Annotation arguments that are resolved in COMPILER_REQUIRED_ANNOTATIONS
phase are resolved again in ANNOTATION_ARGUMENTS phase. If they resolve
to a different symbol, report an error.
KT-56177
If an annotation doesn't specify an explicit use-site target,
previously it was added to both, the primary constructor value parameter
and the property in the FIR. Then, in FIR2IR, only the "correct" one was
added to the IR. Move up the deduplication logic into the frontend.
^KT-56177 Fixed
```
open class Base<T> {
fun foo(): T = ...
}
class Derived<T> : Base<T> {
override fun foo(): T = ...
}
```
In intersection scope of type `Base<T> & Other<R>` we should create
intersection override based on `Base.foo(): T` and `Derived.foo(): R`
at the same time, despite the fact that `Derived.foo` actually directly
overrides `Base.foo`
^KT-56722 Fixed
Now all tests with `Fir` in name are named accordingly to parser which
is used in them -- `FirPsi` or `FirLightTree`. This is needed to keep
consistency between different types of tests, because there is no
single default in parser mode between different scenarios of using FIR
In K1, we have the rules like:
- if there's explicit type of a property, then use it
- if there's an initializer, obtain its expression-type
- Otherwise, use getter's return type
The case when getter's type is implicit is handled at
FirDeclarationsResolveTransformer.transformProperty
^KT-56707 Fixed
The commit is based on b09561c3c3
Co-authored-by: Dmitriy Novozhilov <dmitriy.novozhilov@jetbrains.com>
^KT-40904 Fixed
^KT-55177 Fixed
Review: https://jetbrains.team/p/kt/reviews/8731
True negative test already exist:
`compiler/testData/diagnostics/tests/multiplatform/hmpp/multiplatformCompositeAnalysis/intermediateWithActualAndExpect.kt`
In scope of KT-40904 KT-55177
Review: https://jetbrains.team/p/kt/reviews/8731
In order to make it possible to run hmpp tests with
`ENABLE_MULTIPLATFORM_COMPOSITE_ANALYSIS_MODE`, a dependency manager
must be implemented. This commit implements some basic dependency
manager. I'm not sure in its correctness because
`CommonDependenciesContainer` is an awkward API, but this dependency
manager works for my cases.
Why: I need hmpp + `ENABLE_MULTIPLATFORM_COMPOSITE_ANALYSIS_MODE`
infrastructure to cover `ExpectActualInTheSameModuleChecker` (I will add
the checker in the next commit) with tests. The checker couldn't be
covered with regular hmpp tests because regular hmpp tests and
`K2MetadataCompiler` run the compiler in a different way.
Contrary, `ENABLE_MULTIPLATFORM_COMPOSITE_ANALYSIS_MODE` runs the compiler in a
way `K2MetadataCompiler` does it.
I moved some tests from hmpp to `hmpp/multiplatformCompositeAnalysis`
because otherwise the tests would fail after I implement
`ExpectActualInTheSameModuleChecker` in the next commit.
Also the descriptor dumps were changed
(intermediateWithActualAndExpect.txt and
sealedInheritorsInComplexModuleStructure.txt). It happened because now
common source sets are no longer "squashed" into a single source set but
rather correct dependencies between source sets are established. And
each source set is analyzed separately (exactly like in
K2MetadataCompiler)
- KT-56505 occurred because `source.getChild(KtNodeTypes.MODIFIER_LIST)`
returns any modifier list in the subtree of the source element, not
necessarily the modifier list belonging to the checked element.
`depth = 1` restricts the search to the modifier list belonging to the
checked element itself.
- For example, given `f1` from KT-56505, `getChild` would return the
modifier list of `public var foo = 0`. Because it contains a
visibility modifier, `f1` wasn't marked with
`NO_EXPLICIT_VISIBILITY_IN_API_MODE`.
^KT-56505 fixed
- In explicit API mode, the `public` visibility modifier is not
redundant unless a declaration is hidden by a container. The
`REDUNDANT_VISIBILITY_MODIFIER` diagnostic is now not reported in such
cases.
^KTIJ-24485 fixed
- `REDUNDANT_MODIFIER_FOR_TARGET` is already reported for `open`
interfaces, but for consistency and IDE support, the compiler now
reports `REDUNDANT_MODALITY_MODIFIER` as well via the extended
checker.
- `REDUNDANT_MODIFIER_FOR_TARGET` cannot be disabled for this case
because it's reported via a basic checker while
`REDUNDANT_MODALITY_MODIFIER` is reported via an extended checker.
- Rename `implicitModality` to `redundantModalities` and return a set of
modalities. The idea of a *single* implicit/redundant modality doesn't
stand up to scrutiny. For example, for interfaces with `ABSTRACT`
implicit modality, `OPEN` is also a redundant modifier. But this is
not necessarily the case for all kinds of declarations.
Hypothetically, if it was possible to declare a class that is abstract
by default, adding an `open` modifier to that class would not be
redundant, as it would make the class instantiable.
It's been introduced in the previous commit
("K2: Simplify handling mixed smartcast vs. original candidates")
Because previously, it was assumed wrongly that each next level of
ConeCallConflictResolver filter out the candidates that are 100% less
applicable/specific, but the main one (ConeOverloadConflictResolver)
either leaves the single candidate or the whole same set, thus at
FilteringOutOriginalInPresenceOfSmartCastConeCallConflictResolver
we've got 4 candidates only two of which we might filter out.
The idea is that when we have successful candidates both from smart cast
and original type, we should discriminate in the favor of former ones.
While this problem (see kt55722.kt) existed before this branch is merged,
initially it was recognized on FP Ultimate when we stopped assuming
captured types from the same projections as equal (see kt55722Initial.kt).
^KT-55722 Fixed
^KT-55024 Fixed
^KT-56283 Related
^KT-56310 Related
This is mostly a revert of 2f61a2f56f
There, we erroneously assumed that we may take captured types as equal
if they are based on the same-typed projections.
Each instance of capturing defines its own captured type,
that should not be equal to any other type captured in other place.
Initial motivation was brought by FP Ultimate, where a piece of code
from the new test was found that started working differently after
recent changes.
The most obvious consequence is the change in addAllProjection.fir.kt:
one cannot use an instance as an argument when expected type
is captured type based on the same instance.
Otherwise, it would lead to CCE if we allowed to put arbitrary charsequences
to the list that initially was a MutableList<String>
All other test data changes (but addAllProjection.fir.kt and differentCapturedTypes.kt)
are irrelevant and will be fixed in the subsequent commits
Namely, do not choose `Nothing?` result type when fixing a variable
that has other constraints besides the ones that came from
the relevant type parameter's upper bounds.
See more details in KT-55691.
In K1, the case from specialCallWithMaterializeAndExpectedType.kt
was working (inferred to String?) just because the branches
were analyzed independently with `String?` expected type.
This change became necessary after the previous commit when we united
inference subsystems for if/when branches (see motivation there).
NB: For K1, the behavior is left the same, but the code
was refactored a bit.
^KT-55691 Fixed
^KT-56448 Fixed
Otherwise, it leads to branches inference run fully independent,
while there are cases when it's necessary to flow type information from
one of the branch to another (see the new test).
NB. In K1, it worked differently: if branches were inferred altogether
only for Any/Any? expect types (otherwise they're analyzed independently)
See foo2/foo4 in the test.
To avoid breaking change we need to support foo1/foo3, but we're trying
not to have some special rule for Any, so we've got a new resolution mode
that provides expect type, but doesn't require full completion.
^KT-45989 Fixed
^KT-56563 Fixed
^KT-54709 Related
For change in specialCallWithMaterializeAndExpectedType.kt
At first, see at KT-36776
Long time ago, it's been decided that if/when resolution
should look similar to similar "select()" calls,
but it's a breaking change (see KT-36776), and we were ready for that back then.
But then, there were too many broken cases found, thus we reverted it at
100a6f70ca
But probably, it would be better to try to infer `String?`
instead of `Nothing?` (see next commits)
Note that change in specialCallWithMaterializeAndExpectedType.kt
will be addressed in later commits, too
When expected type is known, use it as expected type for branch bodies.
While it indeed becomes different from the usual select call resolution,
where expected type is applied only after completion starts,
it helps to support, e.g. callable references resolution just as powerful
as it was in K1.
Also, in some cases where diagnostics have been changed, they become
a bit more helpful since they are reported closer
to the problematic places
cannotCastToFunction.kt test has been removed because it relied
on the case erroneously supported by the hack removed from
the FirCallResolver in this commit.
^KT-45989 Fixed
^KT-55936 Fixed
^KT-56445 Fixed
^KT-54709 Related
^KT-55931 Related
Beside some corner cases, it's already prohibited in K1 because
adaptation have a bit strange nature
(they don't represent any existing real function exactly)
^KT-55137 Fixed
This flag is true by default but is set to false for
- Java methods and constructors
- interface delegation methods that delegate to Java
The NAMED_ARGUMENTS_NOT_ALLOWED logic is mostly refactored to use the
new flag though some custom logic remains for determining the correct
message and to work around a corner case with fake overrides.
The flag is (de)serialized from/to metadata. For backward compatibility
with K1, delegated methods to Java types are deserialized as stable.
^KT-40480 Fixed
Kirill Rakhman