Sometimes, state-machine, generated in inline functions with
crossinline parameter, is transformed, since all usages should be
renamed.
However, this is wrong: in this case, we will have state-machine
inside state-machine.
This fix addresses the issue.
#KT-25893 Fixed
Before this commit, function with same name but different signature
wasn't counted as name conflict, now it is - just because that exact
check in this place could be too complex and error-prone
Enhancement for KT-27539
Since at least in gradle-imported projects, the template could be found
in several modules, some of them - incomplete (parent/buildscript module)
the previous logic of filtering for already processed templates and roots
is incorrect, and therefore removed.
Also fixes the exception in our project "cannot load script definition"
Fix 'java.lang.AssertionError: PSI/document/model changes are not allowed during highlighting'
For script when there are some new dependencies we are restarting highlighting (see ScriptDependenciesCache)
In non-gradle project kotlin-stdlib-sources.jar contains sources for common and jvm part, so both (expect and actual) declarations are present in the same library sources scope.
We are moving expect declarations in the bottom of this list to avoid navigation on expect declaration when actual declaration is present in the same scope
^KT-28398 Fixed
In the desugaring for compound assignment to a collection element,
argument expression 'i' is mapped to value parameters 'iG' and 'iS' of
corresponding 'get' and 'set' operators.
In general, these value parameters can have different indices.
This requires extra machinery in argument generation - that is, to be
able to generate a particular expression argument using an arbitrary
callback. In the vast majority of the cases this callback will just use
the corresponding StatementGenerator to generate IR subtree for the
provided expression. In case of 'get' and 'set' operator calls for an
augmented assignment expression this will map corresponding argument
expressions to pregenerated temporary variables.
Thus, in the following context:
```
class A
operator fun A.get(vararg xs: Int) = 0
operator fun A.set(i: Int, j: Int, v: Int) {}
```
statement `a[1, 2] += 3` will be desugared as (in a really pseudo
Kotlin):
```
{
val tmp_array = a
val tmp_index0 = 1
val tmp_index1 = 2
tmp_array.set(
i = tmp_index0,
j = tmp_index1,
v = tmp_array.get(xs = [tmp_index0, tmp_index1]).plus(3)
)
}
```
Update jcabi-aether dependency from 0.10.1 to 1.0-SNAPSHOT. In order to
do that, add https://oss.sonatype.org/content/repositories/snapshots to
the repositories in tools/kotlin-script-util/build.gradle.kts
Script dependencies are resolved using jcabi-aether. Unfortunately "stable"
release (0.10.1) does not support proxies. Therefore, script dependencies
are not pulled behind a proxy. Issue is fixed in SNAPSHOT release of
jcabi-aether.
FIXED: KT-22363
Report about unsupported targets once per target type, show presentable
chunks list and don't show more then 5 chunks.
Example: "Native is not yet supported in IDEA internal build system.
Please use Gradle to build a, b, c, d, e and 10 other (enable 'Delegate
IDE build/run actions to Gradle' in Settings)."
#KT-26980 Fixed
#KT-28316 Fixed
1. Checking for COMPILE_DAEMON_CUSTOM_RUN_FILES_PATH_FOR_TESTS was moved to COMPILE_DAEMON_DEFAULT_RUN_DIR_PATH.
Looks like COMPILE_DAEMON_CUSTOM_RUN_FILES_PATH_FOR_TESTS was introduced incorrectly in 220fab0d3f.
Checking of this property was added in DaemonOptions.runFilesPathOrDefault, while DaemonOptions.runFilesPath was internally used in runFilesPathOrDefault and in many other places.
For example DaemonOptions.runFilesPath used to pass this option to daemon server.
So daemon was started with runFilesPath that ignores COMPILE_DAEMON_CUSTOM_RUN_FILES_PATH_FOR_TESTS.
2. JpsKotlinCompilerRunner._jpsCompileServiceSession was leaked between tests.
Fixed by extracting @TestOnly releaseCompileServiceSession() and calling it in tests tearDown()
3. The result of compileWithDaemon was ignored in compileWithDaemonOrFallback.
So, the fallback was never called, and the FAIL_ON_FALLBACK_PROPERTY was actually was never worked.
This was fixed. Also the message was improved to make it easier to find the original fail cause.
Svyatoslav Kuzmich