Just wondering... Does anyone else feel that "syntax" around lifetimes and
type bounds is becoming very unwieldy? I am referring to this PR:
https://github.com/mozilla/rust/pull/11768. Is using dummy struct fields
for specifying custom type bounds on the struct itself really the best we
can do in a brand-new language?
At least for the lifetimes, Rust could extend existing syntax and allow
specifying them on by-value returns objects, so that, for example, this:
pub trait MutableVector<'a, T> {
fn mut_iter(self) -> MutItems<'a, T>;
...
}
could be written as:
pub trait MutableVector<'a, T> {
fn mut_iter(self) -> 'a MutItems<T>;
...
}
This also makes the intent much more clear. Currently, one would have to
dig into the definition of MutItems<'a,T> to figure out that the lifetime
parameter 'a is used to create a dummy borrow field into the vector, so
that the whole iterator is then treated as a mutable borrow. This feels
very convoluted, if you ask me.
On a slightly different note, is there a strong reason for having to name
lifetime parameters explicitly? Could we simply re-use actual parameter
names prefixed with ' as their lifetimes? The above could then be reduced
to this:
pub trait MutableVector<T> {
fn mut_iter(self) -> 'self MutItems<T>;
...
}
This used to be valid syntax, btw, though it worked because 'self lifetime
was special, IIRC.
Note that I am not proposing to eliminate explicitly named lifetimes
entirely,- there are still cases when they are needed. But as far as I've
seen, those are few and far between.
Thanks for reading,
Vadim
_______________________________________________
Rust-dev mailing list
Rust-dev@mozilla.org
https://mail.mozilla.org/listinfo/rust-dev
