Commit Graph

348 Commits

Author SHA1 Message Date
Alexander Udalov ddfa94e7e9 Support Nothing type in typeOf
The proper support will come in KT-15518, but that would be a breaking
change even for stable Kotlin without kotlin-reflect. Before that issue
is fixed, represent Nothing in types with the Void class, and use a flag
in the no-reflect implementation to remember that it's not actually the
Void class itself.

 #KT-39166 Fixed
2021-07-09 14:40:05 +02:00
Victor Petukhov c3a5a7754d Use proper applicability for constraint warnings
^KT-47316 Fixed
2021-07-09 14:07:01 +03:00
Ilya Muradyan 787ce6335c Change nestedClasses logic to accept classes with nested type aliases
#KT-47650 fixed
2021-07-08 21:04:53 +03:00
Andrey Zinovyev 9fad55d551 [FIR] Use @DeprecatedSinceKotlin in the resolution 2021-07-07 16:19:27 +03:00
Alexander Udalov 113632c49c Psi2ir: load single-value Java array annotation arguments correctly
#KT-47467 Fixed
2021-07-05 21:49:10 +02:00
pyos c3a91efea3 JVM_IR: fix primitive comparison optimizations
1. the `primitive == object?.something` fusion should not apply to
    `primitive.equals(object?.something)` because it can't;

 2. coercions to Int are there for a reason - don't remove them;

 3. better optimize `primitive == object?.something` -- the result
    should be subject to if-null fusion, so it needs to have a specific
    pattern that resembles safe calls.

 #KT-47597 Fixed
2021-07-05 18:13:09 +03:00
Dmitriy Novozhilov ae608ea67f [FE 1.0] Always create return value for CallInstruction
This fixes missing `USED_AS_EXPRESSION` recordings
^KT-47527 Fixed
2021-07-02 17:55:20 +03:00
Dmitry Petrov 584310615c JVM_IR KT-47499 use proper signature for 'clone' intrinsic 2021-07-01 22:10:47 +03:00
Alexander Udalov ae07127f08 JVM IR: support raw types in typeOf 2021-07-01 19:33:56 +02:00
Alexander Udalov 0cb905a4b1 Support mutable collection types in typeOf
flexibleTypes_1_6.kt is fixed for JVM IR in a subsequent commit.

 #KT-35877 Fixed
2021-07-01 19:33:55 +02:00
Alexander Udalov 6e975b3498 Support flexible types internally in typeOf
#KT-45066 Fixed
2021-07-01 19:33:55 +02:00
Roman Artemev 5a284de2d4 add tests for private signatures
regenerate tests

add test for EnumEntry's argument declarations

add another test

regenerate test

fix test a bit
2021-07-01 17:40:27 +03:00
Dmitry Petrov 37b05cd9c2 JVM_IR KT-47120 treat outer name as case-insensitive for local classes 2021-06-30 14:04:22 +03:00
Dmitry Petrov 68c3f30aa7 JVM_IR KT-47510 indy callee parent is a package fragment, not a class 2021-06-29 17:00:31 +03:00
Dmitry Petrov 768afc5ba4 JVM_IR KT-47492 fix 'for' loop generation 2021-06-28 21:07:54 +03:00
Mikhael Bogdanov 5f62b72c82 Properly process big types 2021-06-26 06:10:16 +02:00
Dmitry Petrov 1298ba431b JVM_IR KT-47449 handle star projection arguments in default lambda types 2021-06-25 20:42:22 +03:00
pyos 26e3237b8c JVM_IR: never rename public/protected fields
And if that causes a platform declaration clash, that's not a problem
that can be solved without breaking the ABI anyway.

 #KT-47412 Fixed
2021-06-24 17:41:06 +03:00
Denis.Zharkov 4892ad42b9 FIR: Do not mark not found classes' based types from Java as error
The same happens in FE1.0
2021-06-24 12:17:45 +03:00
Denis.Zharkov 201dded237 FIR: Fix serialization of delegated members
^KT-47413 Relates
2021-06-24 12:17:43 +03:00
Dmitry Petrov 0104b1275f JVM_IR KT-47398 handle @EnhancedNullability String subject as in 1.0 2021-06-23 20:11:24 +03:00
Dmitry Petrov c26d71c4ef JVM KT-47365 add box test 2021-06-23 20:11:23 +03:00
Dmitry Petrov c77884f067 Refactor SAM type handling, replace non-approximated arguments with * 2021-06-22 21:13:56 +03:00
Victor Petukhov 4aeabb6b0f Use upper bound aware type approximator for intersection types inside sam types in contravariant positions to build proper types in terms of subtyping 2021-06-22 21:13:55 +03:00
Dmitry Petrov c19792e7c5 JVM_IR KT-47326 downcast field receiver on JvmField lowering 2021-06-21 19:17:54 +03:00
pyos cf660cf24a JVM_IR: expect unboxed return value from suspend default stubs
#KT-47206 Fixed
2021-06-21 18:53:31 +03:00
Dmitry Petrov ce107d06d4 JVM_IR add test for KT-47300 2021-06-17 21:40:30 +03:00
pyos 6c1a5e1211 JVM: make inline function argument processing a bit shorter 2021-06-16 12:13:32 +02:00
Mads Ager 6b6ba77869 [JVM_IR] Trim the expression test string for null assertions.
^KT-47166 Fixed.
2021-06-11 21:54:04 +03:00
Dmitry Petrov bf077b893c PSI2IR KT-47245 handle declaration statements in control structures 2021-06-11 16:42:31 +03:00
pyos 1f9db7cf25 FE: properly clear the deferred generator stack
Before this commit the stack wasn't cleared for read variable case.
#KT-47168 Fixed
2021-06-11 15:56:20 +03:00
Andrey Zinovyev ddc9c1bd1b [FIR] Fix extension property reference resolution
On type with star projection
#KT-44372 Fixed
2021-06-09 15:15:28 +03:00
Dmitry Petrov 81b09ca09f KT-46267 JVM_IR don't generate unnecessary ATHROW in lateinit var read 2021-06-07 20:02:24 +03:00
pyos 0bd8d16fe2 JVM_IR: fix Result types in continuations & remove more codegen hacks
#KT-47129 Fixed
2021-06-07 09:57:42 +02:00
Dmitry Petrov 7fd033adae JVM_IR KT-47073 use type parameter upper bound for default value 2021-06-04 20:54:01 +03:00
Victor Petukhov 1e5998e0ba Add constraints between two different stub variables instead of creating constraint error due to T1 isn't subtype of T2
^KT-44241 Fixed
2021-06-03 19:18:09 +03:00
Victor Petukhov 19c07e048a Don't check suitability of a builder inference call if unrestricted builder inference is enabled
^KT-42139 Fixed
2021-06-03 17:28:40 +03:00
Victor Petukhov 124a14c8df Use separate constraint position during call substitution as part of inferring postponed type variables
^KT-47052 Fixed
^KT-47082 Fixed
2021-06-03 12:37:17 +03:00
Ilmir Usmanov 7a2ecc58d4 Minor. Regenerate tests 2021-06-03 12:35:26 +03:00
pyos d5d3d9f112 Add regression tests for KT-46890 and other stuff 2021-06-03 12:35:25 +03:00
Dmitry Petrov 7a43c2de79 JVM remove dead code during constant condition elimination
This avoids an extra call to 'analyze', which is rather costly.

Update debugger testData: Constant condition
elimination now performs DCE more consistently.
2021-06-03 00:08:27 +03:00
Mads Ager 0b5ec5535c [JVM_IR] Add String.plus(Any) to intrinsics list.
This filters it out for use in SAM wrapper generation with
invoke-dynamic.

The intrinsic is not actually used for code generation so
it is added to the list of intrinsics that should have been
lowered away before codegen.

^KT-45779 Fixed
2021-06-03 00:04:47 +03:00
Mads Ager f846dd8ea2 [JVM_IR] Avoid using special names in static inline class replacements.
Fixes KT-46952
2021-06-01 16:22:14 +02:00
Dmitry Petrov 9091ca7b51 JVM_IR KT-46864 handle "unused" LDC instructions properly 2021-05-31 21:12:35 +03:00
Mads Ager 886ce055f5 [JVM_IR] Do not use invoke-dynamic for targeting inline-only methods.
Fixes https://youtrack.jetbrains.com/issue/KT-46962
2021-05-31 18:11:50 +03:00
pyos 5c2753b5d1 JVM_IR: use substituted return type in function reference invoke
#KT-46982 Fixed
2021-05-31 16:01:22 +03:00
Alexander Udalov 96f0b53761 IR: get overridden directly from IrProperty when resolving fake override
The previous way of getting them either via getter or setter failed on
Java properties which only have a backing field. Now that IrProperty has
overriddenSymbols (after 53c1de172f), it makes sense to use it directly
instead. Use it only in SyntheticAccessorLowering though to avoid
breaking Kotlin/Native (see KT-47019).

 #KT-46900 Fixed
2021-05-28 17:24:20 +02:00
Victor Petukhov bd7fb56a24 Fix builder inference tests 2021-05-28 15:36:23 +03:00
Victor Petukhov fd5169186e Update test for the builder inference and add new ones 2021-05-28 15:36:21 +03:00
pyos 34878d17eb JVM: be more careful when removing unused constants
1. if an argument of a `pop` cannot be removed, then all other potential
   arguments of that `pop` can't be removed either, and the same applies
   to other `pop`s that touch them;
2. the same is true for primitive conversions, but this is even trickier
   to implement correctly, so I simply did the same thing as with
   boxing operators: replace the conversion itself with a `pop` and keep
   the argument as-is.

Somehow this actually removes *more* redundant primitive type conversions
than the old code in a couple bytecode text tests, so I've patched them
to kind of use the value, forcing the instructions to stay.

 #KT-46921 Fixed
2021-05-27 12:24:22 +02:00