The primary constructor of a class needs to be the first subgraph of the
class control-flow graph. Based on the Kotlin specification, class
initialization order goes first primary constructor, in-place
declarations (properties and init blocks), and then secondary
constructors. If the class doesn't have a primary constructor, then it
is just skipped in the order.
Unfortunately, the class control-flow graph had in-place declarations
first and then all constructors. Instead, we should treat the primary
constructor as the first in-place declaration, and then continue with
the existing processing as secondary constructors. This will guarantee
that super constructor calls have the correct property initialization
information.
^KT-65093 Fixed
There might be several pathes out from the function from CFG point of view:
- normal execution path
- exceptional execution path
For verification of callsInPlace contract it's needed to check only
the normal path
^KT-63414 Fixed
There are many complications with the current design of passing data
from within in-place lambdas to surrounding code. Solving these
complications will involve more time to investigation than is available
within the K2 release. So we are disabling passing type statement
information from lambdas for the time being until more time can be
devoted to a more complete solution.
^KT-60958 Fixed
^KT-63530 Fixed
Previously, there was a contract that each callable symbol in the chain
of `processDirectOverriddenWithBaseScope` will be unique. And if some
symbol is accessible from multiple scopes, then only last of them will
be returned as a component of `MemberWithScope`. So there actually was
no such thing as "pair of two different MemberWithScope with the same
symbol and different scopes"
After the change of `processDirectOverriddenWithBaseScope` contract (see
previous commits) each scope returns `MemberWithScope` for some symbol
with the previous scope in the hierarchy even if it contains the same
symbol as the current scope
So now scope is actually a part of `MemberWithScope`, which should be
considered as part of equality. Otherwise, we can skip some part of
the overridden hierarchy, because we will start to consider symbol as
visited after its first occurrence (in opposite to the previous behavior,
when only the last scope was returned)
In order to properly analyze top-level property initialization, a
control-flow graph must be created for FirFiles. This change adds the
foundation for the file CFG and updates body resolve to create the CFG.
Checking the CFG for proper initialization is separated into a following
change to ease code review.
KT-56683
This prevents `FirConflictsExpressionChecker.kt`
from missing conflicting local functions. It used
to due to inconsistencies in assigning `<local>`,
and this commit makes it a bit more
straightforward.
The change in KtClassTypeQualifierRenderer
prevents `FirOverrideImplementTest.testLocalClass`
from failing in `intellij`. It didn't fail for
callables, because `KtCallableSignatureRenderer`
doesn't try to render packages.
^KT-59186 Fixed
Also fix graphs for enums with specialized entries - since we don't
create property subgraphs for FirEnumEntry, there is no body to insert
AnonymousObjectEnterNode, AnonymousObjectExitNode, and
AnonymousObjectExpressionExitNode into.
Interpretation: a graph A is a subgraph of B if information available at
nodes of A depends on the paths taken in B. For example, local classes
are subgraphs of a graph in which they are declared, and members of
those classes are subgraphs of the local class itself - because these
members can reference captured values.
Consequences:
* if graph G is a subgraph of node N, then G is a subgraph of N's
owner;
* `ControlFlowAnalysisDiagnosticComponent` will only visit root graphs;
* `graph.traverse` will ignore subgraph boundaries, as if all subgraphs
are inlined into one huge root graph;
* if a control flow checker needs information from a declaration to
which a graph is attached, it must look at subgraphs explicitly.
For example, consider the `callsInPlace` checker. When a function
has a `callsInPlace` contract and a local declaration, the checker must
visit that local declaration to ensure it does not capture the allegedly
called-in-place argument - hence `graph.traverse` will look at the
nodes. However, the local declaration can also be a function with its
own `callsInPlace` contracts, so the checker should also run for it in
isolation. If that sounds quadratic, that's because unfortunately it is.
This removes the need for hacks around the order in which function
call arguments are visited, fixes called-in-place lambda arguments
for augmented assignment operators, and makes CFG dumps a bit prettier.
Quick quiz:
Q: In a CFG, what does `a -> b -> c -> d` mean?
A: `a`, then `b`, then `c`, then `d`.
Q: In a CFG, what does `a -> b -> d; a -> c -> d` mean?
A: `a`, then `b` or `c`, then `d`.
Q: So how do you encode "a, then (b, then c) or (c, then b), then d`?
A: You can't.
Problem is, you need to, because that's what `a; run2({ b }, { c }); d`
does when `run2` has a contract that it calls both its lambda arguments
in-place: `shuffle(listOf(block1, block2)).forEach { it() }` is a
perfectly valid implementation for it, as little sense as that makes.
So that's what union nodes solve. When a node implements
`UnionNodeMarker`, its inputs are interpreted as "all visited in some
order" instead of the normal "one of the inputs is visited".
Currently this is used for data flow. It *should* also be used for
control flow, but it isn't. But it should be. But that's not so easy.
BTW, `try` exit is NOT a union node; although lambdas in one branch can
be completed according to types' of lambdas in another, data does not
flow between the branches anyway (since we don't know how much of the
`try` executed before jumping into `catch`, and `catch`es are mutually
exclusive) so a `try` expression is more like `when` than a function
call with called-in-place-exactly-once arguments. The fact that
`exitTryExpression` used `processUnionOfArguments` in a weird way
should've hinted at that, but now we know for certain.
The code was already duplicated between FirDataFlowAnalyzer and
FirReturnsImpliesAnalyzer, so might as well use the latter to slightly
speed up the former.
Make smart-casts non-transparent expression without delegation
to underlying FirQualifiedAccessExpression, as children delegation in
fir tree has unclear semantics
Remove two different kinds of tree nodes for smart-casts
Consider the following code:
```
fun test(a: List<String>) {
a.first()
}
```
The dispatch receiver type of `first` in this case is `List<T>` before
this change. After this change, it's `List<String>`.
In addition, this change also replace the dispatch receiver type with
the more specific type if available. For example, consider the following
```
class MyList: ArrayList<String>()
fun test(a: MyList) {
a.get(0)
}
```
The dispatch receiver type of `get` is `MyList`, instead of
`ArrayList<String>`. That is, a fake override is created in this case.
The fix is a bit hacky, but it's very simple. In addition, it still does
not handle the case where the receiver is a lambda function. But such
case seems to be fairly rare in practice.
Previously unsafe call is reported as part of InapplicableWrongReceiver.
This makes it difficult for the downstream checkers to report different
diagnostics.
Brian Norman