Tom Nicholas

@varchasgopalaswamy if you can get uncertainty support working in xarray in a general and interoperable way, people would **love** that! I would expect it to be a pretty big project...

> We're thinking of adopting xarray throughout our code, which requires support for uncertainty propagation, so I have a vested interest in getting this to work! This is how every...

cc'ing @dcherian as a cf-xarray pro

I was wondering whether `pint.wraps` should be able to handle any object that defines pint-like methods and attributes, a question that's related to https://github.com/hgrecco/pint/issues/1099. That seems like a lot of...

> If I understand wraps correctly, it makes sure that the input data is in the specified units, strips them and attaches the units specified as return value to the...

> the disadvantage of a new kwarg is that it could shadow a variable (if we choose the name carefully enough that might not be as much of an issue,...

> What do you think about adding a new decorator that does both? It could share all internals @keewis I don't know if I'm being dense but I actually can't...

> if you put the pre/postprocessing into separate functions, the new decorator can call those Yes, but I think you still need to duplicate all the triple wrapping code, like...

> wouldn't it be possible to do this instead: Yeah, but that doesn't really solve my question - it still duplicates `def wrapper` (and for a decorator with arguments there...

> If I understand wraps correctly, it makes sure that the input data is in the specified units, strips them and attaches the units specified as return value to the...