Aleksey Kladov
739babc391
Move Namespace enum closer to usage
2019-11-04 23:02:35 +03:00
Aleksey Kladov
872ac566bf
push name down to hir_expand
2019-10-30 18:56:20 +03:00
Aleksey Kladov
f8ddef875a
remove forward pointer for name
2019-10-30 17:43:14 +03:00
Aleksey Kladov
c9cd6aa370
Move ids to hir_def crate
2019-10-30 13:11:38 +03:00
Aleksey Kladov
a136cc0653
introduce ra_hir_def
2019-10-30 12:27:54 +03:00
Florian Diebold
632a22ae62
Upgrade Chalk, make use of TypeName::Error variant
2019-10-29 13:01:33 +01:00
Florian Diebold
b342ee5616
Upgrade Chalk (without using its dyn/impl Trait support)
2019-10-26 15:11:39 +02:00
Shotaro Yamada
965ca0d271
.collect()
directly into Arc<[T]>
2019-10-14 14:25:05 +09:00
kjeremy
4ba2c5b2de
Chalk update to simplified IR
2019-10-10 14:51:50 -04:00
Florian Diebold
daaf46177e
Add SubstsBuilder
...
+ further refactoring.
2019-09-26 23:03:39 +02:00
Florian Diebold
d0d1a202d4
Fix retrieval of the Fn traits
...
I used the wrong query, so it only used the lang items from the respective
crate...
2019-09-25 20:04:37 +02:00
Florian Diebold
6a86706650
Implement the call argument checking order hack for closures
2019-09-24 23:05:12 +02:00
Florian Diebold
3b06faad26
Make closures impl closure traits
2019-09-24 23:05:12 +02:00
Florian Diebold
619a8185a6
Give closures types
2019-09-24 23:05:12 +02:00
Florian Diebold
7bb6fdcf52
Upgrade Chalk again
2019-09-24 22:29:52 +02:00
Florian Diebold
6df9c4035a
Upgrade Chalk
2019-09-23 20:36:15 +02:00
Florian Diebold
bc905d202c
Handle projection types from Chalk
2019-09-22 20:08:46 +02:00
Florian Diebold
a040fde3ae
Remove TraitItem and ImplItem in favor of AssocItem
2019-09-17 19:47:45 +02:00
Florian Diebold
a61615c955
Upgrade Chalk
...
... and remove Ty::UnselectedProjection. It'll be handled differently.
2019-09-14 10:04:56 +02:00
Niko Matsakis
3f9996d18a
also make "unknown" case non-enumerable
2019-09-09 15:30:26 -04:00
Niko Matsakis
65582933cf
make all traits non-enumerable
...
As discussed on Zulip, this actually matches the present behavior of
rustc.
2019-09-09 15:24:24 -04:00
Florian Diebold
9db34eec20
Fix Chalk environments
...
The clauses need to be wrapped in `FromEnv` clauses for elaboration (i.e.
things like inferring `T: Clone` from `T: Copy`) to work correctly.
2019-09-07 16:30:37 +02:00
Florian Diebold
60bdb66ef2
Lower bounds on trait definition, and resolve assoc types from super traits
2019-09-07 14:31:43 +02:00
Florian Diebold
b8c1e402fa
Make type walking infrastructure a bit nicer
...
If/when we switch to using Chalk's Ty, we'll need to replace this by its `Fold`
trait, but I didn't want to import the whole thing just yet.
2019-09-03 14:00:35 +02:00
Florian Diebold
741e350d4b
Add support for associated type bindings (where Trait<Type = X>
)
2019-09-03 14:00:35 +02:00
Florian Diebold
16a7d8cc85
Add impl Trait
and dyn Trait
types
...
- refactor bounds handling in the AST a bit
- add HIR for bounds
- add `Ty::Dyn` and `Ty::Opaque` variants and lower `dyn Trait` / `impl Trait`
syntax to them
2019-08-22 19:33:00 +02:00
Florian Diebold
5af9691dc9
Handle placeholder assoc types when Chalk produces them
2019-08-12 21:43:00 +02:00
Florian Diebold
6265497523
Normalize associated types during inference
2019-08-12 21:43:00 +02:00
Florian Diebold
6cfdfdecba
Add representations of associated types
...
This adds three different representations, copied from the Chalk model:
- `Ty::Projection` is an associated type projection written somewhere in the
code, like `<Foo as Trait>::Bar`.
- `Ty::UnselectedProjection` is similar, but we don't know the trait
yet (`Foo::Bar`).
- The above representations are normalized to their actual types during type
inference. When that isn't possible, for example for `T::Item` inside an `fn
foo<T: Iterator>`, the type is normalized to an application type with
`TypeCtor::AssociatedType`.
2019-08-12 21:43:00 +02:00
Florian Diebold
dcb0c02e3a
Remove blacklist, instead mark Sized
as non-enumerable
...
This seems to be enough to prevent hanging in rust-analyzer, Chalk and the rustc
repo.
2019-07-15 08:49:55 +02:00
Florian Diebold
e2bce9e7eb
Cargo update, including updating Chalk
2019-07-14 18:22:47 +02:00
Florian Diebold
0a20770f46
Some renamings for clarity
2019-07-14 18:22:47 +02:00
Florian Diebold
9afbf2dff4
Unify normalize
and implements
to simplify code
2019-07-08 21:47:37 +02:00
Florian Diebold
15862fc041
Use environment for associated type normalization as well
2019-07-08 21:20:17 +02:00
Florian Diebold
b1b12072ed
Start handling environment in trait resolution
...
I.e. if we are inside a function with some where clauses, we assume these where
clauses hold.
2019-07-08 21:20:17 +02:00
Florian Diebold
638100dc8b
Refactor a bit & introduce Environment struct
2019-07-08 21:20:17 +02:00
Florian Diebold
f854a29c9a
Make EnumVariant a GenericDef and simplify some code
2019-07-06 17:43:13 +02:00
Aleksey Kladov
1834bae5b8
allow rustfmt to reorder imports
...
This wasn't a right decision in the first place, the feature flag was
broken in the last rustfmt release, and syntax highlighting of imports
is more important anyway
2019-07-04 23:09:09 +03:00
Aleksey Kladov
a198d78bd1
cache chalk queries
...
This gives a significant speedup, because chalk will call these
functions several times even withing a single revision. The only
significant one here is `impl_data`, but I figured it might be good to
cache others just for consistency.
The results I get are:
Before:
from scratch: 16.081457952s
no change: 15.846493ms
trivial change: 352.95592ms
comment change: 361.998408ms
const change: 457.629212ms
After:
from scratch: 14.910610278s
no change: 14.934647ms
trivial change: 85.633023ms
comment change: 96.433023ms
const change: 171.543296ms
Seems like a nice win!
2019-06-26 12:54:13 +03:00
Florian Diebold
30647cd195
Some more cleanup
2019-06-15 18:36:58 +02:00
Florian Diebold
49489dc20c
Add basic infrastructure for assoc type projection
2019-06-15 18:21:23 +02:00
Alan Du
ecd420636e
Fix clippy::single_match
2019-06-04 18:05:07 -04:00
Florian Diebold
7fda874dd4
Blacklist some traits from being considered in where clauses
...
For Send/Sync/Sized, we don't handle auto traits correctly yet and because they
have a lot of impls, they can easily lead to slowdowns. In the case of
Fn/FnMut/FnOnce, we don't parse the special Fn notation correctly yet and don't
handle closures yet, so we are very unlikely to find an impl.
2019-05-12 20:23:57 +02:00
Florian Diebold
d6dc75f9f2
Handle auto traits & negative impls
...
We don't pass field types to Chalk yet though, so the auto trait inference won't
be correct.
2019-05-11 16:21:20 +02:00
Florian Diebold
7744cd41e2
Reduce Chalk max_size parameter, add test for slow case
2019-05-11 16:21:20 +02:00
Florian Diebold
d8cd0e36f5
Handle Chalk conversion for FnDef
2019-05-11 16:21:20 +02:00
Florian Diebold
58b68966bf
Handle resolution errors in where clauses
...
This is slightly hacky, but maybe more elegant than alternative solutions: We
just use a hardcoded Chalk trait ID which we special-case to have no impls.
2019-05-11 16:21:20 +02:00
Florian Diebold
50bbf9eb09
Handle where clauses in trait solving
2019-05-11 16:21:20 +02:00
Florian Diebold
a4eb1a546c
Differentiate Tuple / FnPtr type constructors by cardinality
...
This is necessary because Chalk (reasonably) expects each 'struct' to know how
many type parameters it takes.
2019-05-04 19:11:21 +02:00
Florian Diebold
f43e69a64e
Handle recursive types in canonicalization
2019-05-04 18:25:36 +02:00