Reimplement the witness matching logic used for generic requirements
so that it properly models the expectations required of the witness,
then captures the results in the AST. The new approach has a number of
advantages over the existing hacks:
* The constraint solver no longer requires hacks to try to tangle
together the innermost archetypes from the requirement with the
outer archetypes of the context of the protocol
conformance. Instead, we create a synthetic set of archetypes that
describes the requirement as it should be matched against
witnesses. This eliminates the infamous 'SelfTypeVar' hack.
* The type checker no longer records substitutions involving a weird
mix of archetypes from different contexts (see above), so it's
actually plausible to reason about the substitutions of a witness. A
new `Witness` class contains the declaration, substitutions, and all
other information required to interpret the witness.
* SILGen now uses the substitution information for witnesses when
building witness thunks, rather than computing all of it from
scratch. ``substSelfTypeIntoProtocolRequirementType()` is now gone
(absorbed into the type checker, and improved from there), and the
witness-thunk emission code is simpler. A few other bits of SILGen
got simpler because the substitutions can now be trusted.
* Witness matching and thunk generation involving generic requirements
and nested generics now works, based on some work @slavapestov was
already doing in this area.
* The AST verifier can now verify the archetypes that occur in witness substitutions.
* Although it's not in this commit, the `Witness` structure is
suitable for complete (de-)serialization, unlike the weird mix of
archetypes previously present.
Fixes rdar://problem/24079818 and cleans up an area that's been messy
and poorly understood for a very, very long time.
When performing the occurs check, look for the *representative* of the
type variable we're about to bind, rather than the type variable
itself. Fixes rdar://problem/26845038, SR-1512, SR-1902, SR2635,
SR-2852, and SR-2766.
We've been performing the "occurs" check when computing potential
bindings for type variables, but we weren't actually performing the
check for bindings that *must* occur. Perform the occurs check before
binding type variables, which fixes a few crashers and is far more principled.
Note that this obviates the need for tracking the type variables we've
substituted in simplifyType(), so simplify that as well.
Fixes rdar://problem/27879334 / SR-2351.
When we're creating a new constraint because we couldn't solve it (and
need to record the result), do so without trying to simplify it yet
again. It's just wasted work.
ConstraintSystem::addConstraint() is no longer used to simplify
existing constraints, nor does it have to deal with "externally
solved" constraints. Remove those parameters and simplify the code. NFC
When adding constraints into the constraint system, don't immediately
allocate a Constraint and add it via the most-general
addConstraint(). Instead, go through a more specific entrypoint (e.g.,
addValueMemberConstraint, addRestrictedConstraint, etc.), so we can
start phasing out the general "add an already-formed constraint"
function. NFC
There's no point in adding the constraints known on an associated type
to the constraint system, because they're implied by the protocol
constraint itself.
We had a few places that were performing ad hoc variants of
ConstraintSystem::getFixedTypeRecursive(); simplify it's interface so
we can use it everywhere consistently. Fixes rdar://problem/27261929.
Similar to “isTypeParameter,” this new entry point determines whether the type is a type variable or a nested type of a type thereof. The latter case isn’t actually formed yet, so this is NFC staging the trivial bits of this change.
In most places where we were checking "is<ErrorType>()", we now mean
"any error occurred". The few exceptions are in associated type
inference, code completion, and expression diagnostics, where we might
still work with partial errors.
Type::subst()'s "IgnoreMissing" option was fairly unprincipled, dropping
unsubstituted types into the resulting AST without any indication
whatsoever that anything went wrong. Replace this notion with a new
form of ErrorType that explicitly tracks which substituted type caused
the problem. It's still an ErrorType, but it prints like the
substituted type (which is important for code completion) and allows
us to step back to the substituted type if needed (which is used by
associated type inference). Then, allow Type::subst(), when the new
UseErrorTypes flag is passed, to form partially-substituted types that
contain errors, which both code completion and associated type
inference relied on.
Over time, I hope we can use error-types-with-original-types more
often to eliminate "<<error type>>" from diagnostics and teach
Type::subst() never to return a "null" type. Clients can check
"hasError()" to deal with failure cases rather than checking null.
There's a bit of a hack to deal with generic typealiases, but
overall this makes things more logical.
This is the last big refactoring before we can allow constrained
extensions to make generic parameters concrete. All that remains
is a small set of changes to SIL type lowering, and retooling
some diagnostics in Sema.
Now that TupleTypeElts are simpler in Swift 3 (though they're about to
become more complicated for other reasons), most of the cases where we
are explicitly constructing ones are really just plain copies or can
otherwise use existing helper functions.
NFC
Now that the previous patches have shaken out implicit assumptions
about the order of generic requirements and substitutions, we can
make a more radical change, dropping redundant protocol requirements
when building the original generic signature.
This means that the canonical ordering and minimization that we
used to only perform when building the mangling signature is done
all of the time, and hence getCanonicalManglingSignature() can go
away.
Usages now either call getCanonicalSignature(), or operate on the
original signature directly.
Suppose you have this protocol:
protocol P {
typealias A = Int
typealias B = Self
}
Clearly, 'P.B' does not make sense, because then the type parameter
would "leak out" of the existential container. However, 'P.A' is totally
fine, it just means 'Int'.
Previously, we would allow 'P.A' in type context, and diagnose on 'P.B'.
However, due to an oversight, neither one was allowed in expression
context, so for example you could not write 'P.A.self', even though
that should just mean 'Int.self'.
Fix this by generalizing performMemberLookup(), and fix up some
diagnostics to be more specific when something is wrong -- we want
to avoid talking about typealiases as 'static members', since that
doesn't really make much sense.
Fixes <https://bugs.swift.org/browse/SR-2314>.
One last bit of SE-0072. We shouldn't fall back to bridged classes in the absence of type context for literals anymore. By itself, this kind of hoses the use of literals with NS types, but I think we can get most of the QoI back with overlay changes I plan to propose following this.
Extend the handling of function reference kinds to member references
(e.g., x.f), and therefore the logic for stripping argument labels. We
appear to be stripping argument labels from all of the places where it
is required.
When referencing a function in the type checker, drop argument labels
when we don't need them to type-check an immediate call to that
function. This provides the semantic behavior of SE-0111, e.g.,
references to functions as values produce unlabeled function types,
without the representational change of actually dropping argument
labels from the type system.
At the moment, this only works for bare references to functions. It
still needs to be pushed through more of the type checker and more AST
nodes to work in the general case.
Keep this work behind the frontend flag
-suppress-argument-labels-in-types for now.
Rather than synthesizing a global operator '==' for Equatable enums,
synthesize a member operator, which is more idiomatic and much
cleaner.
To make sure that these synthesized operators can actually be found,
start considering operator requirements in protocols
more generally in the type checker, so that, e.g., "myEnum == myEnum"
will type-check against Equatable.== and, on successful type-check,
will call the (newly-synthesized) witness for '=='. This both makes it
easier to make sure we find the operators in, e.g., complex multi-file
and lazy-type checking scenarios, and is a step toward the
type-checking improvements described in SE-0091.
Allow 'static' (or, in classes, final 'class') operators to be
declared within types and extensions thereof. Within protocols,
require operators to be marked 'static'. Use a warning with a Fix-It
to stage this in, so we don't break the world's code.
Protocol conformance checking already seems to work, so add some tests
for that. Update a pile of tests and the standard library to include
the required 'static' keywords.
There is an amusing name-mangling change here. Global operators were
getting marked as 'static' (for silly reasons), so their mangled names
had the 'Z' modifier for static methods, even though this doesn't make
sense. Now, operators within types and extensions need to be 'static'
as written.
change includes both the necessary protocol updates and the deprecation
warnings
suitable for migration. A future patch will remove the renamings and
make this
a hard error.
There are many places where we do the 'if inside a protocol, get the
Self type parameter, otherwise, use the declared type' dance.
We actually have really handy utility methods that encapsulate this,
so let's use them more.
The underlying type can now refer to generic parameters from an
outer context, and we allow qualified and unqualified access to
such typealiases.
One problem remains, with specializations of generic typealiases
in expression parsing context, marked with FIXME in the test.
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.
Now that ConstraintSystem::openGeneric() is the only remaining
caller of this function, inline it in there and open-code the
logic.
Also, add a hack for opening nominal types contained inside
protocol types. This is invalid, but we should not crash, so
bind the type variable for the protocol 'Self' type to the
'Self' archetype, since it will not be equated with anything
otherwise.
- Change openGeneric() to take two DeclContexts, one is the generic
context for the signature and the second is the generic context
containing the declaration being opened
- This allows us to clean up the logic around skipProtocolSelfRequirement;
instead of testing both the DeclContext and its parent, we know
exactly what DeclContext to test. Also, use the right Self type here,
instead of always using (0, 0)
- Now that we have the right DeclContexts handy, we can move the
getGenericTypeContextDepth() call into openGeneric(), simplifying
callers
- Now that openType() no longer opens generic signatures, it takes fewer
parameters
We were failing to create an unavailable TypeAlias for the old name
in the case the renamed type was generic, leading to poor diagnostics.
Also, Sema resolves generic TypeAliases very early, while building
a Type from a TypeRepr -- this means the unavailable/deprecated
check runs too late to catch generic TypeAlises.
Add a hack where we preserve a reference to the original TypeAliasDecl
by way of constructing a SubstitutedType which desugars to the
replacement type, rather than resolving the replacement type
directly, so that the availability check can pick it up.
A better fix for this would be to introduce a BoundGenericAliasType
sugared type, but that's a bigger change that can come later.
Fixes <rdar://problem/26206263>.
The verifier now asserts that Throws, ThrowsLoc and isBodyThrowing()
match up.
Also, add /*Label=*/ comments where necessary to make the long argument
lists easier to read, and cleaned up some inconsistent naming conventions.
I caught a case where ClangImporter where we were passing in a loc as
StaticLoc instead of FuncLoc, but probably this didn't affect anything.
Simplify and improve the checking of @objc names when matching a
witness to a requirement in the @objc protocol. First, don't use
@objc-ness as part of the initial screening to determine whether a
witness potentially matches an @objc requirement: we will only reject
a potential witness when the potential witness has an explicit
"@nonobjc" attribute on it. Otherwise, the presence of @objc and the
corresponding Objective-C name is checked only after selecting a
candidate. This more closely mirrors what we do for override checking,
where we match based on the Swift names (first) and validate
@objc'ness afterward. It is also a stepping stone to inferring
@objc'ness and @objc names from protocol conformances.
Second, when emitting a diagnostic about a missing or incorrect @objc
annotation, make sure the Fix-It gets the @objc name right: this might
mean adding the Objective-C name along with @objc (e.g.,
"@objc(fooWithString:bar:)"), adding the name to an
unadorned-but-explicit "@objc" attribute, or fixing the name of an
@objc attribute (e.g., "@objc(foo:bar:)" becomes
@objc(fooWithString:bar:)"). Make this diagnostic an error, rather
than a note on a generic "does not conform" diagnostic, so it's much
easier to see the diagnostic and apply the Fix-It.
Third, when emitting the warning about a non-@objc near-match for an
optional @objc requirement, provide two Fix-Its: one that adds the
appropriate @objc annotation (per the paragraph above), and one that
adds @nonobjc to silence the warning.
Part of the QoI improvements for conformances to @objc protocols,
rdar://problem/25159872.
Simplify and improve the checking of @objc names when matching a
witness to a requirement in the @objc protocol. First, don't use
@objc-ness as part of the initial screening to determine whether a
witness potentially matches an @objc requirement: we will only reject
a potential witness when the potential witness has an explicit
"@nonobjc" attribute on it. Otherwise, the presence of @objc and the
corresponding Objective-C name is checked only after selecting a
candidate. This more closely mirrors what we do for override checking,
where we match based on the Swift names (first) and validate
@objc'ness afterward. It is also a stepping stone to inferring
@objc'ness and @objc names from protocol conformances.
Second, when emitting a diagnostic about a missing or incorrect @objc
annotation, make sure the Fix-It gets the @objc name right: this might
mean adding the Objective-C name along with @objc (e.g.,
"@objc(fooWithString:bar:)"), adding the name to an
unadorned-but-explicit "@objc" attribute, or fixing the name of an
@objc attribute (e.g., "@objc(foo:bar:)" becomes
@objc(fooWithString:bar:)"). Make this diagnostic an error, rather
than a note on a generic "does not conform" diagnostic, so it's much
easier to see the diagnostic and apply the Fix-It.
Third, when emitting the warning about a non-@objc near-match for an
optional @objc requirement, provide two Fix-Its: one that adds the
appropriate @objc annotation (per the paragraph above), and one that
adds @nonobjc to silence the warning.
Part of the QoI improvements for conformances to @objc protocols,
rdar://problem/25159872.
a generic function type during constraint solving, as opposed to
checking a bunch of implicit things that we already know. This
should significantly improve the efficiency of checking uses of
generic APIs by reducing the total number of type variables and
constraints.
It is becoming increasingly funny to refer to this minimized generic
signature as the "mangling" signature.
The test changes are kind of a wash: in one case, we've eliminated
a confusing extra error, but in another we've caused the confusing
extra error to refer to '<<error type>>'. Not worth fighting right
now. The reference-dependencies change is due to not needing to
pull in all of those associated types anymore, which seems correct.
to remap an unbound generic type into a bound generic type. This shares
a common codepath to make way for future progress.
This exposed a case where we'd produce invalid ASTs in some testcases
in the validation tests because started asserting on the invalid code.
Solve this by detecting the problem and producing a circularity error.
Overall, NFC.
My previous commit here didn’t work correctly for nested tuples, both
because it didn’t recurse into them to propagate access kind correctly
and because an outer TupleIndex overload (when indexing into the nested
tuple) could still be expecting an lvalue type.
This fix is much better. ConstraintSystem::resolveOverload now
correctly always expects rvalue types from rvalue tuples. And during
applyMemberRefExpr, if the overload expects an rvalue but the tuple
contains lvalues, coerceToType() correctly does any recursive munging
of the tuple expr required.
