https://github.com/swiftlang/swift/pull/72659 turned out to have some
source compatibility fallout that we need to fix. Instead of introducing
yet another brittle compatibility hack, stop emitting errors about a
missing `any` altogether until a future language mode.
Besides resolving the compatibility issue, this will encourage
developers to adopt any sooner and grant us ample time to gracefully
address any remaining bugs before the source compatibility burden
resurfaces.
A subsequent commit adds a diagnostic group that will allow users to
escalate these warnings to errors with `-Werror ExistentialAny`.
Some editors use diagnostics from SourceKit to replace build issues. This causes issues if the diagnostics from SourceKit are formatted differently than the build issues. Make sure they are rendered the same way, removing most uses of `DiagnosticsEditorMode`.
To do so, always emit the `add stubs for conformance` note (which previously was only emitted in editor mode) and remove all `; add <something>` suffixes from notes that state which requirements are missing.
rdar://129283608
The original bug was a crash-on-invalid with a missing '}', but it
actually exposed a bug with nested protocols (SE-0404) and another
long-time bug.
- Whatever we do, we should skip this for protocols because their 'Self'
parameter is not bound from context.
- getTrailingWhereClause() is not the right proxy for "has a generic
signature different than its parent", in particular it doesn't
round-trip through serialization. Instead, just compare generic
signatures for pointer equality in the early return check.
The second change is source-breaking because it was possible to
write a nested type with a `where` clause and use it contradicting
its requirements across a module boundary.
Fixes rdar://113103854.
The previous warning "must be explicitly marked as 'any'" isn't clear if
you don't already know the feature, this new phrasing should make it
clearer by including the correct spelling of the type as an example.
Fixes rdar://90384448
If you have a class with an (invalid) protocol nested inside:
class OuterGenericClass<T, U> {
protocol InnerProtocol : OuterGenericClass {}
}
The reference to 'OuterGenericClass' in the protocol's inheritance
clause is not "in context", and should not have inferred generic
arguments <T, U>, because the protocol does not inherit the generic
arguments of the outer scope.
Previously, this would trigger an assertion in the requirement
machine, because the generic parameter τ_0_1 is not a valid generic
parameter in the protocol's generic signature.
The following regression test added for this feature is not passing:
Swift(linux-x86_64) :: decl/protocol/protocols_with_self_or_assoc_reqs_executable.swift
with a compiler crash happening during SILFunctionTransform "Devirtualizer".
Reverting to unblock CI.
This reverts commit f96057e260, reversing
changes made to 3fc18f3603.
There were two problems here:
- isUnsupportedMemberTypeReference() checked if the immediate parent type was
an unbound generic type, but did not check for the parent of the parent, etc.
- The caller of isUnsupportedMemberTypeReference() had to re-check the
various invalid conditions to determine what diagnostic to emit, and if
none of those conditions matched it would just return an ErrorType without
emitting a diagnostic.
Fix both of these by having isUnsupportedMemberTypeReference() return an
enum indicating what went wrong, and handle more cases.
Fixes <rdar://problem/67292528>.
Although such functionality is not yet supported we have to
mirror AST lookup and add such members into results, otherwise
there is a risk of inner/outer results mismatch.
Witness matching is a source of a lot of ad-hoc cycles, and mixes the
logic that performs resolution, caching, validation, and cycle detection into one
place. To make matters worse, some checkers kick off other checks in
order to cache work for further declarations, and access an internal
cache on their subject conformance for many requirements at once, or
sometimes just one requirement.
None of this fits into the request evaluator's central view of the
caching. This is further evidenced by the fact that if you attempt to
move the caching step into the evaluator, it overcaches the same
witness and trips asserts.
As a start, define requests for the resolution steps, and flush some
hacks around forcing witness resolution. The caching logic is mostly
untouched (the requests don't actually cache anything), but some cycle
breaking is now handled in the evaluator itself. Once witness matching
has been refactored to cache with the evaluator, all of these hacks can
go away.
My urge to destroy the LazyResolver outweighs the compromises here.
In preparation for checkEnumRawValues being turned into a request, move the common diagnostics to the decl checker so they're always emitted at the right time.
Computing the generic signature changes the way that cycles appear in the compiler. For now, break these cycles. We should investigate each place where hasComputedGenericSignature() is used in service of breaking cycles. See rdar://55263708
Existential types don't support nested types so let's not
allow the lookup to make such access, with only exception
for typealias or associated types without type parameters,
they are currently allowed.
Resolves: rdar://problem/38505436
(...is constrained to be a subtype of another)
Previously the compiler would just mark the entry in the inheritance
clause invalid and move on without emitting any errors; in certain
circumstances in no-asserts builds this could actually lead to
everything working "correctly" if all conforming types happened to
pick the same concrete type for both associated types. In Swift 4 this
can actually be enforced with a same-type requirement, which will
guarantee that the two associated types are the same even in generic
contexts.
This fix avoids assertions and crashes, but the diagnostic is still
incorrect, and in the simple case of the inheritance clause it's
redundant. Doing something better and possibly even downgrading it to
a warning in Swift 3 mode is tracked by rdar://problem/32409449.
Initial patch by Slava, fixed up by me.
This was added at some point with the associated type where
clause work, but it appears to be unnecessary, because all
the tests passed with this removed.
It also introduced a source compatibility issue where we
stopped accepting typealiases to protocol compositions in
protocol inheritance clauses.
Add a test for this case too, since it wasn't tested before.
Fixes <https://bugs.swift.org/browse/SR-4855>.
When performing a name lookup from inside of a protocol
or extension, skip directly to the source file context
when we are done visiting the protocol or extension.
Otherwise, if we have invalid code where the protocol
or extension is nested inside another type, we might
find a member whose type contains generic parameters
of the outer type; these parameters will not resolve,
since we do not model protocols or extensions nested
inside generic contexts (yet?).
This supercedes an earlier workaround for a similar
issue; the new workaround fixes more crashes.
This is needed to avoid crasher regressions with an
upcoming patch.
Previously we prohibited unbound generics in the underlying
type of a typealias, but due to an oversight the check was
not performed when resolving a nested type.
So this worked:
struct Outer { struct Inner<T> {} }
typealias OuterInner = Outer.Inner
let _: OuterInner<Int> = Outer.Inner<Int>()
However it was easy to cause a crash this way by stating an
unbound generic type where one was not expected. Also,
unqualified types in a typealias did not get this treatment,
so the following did not work:
typealias MyOptional = Optional
Formalize the old behavior by allowing unbound generic types
in the underlying type of a typealias, while otherwise
prohibiting unbound references to nested types.
There's no need to walk up from a function or type context -- if
no generic signature (or environment) is set, the parent won't have
one, either, and if we're in the middle of validating the child
context, using the parent's signature or environment to resolve
dependent types is just wrong.
Consider this code:
struct A<T> {
struct B {}
struct C<U> {}
}
Previously:
- getDeclaredType() of 'A.B' would give 'A<T>.B'
- getDeclaredTypeInContext() of 'A.B' would give 'A<T>.B'
- getDeclaredType() of 'A.C' would give 'A<T>.C'
- getDeclaredTypeInContext() of 'A.C' would give 'A<T>.C<U>'
This was causing problems for nested generics. Now, with this change,
- getDeclaredType() of 'A.B' gives 'A.B' (*)
- getDeclaredTypeInContext() of 'A.B' gives 'A<T>.B'
- getDeclaredType() of 'A.C' gives 'A.C' (*)
- getDeclaredTypeInContext() of 'A.C' gives 'A<T>.C<U>'
(Differences marked with (*)).
Also, this change makes these accessors fully lazy. Previously,
only getDeclaredTypeInContext() and getDeclaredIterfaceType()
were lazy, whereas getDeclaredType() was built from validateDecl().
Fix a few spots where the return value wasn't being checked
properly.
These functions return ErrorType if a circularity was detected via
the generic parameter list, or if the extension did not resolve.
They return Type() if the extension cannot be resolved *yet*.
This is pretty subtle, and I'll need to do another pass over
callers of these functions at some point. Many of them should be
moved over to use getSelfInContext(), getSelfOfContext() and
getSelfInterfaceType() instead.
Finally, this patch consolidates logic for diagnosting invalid
nesting of types.
The parser had some code for protocols in bad places and bad things
inside protocols, and Sema had several different bail-outs for
bad things in protocols, nested generic types, and stuff nested
inside protocol extensions.
Combine all of these into a single set of checks in Sema. Note
that we no longer give up early if we find invalid nesting.
Leaving decls unvalidated and un-type-checked only leads to
further problems. Now that all the preliminary crap has been
fixed, we can go ahead and start validating these funny nested
decls, actually fixing some crashers in the process.
