Merge #2396

2396: Switch to variant-granularity field type inference r=flodiebold a=matklad

r? @flodiebold 

Previously, we had a `ty` query for each field. This PR switcthes to a query per struct, which returns an `ArenaMap` with `Ty`s. 

I don't know which approach is better. What is bugging me about the original approach is that, if we do all queries on the "leaf" defs, in practice we get a ton of queries which repeatedly reach into the parent definition to compute module, resolver, etc. This *seems* wasteful (but I don't think this is really what causes any perf problems for us). 

At the same time, I've been looking at Kotlin, and they seem to use the general pattern of analyzing the *parent* definition, and storing info about children into a `BindingContext`. 

I don't really which way is preferable. I think I want to try this approach, where query granularity generally mirrors the data granularity. The primary motivation for me here is probably just hope that we can avoid adding a ton of helpers to a `StructField`, and maybe in general avoid the need to switch to a global `StructField`, using `LocalStructFieldId` most of the time internally. 

For external API (ie, for `ra_ide_api`), I think we should continue with fine-grained `StructField::ty` approach, which internally fetches the table for the whole struct and indexes into it.

In terms of actual memory savings, the results are as follows:

```
This PR:
   142kb FieldTypesQuery (deps)
    38kb FieldTypesQuery

Status Quo:
   208kb TypeForFieldQuery (deps)
    18kb TypeForFieldQuery
```

Note how the table itself occupies more than twice as much space! I don't have an explanation for this: a plausible hypothesis is that single-field structs are very common and for them the table is a pessimisation. 

THere's noticiable wallclock time difference.

Co-authored-by: Aleksey Kladov <[email protected]>

1 files changed, 6 insertions, 3 deletions

diff --git a/crates/ra_hir/src/ty/infer/pat.rs b/crates/ra_hir/src/ty/infer/pat.rs
index c125ddfbc..641d61e87 100644
--- a/crates/ra_hir/src/ty/infer/pat.rs
+++ b/crates/ra_hir/src/ty/infer/pat.rs
@@ -27,10 +27,11 @@ impl<'a, D: HirDatabase> InferenceContext<'a, D> {
 
         let substs = ty.substs().unwrap_or_else(Substs::empty);
 
+        let field_tys = def.map(|it| self.db.field_types(it.into())).unwrap_or_default();
         for (i, &subpat) in subpats.iter().enumerate() {
             let expected_ty = def
                 .and_then(|d| d.field(self.db, &Name::new_tuple_field(i)))
-                .map_or(Ty::Unknown, |field| field.ty(self.db))
+                .map_or(Ty::Unknown, |field| field_tys[field.id].clone())
                 .subst(&substs);
             let expected_ty = self.normalize_associated_types_in(expected_ty);
             self.infer_pat(subpat, &expected_ty, default_bm);
@@ -56,10 +57,12 @@ impl<'a, D: HirDatabase> InferenceContext<'a, D> {
 
         let substs = ty.substs().unwrap_or_else(Substs::empty);
 
+        let field_tys = def.map(|it| self.db.field_types(it.into())).unwrap_or_default();
         for subpat in subpats {
             let matching_field = def.and_then(|it| it.field(self.db, &subpat.name));
-            let expected_ty =
-                matching_field.map_or(Ty::Unknown, |field| field.ty(self.db)).subst(&substs);
+            let expected_ty = matching_field
+                .map_or(Ty::Unknown, |field| field_tys[field.id].clone())
+                .subst(&substs);
             let expected_ty = self.normalize_associated_types_in(expected_ty);
             self.infer_pat(subpat.pat, &expected_ty, default_bm);
         }