Commit Graph

9699 Commits

Author SHA1 Message Date
Pavel Yaskevich
2ad754d470 Merge pull request #20152 from xedin/add-autoclosure-flag-to-param-decl
[AST/Sema] Associate `@autoclosure` flag with parameter declaration
2018-11-10 14:53:31 -08:00
Pavel Yaskevich
c0c605295d Merge pull request #20495 from theblixguy/fix/SR-9213
[Sema] [CSDiag] Fixes incorrect diagnostic when using keypath with tuples
2018-11-10 14:03:04 -08:00
Pavel Yaskevich
ed84f18686 [AST] NFC: Remove a FIXME related to ParamDecl::IsAutoClosure field 2018-11-10 11:59:29 -08:00
Pavel Yaskevich
bce1ba5f37 [AST] Remove @autoclosure flag from function type ExtInfo 2018-11-10 11:59:29 -08:00
Pavel Yaskevich
a1df238d47 Associate @autoclosure only with parameter types
Mangled implicit closures shouldn't have `@autoclosure` flag
associated with them, because their type is just a regular
function type.
2018-11-10 11:59:28 -08:00
Pavel Yaskevich
d9504ef919 [TypeChecker] Add typeCheckParameterDefault method
`typeCheckParameterDefault` is used to type-check default value
expression associated with a given parameter. It makes it possible
to look-through `@autoclosure` function to use its result as
contextual type, and later build implicit autoclosure expression
if needed.
2018-11-10 11:59:28 -08:00
Pavel Yaskevich
53417f4aba [AST] Propagate @autoclosure flag to parameter decl and type flags 2018-11-10 11:59:28 -08:00
Pavel Yaskevich
4c1885b51b [AST] Register @autoclosure as parameter decl flag
It should become a declaration attribute instead,
but it can't be done at the moment because `@autoclosure`
is a type attribute.
2018-11-10 11:58:49 -08:00
John McCall
44e0f44040 Merge pull request #20493 from rjmccall/keypath-compiler-abi
Change the compiler ABI of keypaths.
2018-11-10 14:18:44 -05:00
Suyash Srijan
ba227073ce Removes 'swift' from diagnostic name 2018-11-10 18:44:02 +00:00
Suyash Srijan
15ebc07421 Fixes incorrect diagnostic when using keypath with tuples 2018-11-10 17:19:09 +00:00
John McCall
3e5165d1ab Change the compiler ABI of keypaths.
Previously, the stdlib provided:

- getters for AnyKeyPath and PartialKeyPath, which have remained;

- a getter for KeyPath, which still exists alongside a new read
  coroutine; and

- a pair of owned mutable addressors that provided modify-like behavior
  for WritableKeyPath and ReferenceWritableKeyPath, which have been
  replaced with modify coroutines and augmented with dedicated setters.

SILGen then uses the most efficient accessor available for the access
it's been asked to do: for example, if it's been asked to produce a
borrowed r-value, it uses the read accessor.

Providing a broad spectrum of accessor functions here seems acceptable
because the code-size hit is fixed-size: we don't need to generate
extra code per storage declaration to support more alternatives for
key paths.

Note that this is just the compiler ABI; the implementation is still
basically what it was.  That means the implementation of the setters
and the read accessor is pretty far from optimal.  But we can improve
the implementation later; we can't improve the ABI.

The coroutine accessors have to be implemented in C++ and used via
hand-rolled declarations in SILGen because it's not currently possible
to declare independent coroutine accessors in Swift.
2018-11-10 02:08:04 -05:00
Slava Pestov
c7338d06ca AST: Remove owning addressors 2018-11-09 20:49:44 -05:00
Arnold Schwaighofer
fb7b223ba2 Add support to modify chaining behavior of dynamic replacements
Default to not chain dynamic replacements: Only one replacement and the
original implementation are active.
2018-11-09 13:17:09 -08:00
Slava Pestov
a1ae75b2ba Merge pull request #20455 from slavapestov/fix-misc-regressions
Fix a handful of regressions
2018-11-09 13:20:44 -05:00
Dan Zheng
2a4e1b83fd Implement @dynamicCallable. (#20305)
* Implement dynamically callable types (`@dynamicCallable`).

- Implement dynamically callable types as proposed in SE-0216.
  - Dynamic calls are resolved based on call-site syntax.
  - Use the `withArguments:` method if it's defined and there are no
    keyword arguments.
  - Otherwise, use the `withKeywordArguments:` method.
- Support multiple `dynamicallyCall` methods.
  - This enables two scenarios:
    - Overloaded `dynamicallyCall` methods on a single
      `@dynamicCallable` type.
    - Multiple `dynamicallyCall` methods from a `@dynamicCallable`
      superclass or from `@dynamicCallable` protocols.
  - Add `DynamicCallableApplicableFunction` constraint. This, used with
    an overload set, is necessary to support multiple `dynamicallyCall`
    methods.
2018-11-09 09:49:14 -08:00
John McCall
e741147d70 Merge pull request #20459 from rjmccall/key-path-lvalue-emission
Always use the l-value logic for emitting key path applications
2018-11-09 12:08:27 -05:00
Arnold Schwaighofer
aae2c3e870 Merge pull request #20440 from aschwaighofer/proper_error_message
Replace assert by proper error message
2018-11-09 08:58:17 -08:00
Arnold Schwaighofer
c83f63855b Merge pull request #20428 from aschwaighofer/private_imports
Add @_private(from: "SourceFile.swift") imports
2018-11-09 07:18:25 -08:00
John McCall
c0285a744c Always use the l-value logic for emitting key path applications.
Not NFC because it also fixes an evaluation order bug (and reorders
some less-important stuff): the key-path expression needs to be
evaluated immediately during formal evaluation and cannot be delayed
until start-of-access.
2018-11-09 02:42:17 -05:00
Azoy
92a12c0f0a improve diagnostic for inferred uninhabited 2018-11-09 00:21:15 -06:00
Azoy
18d2cc5b01 Uninhabited downgrade to warning 2018-11-09 00:21:14 -06:00
Slava Pestov
f49e0ba330 AST: Remove unused DeclContext and GenericEnvironment from mangling 2018-11-09 00:44:29 -05:00
swift-ci
4fb4cb058d Merge pull request #19618 from apple/marcrasi-static-assert 2018-11-08 21:33:35 -08:00
Arnold Schwaighofer
6cc3d377ff Refactor part2 2018-11-08 18:16:17 -08:00
Arnold Schwaighofer
9feb75806a Refactor 2018-11-08 17:50:25 -08:00
Jordan Rose
e7dd1c17b2 Merge pull request #20384 from jrose-apple/witless
Require @usableFromInline on associated type witnesses
2018-11-08 16:36:48 -08:00
Jordan Rose
d3edf7c9ff [SILGen] Always serialize witness tables in non-resilient modules (#20390)
...even if the conforming nominal type is resilient. It's the owner
of the conformance whose resilience matters.

I also factored this part out into a separate check at the AST level
so we can tweak it, and also so I can use it to (slightly) speed up
compiling a resilient swiftinterface.
2018-11-08 15:34:00 -08:00
Arnold Schwaighofer
bfb66422f3 Replace assert by proper error message 2018-11-08 13:30:18 -08:00
Arnold Schwaighofer
e4f4dfcf84 Address feedback 2018-11-08 11:13:42 -08:00
Arnold Schwaighofer
963c64e3e7 Add @_private(from: "SourceFile.swift") imports
A module compiled with `-enable-private-imports` allows other modules to
import private declarations if the importing source file uses an
``@_private(from: "SourceFile.swift") import statement.

rdar://29318654
2018-11-08 08:00:47 -08:00
John McCall
1aeae263f1 Merge pull request #20403 from rjmccall/borrowed-attr-read-accessor
Add a @_borrowed attribute to force the use of an opaque read accessor
2018-11-08 00:13:50 -05:00
Marc Rasi
bf18697b4f parsing, typechecking, and SILGen for #assert
`#assert` is a new static assertion statement that will let us write
tests for the new constant evaluation infrastructure that we are working
on. `#assert` works by lowering to a `Builtin.poundAssert` SIL
instruction. The constant evaluation infrastructure will look for these
SIL instructions, const-evaluate their conditions, and emit errors if
the conditions are non-constant or false.

This commit implements parsing, typechecking and SILGen for `#assert`.
2018-11-07 16:34:17 -08:00
John McCall
e01573dfdb Add a @_borrowed attribute to force the use of an opaque read accessor.
This is strawman syntax pending a proposal; I pitched some ideas here:
  https://forums.swift.org/t/value-ownership-when-reading-from-a-storage-declaration/15076
2018-11-07 18:27:32 -05:00
Slava Pestov
564695d9bf AST: Remove DeclContext::isExtensionContext()
It's identical in every way to isa<ExtensionDecl>(), so
make code less confusing by only having one way of doing
that check.
2018-11-07 17:34:56 -05:00
Jordan Rose
f33bf67dc9 Require @usableFromInline on associated type witnesses
...when the protocol and the conforming type are not both public but
are both public-or-usableFromInline. It's possible to write inlinable
functions that depend on these types:

    public protocol HasAssoc {
      associatedtype Assoc
    }
    public func getAssoc<T: HasAssoc>(_: T) -> T.Assoc

    @usableFromInline struct Impl: HasAssoc {
      @usableFromInline typealias Assoc = Int
    }

    @inlinable func test() {
      let x: Int = getAssoc(Impl())
    }

rdar://problem/43824052
2018-11-06 17:00:46 -08:00
Arnold Schwaighofer
b102c7f6b4 Parser/Sema/SILGen changes for @_dynamicReplacement(for:)
Dynamic replacements are currently written in extensions as

extension ExtendedType {
  @_dynamicReplacement(for: replacedFun())
  func replacement() { }
}

The runtime implementation allows an implementation in the future where
dynamic replacements are gather in a scope and can be dynamically
enabled and disabled.

For example:

dynamic_extension_scope CollectionOfReplacements {
  extension ExtentedType {
    func replacedFun() {}
  }

  extension ExtentedType2 {
    func replacedFun() {}
  }
}

CollectionOfReplacements.enable()
CollectionOfReplacements.disable()
2018-11-06 09:58:36 -08:00
Arnold Schwaighofer
52c1903e54 Add SIL support for [dynamic_replacement_for: ] functions 2018-11-06 09:58:28 -08:00
Arnold Schwaighofer
c158106329 Allow dynamic without @objc in -swift-version 5
Dynamic functions will allow replacement of their implementation at
runtime.
2018-11-06 09:53:21 -08:00
Pavel Yaskevich
0e9e14e4a3 Merge pull request #20336 from xedin/rdar-45819956
[CSDiagnostics] Diagnose attempt to pass `inout` argument to a subscript
2018-11-05 19:48:52 -08:00
Pavel Yaskevich
37ba52c03d [CSDiagnostics] Diagnose attempt to pass inout argument to a subscript
Tweak `PreCheckExpression` to diagnose calls to subscript with
`&` arguments, which is not allowed.

Resolves: rdar://problem/45819956
2018-11-05 16:15:59 -08:00
Michael Ilseman
6636815568 Merge pull request #20315 from milseman/utf8string
[String] Use a UTF-8 representation for native strings
2018-11-05 14:55:02 -08:00
Michael Ilseman
95ef4bc3a8 [String] Emit literals as UTF-8 rather than UTF-16 2018-11-04 10:42:40 -08:00
Doug Gregor
1efe2bca21 [Mangling] Introduce mangling for protocol conformances.
Introduce complete mangling for references to protocol conformances:

* Mangle requirements of conditional conformances when present.
* Mangle conformance access paths for generic environment-dependent
  conformances.
* Abstract protocol conformance references so we can introduce
  symbolic references for them.
2018-11-02 22:35:34 -07:00
Brent Royal-Gordon
9bd1a26089 Implementation for SE-0228: Fix ExpressibleByStringInterpolation (#20214)
* [CodeCompletion] Restrict ancestor search to brace

This change allows ExprParentFinder to restrict certain searches for parents to just AST nodes within the nearest surrounding BraceStmt. In the string interpolation rework, BraceStmts can appear in new places in the AST; this keeps code completion from looking at irrelevant context.

NFC in this commit, but keeps code completion from crashing once TapExpr is introduced.

* Remove test relying on ExpressibleByStringInterpolation being deprecated

Since soon enough, it won’t be anymore.

* [AST] Introduce TapExpr

TapExpr allows a block of code to to be inserted between two expressions, accessing and potentially mutating the result of its subexpression before giving it to its parent expression. It’s roughly equivalent to this function:

  func _tap<T>(_ value: T, do body: (inout T) throws -> Void) rethrows -> T {
    var copy = value
    try body(&copy)
    return copy
  }

Except that it doesn’t use a closure, so no variables are captured and no call frame is (even notionally) added.

This commit does not include tests because nothing in it actually uses TapExpr yet. It will be used by string interpolation.

* SE-0228: Fix ExpressibleByStringInterpolation

This is the bulk of the implementation of the string interpolation rework. It includes a redesigned AST node, new parsing logic, new constraints and post-typechecking code generation, and new standard library types and members.

* [Sema] Rip out typeCheckExpressionShallow()

With new string interpolation in place, it is no longer used by anything in the compiler.

* [Sema] Diagnose invalid StringInterpolationProtocols

StringInterpolationProtocol informally requires conforming types to provide at least one method with the base name “appendInterpolation” with no (or a discardable) return value and visibility at least as broad as the conforming type’s. This change diagnoses an error when a conforming type does not have a method that meets those criteria.

* [Stdlib] Fix map(String.init) source break

Some users, including some in the source compatibility suite, accidentally used init(stringInterpolationSegment:) by writing code like `map(String.init)`. Now that these intializers have been removed, the remaining initializers often end up tying during overload resolution. This change adds several overloads of `String.init(describing:)` which will break these ties in cases where the compiler previously selected `String.init(stringInterpolationSegment:)`.

* [Sema] Make callWitness() take non-mutable arrays

It doesn’t actually need to mutate them.

* [Stdlib] Improve floating-point interpolation performance

This change avoids constructing a String when interpolating a Float, Double, or Float80. Instead, we write the characters to a fixed-size buffer and then append them directly to the string’s storage.

This seems to improve performance for all three types, but especially for Double and Float80, which cannot always fit into a small string when stringified.

* [NameLookup] Improve MemberLookupTable invalidation

In rare cases usually involving generated code, an overload added by an extension in the middle of a file would not be visible below it if the type had lazy members and the same base name had already been referenced above the extension. This change essentially dirties a type’s member lookup table whenever an extension is added to it, ensuring the entries in it will be updated.

This change also includes some debugging improvements for NameLookup.

* [SILOptimizer] XFAIL dead object removal failure

The DeadObjectRemoval pass in SILOptimizer does not currently remove reworked string interpolations as well as the old design because their effects cannot be described by @_effects(readonly). That causes a test failure on Linux. This change temporarily silences that test. The SILOptimizer issue has been filed as SR-9008.

* Confess string interpolation’s source stability sins

* [Parser] Parse empty interpolations

Previously, the parser had an odd asymmetry which caused the same function to accept foo(), but reject “\()”. This change fixes the issue.

Already tested by test/Parse/try.swift, which uses this construct in one of its throwing interpolation tests.

* [Sema] Fix batch-mode-only lazy var bug

The temporary variable used by string interpolation needs to be recontextualized when it’s inserted into a synthesized getter. Fixes a compilation failure in Alamofire.

I’ll probably follow up on this bug a bit more after merging.
2018-11-02 19:16:03 -07:00
Slava Pestov
4fb5ccf40e AST: Rename RemoteASTTypeBuilder to ASTBuilder 2018-11-02 18:19:40 -04:00
Slava Pestov
9c50867370 Move RemoteAST's RemoteASTTypeBuilder to lib/AST/ASTDemangler.cpp 2018-11-02 18:19:40 -04:00
John McCall
abdba1d3f4 Change the integer-literal type from Int2048 to IntLiteral.
Part of SR-290.
2018-10-31 23:14:58 -04:00
Slava Pestov
9499d1d346 Merge pull request #20173 from slavapestov/weak-linked
IRGen: Use @_weakLinked to test backward deployment of resilient protocols
2018-10-31 21:52:58 -04:00
Slava Pestov
57979d1f7c IRGen: Use @_weakLinked to test backward deployment of resilient protocols
Get the attribute working for more link entity kinds, which addresses
all the FIXME:s in the original test case.

Now the protocol resilience tests can be updated to use @_weakLinked
for all newly-added protocol requirements and default implementations.

This allows the tests to pass in the backward deployment test scenario
as well.

Eventually this will be based on availability instead of a special
attribute.

This completes <rdar://problem/29888071>.
2018-10-31 19:55:01 -04:00