(One representation of) the new gpio::Pin/spi::Spi types appears to kill rust-analyzer

[ryankurte]

running rustc 1.61.0, if you happen to try to alias a Pin into an SPI device, rust-analyzer explodes...

type Spi1 = Spi<
        SPI1, 
        (
            Pin<'A', 5, Alternate<5> >,
            Pin<'A', 6, Alternate<5> >,
            Pin<'A', 7, Alternate<5> >,
        ),
        TransferModeNormal,
    >;

tested with rust-analyzer prerelease via vscode and nighly from rustup. this appears to be a reproduction of rust-lang/rust-analyzer#10390, with a suggestion this can be avoided by altering the ordering constraint to types before consts, eg. Pin<MODE = Input<Floating>, const P: char, const N: u8> instead of Pin<const P: char, const N: u8, MODE = Input<Floating>>.

i am not sure whether where is an alternate representation that would not trigger this, or if it is worth changing here or waiting for a fix, but, figured i'd leave some footprints for anyone else that runs into the following errors:

thread '<unnamed>' panicked at 'index out of bounds: the len is 1 but the index is 2', /home/runner/.cargo/registry/src/github.com-1ecc6299db9ec823/chalk-ir-0.76.0/src/fold/subst.rs:60:19
stack backtrace:
   0: rust_begin_unwind
             at /rustc/9d1b2106e23b1abd32fce1f17267604a5102f57a/library/std/src/panicking.rs:498:5
   1: core::panicking::panic_fmt
             at /rustc/9d1b2106e23b1abd32fce1f17267604a5102f57a/library/core/src/panicking.rs:116:14
   2: core::panicking::panic_bounds_check
             at /rustc/9d1b2106e23b1abd32fce1f17267604a5102f57a/library/core/src/panicking.rs:84:5
   3: <chalk_ir::fold::subst::Subst<I> as chalk_ir::fold::Folder<I>>::fold_free_var_ty
   4: <chalk_ir::Ty<I> as chalk_ir::fold::SuperFold<I>>::super_fold_with
   5: chalk_ir::fold::boring_impls::<impl chalk_ir::fold::Fold<I> for chalk_ir::GenericArg<I>>::fold_with
   6: <smallvec::SmallVec<A> as core::iter::traits::collect::Extend<<A as smallvec::Array>::Item>>::extend
...
note: Some details are omitted, run with `RUST_BACKTRACE=full` for a verbose backtrace.

WARN [12/03/2022, 9:42:00 pm]: LSP request failed {
  method: 'rust-analyzer/inlayHints',
  param: {
    textDocument: { uri: 'file:///home/ryan/projects/dsf-sensor/src/pub2.rs' }
  },
  error: [fo [Error]: server panicked: index out of bounds: the len is 1 but the index is 2
  	at Yn (/home/ryan/.vscode/extensions/matklad.rust-analyzer-0.3.963/out/main.js:7:223)
  	at Ln (/home/ryan/.vscode/extensions/matklad.rust-analyzer-0.3.963/out/main.js:6:6732)
  	at Immediate.<anonymous> (/home/ryan/.vscode/extensions/matklad.rust-analyzer-0.3.963/out/main.js:6:6579)
  	at processImmediate (internal/timers.js:461:21)] {
    code: -32603,
    data: undefined
  }
}

[burrbull]

Sadly.
You could use released version of f4xx-hal until RA fixed.

[burrbull]

@ryankurte
Is there error if use just:

use hal::gpio::{PA5, PA6, PA7};

type Spi1 =
    Spi<SPI1, (PA5<Alternate<5>>, PA5<Alternate<5>>, PA7<Alternate<5>>), TransferModeNormal>;

[ryankurte]

You could use released version of f4xx-hal until RA fixed.

unfortunately i'm too far along the bleeding edge to back down at this point 🤣

Is there error if use just:

brilliant, that resolves it, thanks! also these changes are -super cool-, having a great experience with the updated HAL, hugely appreciate all your effort stm32-rs folks ^_^

[burrbull]

@ryankurte Is there error if use just:

use hal::gpio::{PA5, PA6, PA7};

type Spi1 =
    Spi<SPI1, (PA5<Alternate<5>>, PA5<Alternate<5>>, PA7<Alternate<5>>), TransferModeNormal>;

Now it's possible also

use hal::gpio::{PA5, PA6, PA7, AF5};

type Spi1 = hal::spi::Spi1<(PA5<AF5>, PA6<AF5>, PA7<AF5>)>;

[ryankurte]

ooh nice! looks like y'all are having some real fun with const types ^_^

thanks for all the help, seems like there's lots of good workarounds