Hi to all,
I have been thinking lately about hyperoperators, and particulary about its similarity with RFC 207 (Arrays: Efficient Array Loops) For the ones that don't have the RFC in mind, I copy its abstract: >This RFC proposes a notation for creating efficient implicit >loops over multidimensional arrays. It introduces the notation >|i for an index iterator for arrays, allowing temporary >multidimensional arrays to be created on the fly. My point is that what hyperoperators express (aplying operators and methods elementwise) is just a subset of the functionality provided by RFC 207, and that the former is both simpler and more powerful. Maybe i should better explain myself with an example. @arr3 = @arr1 ^+ @arr2 @arr4 = $v ^* @arr1 $sum ^=+ @arr1 Could become: @arr3 = @arr1[|i] + @arr2[|i] # also @arr3[|i] = @arr1[|i] + @arr2[|i] @arr4 = $v + @arr1[|i] $sum =+ @arr1[|i] As an alternative and without real thought on it, i would suggest maybe a different notation (more similar to Apocalypse's one) for this RFC. @arr3 = @arr1[^i] + @arr2[^i] # also @arr[^i] = @arr1[^i] + @arr2[^i] @arr4 = $v * @arr1[^i] $sum =+ @arr1[^i] @lengths_array = @arr1[^i].length() I turns out also that RFC's semantics allows much more powerful things like: @mat3 = @mat1[^i,^j] * @mat2[^j,^i] It looks to me that hyperoperators are just a partial solution to a bigger problem, and that the cost (a whole family of new operators) is way too big for its value. On the other hand, a RFC 207-like notation adds a lot more power with a single addition to the language. Just my 5 cents ;-) Angel Faus [EMAIL PROTECTED]
