It's necessary because even for stable a.b.c.d we can't guarantee that
this reference will always point to the same symbol because
different capture type instantiations generate different scopes
with different resulting symbol instances.
For example, NEW_INFERENCE_NO_INFORMATION_FOR_PARAMETER
It became especially relevant after 0e84bf2053
that together with later commits bring a lot of unnecessary
NEW_INFERENCE_NO_INFORMATION_FOR_PARAMETER diagnostic
For the following example, when we run the reference shortener, it
drops `a.b.c` qualifier, because it matches "FOURTH".
```
package a.b.c
fun <T, E, D> foo(a: T, b: E, c: D) = a.hashCode() + b.hashCode() + c.hashCode() // FIRST
fun <E> E.foo() = hashCode() // SECOND
object Receiver {
fun <T, E, D> foo(a: T, b: E, c: D) = a.hashCode() + b.hashCode() + c.hashCode() // THIRD
fun foo(a: Int, b: Boolean, c: String) = a.hashCode() + b.hashCode() + c.hashCode() // FOURTH
fun test(): Int {
fun foo(a: Int, b: Boolean, c: Int) = a + b.hashCode() + c // FIFTH
return <expr>a.b.c.foo(1, false, "bar")</expr>
}
}
```
As shown in the above example, when SHORTEN_IF_ALEADY_IMPORTED option is
given from a user, the reference shortener has to check whether it can
drop the qualifier without changing the referenced symbol and if it is
possible to do that without adding a new import directive, it deletes
the qualifier.
It needs two steps:
1. Collect all candidate symbols matching the signature e.g., function
arguments / type arguments
2. Determine whether the referenced symbol has the highest reference
priority when we drops the qualifier depending on scopes
This commit uses `AllCandidatesResolver(shorteningContext.analysisSession.useSiteSession).
getAllCandidates( .. fake FIR call/property-access ..)` for step1.
For step2, we use a heuristic based on scopes of candidates. If a
candidate symbol is under the same scope with the target expression, it
has a `FirLocalScope` which has the high priority. So when we have a
candidate under a `FirLocalScope` and the actual referenced symbol is
different from the candidate, we must avoid dropping its qualifier
because the shortening will change its semantics i.e., reference.
The order of scopes depending on their scope types is:
1. FirLocalScope
2. FirClassUseSiteMemberScope / FirNestedClassifierScope
3. FirExplicitSimpleImportingScope
4. FirPackageMemberScope
5. others
Note that for "others" the above rule can be wrong. Please update it if
you find other scopes that have a priority higher than the specified
scopes.
One of non-trivial parts is the priority among multiple
FirClassUseSiteMemberScope and FirNestedClassifierScope. They are
basically scopes for class declarations. We decide their priorities
based on the distance of class declaration from the target expression.
Note that we take a strict approach to reject all false positive. For
example, when we are not sure, we don't shorten it to avoid changing its
semantics.
TODO: One corner case is handling receivers. We have to update
```
private fun shortenIfAlreadyImported(
firQualifiedAccess: FirQualifiedAccess,
calledSymbol: FirCallableSymbol<*>,
expressionInScope: KtExpression,
): Boolean
```
The current implementation cannot handle the following example:
```
package foo
class Foo {
fun test() {
// It references FIRST. Removing `foo` lets it reference SECOND.
<caret>foo.myRun {
42
}
}
}
inline fun <R> myRun(block: () -> R): R = block() // FIRST
inline fun <T, R> T.myRun(block: T.() -> R): R = block() // SECOND
```
Tests related to TODO:
- analysis/analysis-api/testData/components/referenceShortener/referenceShortener/receiver2.kt
- analysis/analysis-api/testData/components/referenceShortener/referenceShortener/receiver3.kt
Class exit node should only unify data flow if it is an anonymous
object, in which case it only has one control flow input. Coversely,
any other class exit node has normal merge semantics for its control
flow inputs, but it won't have any data flow inputs.
If you care about direction, then you also care about labels.
3 out of 6 places that use `traverse` are incorrect, by the way.
Node order does not correspond to scoping boundaries.
UncaughtExceptionPath label out of a finally block matches every label
that is not handled by another edge, and a labeled edge from the middle
of a finally block aborts the jump and so should merge all available
data.
They are only used in one place that can just as well use kinds.
Especially considering that "the one place" used them incorrectly and
would not attach local functions in property accessors as subgraphs.
function enter -> default 1 -> default 2 -> rest of function
\----------^ \----------^
This probably has no effect (in non-stupid code, at least), but it makes
graph construction more architecturally correct (now value parameters'
subgraphs get attached to a node).
The receiver of the provideDelegate call is the same FirExpression as
the delegate itself, so there's only one copy of the nodes in the first
place; trying to remove subgraphs completely detaches objects inside it
from the parent graph, which is not great for checkers.
Note that currently if provideDelegate is not selected, there will be a
stray FunctionCallExit node in the control flow graph. This commit *does
not change that*. It has been there for a while. Don't @ me. I'll try to
fix that. No promises.
While it is theoretically useful to know that `{ while(true) {} }`
returns Nothing, CFG node deadness is not precise enough to do that: if
the entire lambda is dead, it's no longer possible to find out whether
the loop is terminating. Besides, `while (true)` and `if (true)` are
pretty much the only constructs like that anyway.
Note that this commit does not affect resolution for lambdas that end in
a Nothing-returning expression, e.g. `throw`.
`candidateSymbol` has any reasonable meaning only for references with
not completed candidate, so this property is moved from FirNamedReference
to new node FirNamedReferenceWithCandidateSymbol, which has real
implementation only in :resolve module (`FirNamedReferenceWithCandidate`)
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.
The delegate is resolved in context-dependent mode and thus can be an
incomplete call; if there is no `provideDelegate` method to complete it,
the result is effectively `val x$delegate = y.id()` where `id` is
`fun <T> id(x: T) = x`, except we don't get a real node for `id` so the
DFA edges from lambdas in `y` go who knows where.
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 distinction is similar to persistent data structures and their
builders. Only the MutableFlow can be passed to LogicSystem for
modification; when it's ready, it can be converted into PersistentFlow
and attached to a CFG node.
The result is that the API is cleaner, the implementation is a bit more
neat, and hopefully the use of PersistentHashMap.Builder improves
performance a little. Also the node-to-flow map can now be removed in
favor of just storing PersistentFlow inside a node, seeing as it's
explicitly immutable and all that.
If a certain type statement is true on loop entry and all continue
paths, then it is also true on exit if the condition did not reassign
the variable.
^KT-7676 tag fixed-in-k2
Dmitriy Novozhilov