pdf"):
        pdf_reader = PdfReader(open(os.path.join(folder_path, pdf_file), "rb"))
        pdf_writer = PdfWriter()

        for page_num in range(len(pdf_reader.pages)):
            pdf_page = pdf_reader.pages[page_num]
            
            # Only apply watermark to the target page
            if page_num == page_number - 1:  # convert to 0-based index
                # Get the actual page size of the source PDF page
                page_width = float(pdf_page.mediabox.width)
                page_height = float(pdf_page.mediabox.height)
                
                # Create a canvas with the same size as the source page
                packet = io.BytesIO()
                can = canvas.Canvas(packet, pagesize=(page_width, page_height))
                
                # Optional: Make sure watermark is drawn on top (set transparency if needed)
                # can.setFillAlpha(0.3)  # uncomment to make watermark semi-transparent
                can.drawImage(image_path, x, y)
                can.save()
                
                # Move to the start of the BytesIO buffer
                packet.seek(0)
                watermark_pdf = PdfReader(packet)
                
                # Merge the watermark ON TOP of the original page
                pdf_page.merge_page(watermark_pdf.pages[0], over=True)
            
            # Add the (possibly modified) page to the writer
            pdf_writer.add_page(pdf_page)
        
        with open(os.path.join(output_folder, pdf_file), "wb") as output_file:
            pdf_writer.write(output_file)

print("Watermarks added successfully to the specified pages!")

What Changed? #

1. Reordered the page addition: Now we modify the page first (if it's the target page), then add it to the writer. This ensures the watermarked page is saved.
2. Dynamic page size: We fetch the actual dimensions of the source PDF page using mediabox, so the canvas matches exactly—no more off-screen watermarks.
3. Explicit top-layer merge: Added over=True to merge_page() to ensure the watermark is drawn on top of the original content.
4. Optional transparency: Added a commented line to make the watermark semi-transparent (super useful for text/watermark images that shouldn't block content).

Additional Tips #

• Test with a small, simple PDF first to verify the watermark position.
• If your watermark is still missing, check the x/y coordinates—remember PDF uses a coordinate system where the origin (0,0) is the bottom-left corner of the page.
• Ensure your watermark image is not too large for the page dimensions.

内容的提问来源于stack exchange，提问作者Przemysław Mateusz Buzałek