bbbdfd25d9
KT-4820: Added new tests, introduced functionality to detect overridden functions, other minor fixes
2014-05-09 05:48:50 -04:00
35c5a62b11
KT-4820: Intention to remove an unused index variable from a for loop.
2014-05-09 05:48:50 -04:00
97010c7d7f
KT-4820: Intention to introduce an index variable to a for loop over an ordered collection via "withIndices()."
2014-05-09 05:45:30 -04:00
399c9c1175
JS backend: fixed compiler crash when use compareTo in infix call.
...
#EA-54661 fixed
2014-05-08 19:57:28 +04:00
a578fc6762
JS backend: fixed infix calls.
...
#KT-3998 fixed
#EA-56241 fixed
2014-05-08 19:57:28 +04:00
898275e658
JS backend: added tests for infix calls.
...
#KT-3998 in progress
#EA-56241 in progress
2014-05-08 19:57:28 +04:00
235b03c407
Minor in JS backend: added useful message to assert(for EA-48931) and drop unnecessary code.
2014-05-08 19:57:27 +04:00
814b558bcf
Inspection to report negated binary expressions that can be simplified
2014-05-08 19:02:58 +04:00
22e21b4db0
Unextend InlineCodegen from ParentCodegenAware
2014-05-08 18:42:35 +04:00
13728e1598
Minor, pull non-null getParentContext() to MethodContext
2014-05-08 17:48:40 +04:00
5fe7bfdd83
Move 'earlierScripts' from ScriptCodegen to ScriptContext
...
To get rid of a questionable usage of getParentCodegen() in ExpressionCodegen
2014-05-08 17:48:40 +04:00
6bd46d38f5
Add missing copyrights
2014-05-08 17:38:51 +04:00
271f2f562c
Delete unused parameter in SamWrapperCodegen
2014-05-08 17:21:11 +04:00
8041cac9dd
Remove GenerationStateAware
2014-05-08 16:45:16 +04:00
985e9a5f31
Make JvmFunctionImplTypes lazy
...
It shouldn't compute all FunctionImpls at least until first asked
2014-05-08 16:45:16 +04:00
3a0aac4857
Simplify type mapping logic in CodegenBinding
...
- inline asmType to calling getAsmType, which does something more
- refactor getJvmInternalName to use getAsmType as well
- simplify getAsmType and fix a probable bug in mapping singletons nested in
enums (which wasn't reproduced, though a test is added)
- delete unnnecessary ASM_TYPE recording for enum entries in
CodegenAnnotatingVisitor
2014-05-08 16:45:15 +04:00
372bf17982
Remove BindingTraceAware, TypeMapperAware, BindingContextAware
...
- delete suspicious caching logic from CodegenBinding.getAsmType(), which makes
JetTypeMapper now depend only on BindingContext
- inline all of these classes into GenerationStateAware and the latter into
JetTypeMapper
2014-05-08 16:45:15 +04:00
fb54ecf443
Rename FunctionTypesUtil -> JvmFunctionImplTypes
...
It's not an util class anymore
2014-05-08 16:45:15 +04:00
c7277250d9
Use ReflectionTypes instead of KotlinBuiltIns in type checker
2014-05-08 16:45:15 +04:00
6004c6cefe
Introduce ReflectionTypes, use it in JVM back-end
...
Make FunctionTypesUtil a component
2014-05-08 16:45:14 +04:00
3ac89030c7
Make CodegenBinding.ASM_TYPE a collective slice
...
Change Type->JetFile mapping logic in PsiCodegenPredictor to be less hackish,
but probably the compiler will start requiring a little more memory due to a
collective slice multimap
2014-05-08 16:45:14 +04:00
d737932bc7
Remove usages of CodegenBinding.initTrace
...
This is a low level JVM back-end detail; clients should be using
GenerationState instead
2014-05-08 16:45:14 +04:00
d42598a935
Move default inline flag to JVM back-end
...
- other back-ends are likely to have its own value of this flag
- get rid of DEFAULT_INLINE_FLAG_FOR_TEST
- the main goal was to simplify the default GenerationState constructor, which
now always uses default inline flag value
2014-05-08 16:45:14 +04:00
9679e3ce4a
insert type intention now only applicable in function names (KT-4950)
...
#KT-4950 Fixed
2014-05-08 14:11:21 +04:00
442f47bbe5
Removed unused constructor
2014-05-07 22:50:23 +04:00
dab6045f6a
Fixed Code completion does not work inside function variable invocation
...
#KT-4978 Fixed
2014-05-07 22:50:23 +04:00
7bb84c139d
Smart completion: fixed logic in elvis operator
2014-05-07 22:50:23 +04:00
ab563bd891
Added utility extension methods to not use TypeUtils
2014-05-07 22:50:22 +04:00
a65bd1a74d
Added test data for future use
2014-05-07 22:50:22 +04:00
1b034c5fc3
Smart completion works for block's result
2014-05-07 22:50:22 +04:00
45f518a907
Smart completion after ?: works incorrectly
...
#KT-4926 Fixed
2014-05-07 22:50:22 +04:00
5e34791e68
Smart completion should not suggest variants for private signatures
...
#KT-4899 Fixed
2014-05-07 22:50:22 +04:00
57c41ce0b0
Smart completion: added test for KT-4929 fixed by svtk
2014-05-07 22:50:22 +04:00
0920ab2c25
Fixed mock jdk
2014-05-07 22:50:21 +04:00
e7c87b2391
Minor rename
2014-05-07 22:50:21 +04:00
d781626201
Smart completion: only overloads with all previous arguments matched are considered
2014-05-07 22:50:21 +04:00
717977af61
No more java lookup elements in code completion (except for one case)
2014-05-07 22:50:21 +04:00
33de0e8393
Smart completion for then and else branches of if
...
#KT-4910 Fixed
2014-05-07 22:50:20 +04:00
7aa11a4207
Smart completion: auto-cast to non-null type is not recognized for field from constructor
...
#KT-4906 Fixed
2014-05-07 22:50:20 +04:00
3c75e9ec28
Smart completion: "null" when nullable value expected
2014-05-07 22:50:20 +04:00
32405e57e6
Smart completion: checking constructor's visibility + true and false for Boolean
...
#KT-4916 Fixed
2014-05-07 22:50:20 +04:00
9ae59db329
Smart completion: more correct Tab key handling
...
#KT-4954 Fixed
2014-05-07 22:50:20 +04:00
5f931be39b
Smart completion after == and !=
...
#KT-4912 Fixed
2014-05-07 22:50:19 +04:00
74da6929a6
Update to IDEA 135.815 EAP
2014-05-07 21:12:56 +04:00
33fd82cf45
New Intention: Replace for loop with forEach
...
Replaces an expression of the form “for (i in list) { … }” with an
expression of the form “list.forEach { i -> … }”
2014-05-07 14:01:37 +04:00
dbb28c571b
New Intention Action: Replace a call to forEach function with for loop.
...
Replaces an expression of form x.forEach { … } with for (a in x) { … }.
2014-05-07 14:00:50 +04:00
bd056d037f
Init slice debug names for CodegenBinding
2014-05-07 03:30:24 +04:00
02e805b06e
Minor, rename accidentally named "context1" back to "context"
2014-05-07 03:30:24 +04:00
68313172c2
Don't use KFunctionN types in smart completion
...
FunctionN should be enough
2014-05-07 03:30:14 +04:00
c2594d7695
Delete "println" from production code in KotlinResolveCache
2014-05-07 03:28:48 +04:00
Ross Hanson