So I am trying to make a tool that basically does what iLovePDF does (cut a PDF into user defined sets of pages, and then compresses them to a user defined level).

In my code I have the following snippet:

# Compression (in the writer)
for page in writer.pages:
    if compress:
        if compress_level>1:
            for img in page.images:
                if img.image.mode != "RGBA":
                    img.replace(img.image, quality=95)
        page.compress_content_streams()

However, for two images I get the following warning:

image and mask size not matching: {'/Type': '/XObject', '/Subtype': '/Image', '/Width': 2, '/Height': 2, '/ColorSpace': ['/Indexed', '/DeviceRGB', 1, b'x00x00x00xffxffxff'], '/BitsPerComponent': 1, '/Interpolate': False, '/SMask': IndirectObject(73, 0, 1823765420048)}

and

image and mask size not matching: {'/Type': '/XObject', '/Subtype': '/Image', '/Width': 2, '/Height': 2, '/ColorSpace': ['/Indexed', '/DeviceRGB', 1, b'x00x00x00xffxffxff'], '/BitsPerComponent': 1, '/Interpolate': False, '/SMask': IndirectObject(76, 0, 1823765420048)}

Here are the images before the compression:

and

And here after the compression:

and

I’ve not been able to find a good fix for this. I’ve seen some things with converting RGBA to RGB, but to no avail.

As you can see in the code, I don’t mind skipping the image replacing if this problem occurs (which I already did with RGBA images), but I don’t know what to catch, as img.replace() doesn’t return anything (at least not when I tried), and I also am not sure I know what code to insert even if I could catch it.

Lastly, if you know of some much easier/better way to do this whole thing, that’s also a welcome answer.