- `LibraryBinary` should not contain any decompiled files, as we want
FIR symbols in tests to be provided from indexed stubs or class files,
but definitely not from decompiled PSI. This brings `LibraryBinary`
much closer to the behavior of binary libraries in the IDE.
- Some tests may still require access to a decompiled file, for example
when trying to test `getOrBuildFir` for some `KtElement` coming from a
library. This commit introduces `LibraryBinaryDecompiled`, which
does contain decompiled files.
- We don't really need `LibraryBinary` as a main test module kind
anymore, since tests generally want to access some main `KtFile`.
Hence, test configurators for `LibraryBinary` have been turned into
configurators for `LibraryBinaryDecompiled`.
- An alternative would be decompiling files on demand, but this is not
currently feasible because the Standalone API doesn't reconcile stubs
with decompiled PSI, like the IDE does automatically. (For the same
declaration, the stub and the PSI will have a different identity.) As
long as there is no support for this, we'll have to rely on a separate
test module kind.
^KT-65960
When a symbol X from a binary library A uses another symbol Y from
another binary library B, `LLFirDependenciesSymbolProvider` cannot
resolve Y for X. This is because the existing
LLFirAbstractSessionFactory passes symbol providers for only builtin
libraries to `LLFirDependenciesSymbolProvider` even when the session has
dependencies. As a result, when `LLFirDependenciesSymbolProvider`
searches a symbol, it can find only symbols from the builtin libraries,
but it cannot find a symbol from libraries other than builtin libraries.
This happens only for the binary libraries.
^KT-65240 Fixed
This commit adds code to check whether a deserialized cone type is a
special function type kind or not when resolving the type of a lambda
expression (anonymous function). If it is a special function kind, it
sets the type of lambda based on the special function kind.
^KT-64994 Fixed
The test infrastructure for analysis supports binary module tests, but
the binary build does not use another binary module as a dependency when
it passes the class path. As a result, each binary module build does not
work when they have dependency on each other.
This commit fixes the issue by
1. Topological sort in the order of dependency graph for test modules.
2. Pass module paths as extra class paths when they have dependency on
each other.
^KT-64994
The base class in scripting considered obsolete and therefore supported
via some ad-hoc mechanisms. In particular parameters to the base class
c-tor are passed via script provided properties. But in combination
with the resolution logic, this leads to issues described in KT-60452
This commits filters out such parameters from script resolution
scope and avoids this problem for now.
Bot it should be noted that proper diagnostics for properties shadowing
should still be implemented - see #KT-65809
#KT-60452 fixed
Before this commit, the expect-actual resolver could find expects in
regular dependencies
Note: The appeared `VIRTUAL_MEMBER_HIDDEN` in
compiler/fir/analysis-tests/testData/resolveWithStdlib/multiModule/FakeOverrides.kt
isn't caused by my change. It's caused by fixing the testData dependency
syntax notation.
The testData improperly used regular dependency syntax notation, while
it should have been using dependsOn
Before:
Regular dependency syntax notation
// MODULE: androidMain(commonMain)
After:
dependsOn dependency syntax notation
// MODULE: androidMain()()(commonMain)
In some cases we transform flexible type into non-flexible during
enhancement, but don't add `@EnhancedNullability` attribute to them,
which breaks consistency with K1 on IR level
This commit fixes it
^KT-65302
Even though SO may not be correct
overrides sometimes, it feels more
natural to treat fake overrides as...
well, "overrides". And without it
we'd need to make the code in
`FirOverrideChecker` less intuitive.
Make parcelize work consistently with
`FirOverrideChecker`.
Actually, there is a philosophical question about
what to do when a plugin generates a class
in a new package, and this class is red code
(for instance, doesn't implement an abstract
member from a supertype). There's no source
to report such an error, but we probably do
want to run checkers to avoid trying to
compile red code to binaries (because it may
crash in backend, or it may silently work).
The root cause of the exception is that we missed such an element, and
it led to unresolved declaration during iteration over file declarations
^KT-65562 Fixed
^KT-61798 Fixed
Review: https://jetbrains.team/p/kt/reviews/14230/timeline
Q: How resolved modality of an accessor affects expect/actual matching
algorithm of properties?
A: It's not expect/actual matching of properties, but expect/actual
matching of properties setter. Declarations (setters in our case) in
actual are allowed to have wider visibility but only if the
appropriate expect is `final`. Before this commit, all property
accessors were always `final` by mistake. After this commit, they
have the modality derived from their containingProperty.
Related problem: PositionStrategy for ACTUAL_WITHOUT_EXPECT can be
improved KT-65720
The fix in `resolveModality` is similar to how `resolveVisibility` works
Related tests:
- ConfigurationCacheForAndroidIT.testKotlinAndroidProject
- FirIdeNormalAnalysisSourceModuleCombinedDeclaredMemberScopeTestGenerated.testEnumClassWithAbstractMembers
- FirIdeNormalAnalysisSourceModuleCombinedDeclaredMemberScopeTestGenerated.testEnumEntryInitializerWithOverriddenMember
- FirIdeNormalAnalysisSourceModuleDeclaredMemberScopeTestGenerated.testDelegateInterfaceLibrary
- FirIdeNormalAnalysisSourceModuleDeclaredMemberScopeTestGenerated.testEnumClassWithAbstractMembers
- FirIdeNormalAnalysisSourceModuleDeclaredMemberScopeTestGenerated.testEnumEntryInitializerWithOverriddenMember
- FirIdeNormalAnalysisSourceModuleMemberScopeTestGenerated.testEnumClassWithAbstractMembers
- FirIdeNormalAnalysisSourceModuleMemberScopeTestGenerated.testEnumEntryInitializerWithOverriddenMember
- FirIdeNormalAnalysisSourceModuleSymbolByPsiTestGenerated.testEnumEntryOverride
- FirIdeNormalAnalysisSourceModuleSymbolByPsiTestGenerated.testEnumValueMember
- FirStandaloneNormalAnalysisSourceModuleCombinedDeclaredMemberScopeTestGenerated.testEnumClassWithAbstractMembers
- FirStandaloneNormalAnalysisSourceModuleCombinedDeclaredMemberScopeTestGenerated.testEnumEntryInitializerWithOverriddenMember
- FirStandaloneNormalAnalysisSourceModuleDeclaredMemberScopeTestGenerated.testEnumClassWithAbstractMembers
- FirStandaloneNormalAnalysisSourceModuleDeclaredMemberScopeTestGenerated.testEnumEntryInitializerWithOverriddenMember
- FirStandaloneNormalAnalysisSourceModuleMemberScopeTestGenerated.testEnumClassWithAbstractMembers
- FirStandaloneNormalAnalysisSourceModuleMemberScopeTestGenerated.testEnumEntryInitializerWithOverriddenMember
- DiagnosticCompilerTestFE10TestdataTestGenerated*testChangeSetterVisibilityInOverride
This fixes a bunch of missing overridden symbols in IR.
This is also required for fixing KT-59921 in the following commit
where we need to keep all overridden symbols of intersection overrides
so that we can enhance them properly.
#KT-57300 Fixed
#KT-57299 Fixed
#KT-59921
#KT-57300
#KT-62788
#KT-64271
#KT-64382
NB: in order to produce correct IR origins, the source element kinds for
some FIR elements has been changed. As a side effect, mapping PSI to FIR
slightly changed: namely, for `a[b]++`, `a[b]` used to be mapped on
`set` call or callable reference, but now it is mapped on `get` call.
^KT-61891: Fixed
^KT-64387: Fixed
When encountering multiple available same named types,
resolve all of them together with corresponding extensions.
This aligns well with the current resolve behavior of classes
with the same name (see `SameNameClassesFromStarImports.kt`)
Also, add extra test for resolving the ambiguous type without
the extension function. The behavior is the same - resolve to the
both classes.
Render caret names in resolve testData, so that we don't have to guess
the indices of the carets;
it makes the multi-caret testData less error-prone.
Relevant for KT-65152 and KT-62695
The solution is not complete, since resolution of KDocs is not
fully specified and is rather tricky.
Many more corner cases are to be expected.
This solution should at least give some groundwork for the future
improvements/fixes of the resolver.
^KT-62695 Fixed
Make sure that, when ambiguous, both the package
and the callable/class/property are resolved from
KDoc reference.
Do a slight refactoring of `KDocReferenceResolver`
This behavior is also true for K1
Allow multiple carets to be used in reference resolve tests.
Adjust the testdata for the only test case which has actually
had multiple carets in it.
Required for KT-65152 and KT-62695
Instead of relying on UNRESOLVED_REFERENCE directive for the whole file,
just render the unresolved reference right in the .txt testdata
This will allow to introduce multi-caret resolve testdata more naturally
Required for KT-65152 and KT-62695
N.B. This implementation does not 100% prevent conflicts or resolve
alterations when shortening functions.
To guarantee that,
we would need to carefully consider all the references in the file,
and to check whether they have changed their resolve in the presence
of a new import.
This is not trivial and will be approached separately under a different
task.
^KTIJ-27841 Fixed
Marco Pennekamp