SyntheticAccessorLowering was initially implemented under the assumption
that any access to an invisible declaration will cause an accessor to be
generated _in the same file_. Moreover, it's declared in the group of
phases that are performed by file.
But this assumption is incorrect for constructors which need to be
hidden (those which take parameters of inline class types), since such
constructor is public and can be called from anywhere. In this case,
SyntheticAccessorLowering actually generated a new accessor for the
hidden constructor for each (!) source file where that constructor is
called, which led to ClassFormatError because of the class file having
multiple methods with the same signature. The internal `functionMap`
cache didn't help because it's not shared among phase instances for
different files (well, it helped to generate not more than one accessor
per usage-file).
In this change, we use the global cache, stored in JvmBackendContext,
for accessors to hidden constructors. Note that after this change, calls
to hidden constructors are always transformed to the corresponding
accessor in SyntheticAccessorLowering right away, but that accessor
might be orphaned for a while (not declared in any parent's
declarations). Only when SyntheticAccessorLowering encounters the
original constructor which needs to be hidden, it adds the accessor
beside it.
The test is sensitive to the file order, so both variants are added.
Currently FirThisReceiverExpression of instance methods are translated
to references of the class' thisReceiver,
not the method's dispatch receiver,
which causes problems with IrFrameMap::typeOf,
as the class' thisReceiver is not in the typeMap.
This commit translates non-qualified "this" references of
instance methods to references of the methods' dispatch receiver.
Whenever we want the default 0/null value for a type we need to
check if it is a non-nullable inline class type and produce
the right value for the underlying type.
For example, synthetic `$annotations` methods for properties were
previously mangled to `$annotations-...`, which breaks annotation
loader, and fails an assert in ClassCodegen.generateMethod.
This was missing for IrFunctionReferences, which caused problems when
copying default parameters before CallableReferenceLowering in the
JVM_IR backend.
1. Scheme of capturing local variables not touched
2. Lowered local functions are transposed to the nearest class (including local) or file
3. Local classes are also transpose to the nearest class (including local) or file
In SourceCompilerForInline we could enter ERASED_INLINE_CLASS context
from containing declaration context. That broke codegen context
hierarchy invariants assumed in accessor generation.
In JDK 9, Class.simpleName changed behavior for local/anonymous Kotlin
classes (see KT-23072), this is why we now check for both variants of
the name in tests. Also, the format of annotation arguments changed a
little, where float parameters no longer have the trailing "f", and
class literals are rendered with ".class" at the end
For the inline class:
```
inline class IC(val x: Int)
```
Type (IC..IC?) should be mapped to the wrapper `IC`
because it can hold object and also because it does so for primitives
#KT-28983 Fixed
Mikhail Zarechenskiy