It is difficult to implement properly tail calls for a real coroutine (when there
are other non tail calls), because of continuation interception semantics. And the
benefits aren't clear at this point, so let's turn it off for now.
Passing `EXTENSION_RECEIVER` when processing `noReceiver`
looks like a mistake in general. This change is backed
by the `hidesMembers` and
`memberWithHidesMemberAnnotationVsMemberWithout` tests.
The exact reason with `memberWithHidesMemberAnnotationVsMemberWithout`
is that it first checks `@HidesMembers` candidates,
only takes the `kotlin/collections/Iterable<T>.forEach`,
but then yields `InapplicableWrongReceiver`,
because `explicitReceiverKind = EXTENSION_RECEIVER`
(which is strange, because we really don't have an explicit receiver).
Then we visit the same scope once more (now for all candidates)
and take 2 functions:
- `kotlin/collections/Iterable<T>.forEach`
- `kotlin/sequence/Sequence<T>.forEach`
...and they both result in `RESOLVED`,
because this time `explicitReceiverKind = NO_EXPLICIT_RECEIVER`.
This change ensures the first candidate we see
while checking `@HidesMembers` is taken as `RESOLVED`.
^KT-55503 Fixed
Remove code generation for external fake overrides
Fake overrides are resolved to real declaration on call sites
This also removes generation of unused
"equals", "hashCode" and "toString" methods.
Avoid using hashes of IrType in generated code
because they are unstable
Merge-request: KT-MR-8658
Merged-by: Svyatoslav Kuzmich <svyatoslav.kuzmich@jetbrains.com>
`@JsFun("code")` is now executed inside a JS function which brings its
`arguments` into the scope shadowing D8's CLI `arguments`.
Accessing arguments through `globalThis` fixes the problem.
Merge-request: KT-MR-8630
Merged-by: Svyatoslav Kuzmich <svyatoslav.kuzmich@jetbrains.com>
These type parameters where used in function parameters,
but since suspendImpl is static function, it has no access
to class type parameters. Solution is to copy them to
the function itself.
#KT-55125 Fixed
Before:
- The cinterop task's output files were added as file dependency
to the declaring target's apiElementsConfiguration
- A 'copy' task was created for the cinterop artifact that was added
to the apiElements configuration as artifact (and therefore also
to the apiElements-published). This copy task placed the
same klib into the 'libs' folder
- Before passing libraries to the compiler the, all files
that are likely to be the result of the 'copy' task (in any libs dir) were
filtered
After:
- The cinterop task's output files are *not* added as file dependency
- The cinterop task's output is added as artifact directly to
the 'apiElements' configuration
- There is no more need for a copy task
- There is no more need for manually filtering the 'libs' artifacts
Tested:
project to project dependencies,
project to project dependencies (with KGP based dependency resolution)
project to repository
project to repository (with KGP based dependency resolution)
^KT-37051 Verification Pending
This commit fixes:
```
./gradlew :kotlin-native:backend.native:compileCompilerKotlin -PdeployVersion=1.9.0-dev-900
```
`-PdeployVersion=1.9.0-dev-900` is important for reproducibility (it's
not necessarily `1.9.0-dev-900` any version starting with `1.9.0-dev`
will reproduce the problem)
FunctionalTypeKind can be used in FE 1.0 too, so there is no need to
keep both classes. Also, removal of FunctionClassKind simplifies work
with FunctionalTypeKind in common code, like Analysis Api
This is needed because now set of available synthetic functional interfaces
depends on session, because each module can contain different plugins,
which provide different kinds of functional types
This is needed to provide an ability to extend different kinds of
functional types
Also, cleanup and rename utilities related to functional types to avoid
possible confusions
Being disabled by default
and not well-documented, these functions cause confusion among early
adopters as to why their code don't work properly.
Assert APIs need a proper design across Kotlin platforms.
Since APIs are not available in common code and K/JS, it is premature
to have such a general feature in a new experimental platform.
Compiler tests:
* Mute tests that rely on assert.
* Replace JVM-specific assert calls with require calls and unmute passed K/JS tests.
Merge-request: KT-MR-8636
Merged-by: Svyatoslav Kuzmich <svyatoslav.kuzmich@jetbrains.com>
- `FirTypeResolverImpl.resolveSymbol` cannot simply assert that a type
parameter only has a single qualifier, because code may be fed to the
compiler where a type parameter is the start of a type chain (see for
example the added `typeParameterChainInReturnType` test).
- The fix assumes that any multi-qualifier type parameter trivially
resolves to `null`, because such a chained type cannot exist.
^KT-56212 fixed
^KTIJ-24083 fixed
Alexander.Likhachev