fix(#37)

[kingwingfly]

Objective

fixes #37

Solution

At first, I feel this part is strange:

    let len = 1 + input.end - input.start;
    let ident_tuples = (0..=len)
        .map(|i| {
            let idents = input
                .idents
                .iter()
                .map(|ident| format_ident!("{}{}", ident, i));
            to_ident_tuple_enumerated(idents, input.idents.len())
        })
        .collect::<Vec<_>>();

If end = 2, start = 0, then len = 3. But ident_tuples has 4 elements, while we only need 2.

I make changes to it first, and then #37 just fixed.

Testing

• Did you test these changes? If so, how?

RUSTFLAGS="--cfg docsrs" cargo +nightly expand --example all_tuples

And got

#[allow(unused_variables)]
#[allow(non_snake_case)]
#[allow(clippy::unused_unit)]
impl Foo for () {
    const FOO_HARDER: bool = true;
    fn foo() -> bool {
        true
    }
}
#[allow(unused_variables)]
#[allow(non_snake_case)]
#[allow(clippy::unused_unit)]
#[doc(fake_variadic)]
///This trait is implemented for tuples down to 0 up to 15 items long.
impl<F: Foo> Foo for (F,) {
    const FOO_HARDER: bool = true && F::FOO_HARDER;
    fn foo() -> bool {
        true
    }
}
#[allow(unused_variables)]
#[allow(non_snake_case)]
#[allow(clippy::unused_unit)]
#[doc(hidden)]
impl<F0: Foo, F1: Foo> Foo for (F0, F1) {
    const FOO_HARDER: bool = true && F0::FOO_HARDER && F1::FOO_HARDER;
    fn foo() -> bool {
        true
    }
}

• Are there any parts that need more testing?
  I open doc with this cmd

RUSTFLAGS="--cfg docsrs" cargo +nightly doc --open --example all_tuples

But the fake_variadics is not right. I believe it's due to I missed some flags and it'll be right when publishing onto cratesio.
However, I'm not very sure.

• How can other people (reviewers) test your changes? Is there anything specific they need to know?
  Use correct command to check cargo doc result
  Mine is incorrect:

RUSTFLAGS="--cfg docsrs" cargo +nightly doc --open --example all_tuples

By the way,

• I change doc_auto_cfg to doc_cfg, since they are merged since Rust nightly 1.92. And docs.rs uses nightly Rust. Anyway, this crate has no features gate, there's no effect I think.

[github-actions]

Welcome, new contributor!

Please make sure you've read our contributing guide and we look forward to reviewing your pull request shortly ✨

[BenjaminBrienen]

Does this also fix #42?

[BenjaminBrienen]

Please add a test case to make sure that if the input is 16 to 20, then it is not implemented for 0..=15. I have a suspicion.

[kingwingfly]

I first add a test 3..15, then it conflicts with the original 1..15, so I believe 16..20 doesn’t impl for 1..15. 1..15and 15..20 are in the same block in doc, if 15..20 implemented for 1..15 again, it would not compile. Maybe current test is enough...

EDIT: I only added test in all_tuples and forgot others...

[kingwingfly]

I think it happens to fix #42 too. Because in this PR, only format idents needed (i.e. if all_tuples!(foo, 0, 2, T), only [T1, T2] and up to 2). There's no chance to generate redundant impl block. The original len = 2 + 1 - 0 = 3, it's then incorrect.