When testing KeyPathApplication constraints, we would keep going after rejecting a concrete KeyPath application by trying PartialKeyPath and AnyKeyPath, even though that's not what we want, since any key path application expression can type check with an AnyKeyPath. We would then miscompile by building the AST such that we applied the mismatched key path expression directly to the base. We also didn't handle expressions where the base was a subtype of the key path's base type correctly—the conversion means the base can't be written through in this situation, and we hardcoded the concrete-to-existential case instead of handling general conversions. Fix these problems, and add an AST verifier for KeyPathApplicationExprs to help catch problems in the future. Fixes SR-6300 | rdar://problem/35368903.
We can wind up in a situation where the base of a keypath
application is concrete but is being applied to a keypath
with an existential source type. Coerce the base to the
intended type if necessary when building the final subscript
expression.
Resolves SR-5878.
Special DeclNames represent names that do not have an identifier in the
surface language. This implies serializing the information about whether
a name is special together with its identifier (if it is not special)
in both the module file and the swift lookup table.
Update error messages to mention the invalid character.
Improve the diagnostic of floating point exponents.
Add tests for error messages when parsing floating point exponents.
Update existing tests for new error messages.
Rephrased error message to indicate which character is unexpected.
Provide error message variations when parsing binary, octal, decimal (default), and hexadecimal integer literals.
Look for unexpected digits in binary and octal integer literals.
Look for unexpected letters in hex integer literals.
Resolves: SR-5236 rdar://problem/32858684
We want the type of a KeyPathExpr to be the specific *KeyPath<T, U> subclass appropriate for the literal, with upcasts to a more general contextual type, since we rely on that invariant elsewhere to extract the base and projected value types. Fixes SR-5008 | rdar://problem/32395076.
Calls involving single trailing closure arguments require special
handling because we don't have as much contextual information
about function/argument types as in with regular calls, which means
that diagnosing such situations only by `visitApplyExpr`
yields subpar results.
Resolves: SR-4836.
Previously we would erroneously flag a `var` as not being mutated if the only mutations were through WritableKeyPaths. Fixes SR-5214 | rdar://problem/32599483.
Conditional and forced downcasts enter a constraint that almost
always succeeds; only when applying the solution do we evaluate
the feasability of the cast and determine if it always succeeds,
always fails, or conditionally succeeds. This changes how the
resulting AST is represented and can also emit diagnostics.
If the conditional cast is at this stage determined to always
succeed, we treat it as an unconditional cast, going through
ExprRewriter::coerceToType() to build the AST for the coercion.
However conditional cast constraints don't enter the same
restrictions into the solution as unconditional casts do, so
coerceToType() would fall over if casting a Swift type to a CF
type by first bridging the Swift type to Objective-C.
Get around this by checking for this case explicitly when
lowering a CoerceExpr.
It feels like there's a more fundamental issue here with how
casts are modeled in the constraint solver, but I'm not going
to try understanding that now.
Fixes <rdar://problem/32227571>.
Add an explicit argument label to the subscript declarations used in the
tests so that x[keyPath:] refers to the subscript rather than the
keypath.
Update test output accordingly.
When this score kind was added, the marker for the last score kind was
not updated. As a result, increasing the score for SK_KeyPathSubscript
was not actually having an effect.
Add tests that include a type with subscripts that also take the key
paths as their argument with an identical argument label ("keyPath").
These tests show that we're actually falling back on the keypath
application in many cases despite the score kind specific to keypath
application. I'll open a couple new JIRAs to investigate this
behavior.
The update to SK_LastScoreKind results in fixing a crash in these tests
that happens as a result of an assumption in CSRanking.cpp that if we're
comparing two solutions we end up with decls for overloads to compare,
which isn't the case here due to the keypath application.
Record the initializer type as soon as we have a solution, before
it is applied, and get the type from the constriant system instead
of from the final type checked expression.
Note that the coerceToMaterializable() was unnecessary, since we
always coerce the value to an rvalue type with coerceToType().
Eventually coerceToMaterializable() should go away.
This is mostly NFC, except using the result of simplifyType() rather
than the type of the final expression changes some diagnostics where it
appears we were previously losing sugar.
Also this accidentally fixes a crasher. Unfortunately the underlying
issue is still there (applying a solution has bugs with opened
existentials "leaking" out) -- this merely masks the problem by
getting the initializer type directly from the constriant system.
And make an attempt to use the constraint system's contextualType to bind type information; this makes things better but doesn't seem to be a complete solution for contextually typing key paths.
