Ruth Comer

If it helps, several other SciTools projects have adopted `black`, and I think we’re pretty happy about not having to think about the formatting any more.

According to the docstring, the class expects a collection of geometries, so the tests are passing tuples and lists. It seems reasonable to me that a user might want to...

It looks like `pcolorfast` is not implemented in Cartopy. At least, I can’t find it by searching [geoaxes.py](https://github.com/SciTools/cartopy/blob/main/lib/cartopy/mpl/geoaxes.py).

We already have [concatenate_cube](https://scitools-iris.readthedocs.io/en/latest/generated/api/iris/cube.html?highlight=concatente_cube#iris.cube.CubeList.concatenate_cube).

An instance of this just came up on MO Yammer. In that particular case, the overlapping period had equal data, metadata, etc. So in that case Iris might reasonably be...

I notice that over at https://github.com/ESMValGroup/ESMValCore/issues/1423, @zklaus has argued that such functionality should be added in downstream packages to address specific use-cases. We could maybe improve the error message from...

I think `match` becomes `False` [here](https://github.com/SciTools/iris/blob/3ee7c56bcfea032fc5770f6c9b84b91511441d88/lib/iris/_concatenate.py#L859) if the coordinates are overlapping. If I’ve got that right, could we raise an exception there to definitively say so? (edit: only if `error_on_mismatch`...

Seems that ignoring the masks was deliberate, but only because “the chosen approach fixes the problem [of all-masked data raising errors], is simpler, and reflects numpy behaviour.” #905 Perhaps using...

There is also a relevant thread at https://github.com/numpy/numpy/issues/14624 I think what Iris does in testing is sensible: https://github.com/SciTools/iris/blob/56a97d66b8404f96ef61fe2d52433eb58373f338/lib/iris/tests/__init__.py#L590-L668

:fearful: I think I can see where you're coming from, but from a pure user perspective this might just look like "making us change our code for no good reason"....