- Add a flag to the serialized module (IsEmbeddedSwiftModule)
- Check on import that the mode matches (don't allow importing non-embedded module in embedded mode and vice versa)
- Drop TBD support, it's not expected to work in embedded Swift for now
- Drop auto-linking backdeploy libraries, it's not expected to backdeploy embedded Swift for now
- Drop prespecializations, not expected to work in embedded Swift for now
- Use CMO to serialize everything when emitting an embedded Swift module
- Change SILLinker to deserialize/import everything when importing an embedded Swift module
- Add an IR test for importing modules
- Add a deserialization validation test
For property declarations, the `@_spi` attribute is attached to the VarDecl AST
node, rather than the PatternBindingDecl AST node, so the `isSPI()` query
should take this into account. Failing to do so caused the availability checker
to erroneously require that `@_spi` properties of types in `-library-level api`
libraries have availability annotations.
Resolves rdar://113587321.
Within one module, SPI decls are always visible. Conceptually we want
the same behavior for `@_private` imports where the client pretends it's
part of the same module.
rdar://81240984
llvm::SmallSetVector changed semantics
(https://reviews.llvm.org/D152497) resulting in build failures in Swift.
The old semantics allowed usage of types that did not have an
`operator==` because `SmallDenseSet` uses `DenseSetInfo<T>::isEqual` to
determine equality. The new implementation switched to using
`std::find`, which internally uses `operator==`. This type is used
pretty frequently with `swift::Type`, which intentionally deletes
`operator==` as it is not the canonical type and therefore cannot be
compared in normal circumstances.
This patch adds a new type-alias to the Swift namespace that provides
the old semantic behavior for `SmallSetVector`. I've also gone through
and replaced usages of `llvm::SmallSetVector` with the
`Swift::SmallSetVector` in places where we're storing a type that
doesn't implement or explicitly deletes `operator==`. The changes to
`llvm::SmallSetVector` should improve compile-time performance, so I
left the `llvm::SmallSetVector` where possible.
Eliminate the error message
error: global freestanding macros not yet supported in script mode
by implementing name lookup, type checking, and code emission for
freestanding macros. The key problem here is that, in script mode,
it is ambiguous whether a use of a freestanding macro is an expression
or a declaration. We parse as an expression (as we do within a
function body), which then gets wrapped in a top-level code
declaration.
Teach various parts of the compiler to look through a top-level code
declaration wrapping a macro expansion expression that is for a
declaration or code-item macro, e.g., by recording these for global
name lookup and treating their expansions as "auxiliary" declarations.
Fixes rdar://109699501.
stated in the original source.
If an extension macro can introduce protocol conformances, macro expansion
will check which of those protocols already have a stated conformance in the
original source. The protocols that don't will be passed as arguments to
extension macro expansion, indicating to the macro that it should only add
conformances to those protocols.
Reformatting everything now that we have `llvm` namespaces. I've
separated this from the main commit to help manage merge-conflicts and
for making it a bit easier to read the mega-patch.
This is phase-1 of switching from llvm::Optional to std::optional in the
next rebranch. llvm::Optional was removed from upstream LLVM, so we need
to migrate off rather soon. On Darwin, std::optional, and llvm::Optional
have the same layout, so we don't need to be as concerned about ABI
beyond the name mangling. `llvm::Optional` is only returned from one
function in
```
getStandardTypeSubst(StringRef TypeName,
bool allowConcurrencyManglings);
```
It's the return value, so it should not impact the mangling of the
function, and the layout is the same as `std::optional`, so it should be
mostly okay. This function doesn't appear to have users, and the ABI was
already broken 2 years ago for concurrency and no one seemed to notice
so this should be "okay".
I'm doing the migration incrementally so that folks working on main can
cherry-pick back to the release/5.9 branch. Once 5.9 is done and locked
away, then we can go through and finish the replacement. Since `None`
and `Optional` show up in contexts where they are not `llvm::None` and
`llvm::Optional`, I'm preparing the work now by going through and
removing the namespace unwrapping and making the `llvm` namespace
explicit. This should make it fairly mechanical to go through and
replace llvm::Optional with std::optional, and llvm::None with
std::nullopt. It's also a change that can be brought onto the
release/5.9 with minimal impact. This should be an NFC change.
Macro-generated extensions are hoisted to file scope, because extensions are
not valid in nested scopes. Callers of 'visitAuxiliaryDecls' assume that the
auxiliary decls are in the same decl context as the original, which is clearly
not the case for extensions, and it leads to issues like visiting extension at
the wrong time during SILGen. The extensions are already added to the top-level
decls, so we don't need to visit them as auxiliary decls, and we can type-check
macro-expanded decls at the end of visitation in TypeCheckDeclPrimary.
* Don't invalidate the lookup cache in 'getOrCreateSynthesizedFile()'
Adding a synthesized file itself doesn't introduce any decls. Instead,
we should invalidate the right after the actual declrations are added
in the file
* Remove 'SourceLookupCache::invalidate()' method. It was just used
right before the destruction. It was just not necessary
* Include auxiliary decls in 'SourceLookupCache::lookupVisibleDecls()'
Previously, global symbol completion didn't include decls synthesized
by peer macros or freestanding decl macros
* Include "auxiliary decls" in visible member lookup, and visible local
decl lookup
* Hide macro unique names
rdar://110535113
`lib/swift/host` contains modules/libraries that are built by the host
compiler. Their `.swiftmodule` will never be able to be read, ignore
them entirely.
The module-scope lookup tables use the same code for adding
module-scope declarations as for adding member operators, which are
found via "global" operator lookup. This requires us to expand macros
that can produce members of types, which violates the outside-in
expansion rule described in the proposals.
Stop recording member-producing macros, whether they are peer macros
applied to member declarations or are freestanding declaration macros
within a member context. This re-establishes the outside-in expansion
rule. It also means that member operators introduced by macro
expansion won't be found by global operator lookup, which is a
(necessary) semantic change.
Well, this is fun. Due to the use of the module-scope lookup table to
find operators, we need to carefully filter out any macro-introduced
declarations that *aren't* operators when forming the module-scope
lookup table. Otherwise, we can find macro-introduced static entities
within types... from completely unrelated scopes.
Fixes rdar://109219036.
API development sometimes requires a redesign while supporting early
adopters. Currently this is done by adding @_spi(name) to the API but
that requires adding the attribute in import statements as well, causing
manual overhead of adding and then removing when the redesign is done.
This PR introduces a special spi group name '_' and allows an implicit
spi import of a module containing API attributed with '@_spi(_)'
Resolves rdar://109797632
Avoid parsing the syntax tree up-front, and instead
only parse it when required, which happens when either:
1. ASTGen parsing is enabled (currently disabled
by default)
2. Round trip checking is enabled for a primary
file (enabled by default in a debug build,
except when dep scanning or doing an IDE
operation)
3. We need to evaluate a macro in that file
This change therefore means that we now no longer
need to parse the syntax tree for secondary files
by default unless we specifically need to evaluate
a macro in them (e.g if we need to lookup a member
on a decl with an attached macro). And the same
for primaries in release builds.
rdar://109283847
These metatypes are a gateway to more incorrect
uses of these noncopyable values because we don't
yet have the corresponding runtime support yet.
The other use cases of using metatypes of
noncopyable types in generics is not high enough to
warrant people using them yet.
resolves rdar://106452518
Expression macros ascribed to non-private contexts need private
discriminators so they don't conflict with other uses of the same macro
in other source files.
Thank you, Richard, for noticing this omission!
It doesn't seem like there's any client that's
actually taking advantage of setting it to `false`,
and its default value of `false` is more likely
than not going to cause clients to accidentally
miss comments that they may want. In fact, this
was exactly the case for code completion's brief
field. Finally, the parameter wasn't even
consistently applied, as we would attempt to
deserialize swiftdoc comments even if it were
`false`.
This method was misleading. The majority of callers (all but one!) don't want
to unconditionally treat all locations in any macro expansion buffer the
same way, because the code also must handle nested macro expansions. There
is one part of SourceKit (and possibly others) that really do want to ignore
all macro expansions, but those can be handled within SourceKit / IDE code,
because I don't believe this utility is useful in the frontend.
Teach `getTopLevelDeclsWithAuxiliaryDecls` not to provide extension
declarations, because those are covered by the synthesized file, which
all clients need to walk anyway. Without this, we end up asserting in
TBD generation about duplicate symbol visitation.
Encountered while investigating rdar://108056018.
Whenever we perform a name lookup, we need to make sure to expand
all macros that use `names: arbitrary`, because of course they can
produce... wait for it... arbitrary names, and we don't know what they
are until we instatiate them.
When we look into expanded macros to find declarations for module-scope
name lookup, make sure that we only return declarations whose name
matches the requested name.
We were effectively spamming the results with unrelated names, and
most clients trust the results of this lookup (vs. filtering them
further), so they would get confused. The particular failure for this
test is that type reconstruction would fail because we were looking
for one unique name, but we found many non-matching names, and type
reconstruction bails out rather than try to filter further.
The macro name resolution in the source lookup cache was only looking at
macros in the current module, meaning that any names introduced by peer
or declaration macros declared in one module but used in another would
not be found by name lookup.
Switch the source lookup cache over to using the same
`forEachPotentialResolvedMacro` API that is used by lookup within
types, so we have consistent name-lookup-level macro resolution in both
places.
... except that would be horribly cyclic, of course, so introduce name
lookup flags to ignore top-level declarations introduced by macro
expansions. This is semantically correct because macro expansions are
not allowed to introduce new macros anyway, because that would have
been a terrible idea.
Fixes rdar://107321469. Peer and declaration macros at module scope
should work a whole lot better now.
