* Add @_used and @_section attributes for global variables and top-level functions
This adds:
- @_used attribute that flags as a global variable or a top-level function as
"do not dead-strip" via llvm.used, roughly the equivalent of
__attribute__((used)) in C/C++.
- @_section("...") attribute that places a global variable or a top-level
function into a section with that name, roughly the equivalent of
__attribute__((section("..."))) in C/C++.
SE-390 concluded with choosing the keyword discard rather than forget for
the statement that disables the deinit of a noncopyable type. This commit
adds parsing support for `discard self` and adds a deprecation warning for
`_forget self`.
rdar://108859077
We shouldn't diagnose adding @_predatesConcurrency without considering the context because the attribute itself is for ABI-preserving purposes.
rdar://90738915
Currently, this is staged in as `_forget`,
as part of SE-390. It can only be used on
`self` for a move-only type within a consuming
method or accessor. There are other rules, see
Sema for the details.
A `forget self` really just consumes self and
performs memberwise destruction of its data.
Thus, the current expansion of this statement
just reuses what we inject into the end of a
deinit.
Parsing of `forget` is "contextual".
By contextual I mean that we do lookahead to
the next token and see if it's identifier-like.
If so, then we parse it as the `forget` statement.
Otherwise, we parse it as though "forget" is an
identifier as part of some expression.
This way, we won't introduce a source break for
people who wrote code that calls a forget
function.
This should make it seamless to change it from
`_forget` to `forget` in the future.
resolves rdar://105795731
The attached and freestanding macro attributes use the same parsing
logic and representation, so generalize the "attached" attribute into
a more general "macro role" attribute.
Describe attached macros with the `@attached` attribute, providing the
macro role and affected names as arguments to the macro. The form of
this macro will remain the same as it gains other kinds of attached
macro roles beyond "accessor".
Remove the "accessors" role from `@declaration`, which will be going
away.
Add support for freestanding declaration macros.
- Parse `@declaration` attribute.
- Type check and expand `MacroExpansionDecl`.
Known issues:
- Generic macros are not yet handled.
- Expansion does not work when the parent decl context is `BraceStmt`. Need to parse freestanding declaration macro expansions in `BraceStmt` as `MacroExpansionDecl`, and add expanded decls to name lookup.
This allows us to evolve e.g. token kinds and how attributes are modelled independently between SwiftSyntax and the compiler. It also makes it easier to e.g. add an attribute because you don’t need to create PRs for two repositories.
Make libSyntax depend on swift-syntax: the new home for all
of this infrastructure. This greatly simplifies the addition and
amending of syntax nodes as only the swift-syntax paired with a
swift checkout will need to be changed. This is in contrast to
the existing build flow where a paired PR to both repos must be
made to change anything here.
Note that a paired PR may still be required if the legacy parser
needs to be adjusted in response to syntax nodes changing, but I
anticipate this to be a much more infrequent event now that
the C++ end of libSyntax is deprecated.
SequenceExprSyntax should have odd number elements. Previously 'a as b'
was parsed like:
```
(sequence_expr
(identifier_expr "a"),
(as_expr
'as'
(typeidentifier "b")))
```
So it had even number elements. Now it's parsed
```
(sequence_expr
(identifier_expr "a"),
(unresolved_as_expr 'as')
(type_expr
(typeidentifier "b")))
```
Strange as it sounds, we parse
@foo public
in item position as a decl that has a hanging attribute and access
control modifier. We need to be able to stick... something in the
tree here so we don't just drop these tokens on the floor.
This represents labeled statements as an explicit kind of statement and removes the Labeled trait. Any kind of statement is allowed to be labeled in the tree, but we specifically diagnose the syntax elements that aren't allowed to have labels. This homogenizes the way clients deal with statement labels and also makes parser recovery quite a bit easier in the case where we have a label but no actual statement following it.
When the source code is invalid, this allows us to represent tokens that could not be used to form a valid syntax tree with more fidelity.
This commit does not start using GarbageNodes yet, it just sets everything up for them.
In the future, we only want to attach a source presence to tokens, not to nodes. All concreate nodes which are missing can be represented by creating the node and marking all tokens as missing, but if a syntax node carries a child, that has a base kind (like `Decl`), we can’t decide which concrete node to instantiate. Introduce `MissingDecl` etc. node for this purpose.
rdar://97908258
rdar://97775360
