Dylan DPC f3ab5a66a9

Rollup merge of #102538 - cjgillot:def-span-ctxt, r=fee1-dead ...

Give `def_span` the same SyntaxContext as `span_with_body`.

https://github.com/rust-lang/rust/issues/102217

I'm not sure how to add a test, since the erroneous span was crafted using a proc macro.
The debug assertion in `def_span` will ensure we have the correct behaviour.

2022-10-02 20:42:21 +05:30

..

dep_graph

Allow query system to recover a HirId.

2022-10-01 15:58:42 +02:00

hir

Rollup merge of #102538 - cjgillot:def-span-ctxt, r=fee1-dead

2022-10-02 20:42:21 +05:30

infer

Const unification is already infallible, remove the error handling logic

2022-09-22 08:20:13 +00:00

middle

Refactor rustc lint API

2022-10-01 10:03:06 +00:00

mir

remove cfg(bootstrap)

2022-09-26 10:14:45 +02:00

query

Compute by owner instead of HirId.

2022-10-01 16:22:40 +02:00

thir

Use boxed slices in PatKind.

2022-09-02 07:26:22 +10:00

traits

Neither require nor imply lifetime bounds on opaque type for well formedness

2022-09-21 13:11:54 +00:00

ty

Compute by owner instead of HirId.

2022-10-01 16:22:40 +02:00

util

…

arena.rs

cache collect_trait_impl_trait_tys

2022-09-14 20:50:52 +00:00

error.rs

UPDATE - rename SessionSubdiagnostic macro to Subdiagnostic

2022-09-21 11:39:53 -04:00

lib.rs

remove cfg(bootstrap)

2022-09-26 10:14:45 +02:00

lint.rs

Use a SortedMap instead of a VecMap.

2022-10-01 16:24:30 +02:00

macros.rs

…

metadata.rs

rustc: Parameterize ty::Visibility over used ID

2022-09-07 13:35:41 +04:00

tests.rs

…

thir.rs

remove cfg(bootstrap)

2022-09-26 10:14:45 +02:00

values.rs

Make HandleCycleError an enum instead of a macro-generated closure

2022-09-06 19:26:08 -05:00