The main change here is in `JvmInlineClassLowering.visitFunctionAccess`,
where we now store the substituted return type of the function as the
type of the call expression. Without it, the call could have a
meaningless type, e.g. some `T` which is inaccessible at that place, and
that could backfire in subsequent lowerings in codegen. For example, in
the `stringPlus.kt` test, it would prevent the code in
`FlattenStringConcatenationLowering.isStringPlusCall` from recognizing
and replacing the `String.plus` call, leading to a codegen exception.
Other changes are mostly cosmetics to make the code similar to
`visitFunctionReference`, and preventive optimizations for the case when
the substitution map is empty.
In this commit we change value parameter type of containsAll, removeAll,
retainAll from Java collections. Originally it's Collection<?>,
we change it to Collection<T>
#KT-42340 Fixed
This update should have been done in
4b7d34b537 commit, as it is stated in the
`KtTokens.MODIFIER_KEYWORDS_ARRAY` docs.
Shifting all modifiers by one have led resulted in incorrect modifiers
when stubs were read from the disk. Updating `SOURCE_STUB_VERSION` will
make IDEA recreate all stubs from scratch
^KT-42274 Fixed
The tests need to be able to parse and render CodeMetaInfo for different
platforms. For example, we must draw a run gutter a MAC application on
OSX, but for a Win application, the gutter should not be drawn at the
same time. This commit allows to do such checks. In order to indicate
on which platform it should be rendered, it is necessary to specify the
platform for a specific CodeMetaInfo in the test data in the format
{OSX,Unix,Win}, if the block with platforms is missing, this means
that it should be rendered on each platform.
<!LINE_MARKER{OSX}(...)!> - should only be on OSX
<!LINE_MARKER(...)!> - should be on any platform
The current test framework has a number of shortcomings, such as
different markups for different kinds, those formats can not be
mixed in one test file, lacks composability, re-use of it is complicated
This commit contains the first version of the new test runner
to which tests will be migrated in the future.
The inline class is boxed when we pass it as lambda argument, now we
unbox it. If the underlying type is not Any or Any?, bridge method does
the unboxing.
#KT-32450 Fixed
#KT-39923 Fixed
#KT-32228 Fixed
#KT-40282 Fixed
The disposable which decrements ourProjectCount was a non-capturing
lambda, meaning it had a single identity and could only be registered
in the Disposer tree once. Therefore the application environment
would never be disposed if ourProjectCount ever went above 1.
This helps to reduce peak memory in lowerings/codegen by a lot.
A more robust approach would be to have a separate BindingContext for
each file, and clear each of them after running psi2ir on it. This would
also lower peak memory usage in psi2ir.
Provide a fallback workaround compiler argument
-Xir-do-not-clear-binding-context just in case BindingContext is in fact
used somewhere and it's not caught by tests.
Compute FQ name of the main class right after running the analysis and
before invoking codegen. This is needed because MainFunctionDetector
depends on BindingContext, and JVM IR needs to clear BindingContext as
soon as it's not necessary to reduce peak memory usage, thus breaking
any usages of data from it after the codegen.
Also refactor and use the extracted, but not properly reused previously,
copy of findMainClass in findMainClass.kt.
Note that this replaces NPE in KT-42868 with an UOE.
This map is populated during psi2ir and is only used in
insertImplicitCasts, after which it makes sense to clear it, otherwise
it strongly references a lot of descriptors for calls in other files
which would never be used.
Move `postprocess` into `generateSingleFile` and do those steps on each
file instead. This will allow to create
`GeneratorContext.callToSubstitutedDescriptorMap` for each file and
clear it after its only call site (InsertImplicitCasts) has used it.
Also simplify code a bit.
Dmitry Petrov