ImageGrab can't read images copied to the clipboard from Firefox

[3d-illusions]

Tested on windows, and if images are copied from firefox, imagegrab is unable to detect pixel data. Apparently due to mime type information firefox adds?

• OS: Windows 11
• Python: 3.11.11 (main, Feb 11 2025, 07:53:58) [MSC v.1929 64 bit (AMD64)]
• Pillow: 11.2.1

Please paste here the output of running:

--- PIL CORE support ok, compiled for 11.2.1
*** TKINTER support not installed
--- FREETYPE2 support ok, loaded 2.13.3
--- LITTLECMS2 support ok, loaded 2.17
--- WEBP support ok, loaded 1.5.0
*** AVIF support not installed
--- JPEG support ok, compiled for libjpeg-turbo 3.1.0
--- OPENJPEG (JPEG2000) support ok, loaded 2.5.3
--- ZLIB (PNG/ZIP) support ok, loaded 1.3.1.zlib-ng, compiled for zlib-ng 2.2.4
--- LIBTIFF support ok, loaded 4.7.0
*** RAQM (Bidirectional Text) support not installed
*** LIBIMAGEQUANT (Quantization method) support not installed
*** XCB (X protocol) support not installed
--------------------------------------------------------------------

Code to reproduce:

img = ImageGrab.grabclipboard()

steps to reproduce:

copy image from firefox
use above code to read it.

[radarhere]

I'm unable to replicate on macOS.

Could you copy the image, run this code, and then provide us with the output?

from PIL import Image
print(Image.core.grabclipboard_win32())

[3d-illusions]

It's a bit awkward because I don't actually have firefox installed, but this is a report I've had from all firefox users.

Copilot gives this as the reason. Apparently this is the reason:

🧠 Why PIL.ImageGrab.grabclipboard() Fails with Firefox-Copied Images

The issue lies in how Firefox writes image data to the clipboard, and what ImageGrab expects:

🧩 Clipboard Format Mismatch

• PIL.ImageGrab.grabclipboard() on Windows expects image data in CF_DIB (Device Independent Bitmap) format.

• Firefox, when copying images via navigator.clipboard.write(), typically uses PNG blobs with MIME type image/png.

• This PNG format is not recognized by ImageGrab.grabclipboard() as valid image data on Windows — it doesn’t convert or interpret the MIME type, it looks for raw bitmap data.

✅ Why Other Sources Work

• Browsers like Chrome and Edge, or tools like Snipping Tool, write CF_DIB directly to the clipboard.

• These sources use native Windows clipboard APIs that ImageGrab can read without issue.

🔍 Technical Breakdown

Source | Clipboard Format Used | Works with ImageGrab.grabclipboard()? -- | -- | -- Firefox | image/png (MIME blob) | ❌ No — format not recognized Chrome/Edge | CF_DIB | ✅ Yes Snipping Tool | CF_DIB | ✅ Yes Paint / Photoshop | CF_DIB | ✅ Yes

[3d-illusions]

So perhaps MAC enforces CF_DIB, whereas windows doesn't?

[3d-illusions]

🍎 macOS Uses TIFF as Its Native Clipboard Image Format

When Firefox copies an image to the clipboard on macOS, it typically stores it as a TIFF image, which is the standard image format used by macOS's system clipboard (called the Pasteboard).

🔍 What Happens Under the Hood:

• Firefox writes the image as a PNG blob using the Clipboard API.

• macOS automatically converts or wraps that image into a NSImage object.

• The system then exposes it as TIFF data on the clipboard.

• ImageGrab.grabclipboard() on macOS is designed to look for TIFF data, not CF_DIB — and it can decode that into a PIL image.

🧠 Why This Works on macOS but Not Windows

Platform | Firefox Clipboard Format | System Clipboard Format | ImageGrab Compatibility -- | -- | -- | -- macOS | PNG → NSImage → TIFF | TIFF | ✅ Yes Windows | PNG blob (MIME) | CF_DIB expected | ❌ No

[radarhere]

Looking at the code, I think that we accept both PNG and DIB formats on Windows

Pillow/src/PIL/ImageGrab.py

Lines 132 to 151 in 162836a

	fmt, data = Image.core.grabclipboard_win32()
	if fmt == "file": # CF_HDROP
	import struct
	o = struct.unpack_from("I", data)[0]
	if data[16] == 0:
	files = data[o:].decode("mbcs").split("\0")
	else:
	files = data[o:].decode("utf-16le").split("\0")
	return files[: files.index("")]
	if isinstance(data, bytes):
	data = io.BytesIO(data)
	if fmt == "png":
	from . import PngImagePlugin
	return PngImagePlugin.PngImageFile(data)
	elif fmt == "DIB":
	from . import BmpImagePlugin
	return BmpImagePlugin.DibImageFile(data)

[radarhere]

Could you ask one of your users to run the code I mentioned? I'm hoping that you had some communication with them in order to get the python3 -m PIL.report output in your original post.

[3d-illusions]

Thanks radhere, I'll modify my software to print that, and hopefully i'll get a report. If in the meantime anyone reading this who has windows 11 and firefox, it would be great if you could test for me and report outcome here 👍