The Bessel integrals required to integrate the source function are pretty painful. I should implement some of the harmonic approaches that work well at small scales, like
[ ] power law decomposition / FFTLog type stuff (Assassi+17)
[ ] quasi-discrete Hankel transform methods may work better for large scales which need ISW
The Bessel integrals required to integrate the source function are pretty painful. I should implement some of the harmonic approaches that work well at small scales, like