While solving a conjunction that represents a (multi-statement) closure
constraint system should use such closure as its declaration context,
otherwise member lookup would produce incorrect results.
It helps to simply handling of outer constrants because they have
to be added to the constraint system before scope is created but
constraint graph have to get updated after to make sure that
incremental binding inference already knows about types inferred
from conjunction.
Iterate over all of the elements one-by-one and make sure that
each results in a single solution, otherwise fail the conjunction step.
Once all of the elements are handled either stop or,
if conjunction step has been performed in isolation,
return all of the outer constraints back to the system
and attempt to solve for outer context - that should
produce one or more solutions for conjunction to be
considered successfully solved.
Currently all `ComponentSteps` created by `DependentComponentSplitterStep` share the same `Solutions` vector. Because of this, the `ComponentStep`s might modify solutions created by previous `ComponentStep`s. Use different `Solutions` vectors for each `ComponentStep` to avoid sharing information between the `ComponentStep`s.
The concrete manifestation in the added test case is that the `Bar` overload gets added to `Solutions`, it’s score gets reduced by its `ComponentStep` original score, then the `Foo` overload gets added to `Solutions` and both solutions have their score decreased by the `OriginalScore` of `Foo`’s `ComponentStep`, causing `Bar`’s score to underflow.
Fixes rdar://78780840 [SR-14692]
`PotentialBindings` lost most of its responsibilities,
and are no longer comparable. Their main purpose now
is binding and metadata tracking (introduction/retraction).
New `BindingSet` type is something that represents a set
of bindings at the current step of the solver.
Create a new namespace - `swift::constraints::inference` and associate
`PotentialBinding` with it. This way it would be possible for constraint
graph to operate on `PotentialBinding(s)` in the future.
disjunction choice that does not introduce conversions, check to see
if known argument types satisfy generic operator conformance requirements
early, and skip the overload choice if any requirements fail.
This helps the solver avoid exploring way too much search space when
the right solution involves a generic operator, but the argument types
are known up front, such as `collection + collection + collection`.
successfully finding a solution by favoring operators already bound
elsewhere.
Favoring existing operator bindings often lets the solver find a solution
fast, but it's not necessarily the best solution.
Previously we could skip default literal or
supertype bindings if we had already found a solution
with fixes, which could lead us to miss bindings
that produce better diagnostics.
Tweak the logic such that we continue exploring if
we're in diagnostic mode.
Resolves SR-12399.
Reverts apple/swift#30006. It caused a regression that we'd like to address before re-landing:
```swift
struct X {
var cgf: CGFloat
}
func test(x: X?) {
let _ = (x?.cgf ?? 0) <= 0.5
}
```
This reverts commit 0a6b444b49.
This reverts commit ed255596a6.
This reverts commit 3e01160a2f.
This reverts commit 96297b7e39.
Resolves: rdar://problem/60185506
Rather than maintaining a linked list of overload
choices, which must be linearly searched each time
we need to lookup an overload at a given callee
locator, use a MapVector which can be rolled back
at the end of a scope.
Remove ResolvedOverloadSetListItem in favor of
using SelectedOverload, which avoids the need to
convert between them when moving from
ConstraintSystem to Solution.
An awful pattern we use throughout the compiler is to save and restore global flags just for little things. In this case, it was just to turn on some extra options in AST printing for type variables. The kicker is that the ASTDumper doesn't even respect this flag. Add this as a PrintOption and remove the offending save-and-restores.
This doesn't quite get them all: we appear to have productized this pattern in the REPL.
There were a few places where we wanted fast testing to see whether a
particular type variable is currently of interest. Instead of building
local hash tables in those places, keep type variables in a SetVector
for efficient testing.
Now that we've moved to C++14, we no longer need the llvm::make_unique
implementation from STLExtras.h. This patch is a mechanical replacement
of (hopefully) all the llvm::make_unique instances in the swift repo.
Introduce the notion of "one-way" binding constraints of the form
$T0 one-way bind to $T1
which treats the type variables $T0 and $T1 as independent up until
the point where $T1 simplifies down to a concrete type, at which point
$T0 will be bound to that concrete type. $T0 won't be bound in any
other way, so type information ends up being propagated right-to-left,
only. This allows a constraint system to be broken up in more
components that are solved independently. Specifically, the connected
components algorithm now proceeds as follows:
1. Compute connected components, excluding one-way constraints from
consideration.
2. Compute a directed graph amongst the components using only the
one-way constraints, where an edge A -> B indicates that the type
variables in component A need to be solved before those in component
B.
3. Using the directed graph, compute the set of components that need
to be solved before a given component.
To utilize this, implement a new kind of solver step that handles the
propagation of partial solutions across one-way constraints. This
introduces a new kind of "split" within a connected component, where
we collect each combination of partial solutions for the input
components and (separately) try to solve the constraints in this
component. Any correct solution from any of these attempts will then
be recorded as a (partial) solution for this component.
For example, consider:
let _: Int8 = b ? Builtin.one_way(int8Or16(17)) :
Builtin.one_way(int8Or16(42\
))
where int8Or16 is overloaded with types `(Int8) -> Int8` and
`(Int16) -> Int16`. There are two one-way components (`int8Or16(17)`)
and (`int8Or16(42)`), each of which can produce a value of type `Int8`
or `Int16`. Those two components will be solved independently, and the
partial solutions for each will be fed into the component that
evaluates the ternary operator. There are four ways to attempt that
evaluation:
```
[Int8, Int8]
[Int8, Int16]
[Int16, Int8]
[Int16, Int16]
To test this, introduce a new expression builtin `Builtin.one_way(x)` that
introduces a one-way expression constraint binding the result of the
expression 'x'. The builtin is meant to be used for testing purposes,
and the one-way constraint expression itself can be synthesized by the
type checker to introduce one-way constraints later on.
Of these two, there are only two (partial) solutions that can work at
all, because the types in the ternary operator need a common
supertype:
[Int8, Int8]
[Int16, Int16]
Therefore, these are the partial solutions that will be considered the
results of the component containing the ternary expression. Note that
only one of them meets the final constraint (convertibility to
`Int8`), so the expression is well-formed.
Part of rdar://problem/50150793.
Move the logic for creating connected components of orphaned
constraints into the connected-components algorithm code, rather than
making it a special part of SplitterStep.
Have the constraint graph's connected-component implementation be more
self-contained, producing a vector containing each of the actual
components (where each is defined by a list of type variables and a list
of constraints). This simplifies the contract with the client
(SplitterStep) and eliminates a bunch of separate mapping steps to
interpret the results.
It also lets us enrich the Component data structure in the future.
Currently finalization e.g. scope reset and solution minimization
is only done if component step had follow-up e.g. type variable or
disjunction step(s), but it should be done if `take` generated any
fixes as well, or component changed score in any way, otherwise
we might miss some solutions with fixes because "best score" haven't
been reset properly.
The ConstraintSystem class is on the order of 1000s of bytes in size on
the stacka nd is causing issues with dispatch's 64k stack limit.
This changes most Small data types which store data on the stack to non
small heap based data types.
When simplifying a function application constraint, check the argument
labels for that application against the disjunction containing the overload
set, disabling any overloads with mis-matching labels. This is staging for
several different directions:
* Eliminating the argument label matching from performMemberLookup, where it
does not belong
* More aggressively filtering the overload set when we have some concrete
information about argument types
* Identifying favored constraints when we have some concrete information
about argument types
At present, the only easily-visible effect of this change is that
we now properly handle argument label matching for non-member functions.
