azalea/kotlin-fork
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
69,466 Commits 1 Branch 0 Tags
09f1764f8225c06789852f3ffe06f6941eda0f3f
Commit Graph

12104 Commits

Author SHA1 Message Date
Victor Petukhov 09f1764f82 Introduce warning for private inline functions which return anonymous objects without specified supertypes (KT-33917) 2020-09-29 10:23:45 +03:00
Victor Petukhov 63d825fa24 Introduce warning for secondary constructor in enums without delegation to primary constructors (KT-35870) 2020-09-29 10:23:45 +03:00
Mark Punzalan 238cc7c257 [FIR] Enable BytecodeText tests for FIR. 
143 out of 767 tests (18.6%) are currently failing.
 2020-09-29 10:21:21 +03:00
Mark Punzalan 69cd729506 [FIR] Enable BlackBoxAgainstJavaCodegen tests for FIR. 
23 out of 156 tests (14.7%) are currently failing.
 2020-09-29 10:21:21 +03:00
Mark Punzalan 348ba3e08c [FIR] Enable BlackBoxInlineCodegen tests for FIR. 
16 out of 638 tests (2.5%) are currently failing.
 2020-09-29 10:21:21 +03:00
Dmitriy Novozhilov cc4f72e032 [FIR] Capture type from type parameter upper bound when needed 2020-09-29 10:17:43 +03:00
Dmitriy Novozhilov fb871a55a3 [FIR] Don't create backing field for property without explicit type 
#KT-41977 Fixed
 2020-09-29 10:17:42 +03:00
Dmitriy Novozhilov 0c0a8f9849 [FIR] Properly pass type attributes for inference of lambda with type variable as expected type 
#KT-41989 Fixed
#KT-37317 Fixed
 2020-09-29 10:17:42 +03:00
Dmitry Petrov d9906ae8da Minor: unmute some bytecode listing tests in JVM_IR 2020-09-28 21:48:14 +03:00
Victor Petukhov 076eacb3d1 Check implicitly inferred Nothing inside special calls properly: compare constructors instead of types 
^KT-41176 Open
 2020-09-28 13:27:02 +03:00
Alexander Udalov ae6d89b100 Minor, mute new test for FIR 2020-09-25 16:50:58 +02:00
pyos bd6ead0467 JVM_IR: check for null when converting unboxed inline classes to strings 
`C?` can be unboxed into `T?` if if wraps a reference type `T!!`, but in
this case `null` is not a valid value to pass to `toString-impl`.

 #KT-42005 Fixed
 2020-09-25 16:50:34 +02:00
Alexander Udalov f8f2be3d9f Minor, improve test on suspend invoke function reference 
Add WITH_RUNTIME to enable it for the old backend, check the generated
function reference methods by starting the coroutine.
 2020-09-25 14:52:46 +02:00
Steven Schäfer 20a5c44e41 JVM IR: Fix types in generated function reference constructor 
For references to suspend functions this fixes an IllegalStateException
in the type mapper, which is what the "suspendFunctionMethodReference"
test is about.
 2020-09-25 14:52:19 +02:00
Jinseong Jeon f3db113a82 FIR2IR: use @Target to put annotations on either property or backing field 2020-09-25 14:45:49 +03:00
Jinseong Jeon dc2226b42f FIR2IR: put @JvmField on field, not property 
#KT-42118 fixed
 2020-09-25 14:45:48 +03:00
Mikhail Glukhikh fe8e3e99ab [FIR Java] Hide function in scope in case it's an accessor by fact 
#KT-42116 Fixed
 2020-09-25 14:44:09 +03:00
Dmitry Petrov 70095712eb Minor: drop new bytecode listing test that passes only in JVM_IR 
Unfortunately, there's no way to mute it in JVM right now.
Corresponding case is tracked as KT-42179, anyway.
 2020-09-25 13:49:53 +03:00
Dmitry Petrov cfd62c15bf JVM_IR KT-36994 don't generate stub if present in superclass 2020-09-25 11:57:33 +03:00
Dmitry Petrov 1adce11257 Add tests for KT-40191 2020-09-25 11:57:32 +03:00
pyos f30e564c90 JVM_IR: do not use fields' superQualifierSymbol to cast the receiver 
It's only for computing the field owner.

 #KT-42131 Fixed
 2020-09-24 18:28:23 +02:00
Dmitriy Novozhilov 4ef57c120f [FIR] Consider variance of type parameters during java type enhancement 
#KT-41940 Fixed
 2020-09-24 15:43:20 +03:00
Dmitriy Novozhilov 0d29d6d361 [FIR] Transform when conditions with boolean expected type 
#KT-42130 Fixed
 2020-09-24 14:59:35 +03:00
Jinseong Jeon 5b136516c4 FIR: utilize checking subtype of functional type and finding invoke symbol 2020-09-24 14:05:52 +03:00
Jinseong Jeon b10466f6a2 FIR: extend suspend conversion to subtype of functional type 2020-09-24 14:05:52 +03:00
Roman Artemev 6670e4b21d [JS IR] Fix callable reference to generic constructor 
- add tests
 - fix KT-42025
 2020-09-24 12:50:47 +03:00
Ilmir Usmanov 4da67bf013 Minor. Unmute tests 2020-09-24 10:08:43 +02:00
Ilmir Usmanov f22f10febb JVM_IR: Unbox inline classes in suspend functions 
if the function can be verified to safely returning unboxed
inline class.
Box the return value on resume path.
 2020-09-24 10:08:41 +02:00
Mads Ager ac742123fd [JVM_IR] Generate line numbers in synthetic bridges. 
The current backend does that for some bridges. We do it consistently
for all bridges.

The line number used differs. For the JVM_IR backend, we use the
line number of the class to which the bridge is added. For the
JVM backend, that does not appear to be the case for bridges
in lambdas. I prefer the line number for a lambda invoke bridge
to be the line for the lambda instead os some surrounding class.
 2020-09-24 08:59:48 +02:00
Steven Schäfer 3a7cc93c4e JVM IR: More precise check for calls to the implementation method in a default stub 2020-09-24 07:53:31 +02:00
Steven Schäfer 111c550f3c JVM IR: More tests for inlining in $default stubs 2020-09-24 07:53:31 +02:00
Steven Schäfer 5e27d9b089 JVM IR: Make inlining in $default stubs compatible with the JVM BE 
The JVM BE inlines calls to the underlying function in a $default stub
verbatim, e.g., without renaming LVT entries or regenerating anonymous
objects. This commit introduces the same behavior in the JVM IR BE.

Fixes KT-36769.
 2020-09-24 07:53:31 +02:00
Dmitry Petrov 1c4567c999 Add tests for KT-40152 2020-09-23 18:14:21 +03:00
Dmitry Petrov cd37301ea2 Minor: TARGET_BACKEND=JVM in JVM-specific tests 2020-09-23 18:14:21 +03:00
Dmitry Petrov 99dbeecc40 Some more bytecode listing tests for JVM_IR 2020-09-23 18:14:20 +03:00
Victor Petukhov 25ea3df5ca Put fix in 9123c4f73baf77f8a50dede6c890c46f5ffafd6c under the inference compatibility flag 2020-09-23 17:15:58 +03:00
Victor Petukhov bfb46befa5 Mark projection of a nullable captured type as not null during simplification constrains with it and a nullable type variable 
^KT-41913 Fixed
 2020-09-23 17:15:57 +03:00
Alexander Udalov 84df996204 JVM IR: fix "step over" for inline function calls in conditions 
Reimplement the same hacky approach used in the old backend (see
cc2fe6b0c6).

Previously, the debugger incorrectly stepped into Collections.kt on
"step over" inline function calls from stdlib like 'any'.

Since `if` and `when` expressions are represented the same way in IR,
the behavior is fixed for both of them. It's not the case in the old JVM
backend, where stepping over `when` conditions still suffers from the
same problem, which the newly added test checks.
 2020-09-23 15:54:15 +02:00
Dmitriy Novozhilov 10e7c0f1b2 [TEST] Fix long literal in testdata broken in 4374c06 2020-09-23 15:22:03 +03:00
Roman Artemev 76ed09482f [IR BE] Fix inline class lowering 
- provide additional type parameters from context when transforming
 constructor into static function.
 - fix KT-41771
 - add test
 2020-09-23 14:49:38 +03:00
pyos 5954db18cb JVM_IR: fix lifting of arguments to object super constructors 
If the arguments are reordered, they can not only be reads of temporary
variables, but also implicit type casts of said reads.

Fixes #KT-42002
 2020-09-23 13:49:16 +02:00
Dmitriy Novozhilov 896103b94b [FIR] Properly resolve implicit invoke calls 
#KT-41990 Fixed
 2020-09-23 14:05:11 +03:00
Dmitriy Novozhilov 40a9bb6eac [FIR] Assume nullable types as good types for ILT approximation 
#KT-41982 Fixed
 2020-09-23 14:05:10 +03:00
Dmitriy Novozhilov 4374c06537 [FIR] Approximate all integer literals which resolved in independent mode 
#KT-42016
 2020-09-23 14:05:10 +03:00
Dmitriy Novozhilov 5efd533f55 [FIR] Desugar unary plus and minus as part of integer literal 
#KT-42016
 2020-09-23 14:05:09 +03:00
Alexander Udalov d48307ec34 JVM IR: do not copy type parameters into suspend lambda classes 
Declarations inside that class are referencing the type parameters of
the containing function anyway, since we don't do any remapping. So the
resulting IR is slightly more correct, and doesn't lead to type
parameter/argument size mismatch error on IrBased/Wrapped-descriptors.

This reverts a part of 01da7f289b, which looks like it was no longer
necessary after 8d0ffa1444.

 #KT-42028 Fixed
 2020-09-23 12:21:37 +02:00
Mikhail Glukhikh 4653a16450 [FIR2IR] Add problematic test with in/out clash during approximation 2020-09-23 09:01:45 +03:00
Mikhail Glukhikh b64b32e06b FIR: keep captured types in substitution scope 2020-09-23 09:00:57 +03:00
Mikhail Glukhikh b7059a3eeb FIR: perform type approximation when completion isn't required 2020-09-23 09:00:57 +03:00
Mikhail Glukhikh 0310272738 FIR: perform more proper type approximation in completion 
Note: test data change is questionable here
 2020-09-23 09:00:57 +03:00