Very nice summary, thanks for posting. One question I had was what should the signature of a function be to receive a generator? For example, if the only method of extrema is extrema(A::AbstractArray), is that too restrictive?
Jared Crean On Tuesday, October 11, 2016 at 1:05:03 PM UTC-4, Stefan Karpinski wrote: > > Since the 0.5 release affects everyone here, I wrote a longish blog post > about what the major changes are: > http://julialang.org/blog/2016/10/julia-0.5-highlights. > > One other change that I left out of the post because it was getting pretty > long and it seems a bit esoteric is that array comprehensions are now > type-inference-independent. That means that the type of the resulting array > only depends on the actual types of values produced, not what the compiler > can prove about the expression in advance. In particular, this means that > comprehensions behave the same way in global scope as in local scope now, > which is a fairly major relief to anyone who's struggled with that. >
