drawFFT_finish vs image.calculate_inverse_fft

[jecampagne]

Hi, in the gsobject.drawFFT there is a chain of 3 functions

• kimage, wrap_size = self.drawFFT_makeKImage(image) : that compute the f(k) fourier-space coefficient w/o the phase
• self._drawKImage(kimage): that update the f(k) including the phases
• and finally self.drawFFT_finish(image, kimage, wrap_size, add_to_image) that compute the Inverse FFT using _galsim.irfft (nb. here self is a Transformation object)

I was wandering what prevent to use the kImage.calculate_inverse_fft() function in place of the last step (drawFFT_finish) ? Thanks

[rmjarvis]

I'm not sure what you mean by prevent here. The two functions have to do slightly different things before and after the actual call to _galsim.irfft. So we couldn't just replace the drawFFT_finish call with calculate_inverse_fft within drawFFT. But if a user wants to do that for some other use case, there isn't anything preventing you from doing so.