These properties have a very specific detail in their behavior, in that
the constructor's name is set to be "<init>". While this is OK for the
IR serialization, this may not always be expected in other cases, and
their rather common names (`name`, `fqNameSafe`) suggested that these
properties could be used in generic contexts.
Change all usages outside IR serialization to use
`IrDeclarationWithName.name` and nullable
`IrDeclarationWithName.fqNameWhenAvailable` instead
Before this commit, java.lang.Class was read via JvmBackendContext,
its module descriptor and its package member scope.
However, this works only for old FE with descriptors and
descriptor-based scopes available.
The synthesized arguments caused the size of default value mask off by
one when it is close to the boundary of Int.SIZE, which in turn
resulted in wrong signature at call sites.
in OUTERCLASS field.
The inliner generates two versions of suspend functions/lambdas in
inline functions: with state-machine and without. The former is used
to call the function from Java or via reflection and have ordinary
name, while the latter is used by inliner and have $$forInline suffix.
The inliner throws the state-machine version away, duplicates
$$forInline version and then call state-machine generator.
If these suspend functions/lambdas are not going to be inlined,
$$forInline version is not generated. However, all objects, which are
used in these suspend functions/lambdas, have $$forInline version
written to OUTERCLASS field. This leads to errors by proguard.
Since they are used in both state-machine version and for-inline ones,
we can simply remove $$forInline suffix from OUTERCLASS field and this
fixes the issue.
#KT-31242 Fixed
This change is to fill the gap between Kotlin Collection
classes(immutable) and Java Collection classes(mutable), to avoid
calling an unsupported operation like remove() on an immutable class in
jvm.
IrConstructorCall gets type parameters from the class in addition to the
constructor declaration. This behavior is already implemented for
ClassConstructorDescriptorImpl, but was not implemented for
WrappedClassConstructorDescriptor, leading to missing type arguments for calls
to constructors generated in a lowering pass.
The comment in the code is correct that EnclosingMethod
attributes should only be generated for local and
anonymous classes. We were generating them for member
classes as well which leads to invalid class files.
With this change I had to mute one more tests. That is
because we lose the parent method and therefore we
see a class as a member class instead of a local class.
With the old descriptor based check that test still
passes.
This change also makes sure that no line numbers are generated
in the wrappers in the JVM_IR backend.
Change-Id: If6c37f8a20894455abddb526039df059e02015a3
Alexander Udalov