Doing so speeds up psi2ir ~2 times, and thus improves total compiler
performance by about 6-8%.
Unless JVM IR is in the mode where linking via signatures is the only
way (-Xserialize-ir, -Xklib), signatures are actually not needed at all,
SymbolTable can use the frontend representation (descriptors for FE1.0,
and hopefully FIR elements for K2) as hash table keys. The only catch is
that since other backends still need to work with signatures, all the
common IR utilities, such as irTypePredicates.kt, need to work correctly
for IR elements both with signatures and without.
Also, introduce a fallback compiler flag -Xlink-via-signatures, in case
something goes wrong, to be able to troubleshoot and workaround any
issues.
#KT-48233
Previously (few commits earlier), it contained two versions
of receiver (lhs) generated separately for each desugaring version
that looked a bit redundant.
Now, at FIR building stage we just don't create desugaring sub-trees,
instead they are being built during bodies transformation and that seems
to be much convenient there, since we don't need to reverse-engineer
get-set-operator version to check if containing calls are successful
(as we just built those calls and retain them)
Semantically, this changes may only change how data flow works
for such statements (see changed compatibilityResolveWithVarargAndOperatorCall.kt)
^KT-50861 Relates
The issue this commit fixes occurs when we have an external interface
implemented by a Kotlin class, if that interface has methods with
varargs.
Kotlin functions expect varargs passed as arrays, but JavaScript code
may be unaware of this convention.
So, when generating a bridge for external interface method
implementaion, we insert some additional logic for extracting varargs
using the JavaScript `arguments` object.
A simplified example:
```kotlin
external interface Adder {
fun sum(vararg numbers: Int): Int
}
class AdderImpl: Adder {
override fun sum(vararg numbers: Int) = numbers.sum()
}
```
For `AdderImpl` we generate the following JS code:
```js
AdderImpl.prototype.sum_69wd7h_k$ = function (numbers) {
return sum(numbers);
};
AdderImpl.prototype.sum = function () {
var numbers = new Int32Array([].slice.call(arguments));
return this.sum_69wd7h_k$(numbers);
};
```
#KT-15223 Fixed
So, treat CHECKCASTs as { POP, Unit } sequences. If the CHECKCAST is
between suspension point and ARETURN, put check for COROUTINE_SUSPENDED
before it and return if the suspension point is suspended.
This is safe, since if the function throws CCE, it will be thrown from
the last state in state-machine and we cannot reenter the function
during resume. So, in case of CHECKCAST throwing CCE the behavior is the
same, whether we have state-machine or not.
We do not need to disable TCO in some cases, as we do for functions,
returning Unit, since in latter case suspend function, returning Unit
might appear as returning non-Unit during resumption due to missing
{ POP, Unit } sequence, which is not executed, since the function is
tail-call. However, in case of functions, returning non-Unit there is
no such concern, since we do not POP result of suspension point, but
rather, return it after CHECKCAST.
#KT-50835 Fixed
In this commit we upgrade FIR builder inference logic from
the compiler version to 1.7. FIR-based compiler now works with
"don't use builder inference" flag always ON and supports switching
the flag "use builder inference only if needed". To do it,
ContraintSystemCompleter (FIR) and KotlinConstraintSystemCompleter (FE 1.0)
are made similar with extracting some common parts into
ConstraintSystemCompletionContext.
Test status: one BB test fails after this commit (KT-49285).
Also we have a crush in DFA logic in FIR bootstrap test and somehow
questionable behavior in FIR diagnostic test. However,
two BB tests were fixed, the 3rd case from KT-49925 were also fixed.
#KT-49925 Fixed
Result of calling `ExpressionCodegen.gen` was used only in two call
sites in the inliner. In all other call sites, we were doing unnecessary
work (and sometimes were even failing) by trying to construct a
KotlinType instance out from an IrType.
Note that even though the code from KT-50617 no longer fails to compile,
the underlying problem is still not solved, since the IrType for foo's
dispatch receiver is constructed incorrectly. The reason is that
SymbolTable links everything by IdSignature, which is identical for
classes with the same FQ name.
#KT-50617 Fixed
Previously enhanced symbols were cached inside SignatureEnhancement,
which is created independently for each enhancement scope. This may
cause creation of multiple enhanced symbols for same java declaration
in presence of multiple scope sessions (mutithread compiler, IDE,
separate scope session for checkers)
^KT-50858 Fixed
This big refactoring is needed to cleanup building of overrides
mappings and prevent creating redundant intersection overrides in
cases when there is no need in them:
```kotlin
interface A {
fun foo()
}
interface B {
fun foo()
}
interface C : A, B {
override fun foo()
}
```
Before this refactoring there was next override tree:
C.foo
intersection override (A.foo, B.foo)
A.foo
B.foo
Also this commit fixes special mapping of overrides in jvm scopes
for declarations which have kotlin builtins in supertypes with
special java mapping rules (collections, for example)
It's safe as not having const is more restrictive, therefore can be
allowed in common. Otherwise, it's not possible to declare an expect
declaration for a platform property with `const` modifier in common
KT-18856
Alexander Udalov