[std.numeric.fft] Arbitrary sized FFT

jmh530 john.michael.hall at gmail.com
Mon Nov 7 15:31:44 UTC 2022


On Saturday, 5 November 2022 at 14:47:40 UTC, RIG_ wrote:
> [snip]
>
> I have thought about using [`fftw`](https://www.fftw.org/), but 
> it feels like doing `fft`s should in the standard library.
>
> [snip]

I would guess that whatever python does is written in C and then 
called from python. For some things, it doesn't make sense to 
reinvent the wheel, and I think fft probably fits. D has the 
ability to call C code. Now, it might be that the API for fftw is 
very C-like. A D wrapper that is more convenient to use might be 
worthwhile as a dub project (not knowing anything about fftw).

With respect to whether it should be in the standard library, it 
might make sense to get it solid as a dub library and then 
propose that for phobos inclusion. This is how sumtype was added. 
It can be difficult to work on the same schedule as phobos.

It would be nice if there were a way to use this with importC, 
but taking a brief look at the fftw github page (with the big 
warnings that you shouldn't check it out unless you know what 
you're doing) suggests that might not be so easy.


More information about the Digitalmars-d mailing list