Remove side-effects from convolution.convolve_fft

[jamienoss]

[src]

# restore NaNs in original image (they were modified inplace earlier)
# We don't have to worry about masked arrays - if input was masked, it was
# copied
array[nanmaskarray] = np.nan
kernel[nanmaskkernel] = np.nan

[jamienoss]

In fact, this is not even correct! nanmaskarray is computed as ~np.isfinite(array) (src), and therefore any inf in the input array or kernel will be incorrectly replaced by NaN.

[pulkit6559]

I'm new to astropy and a beginner, is it okay if i work on this?

[pllim]

@pulkit6559 , please checkout the issues labeled as "good first issue" first. Thank you for your interest!

[jamienoss]

@pulkit6559 I wasn't necessarily intending to work on this. The soln. mainly involves correctly copying the input arrays only when necessary. See #7503

One caveat, I wouldn't start until ##8114 is resolved, which I'm currently working on. convolve_fft() will be refactored in which case starting now might mean doing twice the work for you. Unless you finish first in which case it's twice for me, but I'm cool with that 😉

Also, I should note that I am not a maintainer of this package so have no actual official say when it comes to workload distribution. That being said, it is open src and, canonically, everyone is "free" to work on whatever they want. The only caveat is that someone else might beat you to the punch and/or do a better job. Either way, you are free to try.

[pulkit6559]

@pllim @jamienoss thankyou for the response, i'll try to tackle the easier ones first and then see if i could come up with something for this. :)

[jamienoss]

Sorry @pulkit6559 I had to go ahead and fix this for #8114.