canonical way to get a view of a coordinate with the same shape as the DataArray?

[tom-new]

Suppose I have a Dataset of temperatures on coordinates x, y, and z, and I want to calculate some variable that depends on both temperature and z (say, pressure). In my pre-xarray days, I would have made a meshgrid with x, y, and z to get z in the same shape as temperature so I could use numpy arithmetic on them.

However, dataset.z has shape (4,), and so cannot be broadcast with dataset.temperature. Now, if I am writing the pressure method myself, I can write it such that it uses xarray broadcasting instead of numpy broadcasting. But what if I am using somebody else's method (e.g. scipy.interpolate.RectBivariateSpline.ev)? What should I do in this case, where the method expects the inputs to be numpy broadcastable instead of xarray broadcastable?

Of course, I could make a meshgrid myself like I would have before, but that seems like overkill. ChatGPT suggested dataset["temperature"].coords["z"].broadcast_like(dataset["temperature"]), which works, but I can't help but feel like there should be a simpler way of doing this as it seems like it shouldn't be an uncommon operation.

[tom-new]

Of course, I trusted ChatGPT too blindly. Of course, one can simply do dataset.z.broadcast_like(dataset.temperature). D'oh.