There were a number of mistakes in this test:
• The whole thing was indented one space.
• Some fix-it tests were malformed and therefore not being tested.
• The output checking could in theory allow content before or after the intended content.
- In Sema, don't traverse nested declarations while deducing the opaque return type. This would
cause returns inside nested functions to clobber the return type of the outer function.
- In IRGen, walk the list of opaque return types we keep in the SourceFile already for type
reconstruction, instead of trying to visit them ad-hoc as part of walking the AST, since
IRGen doesn't normally walk the bodies of function decls directly.
Fixes rdar://problem/50459091
Our ad-hoc mechanism for building the signature did not always produce requirements in the order
expected by the rest of the system; using the GSB should ensure we build a valid generic signature.
Fixes rdar://problem/50309983.
If "convertTo" type is an optional let's look through it to see
whether it contains another function type which, if so, would
rule out possibility of missing explicit call.
Resolves: rdar://problem/50438071
Way back in Swift 1 I was trying to draw a distinction between
"overlays", separate libraries that added Swift content to an existing
Objective-C framework, and "the Swift part of a mixed-source
framework", even though they're implemented in almost exactly the same
way. "Adapter module" was the term that covered both of those. In
practice, however, no one knew what "adapter" meant. Bring an end to
this confusion by just using "overlay" within the compiler even for
the mixed-source framework case.
No intended functionality change.
Previously, requests would fail silently by returning an empty struct
in the response.
With this change, responses will properly report fail with the internal
error.
These are testing for bitwise identical results, but don't guarantee that
the buffers being used always have identical alignment. This will result
in small rounding differences when vector codepaths are used for different
elements of some results.
This is partially an underlying bug in Accelerate (which is outside the
scope of this project to fix), and partly a test bug (which we can address
by adopting approximate comparisons here). In the short term, though, I'm
going to disable these.
This test was overfitting what it was trying to test, causing different
targets (and even different assert configurations) to fail to match.
Instead, make this an executable test and ensure that round-tripping
through setting and getting behaves as expected.
Fixes rdar://46486517
Keep track of information that led the module interface loader to reject loading a compiled module, if it needed to fall back to compiling an interface.
rdar://47792754
They aren't normally decl contexts, but if one has an opaque type, we want to be able to record
the property as a context so that we can reconstruct it in RemoteAST.
Fixes a crash in IRGen
TODO: also fix the demangler/remangler part of this mangling change.
Currently it's not a problem because we never demangle such a symbol (it's even not round-trip checked in Mangler::verify).
rdar://problem/50405691
KeyPath dynamic member lookup is limited to what key path itself
could do, so let's detect and diagnose invalid references just
like we do for regular key path expressions.
Resolves: rdar://problem/50376224
