8e07482862
JVM IR: Fix generation of parameterless default constructor
...
The JVM backend does not generate parameterless default constructors for
private constructors and constructors of local, inner, or inline
classes.
2020-01-21 16:34:33 +01:00
5321a6af33
Forbid spread operator in signature-polymorphic calls
...
#KT-35226 Fixed
2020-01-21 15:57:35 +01:00
99294cfdae
Minor, fix test data
...
TypeCastException was unresolved after an accidental change in
f48bdc1fcb
2020-01-21 11:26:21 +01:00
79840a05b2
Fix test expectations.
...
Do not check for occurrences of "this" on the current backend.
I accidentally unified the checking for the two backends
without checking that it worked (used the wrong test suite to
test).
2020-01-21 11:26:02 +01:00
1c527fc159
IR: avoid name clashes between raised local functions.
2020-01-21 12:50:55 +03:00
c0600e6873
JVM IR: minor, mute and adapt noCallAssertions.kt to 1.4
2020-01-20 19:12:59 +01:00
078b934580
Mute/unmute JVM backend tests for language version 1.4
2020-01-20 19:12:59 +01:00
f48bdc1fcb
Fix codegen box tests for language version 1.4
...
Since API version 1.4, NullPointerException is thrown for casts of null
to any type instead of TypeCastException.
2020-01-20 19:12:59 +01:00
6a90dc2efe
Fix bytecode text tests for language version 1.4
...
New null check assertions are generated a bit more efficiently, with a
call to `checkNotNull` instead of IFNONNULL+jump.
2020-01-20 19:12:59 +01:00
1ef5e25c60
JVM_IR: Minor. Update test data
2020-01-20 18:59:00 +01:00
6ede10c1ca
JVM_IR: Minor. Add test for SMAP of inner object/lambda of inline suspend function
2020-01-20 16:00:36 +01:00
877509306d
JVM_IR: Minor. Unmute tests.
2020-01-20 16:00:36 +01:00
891a55d79a
JVM_IR: Fix visibility of inner suspend lambda inside inline function
...
It should be public, otherwise, after inlining, we will get IllegalAccessError.
2020-01-20 16:00:36 +01:00
56c8fdc6c4
JVM_IR: Perform tail-call optimization on IR
...
It is easier to introduce a new lowering so the codegen will emit code for the old
tail-call optimizer to understand. Also, this is more flexible and would allow to
optimize cases, which are now feasible with the old optimizer.
Note, that because of bytecode inlining, we cannot replace the old one, but we cannot
emit code, that is simpler for it to optimize.
2020-01-20 16:00:35 +01:00
9f5b51ed43
JVM_IR: Do not generate fake continuation markers inside inline suspend lambdas
...
Since the markers replace ALOAD 0 as continuations, passed to suspend calls, in
JVM_IR we do not need this, since in JVM_IR all inline lambdas are static
functions.
2020-01-20 16:00:35 +01:00
4d9d62ad12
JVM_IR: Also check attributes for suspendImpls when generating continuation
...
Otherwise, it would not generate continuations for overloads.
2020-01-20 16:00:34 +01:00
d17afddaa9
JVM_IR: Fix bridges generation for suspend functions
...
Use view's signature as jvm signature.
2020-01-20 16:00:34 +01:00
2507e2b526
JVM_IR: Fix parents of $$forInline companions
2020-01-20 16:00:33 +01:00
7167d5f75c
JVM_IR: Generate $$forInline companion for transformed crossroutines
2020-01-20 16:00:33 +01:00
64c1446fbe
JVM_IR: Generate $$forInline companion for suspend inline functions
...
Set isInline flag for $$forInline functions. Otherwise, SMAP will not be generated
for inner objects/lambdas, leading to error in inliner.
2020-01-20 16:00:32 +01:00
4ef2ecf9a9
JVM_IR: Add returns unit marker if suspend call returns unit
2020-01-20 16:00:32 +01:00
5c5635ce20
Fix codegen & bytecode tests after unifying exceptions in JVM backend
...
See KT-22275 for details
2020-01-20 16:36:03 +03:00
1ed7e33f42
JVM_IR: Fix default argument bit mask for methods made static.
...
When called by reflection the bit mask will be generated
discounting dispatch/extension receivers. Make sure that the
interpretation of the bit mask is consistent for direct and
reflective calls.
In addition, this also fixes the modifiers on java 8 parameter
metadata for the dispatch and extension receivers for these
inline class methods.
2020-01-20 12:59:25 +01:00
33a24bfd27
JVM_IR: Avoid using parent name in name of captures.
...
If parent is an anonymous object that could lead to the use
of names such as `$this$no name provided` which is invalid
in dex files.
2020-01-20 10:49:46 +01:00
cb0b44273d
[NI] Check stub types in result type
...
An uninferred parameter stub may leak through calculation of CST(Inv<Uninferred Stub>, Nothing) into a result type.
A stub type in the result type means a type error. So we can afford recalculating
CST with stub-containing types filtered out, since its an error anyway.
This prevents stub types leakages and helps with reporting type error diagnostics.
KT-35914 Fixed
KT-35943 Fixed
2020-01-20 11:11:31 +03:00
58c235e722
Set up tests for fun interfaces for JVM backend for now
2020-01-17 19:37:48 +03:00
8350cf7d14
FIC: rename FunctionConversion -> FunctionalConversion
2020-01-17 19:37:47 +03:00
70094884ca
FIC: support suspend conversions in jvm codegen
2020-01-17 19:36:11 +03:00
3849b5e723
FIC: add codegen tests, adapt previously tests for SAMs
2020-01-17 19:36:09 +03:00
b7c43fc7db
Refactoring: move test folder out of "inline classes" folder
...
It was added there by mistake
2020-01-17 19:36:08 +03:00
0ee977c42e
FIC: Support base version of conversions in JVM backend
...
Proper support for JVM backend will be in the further commmits
2020-01-17 19:36:02 +03:00
55935cc98a
FIC: introduce language feature, modifiers checks and basic tests
2020-01-17 19:36:00 +03:00
437a26684d
NI: Prefer nullable lower bound to flexible one when substitution of type variable is performed and remember flexibility of type parameters based on flexibility of its upper bounds
...
^KT-32435 Fixed
2020-01-17 19:16:09 +03:00
a49ed1eca2
JVM IR: Generate interface delegations for specialized interface defaults
2020-01-17 18:19:26 +03:00
8746d08dd5
JVM IR: Don't skip bridge generation for concrete fake overrides
2020-01-17 18:19:26 +03:00
933ac3a16b
Fix nullable number comparisons in JS
2020-01-17 14:14:59 +03:00
f8341ad7eb
JVM: Update tests for IEEE 754 comparisons
2020-01-17 14:14:59 +03:00
d5ff1047a5
JVM: Fix IEEE 754 comparison generation for boxed numbers
2020-01-17 14:14:58 +03:00
26f435eb90
JVM IR: mute tests which fail after advancing language to 1.4
2020-01-17 12:07:30 +01:00
e7864c7351
Fix inlineCallsStaticMethodFromOtherPackage.kt by disabling related language feature
...
This test is not possible for 1.4, where protected calls from public
inline functions are prohibited.
2020-01-17 12:07:30 +01:00
76b3964e96
Update testdata according to change compiler version to 1.4
2020-01-17 10:33:50 +03:00
6e6f4d0503
IR: close over the set of captured type parameters
2020-01-16 16:37:40 +03:00
9566f329f0
Mute FIR test failures and adjust failing test data due to KT-35948
2020-01-16 15:36:19 +03:00
0f6df5971b
Mute tests failing in JS_IR backend due to KT-35944
2020-01-16 15:36:16 +03:00
ac790bea71
Reorganize diagnostics tests with stdlib and reflect
...
DiagnosticsTestWithStdLib now require stdlib only.
Move test data accordingly.
The only test that required kotlin-reflect is transformed into
codegen box test.
2020-01-16 15:35:58 +03:00
621936e951
Do not generate Throws attribute for delegated members from Kotlin interfaces
...
#KT-31763 Fixed
#KT-35834
2020-01-16 12:43:09 +01:00
5111d1721a
[TEST] Add js runtime to failing box test
2020-01-15 22:12:29 +03:00
04ce10b6c1
[NI] Improve completing callable references with type variable as expected type
...
#KT-32462 Fixed
2020-01-15 22:12:28 +03:00
6fe214d825
JVM IR: fix handling of suspend extension lambdas with captured receiver
...
The problem in the added test was that a suspend lambda was represented
by a function reference with a bound argument for the ObjectRef value,
and the corresponding parameter was not the first parameter of the
referenced local function. This happens because
LocalDeclarationsLowering lifts the local function up and adds a
new parameter for the captured ObjectRef (which is bound at the call
site), but the original receiver parameter remains the first unbound
parameter. So, it's no longer correct to rely on the fact that all bound
parameters of a function reference are located in the beginning of the
parameter list, which was kind of assumed in the `withIndex` call in
`AddContinuationLowering.addCreate`.
2020-01-15 19:42:13 +01:00
6bbfb2ffe9
JVM: add tests for companion object accessor generation
2020-01-15 16:40:06 +03:00
Steven Schäfer