scaramallion
scaramallion
I think originally I created the issue because a user was confused between supplying a raw bytes value for one of the longer O* VRs and supplying the decoded value....
[This was the original issue](https://github.com/pydicom/pydicom/issues/1187) You're right about the SH/string VRs, that should get caught now.
I like just having `bytes` for O*, our typing is already pretty complex and encoding values isn't difficult. It also matches the read type.
Duplicate of #533 😝 Not sure I agree with the function signature, `decode_frame(ds: Dataset, idx: int) -> numpy.ndarray` would be much simpler. But yes, random frame access should be implemented....
But on the other hand, the function signature would have to change if more elements were required in the future (bits stored, for example, has been added as a requirement...
Sorry, been busy. I'll look when I can, but I've got a lot of stuff to get through. I'd really like the functionality, too.
I think this should be a utility function rather than be included with the handlers
Hmm, maybe as a config option? I'm not fond of applying stuff to the pixel data without explicit instructions from users.
With pynetdicom I tried out building the docs using github actions but found the extra steps of having to download the artifacts and view locally was too annoying and ended...
No, this was a couple of months back. It'd make a lot of sense to wait until the online preview is available.