Justus Magin

I've implemented my suggestion to use `Signature.bind_partial` / `Signature.bind` to normalize the parameters, so I think this leaves us with just a few open issues: - we ignore / disallow...

`ipython` seems to call `FormattingQuantity._repr_pretty_` which delegates to both the magnitude and the unit, and since neither the magnitude nor the unit implement `_repr_pretty_`, the result is roughly ```python f"{repr(self.magnitude)}...

this is a side-effect of making the application registry a instance of `ApplicationRegistry`, which solves a lot of inter-library compatibility issues. This still holds: ```python assert ur is pint.get_application_registry().get() ```

I guess the most convenient (and consistent) way would be a separate namespace for constants, as described in https://github.com/hgrecco/pint/issues/1078#issuecomment-950231900

I tried implementing this and noticed that while the proposed format is very easy to use if both magnitude and unit specs are passed, omitting either seems a bit trickier....

shouldn't you implement (and call) `sel` instead of `isel` if you're working in coordinate space?

> do you understand this blackdoc error? so the reason is the position of the triple quotes: with ```python def f(): """ >>> 1 + 2""" pass ``` the extracted...

well, that's just it: there's no way to discriminate between docstrings and multi-line triple-quoted strings without parsing the python code (which is definitely out of scope), so `blackdoc` doesn't even...

if we're fine with the less detailed output of `micromamba info` (no `-a` flag), we could also just replace `conda` with `micromamba`. However, this is probably a bug in the...

if this crashes with both `netcdf4` and `h5netcdf` this might be a bug in the `libhdf5` library. If we can manage to reduce this to use just `h5py` (or `netCDF4`),...