Commit Graph

4068 Commits

Author SHA1 Message Date
Leonid Startsev 527ccaff16 Changes required for serialization plugin to work:
* Support for generation of synthetic nested classes (for implementations
of user-defined serial annotations):

    1. Compiler extenstion points for contributing names and descriptoprs
of nested classes

    2. Control on synthetic class primary constructor visibility

* Public functions for generating initializing expressions for optional
properties
2017-07-28 07:04:57 -07:00
Dmitry Petrov 9e6b706a03 Unwrap property imported from object on property access
#KT-18982 Fixed Target versions 1.1.5
2017-07-28 17:01:04 +03:00
Dmitry Petrov 67336653e0 Generate SAM wrappers only if they are required for a given argument
SAM interface wrapper for an argument is required,
if in the function descriptor for SAM adapter
type for the corresponding value parameter
doesn't match type of the corresponding value parameter
in the original (Java) descriptor.

 #KT-19251 Fixed Target versions 1.1.5
2017-07-28 08:24:46 +03:00
Dmitry Petrov c8bc2d9d16 Minor: code cleanup 2017-07-28 08:24:46 +03:00
Dmitry Petrov c9ad290ad5 Do not store null for temporary in destructuring assignment
#KT-19256 Fixed Target versions 1.1.5
2017-07-27 09:02:26 +03:00
Dmitry Petrov b867c46f72 Generate for-in-range loop as counter loop when possible
If an upper bound is a compile-time constant != Int.MAX_VALUE,
we can generate 'for (i in x..N)' as 'for (i in x until N+1)'.
2017-07-27 09:02:26 +03:00
Dmitry Petrov 7450899b87 Minor: extract get*CompileTimeConstant
Get rid of code duplication.
2017-07-27 09:02:26 +03:00
Mikhael Bogdanov ff676c050f Properly generate inner classes when compiling source for inline
KT-19175: Compiler generates different bytecode when classes are compiled separately or together

 #KT-19175 Fixed
2017-07-26 20:31:14 +02:00
Dmitry Petrov f558b4238c Use counter loop in intrinsic array constructors
#KT-19149 Fixed Target versions 1.1.5
2017-07-24 10:17:30 +03:00
Dmitry Petrov c9d9a8220d Minor: use IntrinsicMethods.INTRINSICS_CLASS_NAME where appropriate 2017-07-24 10:15:27 +03:00
Dmitry Petrov d0c8af5621 Keep track of second receiver type in CallReceiver
#KT-17725 Fixed Target versions 1.1.5
2017-07-24 10:15:27 +03:00
Dmitry Petrov 3f34a97880 Minor: extract CallReceiver to a top-level class, move code around 2017-07-24 10:15:27 +03:00
Mikhail Glukhikh 951e8cd91a Code cleanup: unnecessary local variable applied 2017-07-21 18:58:48 +03:00
Dmitry Petrov 709a7e201f Simplify code for specialized BranchedValues 2017-07-21 08:52:29 +03:00
Dmitry Petrov e1a55e8dec Prefer compact bytecode for primitive/object comparisons with Boolean
'java.lang.Boolean#valueOf' doesn't allocate new objects
(it uses pre-allocated singletons for 'true' and 'false').
2017-07-21 08:52:29 +03:00
Dmitry Petrov 8e9c0294fe Do not box primitives for 'primitive == object'
NB user-defined 'equals' can violate contract for 'Object#equals', e.g.,
it can be asymmetric.
Thus we can't avoid boxing for 'object == primitive'.
2017-07-21 08:52:29 +03:00
Dmitry Petrov fa42f202fa Avoid boxing on 'primitive == boxed' when possible
Similar to 'boxed == primitive' case, different bytecode due to
evaluation order.
2017-07-21 08:52:29 +03:00
Dmitry Petrov 98ee83f4a3 Minor: move condition for boxed==primitive to corresponding isApplicable 2017-07-21 08:52:29 +03:00
Dmitry Petrov 81609e4c6f Generate complex equality comparison only when RHS can have side effects
In "short" version, LHS is always evaluated before RHS,
and RHS may be not evaluated if LHS is null.
So, it makes sense to use "short" version in cases when RHS can't have
side effects (because of more compact bytecode and more opportunities
for code elimination).
2017-07-21 08:52:29 +03:00
Dmitry Petrov 8aacddb9f0 Avoid primitive boxing for 'boxed == primitive' if possible
This makes sense for non-floating-point primitive type
(boolean, char, byte, short, int, long):
floating-point types use specialized versions of 'areEqual'.
2017-07-21 08:52:29 +03:00
Dmitry Petrov 275c758da1 Cleanup and move code around in BranchedValue
- Cleanup inspections
- Move optimized boxed-vs-primitive comparisons to a separate file
2017-07-21 08:52:29 +03:00
Dmitry Petrov 051ac44759 Verify bytecode after mandatory transformations and after optimizations 2017-07-21 08:52:29 +03:00
Alexander Udalov 5128b8a409 Add default implementation for Annotations.findAnnotation 2017-07-20 13:27:34 +03:00
Alexander Udalov cc7ed2ba54 Change map key type in AnnotationDescriptor.getAllValueArguments
Turns out, only the parameter's name is needed at all usages of this
method. Such a map is both easier to use (no need to call
ValueParameterDescriptor.getName) and easier to construct (no need to
resolve annotation class, its constructor, its parameters). In this
commit, only usages have changed but the implementations are still using
the old logic, this is going to be refactored in subsequent commits
2017-07-20 13:25:39 +03:00
Alexander Udalov 9ee4b39e1b Provide default implementation for Annotations.findExternalAnnotation 2017-07-20 13:25:38 +03:00
Denis Zharkov 904c7f9c64 Fix nullable variables spilling in coroutines
See the last comment in testData for clarification

 #KT-18983 Fixed
2017-07-20 10:53:32 +03:00
Dmitry Petrov 891799c853 Generate 'Deprecated' annotation on invisible companion object field
Design decision: if the field for a companion object should have
non-public visibility, generate it with @Deprecated annotation in
language version 1.2.

TODO: generate it with proper visibility in 1.3 and later.

KT-11567 Companion object INSTANCE field more visible than companion object class itself
2017-07-19 17:38:17 +03:00
Roman Elizarov f2b5f37b22 Support inline suspend functions built with compiler version less than 1.1.4/1.2-M1
The error message is removed and is replaced with a code that adapts
inline suspend functions produced by the old compiler with the
suspension markers that new compiler expects.
2017-07-19 08:59:55 +03:00
Nikolay Krasko d282f14eac Don't add double linenumber when return expression is absent (KT-18949)
#KT-18949 Fixed
2017-07-18 14:39:54 +03:00
Nikolay Krasko a41c9d2476 Avoid state machine build for suspend function with single inline call
#KT-18949 Fixed
2017-07-18 14:39:54 +03:00
Nikolay Krasko ce4f923ba0 Replace return with nop to avoid merging line instructions (KT-18949)
Dex ignores subsequent line numbers for same instructions and interprets
instruction after inline as if they were inlined. This makes debugger
behaves as if there's nowhere to stop on line with breakpoint.

This also makes stepping through inline function consistent with
non-inline analog. In both context debugger now stops on '}'.

 #KT-18949 Fixed
 #KT-17120 Fixed
2017-07-18 14:39:54 +03:00
Dmitry Petrov 8c7352e668 Generate constant "" for effectively empty string 2017-07-17 09:18:41 +03:00
Alexander Udalov 2611c7de7e Restore anonymous classes for local delegate metadatas in inline functions
The problem is that now that the local delegated property metadata is in
the $$delegatedProperties array of the containing class, the access to
it from code calling an inline function with a local delegated property
is illegal.

Currently it seems to be a lot of work to support this rather rare case
properly (see the comment in ExpressionCodegen.getVariableMetadataValue)
so we postpone it and return the old behavior of using the anonymous
KProperty subclass for metadata
2017-07-14 15:13:12 +03:00
Alexander Udalov 616d575fb6 Support reflection for local delegated properties
#KT-15222 Fixed
2017-07-14 15:13:12 +03:00
Alexander Udalov db27a885f1 Rework property metadata generation for local delegated properties
Instead of generating an anonymous class for each delegated local
variable, store metadatas in the $$delegatedProperties array of the
containing class, as is done for member properties
2017-07-14 15:04:05 +03:00
Alexander Udalov 9344f7f42f Remove unused and obsolete code in JVM backend 2017-07-14 15:04:05 +03:00
Alexander Udalov 728de91140 Refactor generation of metadata for delegated properties
In CodegenAnnotatingVisitor, store all delegated properties whose
metadata should be generated in each class (identified by ASM Type).
Use the stored information later, both in the $$delegatedProperties
array generation and in the access to it from property's accessor
methods, instead of an heuristical indexOfDelegatedProperty()
2017-07-14 15:04:04 +03:00
Alexander Udalov b97589b33e Remove CodegenBinding.LOCAL_VARIABLE_DELEGATE
Calculate this value at call sites explicitly instead
2017-07-14 15:04:04 +03:00
Alexander Udalov 66ea288be7 Minor, extract fakeDescriptorsForReferences.kt 2017-07-14 15:04:04 +03:00
Dmitry Petrov da404de51e Add comment regarding testData generation to RangeValues.kt 2017-07-12 16:23:36 +03:00
Dmitry Petrov 6a673a03e4 Generate test data for in-expression tests, fix bug with indices 2017-07-12 15:56:29 +03:00
Dmitry Petrov f4d63158cc Take into account JvmName annotation when generating accessors
#KT-17444 Fixed Target versions 1.1.5
2017-07-12 09:37:16 +03:00
Dmitry Petrov efb6756cbc Initialize property metadata array before class body generation
Initialization of companion object members (e.g., delegate properties
using provideDelegate convention) can depend on property metadata array,
which in turn can be initialized before other class members.

 #KT-18902 Fixed Target versions 1.1.5
2017-07-12 09:37:16 +03:00
Dmitry Petrov 8a9707c140 Generate proper initialization for possibly nullable SAM wrapper
#KT-18916 Fixed Target versions 1.1.5
2017-07-12 09:37:16 +03:00
Mikhail Glukhikh dfe2c16bc7 More cleanup: lift return / assignment out 2017-07-10 12:59:58 +03:00
Dmitry Petrov 0ce6bac7eb Generate proper 'compareTo' calls for non-primitive 'a in x .. y'
When we have some custom implementation of Comparable, it's important
that we compare values exactly as 'lowBound <= a && a <= highBound'.

Make sure that evaluation order and compareTo calls match for
optimized and non-optimized case.
2017-07-10 10:51:26 +03:00
Dmitry Petrov aa7db727ba Fold I2L with ICONST_n when beneficial
ICONST_0; I2L -> LCONST_0
ICONST_1; I2L -> LCONST_1
2017-07-10 10:51:26 +03:00
Dmitry Petrov ad80c3cd7f If at least one of the range bounds is "pure", order doesn't matter 2017-07-10 10:51:26 +03:00
Dmitry Petrov 905a16e1df Maintain proper evaluation order for 'a in x .. y'
As of Kotlin 1.0 and 1.1, expression 'a in x .. y' is considered
equivalent to 'x.rangeTo(y).a', and should be evaluated in the following
order:
1. x
2. y
3. a
4. compare x with a
5. compare y with a (if needed)
2017-07-10 10:51:26 +03:00
Dmitry Petrov fc3e9318d9 Handle DUPnXm instructions in PopBackwardPropagationTransformer 2017-07-10 10:51:26 +03:00