Cross-language comparison: function map and similar

[Paul Rubin]

Steve D'Aprano <steve+python at pearwood.info> writes:
> Are there language implementations which evaluate the result of map()
> (or its equivalent) in some order other than the obvious left-to-right
> first-to-last sequential order? Is that order guaranteed by the
> language, or is it an implementation detail?

Haskell just gives back an unevaluated thunk.  The elements are
evaluated in whatever order you happen to use them in.  Since the
evaluation isn't supposed to have observable side effects, there's no
way to tell the order.

There are also parallel versions (Control.Parallel.Strategies.parMap
etc.) and an extension that recognizes "racing stripes" on the square
brackets to make list comprehensions run faster:

   foo = [| sqrt(x) | x <- [1.0 .. 1000.0] |]

parallelizes the calculation and offloads it to a GPU or other vector
processor.  You might also like this old post

 https://donsbot.wordpress.com/2007/11/29/use-those-extra-cores-and-beat-c-today-parallel-haskell-redux/

"map" is conceptually the lifting of a function from a given type, to
the type under the action of some functor.  For historical reasons map
only works on lists while fmap works on arbitrary functors, but they are
in principle the same thing.  Depending on what the functor does, the
whole concept of left-to-right order might be meaningless.