Ville Penttinen
18b0bd9bff
Add const type inference
2019-02-25 10:51:46 +02:00
Florian Diebold
bd8ed644e4
Rename Type => TypeAlias
2019-02-24 21:36:49 +01:00
Florian Diebold
c3c0979561
Add test for recursive type aliases
2019-02-24 20:54:04 +01:00
Florian Diebold
5d72b96988
Implement support for type aliases
2019-02-24 20:54:04 +01:00
Florian Diebold
82fe7b77a3
Refactor associated method resolution a bit and make it work with generics
2019-02-23 23:00:02 +01:00
Florian Diebold
dcfb4ee702
Split ty.rs into several modules
...
It was just getting too big. We now have:
- ty: the `Ty` enum and helpers
- ty::infer: actual type inference
- ty::lower: lowering from HIR to `Ty`
- ty::op: helpers for binary operations, currently
2019-02-23 15:36:38 +01:00
Florian Diebold
6a04d1f292
Fix resolution of associated method calls across crates
...
I think it'll be better to make the path resolution the number of unresolved
segments, not the first unresolved index; then this error could simply not have
happened. But I'll do that separately.
2019-02-23 12:37:29 +01:00
bors[bot]
3d8a0982a1
Merge #866
...
866: Implement basic support for Associated Methods r=flodiebold a=vipentti
This is my attempt at learning to understand how the type inference works by adding basic support for associated methods. Currently it does not resolve associated types or constants.
The basic idea is that `Resolver::resolve_path` returns a new `PathResult` type, which has two variants, `FullyResolved` and `PartiallyResolved`, fully resolved matches the previous behavior, where as `PartiallyResolved` contains the `PerNs<Resolution` in addition to a `segment_index` which contains the index of the segment which we failed to resolve. This index can then be used to continue inference in `infer_path_expr` using the `Type` we managed to resolve.
This changes some of the previous apis, so looking for feedback and suggestions.
This should enable fixing #832
Co-authored-by: Ville Penttinen <villem.penttinen@gmail.com>
2019-02-22 19:58:22 +00:00
Ville Penttinen
39679d499f
Ignore failing test for now
2019-02-22 00:27:22 +02:00
Ville Penttinen
2e7bc905be
Remove Const inference for now, refactor PathResult
2019-02-21 23:57:07 +02:00
Ville Penttinen
816971ebc9
Implement basic support for Associated Methods and Constants
...
This is done in `infer_path_expr`. When `Resolver::resolve_path` returns
`PartiallyResolved`, we use the returned `Resolution` together with the given
`segment_index` to check if we can find something matching the segment at
segment_index in the impls for that particular type.
2019-02-21 12:25:55 +02:00
Florian Diebold
72712b8a42
Fix handling of generics in tuple variants and refactor a bit
...
Also make them display a tiny bit nicer.
Fixes #860 .
2019-02-20 22:48:55 +01:00
bors[bot]
564ab84b78
Merge #852
...
852: Handle != r=flodiebold a=matklad
r? @flodiebold
Co-authored-by: Aleksey Kladov <aleksey.kladov@gmail.com>
2019-02-18 09:13:51 +00:00
Aleksey Kladov
4e8a3f565b
handle != operator
2019-02-18 10:09:44 +03:00
Florian Diebold
b82db68400
Handle tuple structs / enum variants properly in type inference
2019-02-17 20:53:13 +01:00
Florian Diebold
795d718ba1
Unify with the autorefed/autoderefed receiver type during method resolution
2019-02-17 14:44:39 +01:00
Florian Diebold
a1bda3fc08
Handle generic args for method calls
2019-02-16 23:06:41 +01:00
Florian Diebold
65bd9bc3a8
Handle impl generics in method calls
2019-02-16 23:06:41 +01:00
Florian Diebold
ccfc6b11c1
Add a test for impl generics
2019-02-16 23:06:41 +01:00
Florian Diebold
0242acae53
Turn ImplBlock into a copy type just containing IDs
...
This makes it more like the other code model types.
Also make Module::definition_source/declaration_source return HirFileIds, to
make them more like the other source functions.
2019-02-16 22:08:13 +01:00
Florian Diebold
e5f9d54661
Fix another crash
2019-02-12 21:49:54 +01:00
Laurențiu Nicola
7e8527f748
Implement completion for associated items
2019-02-12 12:51:01 +02:00
Aleksey Kladov
60524771fd
replace clone with copy
2019-02-12 13:32:34 +03:00
Aleksey Kladov
fa2e6e0bda
fix obsolete comment
2019-02-12 13:29:27 +03:00
Florian Diebold
c098a3fda5
Add comment and mark
2019-02-09 22:03:01 +01:00
Florian Diebold
a28d4befaf
Fix another crash, and try harder to prevent stack overflows
2019-02-09 21:53:55 +01:00
Florian Diebold
c0c3b37255
Fix another crash found when analyzing rustc
2019-02-09 21:37:30 +01:00
Florian Diebold
f1afc93353
Fix handling of literal patterns
...
Wrap them in a LiteralPat node so they can be distinguished from literal
expressions.
2019-02-09 21:37:30 +01:00
Florian Diebold
7ebde241c0
Fix two crashes found by running inference on all of rustc
2019-02-09 21:37:09 +01:00
Aleksey Kladov
12e3b4c70b
reformat the world
2019-02-08 14:49:43 +03:00
kjeremy
6753051a45
Some clippy cleanups
2019-02-06 15:50:26 -05:00
kjeremy
5227102c12
Test more Self inference
2019-02-04 14:44:06 -05:00
Florian Diebold
6b076f1931
Use new Resolver API in type inference
2019-02-01 22:45:23 +01:00
kjeremy
f0fdc9d5c0
Go To Implementation for Trait
2019-01-31 18:34:52 -05:00
bors[bot]
28fdb8d03c
Merge #701
...
701: Minor type inference tweaks r=flodiebold a=marcusklaas
Pass down expectation for reference expressions and type the guard in match expressions.
I wasn't able to add a test for the former addition because the type variable previously introduced would always resolve to the right type in the things I tried!
Co-authored-by: Marcus Klaas de Vries <mail@marcusklaas.nl>
2019-01-30 20:51:47 +00:00
Marcus Klaas de Vries
13cb4a1b37
Add test for passing on ref expectations
2019-01-30 21:14:26 +01:00
Aleksey Kladov
ef7eb8c257
Use Crate instead of CrateId
2019-01-30 22:37:20 +03:00
Jeremy Kolb
3c17643b30
Go to Implementation for structs and enums
2019-01-29 19:13:02 -05:00
Marcus Klaas de Vries
3daca3eb4d
Infer type of match guard
2019-01-28 23:09:14 +01:00
WizardOfMenlo
9416904d14
Added support for primitive types type inference when using std::ops::Not
2019-01-28 14:52:43 +00:00
Marcus Klaas de Vries
6249989e6c
Process second review
2019-01-27 17:59:21 +01:00
Marcus Klaas de Vries
f6eb44cd9e
Use type information from the turbofish
2019-01-27 17:59:21 +01:00
Marcus Klaas de Vries
0da1e8b2f8
Add a FnSig to Ty::FnDef
2019-01-27 17:59:21 +01:00
Marcus Klaas de Vries
04748a0f16
Add missing ty test snapshot
2019-01-27 17:59:21 +01:00
Marcus Klaas de Vries
aa06893a14
Add type params to FnSignature
2019-01-27 17:59:21 +01:00
Marcus Klaas de Vries
3bd47c0285
First attempt at generic type inference for fns
2019-01-27 17:59:21 +01:00
Florian Diebold
94bbb2418a
Add marks
2019-01-26 23:59:11 +01:00
Florian Diebold
77f92674f9
Handle cycles in type vars
...
This might be the cause of #587 .
2019-01-26 23:48:01 +01:00
Aleksey Kladov
619af1e22c
fix AST for if expressions
...
then is not always a block...
2019-01-27 00:37:11 +03:00
Aleksey Kladov
9457b1f0e6
rename source_file -> parse
2019-01-26 11:51:36 +03:00
Aleksey Kladov
4711cbcace
rename FilesDatabase -> SourceDatabase
2019-01-26 11:20:30 +03:00
Aleksey Kladov
be1a005ebd
fold syntax database into files database
2019-01-26 11:09:39 +03:00
Florian Diebold
5a1385007c
Fix type inference of binops on int/float variables
...
Fixes #651 .
2019-01-25 21:32:49 +01:00
Aleksey Kladov
60a607d33f
new struct id
2019-01-24 17:54:18 +03:00
Aleksey Kladov
ec7ed054e0
Functions use new id scheme
2019-01-24 15:28:50 +03:00
Aleksey Kladov
3ab1519cb2
Change ids strategy
...
this is a part of larghish hir refactoring which aims to
* replace per-source-root module trees with per crate trees
* switch from a monotyped DedId to type-specific ids
2019-01-24 13:29:19 +03:00
Aleksey Kladov
7b901f86cd
move SyntaxPtr to ra_syntax
2019-01-23 17:37:10 +03:00
Jeremy A. Kolb
7c27e6d2b3
Update snaps
2019-01-22 09:51:29 -05:00
Florian Diebold
a252110616
Use insta for ty tests
2019-01-21 22:52:35 +01:00
Florian Diebold
e45f476d56
Fix panic trying to get substs on unknown type
...
Fixes #585 .
2019-01-21 21:42:19 +01:00
Florian Diebold
3c4a4c6ae9
Fix panic on inferring field access on an enum
2019-01-20 18:14:19 +01:00
Florian Diebold
969f588025
Generics -> GenericParams
2019-01-19 18:58:04 +01:00
Florian Diebold
9e4b5ecec4
Make generics work in struct patterns
2019-01-19 16:02:06 +01:00
Florian Diebold
d37bb128ef
Collect generic args in struct variant paths as well
2019-01-19 16:02:06 +01:00
Florian Diebold
cc4562ab6e
Collect generic args in type paths
...
E.g. `let x: A<X>` is handled correctly.
2019-01-19 16:02:06 +01:00
Florian Diebold
688a45e00b
Implement beginnings of generics
...
- add HIR for generic params
- resolve generic params in type paths
- add substitions for ADTs
- insert type variables for substitutions
2019-01-19 16:02:06 +01:00
Florian Diebold
415cdc5210
Add some tests
2019-01-19 15:49:48 +01:00
Marcus Klaas de Vries
44e9a9605b
Fixup annotated bindings
2019-01-19 15:37:26 +03:00
Marcus Klaas de Vries
d48d5b8b6c
Add initial (flawed) implementation of binding annotations
2019-01-19 15:37:26 +03:00
Marcus Klaas de Vries
9433a108cf
Make pattern inference work w/o proper expecations
2019-01-19 15:37:25 +03:00
Marcus Klaas de Vries
be1b4034a5
Infer pattern types for lambdas and loops
2019-01-19 15:37:25 +03:00
Marcus Klaas de Vries
06d16a18f6
Implement match binding type inference and arm unification
2019-01-19 15:37:25 +03:00
Marcus Klaas de Vries
ac216880f5
Implement unlabeled struct field pattern inference
2019-01-19 15:37:25 +03:00
Marcus Klaas de Vries
3340807bd2
Get basic struct pattern type inference working!
2019-01-19 15:37:25 +03:00
Marcus Klaas de Vries
9c2d83a4c8
Add crude implementation of tuplestruct pattern inference
2019-01-19 15:37:25 +03:00
Marcus Klaas de Vries
3b0de53904
Work on type inference for ADT patterns
2019-01-19 15:37:25 +03:00
Marcus Klaas de Vries
5648dcd36e
Implement type inference for tuples and refs
2019-01-19 15:37:25 +03:00
Marcus Klaas de Vries
afaa26636e
Add additional pattern variants
2019-01-19 15:37:25 +03:00
Aleksey Kladov
a2ca03d10b
⬆️ salsa
2019-01-17 14:11:00 +03:00
Hirokazu Hata
0aedd4fb2f
Add Ty::Array to walk_mut
2019-01-17 00:08:53 +09:00
Hirokazu Hata
f66ca1bd2e
Fix test case invalid syntax
2019-01-16 23:39:14 +09:00
Hirokazu Hata
478ce1c8b5
Fix infer array test
2019-01-16 23:37:42 +09:00
Hirokazu Hata
2a3262c757
Add infer array test
2019-01-16 23:35:06 +09:00
Marcus Klaas de Vries
a2b6d3da30
Implement rudimentary type inference for unary operators
2019-01-16 00:04:39 +01:00
Aleksey Kladov
8ba9c2d4ce
remove Cancelable from type inference
2019-01-15 20:54:18 +03:00
Aleksey Kladov
b871062e32
remove Cancelable from Ty
2019-01-15 20:43:37 +03:00
Aleksey Kladov
fafcd103d2
remove Cancelable from ids
2019-01-15 19:18:52 +03:00
Aleksey Kladov
18e9a710cd
remove Cancelable from fn_scopes
2019-01-15 19:04:49 +03:00
Aleksey Kladov
8af9a18660
remove Cancelable from Crate API
2019-01-15 18:33:26 +03:00
Aleksey Kladov
c159e414b4
remove Cancelable from Module API, part 2
2019-01-15 18:30:58 +03:00
Aleksey Kladov
11f3c8afb2
remove Cancelable from source binders
2019-01-15 18:13:49 +03:00
Marcus Klaas de Vries
d67eabb512
Fix type inference for raw (byte) strings
2019-01-14 20:56:14 +01:00
Marcus Klaas de Vries
a9a6a50c75
Fixup tests
2019-01-14 19:30:21 +01:00
Marcus Klaas de Vries
81bc8e4973
don't try to treat arrays and tuples as literals
2019-01-14 13:55:49 +01:00
Marcus Klaas de Vries
1574715be5
Use type variables to determine exact type for ambiguous numeric literals
2019-01-14 13:54:31 +01:00
Marcus Klaas de Vries
5f5dc20d85
Try implementing integer type inference (WIP)
2019-01-14 13:52:57 +01:00
Marcus Klaas de Vries
a6146d35b1
Implement type inference for literals (WIP)
2019-01-14 13:52:55 +01:00
Hirokazu Hata
139da0841d
Fix tuple test case
2019-01-13 22:23:06 +09:00
Hirokazu Hata
829f668bd7
Add infer tuple test
2019-01-13 21:04:03 +09:00
Florian Diebold
1ed7fbfc1b
args -> params
2019-01-12 21:58:16 +01:00
Florian Diebold
5db5f5cc1d
Small improvements from review comments
2019-01-12 21:51:56 +01:00
Florian Diebold
082ef52bcb
Implement basic inherent method resolution
2019-01-12 15:01:19 +01:00
Florian Diebold
1212e59bee
Fix assertion error in unification (hopefully)
...
Currently, all types that we handle during inference need to be resolved as far
as possible at the time. It's maybe too brittle of an invariant; I need to think
how we can do this better. This should fix #484 though, I hope (if
it's the same case as I managed to reproduce).
2019-01-11 22:59:00 +01:00
Marcus Klaas de Vries
978de5cf8b
Implement type inference for enum variants
2019-01-10 14:32:56 +01:00
Aleksey Kladov
e8815b614e
nicer trailing comma handling in types
2019-01-09 19:14:21 +03:00
Marcus Klaas de Vries
0b8fbb4fad
Fix typos in ARCHITECTURE.md and a number of crates
...
specifically: gen_lsp_server, ra_arena, ra_cli, ra_db, ra_hir
2019-01-09 01:17:09 +01:00
Marcus Klaas de Vries
e51d44a2de
Process explicit type hints for str, bool and char
2019-01-07 20:43:41 +01:00
Marcus Klaas de Vries
5d15dd70b0
Tidy up binary operator type inference; add test file
2019-01-07 20:39:23 +01:00
Marcus Klaas de Vries
7b0eaef580
Implement type inference for more binary operators
...
Mostly just for primitive numeric types such as u32 and f64. Not
yet a general solution using trait resolution.
2019-01-07 20:11:31 +01:00
Florian Diebold
7bb279b365
Implement autoderef for field accesses
2019-01-07 14:54:23 +01:00
Florian Diebold
d618b1f2ce
if let -> match
2019-01-07 01:10:29 +01:00
Florian Diebold
71f7d82e45
Introduce ArenaMap
2019-01-07 00:05:19 +01:00
Florian Diebold
cf49a11263
Sort ranges in type inference tests
...
Also rename the files to remove the numbers (they don't serve a purpose now that
there are only the data files).
2019-01-07 00:05:19 +01:00
Florian Diebold
6210e82041
Use HIR Expr for type inference
...
Now we can reuse the type inference inside a function when typing whitespace
etc. :)
2019-01-07 00:05:19 +01:00
bors[bot]
31c1999505
Merge #440
...
440: Implement type inference for boolean operators r=flodiebold a=marcusklaas
Tried implementing the easiest part of https://github.com/rust-analyzer/rust-analyzer/issues/390 . Hope this is somewhat close to what the intent of the issue was. Found it surprisingly easy to find my way around the repository - it's well organized!
Very grateful for any pointers.
Co-authored-by: Marcus Klaas de Vries <mail@marcusklaas.nl>
2019-01-06 21:28:36 +00:00
Marcus Klaas de Vries
82d9a77dad
Touch up type inference for boolean operators
...
Also try to infer its subexpressions and set type expectations
whenever possible.
2019-01-06 22:17:54 +01:00
Florian Diebold
a6f33b4ca5
Add test for invalidation of inferred types when typing inside function
...
This currently fails, but should work once we have hir::Expr.
2019-01-05 22:41:12 +01:00
Marcus Klaas de Vries
4fc233a02e
Implement type inference for boolean operators
2019-01-05 21:28:30 +01:00
Florian Diebold
538147bf94
Resolve the Self type
2019-01-04 19:13:50 +01:00
Florian Diebold
d4db61b9a1
Resolve the self parameter during type inference
2019-01-04 19:12:29 +01:00
Florian Diebold
111126ed3c
Type the self parameter
2019-01-04 19:10:50 +01:00
Florian Diebold
226e31dae9
Add test for self type inference
2019-01-04 18:28:43 +01:00
Florian Diebold
b1590bdf6a
Missing return type means unit, not unknown
2018-12-29 12:08:57 +01:00
Florian Diebold
cfa1de72eb
Implement type variables
...
This will really become necessary when we implement generics, but even now, it
allows us to reason 'backwards' to infer types of expressions that we didn't
understand for some reason.
We use ena, the union-find implementation extracted from rustc, to keep track of
type variables.
2018-12-29 12:04:34 +01:00
Aleksey Kladov
a9f55029b9
introduce known names
2018-12-27 20:26:15 +03:00
Florian Diebold
2870effd5c
Implement reference / pointer types
...
- parse them
- infer types of & and * expressions
2018-12-25 20:36:06 +01:00
Florian Diebold
55c941cd9f
Type field accesses
2018-12-25 15:16:42 +01:00
Florian Diebold
6fcd38cc81
Infer result of struct literals, and recurse into their child expressions
2018-12-25 15:16:42 +01:00
Florian Diebold
4ff1618520
Do name resolution by namespace (types/values)
2018-12-25 15:16:42 +01:00
Florian Diebold
b5b68f2094
Add basic HIR and types for structs/enums
2018-12-25 15:16:42 +01:00
Florian Diebold
4befde1eee
Change inference tests to have one per file
2018-12-24 15:36:54 +01:00
Florian Diebold
655f5bc261
Rename a variable for consistency
2018-12-24 15:19:49 +01:00
Florian Diebold
c85748f5fb
Type the return values of call expressions
2018-12-23 17:29:03 +01:00
Florian Diebold
ef67581104
Resolve paths to defs (functions currently) during type inference
2018-12-23 17:13:11 +01:00
Florian Diebold
93ffbf80c6
Make let statements kind of work
2018-12-23 14:01:50 +01:00
Florian Diebold
515c3bc59b
Cleanup
2018-12-23 13:48:04 +01:00
Florian Diebold
7348f7883f
Add testing infrastructure for type inference
...
- move dir_tests to test_utils for that.
2018-12-23 13:48:04 +01:00
Florian Diebold
3899898d75
Parse integer / float types
2018-12-23 13:48:04 +01:00
Florian Diebold
3ac605e687
Add beginnings of type infrastructure
2018-12-23 13:48:04 +01:00
