Attempting to favor a disjunction choice here
can lead to unhelpful results if there's e.g a code
completion token argument, since it acts as a
placeholder.
rdar://127844278
Although I don't plan to bring over new assertions wholesale
into the current qualification branch, it's entirely possible
that various minor changes in main will use the new assertions;
having this basic support in the release branch will simplify that.
(This is why I'm adding the includes as a separate pass from
rewriting the individual assertions)
Track the implied result exprs in the constraint
system, and allow arbitrary propagation of
implied results down if/switch expression
branches. This is required for allowing implied
results in non-single-expression closures.
Start classifying all potential throw sites within a constraint
system and associate them with the nearest enclosing catch node. Then,
determine the thrown error type for a given catch node by taking the
union of the thrown errors at each potential throw site. Use this to
compute the error type thrown from the body of a `do..catch` block
within a closure.
This behavior is limited to the upcoming feature `FullTypedThrows`.
There are some situations where the solver is able to find a valid
solution only during `salvage` (mostly but not limited to unavailable
declarations), which means that we need to keep running `salvage`
even if the diagnostics are suppressed until the underlying issues
in the normal solving mode are fixed.
One of the issues:
```swift
extension Unmanaged {
@inline(__always)
internal static func passRetained(_ instance: __owned Instance?) -> Self? {
guard let instance = instance else { return nil }
return .passRetained(instance)
}
}
```
`.passRetained(instance)` is ambiguous during normal solving but
is able to find a solution during `salvage` because it attemtps
more type bindings.
Resolves: rdar://119001449
`lookupConformance` request is not cached and constraint solver
performs a lot of them for the same type (i.e. during disjunction
solving), let's try to cache previously performed requests to
see whether additional memory use is worth the performance benefit.
Previously we would only do source ordering for
ClosureExprs, but other conjunctions need to have
their source location taken into account too, in
order to make sure we don't try and type-check e.g
a TapExpr in a second closure before we type-check
the first closure.
Also while here, switch to `std::min_element`
instead of sorting, and treat invalid source
locations as incomparable.
rdar://113326835
Move the contextual type locator onto
ContextualTypeInfo, and consolidate the separate
fields in SyntacticElementTarget into storing a
ContextualTypeInfo. This then lets us plumb down
the locator for the branch contextual type of an
if/switch expression from the initial constraint
generation, rather than introducing it later. This
should be NFC.
This is phase-1 of switching from llvm::Optional to std::optional in the
next rebranch. llvm::Optional was removed from upstream LLVM, so we need
to migrate off rather soon. On Darwin, std::optional, and llvm::Optional
have the same layout, so we don't need to be as concerned about ABI
beyond the name mangling. `llvm::Optional` is only returned from one
function in
```
getStandardTypeSubst(StringRef TypeName,
bool allowConcurrencyManglings);
```
It's the return value, so it should not impact the mangling of the
function, and the layout is the same as `std::optional`, so it should be
mostly okay. This function doesn't appear to have users, and the ABI was
already broken 2 years ago for concurrency and no one seemed to notice
so this should be "okay".
I'm doing the migration incrementally so that folks working on main can
cherry-pick back to the release/5.9 branch. Once 5.9 is done and locked
away, then we can go through and finish the replacement. Since `None`
and `Optional` show up in contexts where they are not `llvm::None` and
`llvm::Optional`, I'm preparing the work now by going through and
removing the namespace unwrapping and making the `llvm` namespace
explicit. This should make it fairly mechanical to go through and
replace llvm::Optional with std::optional, and llvm::None with
std::nullopt. It's also a change that can be brought onto the
release/5.9 with minimal impact. This should be an NFC change.
Previously we would wait until CSApply, which
would trigger their type-checking in
`coercePatternToType`. This caused a number of
bugs, and hampered solver-based completion, which
does not run CSApply. Instead, form a conjunction
of all the ExprPatterns present, which preserves
some of the previous isolation behavior (though
does not provide complete isolation).
We can then modify `coercePatternToType` to accept
a closure, which allows the solver to take over
rewriting the ExprPatterns it has already solved.
This then sets the stage for the complete removal
of `coercePatternToType`, and doing all pattern
type-checking in the solver.
Models `PackExpansionType` as a type variable that can only
bind to `PackExpansionType` and `expansion of` constraint that
connects expansion variable to its pattern, shape types.
* Use fancy arrows (`→`) because they are distinct from and shorter than `->`,
and fancier.
* We have two ways of demarcating locators: `@ <locator>` and `[[<locator>]];`.
Stick to the first, which is shorter and clearer.
* 'attempting type variable' → 'attempting binding'. *Bindings* are attempted,
not type variables.
* `considering ->` → `considering:`. I think a colon is semantically more fit
and makes things easier to read if the considered constraint has arrows in its
description or types. It’s also shorter this way.
