Add a test that checks that Kotlin Continuation wrapping Objective-C
completion handler has strong reference to it, i.e. completion handler
doesn't get reclaimed too early.
Follow-up to 4dcfd38.
Make AtomicReference behave like FreezableAtomicReference with the new MM: not frozen by default, don't require freezing value unless the ref itself is frozen. The behaviour with the old MM is unchanged.
Merge-request: KT-MR-5155
In JS, RegExp can return a match before the specified start index,
if it has matched at that index, but it is in the middle of a surrogate
pair. Account for that when advancing to the next position after
a zero-width match so that it doesn't get to the middle of SP.
The implementation was keeping delayed jobs in std::set sorted
only by the scheduled execution time (in microseconds since epoch).
So two jobs submitted to a worker and having the same scheduled time
were considered equivalent by the set, and one of them got lost.
Fix this by using std::multiset instead of std::set.
The previous implementation didn't log anything, including stdout,
stderr and the command line, and didn't detect non-zero exit code,
making it hard to investigate the test failures.
With `unitSuspendFunctionObjCExport=proper` binary option,
for Unit-returning suspend functions ObjCExport now generates
completion handlers without redundant `KotlinUnit*` result parameter.
So Swift (5.5+) can import these functions as Void-returning async
functions.
^KT-47399 Fixed
The variable generated by IrStatementBuilder.irTemporary doesn't inherit
startOffset and endOffset from the builder. In particular, as a result,
temporary variables generated for elvis operator left operand have
UNDEFINED_OFFSET.
Additionally, ProvisionalFunctionExpressionLowering copies the offsets
of a variable to lowered lambda in the variable initializer. With the
problem described above, this causes invalid debug information in
Kotlin/Native, see KT-49360.
Fix irTemporary by using builder's offsets for the variable.
^KT-49360 Fixed
^KT-48816 Fixed
Native compiler uses lazy IR for declarations provided by cinterop.
The problem: `FakeOverrideBuilder` requests super types during
type checking, accessing `.owner` for them. So if a type and super type
are represented as lazy IR, and lazy IR generation is not enabled yet,
then the super type symbol won't be bound by this moment, and the access
will fail.
This happens in KT-48816: fake override builder tries to access `.owner`
for `IrClassSymbol` of `NSObject` (super type of `NSDate` and `NSUUID`).
Fix this by enabling lazy IR generation before building fake overrides.
There are number of problems with up-to-date checks in tests, leading to
a test binary not being rebuilt or rerun when a developer expects it to.
Things became worse after updating to Gradle 7+:
test binary run tasks are considered up-to-date if the test data
has been changed, even after clean.
Even before Gradle 7+, we've observed other similar undesirable effects:
for example, if the compiler has been changed, tasks building
test binaries are still up-to-date.
There are quite a lot of different tasks related to running the tests
for Native, many of them are misused and thus don't do what we expect.
Carefully fixing each particular kind seems counterproductive and
unreliable.
So instead workaround all possible similar problems by forcing all tasks
in the :kotlin-native:backend.native:tests project to be not up-to-date.
This project should contain only test tasks, so this workaround
seems pretty natural -- it is just "tests aren't up-to-date".
Vyacheslav Gerasimov