Fix json reexports of different items with same name

[GuillaumeGomez]

Fixes #107677.

I renamed from_item_id* functions into id_from_item instead because it makes more sense now. I also simplified the logic around it a bit so that the ids function will now directly pass &clean::Item to id_from_item and the ID will be consistently generated (it caused an issue when I updated the ID for imports).

So now, the big change of this PR: I changed how imports' ID is generated: it now includes the target item's ID at the end of the ID. It's to prevent two reexported items with the same name (but different types).

r? @aDotInTheVoid

[GuillaumeGomez]

Fixed the bad primitive change I made.

[aDotInTheVoid]

@rustbot author

[aDotInTheVoid]

Could this test be expanded a bit more? I'd like to see it checking that their are items for both Foos, and that each name imports both of the foo items (instead of just checking that their are two imports with each name).

[aDotInTheVoid]

Eg:

#![feature(no_core)]
#![no_core]

pub mod nested {
    // @set foo_struct = "$.index[*][?(@.docs == 'Foo the struct')].id"

    /// Foo the struct
    pub struct Foo {}

    // @set foo_fn = "$.index[*][?(@.docs == 'Foo the function')].id"

    #[allow(non_snake_case)]
    /// Foo the function
    pub fn Foo() {}
}

// @ismany "$.index[*][?(@.inner.name == 'Foo' && @.kind == 'import')].inner.id" $foo_fn $foo_struct
// @ismany "$.index[*][?(@.inner.name == 'Bar' && @.kind == 'import')].inner.id" $foo_fn $foo_struct

// @count "$.index[*][?(@.inner.name == 'Foo' && @.kind == 'import')]" 2
pub use nested::Foo;
// @count "$.index[*][?(@.inner.name == 'Bar' && @.kind == 'import')]" 2
pub use Foo as Bar;

[aDotInTheVoid]

Looking at the output (of tests/rustdoc-json/reexport/same_name_different_types.rs), for some reason it's duplicating ID's in the output:

  "index": {
    "0:0:1588": {
      "crate_id": 0,
      "docs": null,
      "id": "0:0:1588",
      "inner": {
        "is_crate": true,
        "is_stripped": false,
        "items": [
          "0:1:1582",
          "0:4-0:2:1584",
          "0:4-0:3:1584",
          "0:4-0:2:1584",
          "0:4-0:3:1584",
          "0:5-0:2:1584",
          "0:5-0:3:1584",
          "0:5-0:2:1584",
          "0:5-0:3:1584"
        ]
      },
      "kind": "module",
      "name": "same_name_different_types"
    },

I'm not really sure whats causing this, but idealy it wouldn't happen, although I'd be willing to merge as is, as long as an issue gets created to follow up.

(Also, jsondoclint should probably catch this)

[GuillaumeGomez]

I think it's better to fix it in another PR since this one is already quite big. We can fix it easily by having a HashSet but I'd like to investigate why it's needed in the first place.

[GuillaumeGomez]

And also: updated.

[aDotInTheVoid]

@bors r+

[bors]

📌 Commit 1ae875f has been approved by aDotInTheVoid

It is now in the queue for this repository.