On Saturday, 19 March 2016 at 08:38:20 UTC, Basile B. wrote:
Otherwise there's something that's pretty in the syntax:
Very much so. My own "toy language" project uses (well, _will_
use once I have more than 5% of a parser) a similar syntax, so it
could just be my own biases talking, but I like it.
However later in the function declaration:
"sum := (x: float, y: float, z: float) -> float {
return x + y + z;
};"
I would expect the same system as for variables:
"sum : float = (x: float, y: float, z: float) {
return x + y + z;
};"
or return type inference:
"sum := (x: float, y: float, z: float) {
return x + y + z;
};"
As far as the second example goes, sum is a function returning
float, not a float variable. I could be wrong, but it appears to
me that using variable-like syntax would make the grammar (or at
least its semantics) less context-free. Plus the () -> type {}
syntax works nicely for lambdas. I'm sure the type-inference
option could be supported, though.
The language has some very interesting ideas. I probably wouldn't
use it for general-purpose programming, though. I like my OOP, at
least when it fits the problem at hand.
