Add public function for resolving callable AST exprs to their HIR equivalents #16705
Conversation
rustbot
added
the
S-waiting-on-review
crates/hir/src/semantics.rs
Outdated
| pub fn resolve_method_call_expr( | ||
| &self, | ||
| method_call_expr: &ast::MethodCallExpr, | ||
| ) -> Option<Function> { | ||
| self.imp.resolve_method_call(method_call_expr) | ||
| } |
There was a problem hiding this comment.
This is already exposed via deref, we should re-export SemanticsImpl from the crate root so it and the deref impl for Semantics gets properly documented
There was a problem hiding this comment.
Oh, gotcha. Removed and #16707 opened for SemanticsImpl.
crates/hir/src/source_analyzer.rs
Outdated
| db: &dyn HirDatabase, | ||
| call: &ast::CallExpr, | ||
| ) -> Option<Callable> { | ||
| self.type_of_expr(db, &call.clone().into())?.0.as_callable(db) |
There was a problem hiding this comment.
This returns the return type as a callable, not the function itself. resolve_call seems unnecessary as you can just do sema.type_of_expr(call_expr.expr()).as_callable() (with Option handling) as is. (method calls are special because there is no nested expression that corresponds to the resolved function)
There was a problem hiding this comment.
Hm, yeah. I saw it used that way, but as a user of the public API without intricate knowledge of ra_ap_hir's internals the type_of_expr is rather non-obvious and thus easy to miss (anecdotally it took me quite a bit of searching before I found the pattern). While it doesn't strictly add any functionality that wasn't available before it makes it discoverable and thus more accessible to crate API users.
There was a problem hiding this comment.
Hmm, I wouldn't really say that that's related to our hir api at all, as this is effectively how rust works! A call expression is a call on some receiver expression, so in my eyes it's only logical to me that you'd query the callable on the receiver. Though I suppose we could have a general expr_as_callable(expr: &ast::Expr) -> Option<Callable> function for discoverability and parity with the method_call one.
There was a problem hiding this comment.
Good point, fn resolve_expr_as_callable it is then! Updated.
regexident
force-pushed
the
resolve-callable-exprs
branch
from
beb80ef
to
84f6663
Compare
regexident
force-pushed
the
resolve-callable-exprs
branch
from
84f6663
to
ed34978
Compare
|
@bors r+ |
|
☀️ Test successful - checks-actions |
1 similar comment
|
☀️ Test successful - checks-actions |
|
👀 Test was successful, but fast-forwarding failed: 422 Changes must be made through a pull request. |
(the PR is motivated by an outside use of the
ra_ap_hircrate that would benefit from being able to walk ahir::Function's AST, resolving callable exprs within to their HIR equivalents)