Provide more information in ObjCReason for the case where we are
@objc because we are a witness to an @objc requirement, by carrying the
@objc requirement in the reason. Use this to eliminate the type checker
parameter from describeObjCReason().
Rather than deferring to the heavyweight validateDeclForNameLookup()
to perform the “get overridden decls” operation, perform a much more
minimal validation that only looks for exact matches when the ‘override’
modifier is present.
The more-extensive checking (e.g., that one didn’t forget an override
modifier) is only performed as part of the “full” type checking of
a declaration, which will typically only be done within the same source
file as the declaration. The intent here is to reduce the amount of
work performed to check overrides cross-file, and limit the dependencies
of the “get overridden decls” operation.
Several kinds of declarations can override other declarations, but the
computation and storage for these “overridden” declarations was scattered in
at least 3 different places, with different resolution paths. Pull them
all together into two bits of LazySemanticInfo in ValueDecl (“have we computed
overrides?” and “are there any overrides?”), with a side table for the
actual list of overrides.
One side effect here is that the AST can now represent multiple overridden
declarations, although only associated type declarations track this
information.
Start using LazyResolver::resolveOverriddenDecl() more consistently, unifying
it with the separate path we had for associated type overrides. All of this
is staging for a move to the request-evaluator for overridden declaration
computation.
Previously, some PBDs weren't being marked implicit even though the associated vars were implicit. PatternBindingDecl::createImplicit will be even nicer when we start parsing the location of the equals token.
The storage kind has been replaced with three separate "impl kinds",
one for each of the basic access kinds (read, write, and read/write).
This makes it far easier to mix-and-match implementations of different
accessors, as well as subtleties like implementing both a setter
and an independent read/write operation.
AccessStrategy has become a bit more explicit about how exactly the
access should be implemented. For example, the accessor-based kinds
now carry the exact accessor intended to be used. Also, I've shifted
responsibilities slightly between AccessStrategy and AccessSemantics
so that AccessSemantics::Ordinary can be used except in the sorts of
semantic-bypasses that accessor synthesis wants. This requires
knowing the correct DC of the access when computing the access strategy;
the upshot is that SILGenFunction now needs a DC.
Accessor synthesis has been reworked so that only the declarations are
built immediately; body synthesis can be safely delayed out of the main
decl-checking path. This caused a large number of ramifications,
especially for lazy properties, and greatly inflated the size of this
patch. That is... really regrettable. The impetus for changing this
was necessity: I needed to rework accessor synthesis to end its reliance
on distinctions like Stored vs. StoredWithTrivialAccessors, and those
fixes were exposing serious re-entrancy problems, and fixing that... well.
Breaking the fixes apart at this point would be a serious endeavor.
Introduce some metaprogramming of accessors and generally prepare
for storing less-structured accessor lists.
NFC except for a change to the serialization format.
With batch mode, other files may have forced lazy properties to get finalized before the implicit constructor is formed. Avoid the order dependency by making the behavior stable regardless of the type-checking phase of lazy declarations. Fixes rdar://problem/40903186.
Inherited designated initializers got the same availability
as the corresponding initialier in the superclass.
However if the superclass was more available than the subclass,
we would generate a diagnostic that a member cannot be more
available than its containing type.
This diagnostic had an unknown source location, since the
location was for a synthesized declaration, causing confusion.
Fixes <https://bugs.swift.org/browse/SR-7881> aka
<rdar://problem/40853731>.
Rather than ASTContext-allocating ConcreteDeclRef’s storage when there is a
non-empty substitution map, put the SubstitutionMap directly in the
ConcreteDeclRef. Simplify the various interfaces along the way.
Replace two prominent uses of SubstitutionList, in ConcreteDeclRef and
Witness, with SubstitutionMap. Deal with the myriad places where we
now have substitution maps and need substitution lists (or vice versa)
caused by this change.
Overall, removes ~50 explicit uses of SubstitutionList (of ~400).
If the base class initializer was '@inlinable public' and the
derived class was '@usableFromInline internal', the synthesized
initializer was '@inlinable internal', which was an error.
The DeclChecker had three possible states:
- IsFirstPass true, IsSecondPass false. This is the 'first pass' for
declarations that appear at the top-level, or are nested inside
top-level types.
- IsFirstPass false, IsSecondPass true. This is the 'second pass' for
declarations that appear at the top-level, or are nested inside
top-level types.
- IsFirstPass false, IsSecondPass false. This was used for (some)
local declarations.
This is unnecessarily confusing. We can eliminate the third state
by calling typeCheckDecl() twice in a few places. This allows
IsSecondPass to be removed entirely since it's now always equal to
!IsFirstPass.
The importer would only synthesize its own memberwise init if the
declaration had a zero initializer, which supresses Sema's automatic
synthesis. Trouble is we cannot rely on Sema synthesizing the
memberwise init in case we deserialize declarations after type
checking.
Some other changes caught this with an assert, so fix the importer
to do the right thing here.
We don't want @_fixed_layout to apply to static properties, or you
get fun consequences like not being able to change static stored
properties defined in extensions of imported types to computed.
This has three principal advantages:
- It gives some additional type-safety when working
with known accessors.
- It makes it significantly easier to test whether a declaration
is an accessor and encourages the use of a common idiom.
- It saves a small amount of memory in both FuncDecl and its
serialized form.
Follow-up for 7c707ce97c. Without this, the declaration would be
accepted, but any uses of the overridden property would be treated as
ambiguous because the property wouldn't really be marked as an
override.
rdar://problem/35900345
Rather than storing contextual types in the type witnesses and associated
conformances of NormalProtocolConformance, store only interface types.
@huonw did most of the work here, and @DougGregor patched things up to
complete the change.
When overriding storage with a forced static dispatch materializeForSet,
the override's materializeForSet should not override the base
materializeForSet.
This is the case where a dynamic property witnesses a protocol
requirement, and Sema synthesizes a materializeForSet for it.
In this case, the synthesized materializeForSet dynamically dispatches
to the dynamic property's getter and setter, and the protocol witness
thunk directly calls the synthesized materializeForSet.
The subclass only needs to override the getter and setter in this case,
since the base class's materializeForSet will already do the right
thing.
In fact, marking it as an override exposes a problem where we cannot
serialize an xref to an imported property's materializeForSet, since
it was not created by the importer.
