Coersion to Unit from error type leads to misleading type mismatches:
"expected <expected lambda return type> found Unit", despite no user-provided Unit / empty lambda.
These diagnostics were collected, but not reported before, and that had been disguising the issue for a while.
KT-34729 Fixed
Fix completion of anonymous functions with expression body without expected type.
Premature completion led to losing type info from outer calls.
Also report type mismatches on empty lambda expressions.
KT-34729 In progress
The problems in JVM IR backend in these modules are fixed by:
* kapt3-cli -- 1c527fc1
* frontend -- 174b3db7
* daemon-common-new -- 6fe214d8 & d17afdda
This option has no effect on the production code. But in a custom
scenario of a local bootstrap, where the compiler is replaced with the
one where JVM IR backend is enabled by default, this option allows to
switch back to the old backend.
All of these arguments (except the one for compiling experimental
coroutines) are supposed to go away as soon as we fix related problems
in the JVM IR backend.
Expression type serves as a part of a key for retrieving smart cast info.
Rewriting expression type with SmartCastManager leads to incorrect processing of following smartcasts for same expression, therefore it was removed.
KT-32507 Fixed
KT-32358 Fixed
It's necessary because of new type substitutor is eager than odl, so
if there is a substitution of type parameter deep inside type arguments
then second substitutor wins against first
#KT-35896 Fixed
Since there's no nullability assertion, null could end up in this list
and cause NPE further at `reportNotAcceptedExperimentalities`. This
could happen if a declaration was annotated with WasExperimental, the
marker was annotated with the new RequiresOptIn, and API version 1.2 was
used.
No new tests are added because 1.2 is obsolete and WasExperimental is
internal.
See the test and the issue for the clarification
After the change 7898922066
the expected type for "a" in expression "foo(a)" is A<E & B<*>>
But we have the original type A<E> and smart-casted enriched type A<B<*>>
(non of them is a subtype of A<E & B<*>>)
and fail in checkTypeInternal when checking types in during completion
^KT-35844 Fixed
Treating special functions for `if`, `when`, `try`, `?:` as not accepting `Nothing` result type is incorrect.
Making so leads to cases with uninferred `Nothing` result type for inner calls and lost data flow info.
Ilya Gorbunov