Merge #1786

1786: Various minor trait improvements r=matklad a=flodiebold

 - lower bounds on trait definition, i.e. super traits
 - use super traits for associated types
 - use traits from where clauses and their super traits for method resolution
 - lower fn-like paths (i.e. `Fn(X, Y) -> Z`)
 - pass the environment to Chalk in the correct way to make elaboration work, i.e. inferring things like `T: Clone` from `T: Copy`. The clauses need to be wrapped in `FromEnv` clauses for that to work, which I didn't do before.
 - add some tests for closure inference already

Co-authored-by: Florian Diebold <[email protected]>

1 files changed, 3 insertions, 2 deletions

diff --git a/crates/ra_hir/src/ty/traits/chalk.rs b/crates/ra_hir/src/ty/traits/chalk.rs
index c201c5e50..8a127efa1 100644
--- a/crates/ra_hir/src/ty/traits/chalk.rs
+++ b/crates/ra_hir/src/ty/traits/chalk.rs
@@ -314,7 +314,8 @@ impl ToChalk for Arc<super::TraitEnvironment> {
                 // for env, we just ignore errors
                 continue;
             }
-            clauses.push(pred.clone().to_chalk(db).cast());
+            let program_clause: chalk_ir::ProgramClause = pred.clone().to_chalk(db).cast();
+            clauses.push(program_clause.into_from_env_clause());
         }
         chalk_ir::Environment::new().add_clauses(clauses)
     }
@@ -636,7 +637,7 @@ pub(crate) fn impl_datum_query(
             _ => None,
         })
         .filter_map(|t| {
-            let assoc_ty = trait_.associated_type_by_name(db, t.name(db))?;
+            let assoc_ty = trait_.associated_type_by_name(db, &t.name(db))?;
             let ty = db.type_for_def(t.into(), crate::Namespace::Types).subst(&bound_vars);
             Some(chalk_rust_ir::AssociatedTyValue {
                 impl_id,