correlation_3d for fixed scale factor instead of using Pk2D about ccl HOT 2 OPEN

@James11222 indeed this is not implemented. The easiest short-term, "hacky" solution for you would be to create a Pk2D that spans more than one scale factor. Since you don't care about evolution, you can probably just repeat the same Pk array at a few values of a (I forget what kind of interpolation we're using here, but I imagine 3 values of a will be enough). Let me know if this works for you.

The longer-term solution would be to allow for Pk2Ds to be generated at a single scale factor (and assume that they are the same at all as) or, perhaps even simpler, to allow for factorisable Pk2Ds, so you can pass the k and a dependence as two different multiplicative factors (this is already implemented at the C level, but not visible to python). If this would be a desirable feature, perhaps best to open a dedicated issue for it.

from ccl.

We should catch the C-level error in any case though, crashing the python interpreter is bad.

from ccl.