The change 9de6df6104 caused tests `requireKotlinCompilerVersion{,K2}`
to fail once the compiler version was changed to 2.0.0-dev-*. The reason
is that when patch version is 0, compiler version is equal to the
language version (which is itself always equal to JvmMetadataVersion).
So if we're replacing JvmMetadataVersion first, we end up with
`$ABI_VERSION$-dev-...` in the output, which is incorrect.
Rewrite FlatteningSequence to use
the same optimization technique as AbstractIterator.
Didn't extend AbstractIterator as storing and reloading a value
from a field is slightly slower compared to the implemented approach.
Use integer constants instead of enum to track AbstractIterator's state:
- for switches with a few cases, integer constants are
faster as there's no need to load ordinal value from
enum and map it into tableswitch index;
- on JVM, JIT-compiler can optimize away state-tracking code if it uses
integer constants, but when it comes to enums and tableswitches,
it can't apply a same set of optimizations to it.
Now there is separate class encapsulating logic about how to build
fakeOverrides and one encapsulating logic of which classes
do need building fake overrides.
This also allows to untie strange inheritance dependencies.
^KT-61934
Previously, it had some internal state that was saved between calls
and needed to be cleared.
This state was eliminated, which makes invariants clearer and using
easier.
^KT-61934
Implement an infrastructure for checking IR before JS klib serialization.
Implement the EXPORTING_JS_NAME_CLASH and EXPORTING_JS_NAME_CLASH_ES checks.
^KT-61710 Fixed
Move the checkFullDiagnosticRender() function from
the JvmBackendDiagnosticsHandler class so it can be reused
in the JsBackendDiagnosticsHandler class.
This is the first step to replace the usages of `Importable`
in the FIR generator with `TypeRef`.
This is a step towards commonizing the code generator between
FIR and IR: KT-61970
Recently, 'KtFirCompilerFacility' started to compile files from
dependent modules separately. Until recently, the code fragment was
in the same module as its context, so they naturally went in the same
compilation round. Now a separate 'KtCodeFragmentModule' is created,
so the code fragment case has to be handled explicitly.
This fixes the following groups of tests:
- FirIdeNormalAnalysisLibrarySourceModuleCompilerFacilityTestGenerated
- K2IdeK2CodeKotlinEvaluateExpressionTestGenerated (in IntelliJ)
FIR implementation already does not have this check in
`createContextDependentCopy`, because it's called not
inside `analyze` call, but before it (see
`analyzeInDependedAnalysisSession`)
^KT-62139 Fixed
This uses the same approach as
INFERRED_TYPE_VARIABLE_INTO_EMPTY_INTERSECTION where we use a visitor
to find a call to a symbol that contains the type variable in question.
#KT-56140 Fixed
'DFANode' is not created for certain elements, such as types or
references. Before the change, only the target element itself was
queried for DFA, making the 'ContextCollector' return an empty map of
smart casts.
^KTIJ-26973 Fixed
Implement the same behavior as in K1 evaluator. Also see
'IRFragmentCompilerCodegen#codegenFactory' in the IntelliJ project.
The commit fixes the following failing tests in the IntelliJ project:
- K2IdeK2CodeKotlinEvaluateExpressionTestGenerated$SingleBreakpoint$CompilingEvaluator$InaccessibleMembers.testPrivateAnnotationCompanionValue
- K2IdeK2CodeKotlinEvaluateExpressionTestGenerated$SingleBreakpoint$CompilingEvaluator$InaccessibleMembers.testSelfMembers
- K2IdeK2CodeKotlinEvaluateExpressionTestGenerated$SingleBreakpoint$CompilingEvaluator$InaccessibleMembers.testOuterMembers
- K2IdeK2CodeKotlinEvaluateExpressionTestGenerated$SingleBreakpoint$Uncategorized.testPrivateMember
- K2IdeK2CodeKotlinEvaluateExpressionTestGenerated$SingleBreakpoint$Uncategorized.testProperties
- K2IdeK2CodeKotlinEvaluateExpressionTestGenerated$SingleBreakpoint$Uncategorized.testPrivatePropertyWithExplicitDefaultGetter
- K2IdeK2CodeKotlinEvaluateExpressionTestGenerated$SingleBreakpoint$Uncategorized.testPrivateClass
Timofey Solonin