Here are the options as far as handling GIF frames with LCTs, transparency, and disposal 0/1:
- Extract frame as they are with transparency .
- Have all colors from previous frame(s) which are in current frame's color table carry over with transparency in holes.
- Expand frames' palettes as needed.
All three would mean currently broken GIF files would display properly, though 3 would need extra work when encoding to put back in transparency and squish the color tables and extra work as far as decoupling GIF frames from the GIF's data.
You can already pass a sequence of ints for disposal - https://pillow.readthedocs.io/en/stable/handbook/image-file-formats.html#saving
Pass a single integer for a constant disposal, or a list or tuple to set the disposal for each frame separately
Not sure what you're asking.
At the moment, the transparency for the current GIF frame is stored in info["transparency"].
When saving, that value is written as the transparency index.
If you're asking for a strategy to specify transparency on an individual frame basis when saving, passing a sequence of ints to info["transparency"] sounds like the most consistent approach.
If using the info dictionary is troublesome to you, be aware that Pillow copies keyword arguments from save() in as the encoderinfo dictionary, and often allows them to override info keys.
Navigating to Settings, the Third Party Packages tab, entering Pillow and hitting Ok, I was able to get it to install
Hey im getting an error (AttributeError: module 'PIL._imaging' has no attribute 'png_encoder') when trying to run img.tobytes("png","RGB") on Pillow version 8.3.1
and Python 3.8.2
EDIT: Im on windows, was told y'all may need it
Full error:
Traceback (most recent call last):
File "C:\Users\ibx34\Documents\programming\Python\teddy-bear\venv\lib\site-packages\PIL\Image.py", line 453, in _getencoder
encoder = getattr(core, encoder_name + "_encoder")
AttributeError: module 'PIL._imaging' has no attribute 'png_encoder'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "C:\Users\ibx34\Documents\programming\Python\teddy-bear\cogs\owner.py", line 245, in _eval
ret = await func()
File "<string>", line 17, in func
File "C:\Users\ibx34\Documents\programming\Python\teddy-bear\venv\lib\site-packages\PIL\Image.py", line 747, in tobytes
e = _getencoder(self.mode, encoder_name, args)
File "C:\Users\ibx34\Documents\programming\Python\teddy-bear\venv\lib\site-packages\PIL\Image.py", line 455, in _getencoder
raise OSError(f"encoder {encoder_name} not available") from e
OSError: encoder png not availableHi. If you look at https://pillow.readthedocs.io/en/stable/reference/Image.html#PIL.Image.Image.tobytes, you will see there is a warning.
This method returns the raw image data from the internal storage. For compressed image data (e.g. PNG, JPEG) use save(), with a BytesIO parameter for in-memory data.
I think this is the trap you have fallen into. "png" is not the name of one of our encoders. Instead, I expect you want to do this -
>>> from PIL import Image
>>> im = Image.new("RGB", (100, 100))
>>> import io
>>> fp = io.BytesIO()
>>> im.save(fp, "PNG")
>>> fp.getvalue()
b'\x89PNG\r\n\x1a\n\x00\x00\x00\rIHDR\x00\x00\x00d\x00\x00\x00d\x08\x02\x00\x00\x00\xff\x80\x02\x03\x00\x00\x004IDATx\x9c\xed\xc1\x01\r\x00\x00\x00\xc2\xa0\xf7Om\x0e7\xa0\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00~\x0cu\x94\x00\x01\xa8P\xf29\x00\x00\x00\x00IEND\xaeB`\x82'
Uh this is somewhat unexpected. A quick look into the code makes me worried wrt trying to implement this myself. And the feature request is quite old too... It might be easier doable for me to use something else (numpy?) as underlying data structure and add R32G32B32 open and save routines on top of that.
BGR;32 does not really seem to exist and might be a 8bit per component format with padding, or?
if (strcmp(mode, "BGR;32") == 0) {
/ EXPERIMENTAL /
/ 32-bit reversed true colour /
im->bands = 1;
im->pixelsize = 4;
im->linesize = (xsize*4 + 3) & -4;
im->type = IMAGING_TYPE_SPECIAL;
https://stackoverflow.com/questions/6444548/how-do-i-get-the-picture-size-with-pil