Commit Graph

64 Commits

Author SHA1 Message Date
Victor Petukhov 559bedf5ae Continue processing not builder inference stub variables in the type checker during computing common super type
^KT-47148 Fixed
2021-06-15 17:40:56 +03:00
Tianyu Geng ce767046eb FIR checkers: report SMARTCAST_IMPOSSIBLE 2021-06-10 16:01:13 +03:00
Denis.Zharkov 6136526a3a FIR: Avoid reporting inference errors from DelegatedPropertyConstraintPosition
All necessary diagnostics have already been reported through the checkers
2021-06-07 15:25:50 +03:00
Denis.Zharkov c420957eac FIR: Report INFERENCE_NO_INFORMATION_FOR_PARAMETER diagnostic 2021-06-07 15:25:45 +03:00
Victor Petukhov 1e5998e0ba Add constraints between two different stub variables instead of creating constraint error due to T1 isn't subtype of T2
^KT-44241 Fixed
2021-06-03 19:18:09 +03:00
Victor Petukhov 124a14c8df Use separate constraint position during call substitution as part of inferring postponed type variables
^KT-47052 Fixed
^KT-47082 Fixed
2021-06-03 12:37:17 +03:00
Dmitriy Novozhilov 796f8e6bce Revert "FIR checkers: report SMARTCAST_IMPOSSIBLE"
This reverts commit 84334b08
2021-06-03 09:48:50 +03:00
Tianyu Geng 84334b087c FIR checkers: report SMARTCAST_IMPOSSIBLE 2021-06-02 13:19:12 +03:00
Victor Petukhov 9fd1cbd2e7 Inject stub type variables of a different builder inference call properly 2021-05-28 15:36:19 +03:00
Victor Petukhov 0c427555cf Support definitely not null stub types and proper subtyping on them 2021-05-28 15:36:19 +03:00
Victor Petukhov 5a11450d77 Split stub types into stub type for subtyping and for builder inference and use them in the proper way 2021-05-28 15:36:17 +03:00
Victor Petukhov ac7b459f2a Do subtyping with stub types properly
1) Return stub type if we are calculating super type between two same stub types
2) Return nullable Any if those stub types are different
2021-05-28 15:36:15 +03:00
Denis.Zharkov faa5e46396 FIR: Fix inference case with flexible captured types
Previsously, errors have been ignored because we ignored errors raised
from the completion phase

See the comment above the createConstraintPartForLowerBoundAndFlexibleTypeVariable
2021-05-20 17:24:28 +03:00
Denis.Zharkov dac5c31993 Minor. Simplify ConstraintSystemUtilContext::createArgumentConstraintPosition 2021-05-20 17:24:23 +03:00
Andrey Zinovyev b1c8669b43 [FIR] Resolve infix calls to infix functions only
Implements INFIX_MODIFIER_REQUIRED diagnostics
2021-05-20 13:46:27 +03:00
Victor Petukhov 2239404085 Extract type preparation during type checking to a separate component 2021-05-18 17:52:23 +03:00
Victor Petukhov 7c62e9aecd Introduce warnings reporting by missed constraints because of incorrect optimization in the constraints processor 2021-04-30 15:46:04 +03:00
Victor Petukhov e110b49cab Revert "Don't stop constraints processing if all type variables have proper equality constraints"
This reverts commit b87c2a15
2021-04-30 15:46:03 +03:00
Victor Petukhov 867d7b5bca Allow type variable fixation into intersection type if it isn't meaningless (i.e. has one or more final classes, or two or more open classes)
^KT-46186 Fixed
2021-04-23 15:12:13 +03:00
Tianyu Geng 6a03f31e50 FIR: add UnsafeCall resolution diagnostics
Previously unsafe call is reported as part of InapplicableWrongReceiver.
This makes it difficult for the downstream checkers to report different
diagnostics.
2021-04-19 15:11:13 +03:00
Mikhail Glukhikh 9b3f1b9b8a Inference: handle Exact constraints with captured types properly
#KT-41818 Fixed
2021-04-15 15:55:41 +03:00
Denis.Zharkov 37f923a98f Do not approximate DefinitelyNotNullType in public declarations
Because since 1.6 they become normal denotable types

^KT-26245 Fixed
2021-04-09 13:06:37 +03:00
Denis.Zharkov e74a0c7ef7 Pass languageVersionSettings to AbstractTypeApproximator 2021-04-09 13:06:35 +03:00
Victor Petukhov 7f2c5cde55 Propagate input type position across incorporation properly
^KT-45719 Fixed
2021-03-29 16:17:31 +03:00
Victor Petukhov d06031ece3 Approximate types when only input types check is done. Otherwise given types are inconsistent with types obtained by ResultTypeResolver
^KT-45714 Fixed
2021-03-29 16:17:30 +03:00
Victor Petukhov 60f2f85be7 Process captured types with type variable inside properly, in the operations related with the type variables fixation 2021-03-19 16:16:11 +03:00
Victor Petukhov 91d2f32a57 Don't lose upper non-expected type constraints to include them to intersection type during finding the result type of the fixing type variable 2021-03-12 14:36:54 +03:00
Victor Petukhov b87c2a15b5 Don't stop constraints processing if all type variables have proper equality constraints
Those type variable may have other constraints after incorporation into which, the constraint error may appear

^KT-42042 Fixed
2021-03-12 14:36:54 +03:00
Victor Petukhov 7f7bb70596 Don't fix a type variable into the intersection type if there is an explicit expected type
^KT-43303 Fixed
^KT-42396 Fixed
^KT-42472 Fixed
2021-03-12 14:36:53 +03:00
Ilya Kirillov af4d300686 Use nullability from approximated local type in AbstractTypeApproximator 2021-02-19 16:44:47 +01:00
Dmitriy Novozhilov 3626008ed2 [Inference] Add ability to approximate local types in AbstractTypeApproximator 2021-02-19 12:47:27 +03:00
Simon Ogorodnik 3909e3c54c Decouple TypeCheckerContext and TypeSystemContext 2021-02-15 18:28:58 +03:00
Victor Petukhov dcad9c84fc Don't fix type variables into Nothing in priority way
^KT-44546 Fixed
2021-02-15 15:12:45 +03:00
Victor Petukhov 9efac8f68b Propagate all annotations during creating simple functional types
^KT-44563 Fixed
2021-01-29 13:50:46 +03:00
Victor Petukhov 977ac7cbe7 Propagate nullness flexibility into the result type from type variables in other constraints during result type finding, to prevent variable fixation to less flexible type
^KT-44540 Fixed
2021-01-27 17:23:15 +03:00
Victor Petukhov f06a5321a6 Approximate definitely not-null types for type parameter's types if they are already not-null (has not-null upper bounds)
^KT-44440 Fixed
2021-01-25 14:17:58 +03:00
Victor Petukhov 0dde5ddd7e Fail calls constraints of which contain uninferred type parameter
^KT-44055 Fixed

It's possible only if there is a callable reference among subcalls which go though the old type inference (and the error for uninferred type parameter wasn't reported)
2020-12-25 11:53:16 +03:00
Victor Petukhov 1926434b18 Report error about uninferred type parameter for some special call' subcalls 2020-12-25 11:53:14 +03:00
Jinseong Jeon 0ea6b32c01 NI: allow lower bound of flexible type for coercion-to-Unit
Example from
box/inference/coercionToUnitForLambdaReturnTypeWithFlexibleConstraint

// FILE: TestJ.java
public class TestJ {
  public static <T> In<T> materialize() { return null; }
}

// FILE: test.kt

class In<in T>

fun <T> inferred(e: In<T>?, l: () -> T): T = l()

fun box() {
  inferred(TestJ.materialize<Unit>(), { null })
}

`materialize` has flexible type, both for `In<T>` and `T`.
When analyzing `{ null }`, collected type constraints include:
ft<Unit?, Unit> <: T (from ft<In<ft<Unit, Unit?>>, In<ft<Unit, Unit?>>?>)

By allowing the lower bound of flexible type, FIR resolution can visit
`{ null }` with the expected type Unit, which will lead to proper
coercion to Unit at the end.
2020-12-15 17:52:48 +03:00
Victor Petukhov 04846ca47a Rework checking constraints by presented OnlyInputTypes annotation in accordance with changed incorporation mechanism 2020-11-25 11:15:23 +03:00
Victor Petukhov 0857b9c9e7 Rethink constraints incorporation
Namely, remove incorporation “otherInsideMyConstraint” to eliminate
constraint system redundancy and produce a potentially very large number
 of constructs.
Instead, introduce not so “spreadable” incorporation during variable
fixation (equality constraint with result type into other constraints).
^KT-41644 Fixed
^KT-42195 Fixed
^KT-42920 Fixed
^KT-42791 Fixed
^KT-41741 Fixed
2020-11-25 11:15:20 +03:00
Victor Petukhov 616e40f879 Reuse equality constraints during simplification in adding constraints 2020-11-25 11:15:19 +03:00
Alexander Udalov b3e79d36df Fix compiler warnings and some inspections 2020-11-03 14:00:53 +01:00
Victor Petukhov 84129098cb Add equality constraints without subtyping
^KT-42195 Fixed
^KT-41741 Fixed
2020-10-21 21:23:13 +03:00
Victor Petukhov c6da2a1138 Reuse built functional types for postponed arguments by expected types and paths from a top level type variable
^KT-42221 Fixed
2020-10-16 10:46:22 +03:00
Pavel Kirpichenkov 39a87435ee [FIR/NI] Refactor type variable gathering from lambda types
Motivation:
- drop getArguments from type context as a duplicate of getArgumentList
- reduce the number of collection allocations in getAllDeeplyRelatedTypeVariables

Additional minor improvements, test data fixes
2020-10-13 11:00:21 +03:00
Pavel Kirpichenkov 712a2ce1ab [FIR] Improved lambda completion: initial implementation
Repeat the logic of KotlinConstraintSystemCompleter in ConstraintSystemCompleter.
Implement additional context operations required for updated lambda completion algorithm.
2020-10-13 11:00:21 +03:00
Pavel Kirpichenkov 5eae6f2f4e [FIR] Move PostponedArgumentInputTypesResolver to resolution.common 2020-10-13 11:00:20 +03:00
Pavel Kirpichenkov 3822a32fce [FIR] Prepare commonization of PostponedArgumentInputTypesResolver 2020-10-13 11:00:20 +03:00
Victor Petukhov 6c1dc43d25 Add check already fixed variables in PostponedArgumentInputTypesResolver during adding constraints on them and using inside a functional type
^KT-42374 Fixed
2020-10-02 18:31:54 +03:00