Improve uninit/zeroed lint
* Also warn when creating a raw pointer with a NULL vtable.
* Also identify `MaybeUninit::uninit().assume_init()` and `MaybeUninit::zeroed().assume_init()` as dangerous.
Remove `InternedString`
This PR removes `InternedString` by converting all occurrences to `Symbol`. There are a handful of places that need to use the symbol chars instead of the symbol index, e.g. for stable sorting; local conversions `LocalInternedString` is used in those places.
r? @eddyb
It's a full conversion, except in `DefKey::compute_stable_hash()` where
a `Symbol` now is converted to an `InternedString` before being hashed.
This was necessary to avoid test failures.
Move to using Box<dyn Fn() -> ...> so that we can let plugins register
state.
This also adds a callback that'll get called from plugin registration so
that Clippy and other tools can register lints without using the plugin
API. The plugin API still works, but this new API is more compatible
with drivers other than rustc.
Access through tcx is fine -- by that point, the lint store is frozen,
but direct access through Session will go away in future commits, as
lint store is still mutable in early stages of Session, and will be
removed completely.
This stops storing the pass objects and instead stores constructor
functions.
The primary effect is that LintStore no longer has any interior
mutability.
Replace early_error and sess.err with bug!, in all cases. If the
compiler we're running with, including plugins, is registering something
twice, that's a (compiler/plugin) programmer error -- we should not try
to be nice at the cost of developer ergononomics (hiding the stacktrace
of the second registration is bad).
This also is basically a static bug in ~all cases so it should not be a
detriment to users, including with plugins.
Almost fully deprecate hir::map::Map.hir_to_node_id
- HirIdify `doctree::Module.id`
- HirIdify `hir::Crate.modules`
- introduce a `HirId` to `DefIndex` map in `map::Definitions`
The only last uses of `hir::map::Map.hir_to_node_id` in the compiler are:
- for the purposes of `driver::pretty` (in `map::nodes_matching_suffix`), but I don't know if we can remove `NodeId`s in there (I think when I attempted it previously there was some issue due to `HirId` not being representable with an integer)
- in `ty::query::on_disk_cache` (not sure about the purpose of this one)
- in `mir::transform::check_unsafety` (only important for error message order)
Any suggestions how to kill these off?
r? @Zoxc
Rollup of 11 pull requests
Successful merges:
- #62984 (Add lint for excess trailing semicolons)
- #63075 (Miri: Check that a ptr is aligned and inbounds already when evaluating `*`)
- #63490 (libsyntax: cleanup and refactor `pat.rs`)
- #63507 (When needing type annotations in local bindings, account for impl Trait and closures)
- #63509 (Point at the right enclosing scope when using `await` in non-async fn)
- #63528 (syntax: Remove `DummyResult::expr_only`)
- #63537 (expand: Unimplement `MutVisitor` on `MacroExpander`)
- #63542 (Add NodeId for Arm, Field and FieldPat)
- #63543 (Merge Variant and Variant_)
- #63560 (move test that shouldn't be in test/run-pass/)
- #63570 (Adjust tracking issues for `MaybeUninit<T>` gates)
Failed merges:
r? @ghost
