Difference between .coords and .coordinates

[mlshapiro]

Description
Question: What is the difference between the .coords on ArrayCoordinates1D and .coordinates properties on Coordinates1D?

From the spec for Coordinates1D:

- `coordinates`: read-only array

From the spec for ArrayCoordinates1D:

Traits:
 - `coords`: array

[jmilloy]

The coordinates is a common read-only property of all Coordinates1d.

ArrayCoordinates1d uses coords trait to maintain its coordinate values, just like UniformCoordinates1d uses start, stop, and step for its coordinate values. Each implements the coordinates property, which of course returns a copy of the coords for ArrayCoordinates1d.

If you want to change the coordinates of an ArrayCoordinates1d, you can use c.coords = [0, 1, 2] but not c.coordinates = [0, 1, 2].

[jmilloy]

We could make the coords private if you want (but probably not start, stop and step).

[mlshapiro]

perfect - thank you for the explanation. I will add this explanation to the spec.

I ended up making some notebooks for each of these classes so that I don't have to keep writing examples as I'm trying to develop. Thats where these questions are coming from, mostly just trying to understand.

[jmilloy]

My understanding was that the developer specs would be abandoned once implemented and eventually removed, in favor of the user guides and docstrings, and that @mpu-creare advised us not to worry about maintaining the dev specs if the implementation deviated. The coordinates user guide and docstrings are not complete (#45, #4) but should be soon; I'd put updates for clarification in those locations.