On 6/11/15 9:33 AM, Dennis Ritchie wrote:
On Thursday, 11 June 2015 at 15:03:39 UTC, Dicebot wrote:
Those are considerably less powerful:
- can only have type arguments
- no variadic argument list support
- no arbitrary condition constraints (thus only partial duck typing
support)
On the other hand they have one important advantage: all type
arguments must comply to one or more trairs and thus bodies of
generics are checked before institation. You are only allowed to call
methods and operations of generic arguments that are defined in
relevan trait. This is huge win for code hygiene compared to D.
Any sort of more advanced meta-programming things can only be done via
AST macros which is currently the biggest downside in my eyes when it
comes to features. Though quite some people like that.
The fact that there is no support variadiŃ arguments, it is really
negative.
It is possible that Walter and Andrei against macro because of this:
macro_rules! o_O {
(
$(
$x:expr; [ $( $y:expr ),* ]
);*
) => {
&[ $($( $x + $y ),*),* ]
}
}
fn main() {
let a: &[i32]
= o_O!(10; [1, 2, 3];
20; [4, 5, 6]);
assert_eq!(a, [11, 12, 13, 24, 25, 26]);
}
It looks disgusting! ;)
Is that actual Rust code that compiles and runs? -- Andrei
