Alexander Piskun

It's possible, but it's not that easy. Later, when I have free time, I will try to implement it.

I probably can do this in next major release, for version `0.10.0` as this will require an API change and rewriting a lot of code(but probably this will make code...

You can use `to_pillow` method or look at it source: ` image = Image.frombytes( self.mode, self.size, self.data, "raw", self.mode, self.stride, )` I think `stride` value will be different from width...

@kevinlinxc ``` if __name__ == '__main__': path = "C:\\Users\\whatever\\myimg.heic" heif_file = pillow_heif.open_heif(path) image = Image.frombytes(heif_file.mode, heif_file.size, heif_file.data, "raw", heif_file.mode, heif_file.stride) image.save("hi.jpg") ``` will be the correct code. if it works,...

yeah, but not so fast, we still in process deciding how will look the deployment process, only after that all fixes will start arriving

> Does it mean that there is no option for AMD GPU to run this? AMD 7900 XT/XTX runs code from this repo without any source modifications and they are...

Does PyTorch detects your card as `cuda`? If yes, then it should be compatible. Something like: ```python >>> import torch print(torch.cuda.is_available()) ```

Even ability to build only musllinux_1_2_armv7l tag will be a great option (Alpine 3.13+) for me.

> That being said, https://nextcloud-talk.readthedocs.io/en/latest/events/#chat-message-sent should be exactly what you need? Wow, thanks, I missed that in docs. We will discuss this in a team, maybe that will be enough...

> That being said, https://nextcloud-talk.readthedocs.io/en/latest/events/#chat-message-sent should be exactly what you need? We definitely can use that, instead of current AppAPI proxy implementation, but what then we can put in `url`...