When using `-experimental-skip-all-function-bodies` we don’t run the `TypeCheckSourceFileRequest` and thus don’t go through the decl checker, which calls `InheritedTypeRequest` on all inheritance clauses. This means that the inherited entries are not populated by the time we serialize the module. Trigger the computation of inherited entries by calling `InheritedTypeRequest` during serialization.
Unfortunately, we can’t use the type returned by `getResolvedType` for the serialization because `getResolvedType` returns an inverted protocol type for suppressed conformances but during serialization, we want to serialize the suppressed type with a `isSuppressedBit`. We thus need to call `getEntry(i).getType()` again to get the type to serialize.
rdar://141440011
These tests are using FileCheck to check the result of diagnostic
formatting in ways that don't match the new formatter. Force the old
formatter or, where possible, generalize so that they match both
formatters.
Specifically:
1. Fix the error message so that when we say you can't have a deinit that a
deinit can be on a noncopyable type along side a class or an actor.
2. Even though we already error on @objc enums and say they cannot be
noncopyable, we did not emit an error on the deinit saying that @objc enums
cannot have a deinit. I put in a nice to have error just to make it even
clearer.
rdar://105855978
rdar://106566054
- Infer type wrappers only from direct (declared on type) protocols
- Inferences from protocol to protocol is not allowed
- If type specifies a type wrapper attribute explicitly it has to
match the one associated with a declared protocol, otherwise the
declaration is going to be rejected as having multiple wrappers.
Introduce a `getTopmostDeclarationDeclContext`
utility to ensure we ensure we don't visit a
`nullptr` DeclContext for an erroneous module
under `-experimental-allow-module-with-compiler-errors`.
Additionally, tweak the insertion location such
that we insert at the start of any attributes
present.
rdar://97267326
7856f2d83d only partially skipped writing
out destructors when they were invalid, ie. it skipped writing the decl
itself but not the records common to all decls. This would cause any
records on the destructor to be applied on the next serialized decl.
Make sure to skip serializing anything to do with the destructor when
it's invalid and does not have a class context.
Allowing errors is attempting to create a module regardless of any
potential failures, so do our best to continue rather than crash (even
if a corresponding build would then crash).
Rather than outputting diagnostics and to stderr, output all the extra
information added when deserialization fatally fails to the pretty stack
trace instead. Since the pretty stack trace is added to crash logs, this
should avoid the dance of requesting the compiler output
- Moves the previous "**** DESERIALIZATION FAILURE ..." output to the
last pretty stack trace line
- Removes the module and compiler version notes added to the fatal
diagnostic
- Adds a new effective compiler version line for all frontend failure.
Somewhat duplicates the line from the driver, but adds in the
effective version
- Adds a new line for the full misc version of the module that failed.
May double up with previous "While reading from ..." lines that are
added in various deserialization methods, but better to have it
twice than not at all
It's currently not obvious in crash reports whether compiling with
errors is enabled or not. Since this option can make previously
"impossible" paths now possible, add a message to both the pretty stack
output and fatal deserialization diagnostics to point out that it is
enabled.
A normal compilation would error before merging modules when there are
MissingMemberDecls, so the missing member case is unreachable. That's
not true when allowing errors though, where we continue regardless. Skip
the missing member instead of crashing.
Resolves rdar://76365694.
There was a change in the diagnostics that isn't in 5.5. The error
message itself doesn't matter too much here, just that there was one.
Simplify the CHECK clause so that it will pass in main and 5.5 to avoid
future conflicts.
When compiling with allow errors, it's possible to have invalid
inherited types - both null and ErrorType.
Cleaned up the tests a little - moved the majority of
Frontend/allow-errors.swift into separate files in
Serialization/AllowErrors and use split_file.py instead of #defines.
Resolves rdar://78048470
