Add a new trait to accept more types in the Val-helper functions

[TheBlckbird]

Objective

• Allow the Val-helper functions to accept more types besides just f32

Fixes #20549

Solution

• Adds a new trait that can be implemented for numbers
• That trait has a method that converts self to f32

Testing

• I tested it using Rust's testing framework (although I didn't leave the tests in, as I don't deem them important enough)

Rust test

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn test_val_helpers_work() {
        let p = px(10_u8);
        assert_eq!(p, Val::Px(10.0));

        let p = px(10_u16);
        assert_eq!(p, Val::Px(10.0));

        let p = px(10_u32);
        assert_eq!(p, Val::Px(10.0));

        let p = px(10_u64);
        assert_eq!(p, Val::Px(10.0));

        let p = px(10_u128);
        assert_eq!(p, Val::Px(10.0));

        let p = px(10_i8);
        assert_eq!(p, Val::Px(10.0));

        let p = px(10_i16);
        assert_eq!(p, Val::Px(10.0));

        let p = px(10_i32);
        assert_eq!(p, Val::Px(10.0));

        let p = px(10_i64);
        assert_eq!(p, Val::Px(10.0));

        let p = px(10_i128);
        assert_eq!(p, Val::Px(10.0));

        let p = px(10.3_f32);
        assert_eq!(p, Val::Px(10.3));

        let p = px(10.6_f64);
        assert_eq!(p, Val::Px(10.6));
    }
}

---

Showcase

// Same as Val::Px(10.)
px(10);
px(10_u8);
px(10.0);

[TheBlckbird]

This is still waiting for a decision on #20549.

And this should probably be mentioned with the helper functions in the release notes. If so, I'll have to look into that.

[alice-i-cecile]

This also needs testing against #20158, to ensure that it plays nice with BSN.

[alice-i-cecile]

I don't understand why we need a dedicated ToValNum trait. Why can't we just use impl Into<Val>?

[TheBlckbird]

You mean:

impl From<i32> for Val {
    fn from(value: i32) {
        Val::Px(value as f32)
    }
}

That would have the issues with having to specify a default unit, see #20517 and @cart's comment: #20517 (comment)

Or am I missing something?

[alice-i-cecile]

Ah, I see. Can you please explain this in a doc comment on ToValNum?

[alice-i-cecile]

LGTM if we can get this working with BSN.

[TheBlckbird]

This also needs testing against #20158, to ensure that it plays nice with BSN.

Ok, so I just switched to that branch and merged these changes locally and experimented with the examples. There don't seem to be any problems.

I'll update the examples in here tomorrow and will do the same for BSN once cart merges main into the feature branch again.

[ickshonpe]

I don't like the extra conversions or the macro, it just seems useless to me.
The use case for this is just we want to be able to type px(10) instead of px(10.), which only needs i32. The times you'd need the other conversions are so rare, I don't see any problem with requiring an explicit cast. It could even hide errors, using i128/u128s for UI coords seems especially fishy.

Practically its not going to make any significant difference though, so I approve anyway with that objection.

[TheBlckbird]

The argument could be made to remove i128/u128 from the impl blocks, especially since the limit for these two is higher than for f32 (afaik), which could potentially lead to weird errors (if someone goes that high for any reason, but they should still decide to cast that themselves explicitly).

Thanks for the feedback

[TheBlckbird]

Sorry for the big commit. I changed all the references to Val:: to just the helper functions using a script and edited some edge cases by hand. The examples all compile without any errors and I quickly looked over the diff, so it should be fine.