This can return ErrorType if the AST is invalid.
A handful of callers handle the ErrorType result, but most don't,
blindly assuming the result is always a nominal type. This resulted
in a crash in at least one test case.
Lift the burden from callers by always returning a nominal type here.
When the CustomAvailability experimental feature is enabled, make it an error
to specify an unrecognized availability domain name. Also, add these
diagnostics to a diagnostic group so that developers can control their behavior
when they are warnings.
Resolves rdar://152741624.
Some functions like memchr are defined both in libc++ and libc.
Including both would result in ambiguous references at the call sites.
This is worked around by an attribute that tells the compiler to prefer
one overload over the others. This attribute was not interpreted by
Swift. As a result, importing both libc and libc++ and calling such
functions resulted in compilation errors due to ambiguous overloads.
This PR modifies the lookup logic to exclude the non-preferred Clang
functions from the overload set whenever a preferred overload is
available.
rdar://152192945
No warnings with minimal checking, warnings with `strict-concurrency=complete` and
if declaration is `@preconcurrency` until next major swift version.
Resolves: rdar://151911135
Resolves: https://github.com/swiftlang/swift/issues/81739
The version remapping for `@backDeployed` regressed due to a bug introduced by
https://github.com/swiftlang/swift/pull/81922.
Also, fix some visionOS tests that have gotten out of date because we don't
seem to be running them in CI.
Resolves rdar://152542983.
NFC *except* that I noticed a bug by inspection where we suppress
`@escaping` when print enum element types. Since this affects
recursive positions, we end up suppressing `@escaping` in places
we shouldn't. This is unlikely to affect much real code, but should
still obviously be fixed.
The new design is a little sketchy in that we're using `const` to
prevent direct use (and allow initialization of `const &` parameters)
but still relying on modification of the actual object. Essentially,
we are treating the `const`-ness of the reference as a promise to leave
the original value in the object after computation rather than a
guarantee of not modifying the object. This is okay --- a temporary
bound to a `const` reference is still a non-`const` object formally
and can be modified without invoking UB --- but makes me a little
uncomfortable.
Reduce the burden on the evaluator's caching mechanism by handling the
unwrapping of a conformance down to its normal protocol conformance
outside of these requests. Thanks, Slava!
Verifying USR mangling adds ~30% overhead to indexing times. Since an incorrect USR mangling doesn't result in a correctness issue at the same level as a miscompile, save those 30% in non-assert builds.
Inference of conformance isolation needs to check whether all of the
witnesses are nonisolated. However, witness checking looks at
conformance isolation. To break this reference cycle, split the
conformance isolation request into two requests: a "raw" request that
looks at explicitly-specified isolation, and the existing one that
also performs inference. The existing one builds on the "raw" one, as
does a separate path for the conformance checker.
Fixes rdar://152461344.
When compiling for visionOS, iOS availability attributes are remapped into the
visionOS availability domain automatically. While the version remapping was
being performed correctly, there was a regression that caused the platform name
to be printed incorrectly in many diagnostics. Whenever an iOS version is
remapped to a visionOS version, availability diagnostics will now present
those versions as visionOS versions instead of iOS versions.
Resolves rdar://146293165.
getContextSubstitutionMap() didn't handle the case where getAnyNominal()
returns a ProtocolDecl. This should not take the "fast path", which is
only suitable for concrete nominals.
This manifested as a crash-on-invalid -- the user probably meant to write
"T.Value: Collection" rather than "T.Value == Collection".
Fixes rdar://151479861.
We would previously unconditionally create the accessor as a throwing
accessor irrespective of the parameter. This seems like an oversight and
results in an incorrect accessor being formed. This was caught while
trying to improve the ClangImporter's handling of aliasing macros.
The migration to `MemberImportVisibility` can be performed mechanically by
adding missing import declarations, so offer automatic migration for the
feature.
Resolves rdar://151931597.
The was never invoked because inaccessibility due to SPI protection level is
always diagnosed before missing imports are diagnosed. The functionality could
therefore not be tested and should be removed.
