#KT-36247 fixed
A lot of testdata changed because significanly less (error) descriptors
are created for unresolved types, so diagnostics became different.
Before this commit, we used same transformers for all modules,
which provoked using same ScopeSession for all modules.
Now we re-create transformers for any new module.
This fixes some problems with incorrect caching in ScopeSession.
NB: this provokes ambiguities in some old FE tests,
mostly they're correct (same as old FE).
Expression will be checked against expected type later.
Theoretically, this is not very good, but it aligns with the old
inference, plus it helps avoiding multiple type mismatch diagnostics.
Because PackageViewDescriptor may consist of several package fragments from
different modules (see LazyPackageViewDescriptorImpl#fragments), we now filter
out fragments from irrelevant modules before rendering them into the .txt
The `@SinceKotlin("X.Y.Z")` annotation now hides a particular declaration from
resolution when the API version specified by the `-api-version` option is
_less_ than X.Y.Z. The comparison is performed as for versions in Maven:
MavenComparableVersion is in fact a copy of
org.apache.maven.artifact.versioning.ComparableVersion.
Also support "!API_VERSION" directive in diagnostic tests
#KT-14298 Fixed
Classifiers annotated with `@Deprecated(level = HIDDEN)` now have smaller
priority in imports, similarly to private classes. For example, if two
classifiers named Foo are imported with a star import and one of them is
deprecated-hidden, the name Foo in the source code is resolved to the other
one.
Also a minor change in multi-module diagnostic tests: do not append newlines
after the last module in the test
#KT-13926 Fixed
If a module is configured in such a way that the same function appears multiple
times via different dependencies, it's not helpful to report "none of the
following functions can be called", listing the same function multiple times
There are two different forms of types intestion:
1. Type parameters with multiple bounds
2. Smart casts
The problem was that when member scope of type intersection contained
effective duplicates and that lead to overload resolution ambiguity in
strange cases like `x.hashCode()`
For first type we do effectively the same thing as when building member
scope for class extending several interfaces: group all descriptors by
both-way-overridability relation and then choose most-specific in each
group.
For smart casts we do basically the same thing but with special
treatments:
1. From all descriptors that _equal_ to most specific we choose
the one that works without smartcast if possible (i.e. we choose first from candidates list)
2. If smart-cast value seems to be unstable we use only member scope
of receiver type + all descriptors from smart cast possible types
that has incompatible signature. If we'd include all of them and
choose one as more specific, and it would lead to false
SMART_CAST_IMPOSIBLE (see test unstableSmartCast.kt)
#KT-3996 Fixed
#KT-10315 Fixed
- locals win
- unary calls to plus/minus are not supported in favor of unaryPlus/unaryMinus
- unqualified nested classes are temporarily reported as unresolved
- function without receiver win against extension function
- explicit import win against star import
This allows to fix some cases when there is a difference between explicit and short reference to default object
Fix shorten reference, optimize imports and import insert helper for default objects
ShortenReferences always transforms default object references to shorter form for now
Fix DescriptorUtils#getFqName() for default objects (affects test data mostly)
Fix DescriptorUtils#getImportableDescriptor()
The fqname of class should be clear from code
Example: can't shorten A.Default.B.Default.C to A.B.C
Also fixes problem when nested class of enum class could be accessed via enum entry
Mikhail Glukhikh