ICE with static methods

[brendanzab]

pub struct Vec2<T> { x: T, y: T }

pub impl<T:Copy Number> Vec2<T> {
    static pure fn zero() -> Vec2<T> {
        let _0 = Number::from(0f);    // FIXME: causes ICE
        Vec2 { x: _0, y: _0 }
    }
}

pub trait Number: NumConv {
    static pure fn from<T:Number>(n: T) -> self;
}

pub impl float: Number {
    static pure fn from<T:Number>(n: T) -> float { n.to_float() }
}

pub trait NumConv {
    pure fn to_float(&self) -> float;
}

pub impl float: NumConv {
    pure fn to_float(&self) -> float { *self }
}

fn main() {
    let _: Vec2<float> = Vec2::zero();
}

 $ RUST_LOG=rustc=1,::rt::backtrace rustc src/vec.rs
rust: task failed at 'index out of bounds: the len is 1 but the index is 1', /usr/local/src/rust/src/librustc/middle/trans/common.rs:1327
error: internal compiler error: unexpected failure
note: the compiler hit an unexpected failure path. this is a bug
note: try running with RUST_LOG=rustc=1,::rt::backtrace to get further details and report the results to github.com/mozilla/rust/issues
rust: task failed at 'explicit failure', /usr/local/src/rust/src/librustc/rustc.rc:483
rust: domain main @0x7f8a0c800010 root task failed
rust: task failed at 'killed', /usr/local/src/rust/src/libcore/task/mod.rs:630

Context:

• https://github.com/bjz/lmath-rs/blob/master/src/num/types.rs#L7
• https://github.com/bjz/lmath-rs/blob/master/src/mat.rs#L351

[brson]

Redefining Number like this makes it work

pub trait Number: NumConv {
    static pure fn from<T:NumConv>(n: T) -> self;
}

With T:NumConv instead of T:Number. Somehow trans is not receiving the proper number of vtables for T.

[brson]

Reduced further

pub trait Number: NumConv {
    static pure fn from<T:Number>(n: T) -> self;
}

pub impl float: Number {
    static pure fn from<T:Number>(n: T) -> float { n.to_float() }
}

pub trait NumConv {
    pure fn to_float(&self) -> float;
}

pub impl float: NumConv {
    pure fn to_float(&self) -> float { *self }
}

fn main() {
    let _: float = Number::from(0.0f);
}

[brendanzab]

Ahh cheers!

At least I now have a good interim solution. :)