=== in comparing with nothing

[jishnub]

This might help with type-inference on older versions of Julia

[codecov]

Codecov Report

Base: 71.97% // Head: 71.98% // Increases project coverage by +0.01% 🎉

Coverage data is based on head (fc18d6c) compared to base (6b777fe).
Patch coverage: 100.00% of modified lines in pull request are covered.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #364      +/-   ##
==========================================
+ Coverage   71.97%   71.98%   +0.01%     
==========================================
  Files          79       79              
  Lines        8311     8311              
==========================================
+ Hits         5982     5983       +1     
+ Misses       2329     2328       -1     

Impacted Files	Coverage Δ
src/LinearAlgebra/helper.jl	64.23% <100.00%> (ø)
src/Multivariate/ProductFun.jl	49.44% <0.00%> (+0.55%)	⬆️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[dlfivefifty]

I would suggest just using isnothing(xst)

[jishnub]

isnothing(x) is equivalent to x === nothing on recent Julia versions, but from what I remember, the === nothing used to be the recommended style on older versions. I can't remember if they are equivalent on v1.6, so I went with the one that is known to be better on old versions.

[dlfivefifty]

If isnothing isn't the recommended style then its weird the function even exists...

[jishnub]

I think the idea was that custom Nothing types may extend isnothing for their types, which wouldn't be possible with === nothing. However, the performance issue was unfortunate, which is why it was recommended to use === nothing if not working with custom types. At least on recent Julia versions, all of isnothing, === nothing and == nothing are equivalent, so that's one less thing to remember.