>>62
not even FFOC fucks up that badly. it really only has two "unexpected" coercion cases: number -> string and string -> number, which only happen in the context of its separate concatenation and addition operators so there's no question as to what
"1" + 2 or
1 .. 2 means. although, the BDFLs are considering removing these coercions anyway
>= {3} * 10 -- error: attempt to perform arithmetic on a table value
>= {1, 2, 3} * 10 -- ditto
>= {1, 2, 3} + {4, 5, 6} -- ditto
FFOC isn't even considered an OO language and yet it still has a mechanism to handle its
self value.
x:dosomething(1, 2, 3) -- translates to:
x.dosomething(x, 1, 2, 3) -- with `x' being evaluated only once.
x.dosomething(y, 1, 2, 3) -- Function.prototype.call, anyone?
needless to say, FFOCfags are not impressed by javashit in the least. it's only now getting shit like block scoping and correct handling of lexical scope.
i mean hell, they chose generators over full coroutines! they even used keywords! meanwhile i can go and cripple them on a per-function basis, since i can just write different wrappers over the coroutine library and insert them into function environments as needed.
js is the future