llvm-project/clang/lib/Sema/SemaDeclCXX.cpp

7592 lines
295 KiB
C++
Raw Normal View History

//===------ SemaDeclCXX.cpp - Semantic Analysis for C++ Declarations ------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file implements semantic analysis for C++ declarations.
//
//===----------------------------------------------------------------------===//
#include "clang/Sema/SemaInternal.h"
#include "clang/Sema/CXXFieldCollector.h"
#include "clang/Sema/Scope.h"
#include "clang/Sema/Initialization.h"
#include "clang/Sema/Lookup.h"
#include "clang/AST/ASTConsumer.h"
#include "clang/AST/ASTContext.h"
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
#include "clang/AST/CharUnits.h"
#include "clang/AST/CXXInheritance.h"
#include "clang/AST/DeclVisitor.h"
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
#include "clang/AST/RecordLayout.h"
#include "clang/AST/StmtVisitor.h"
#include "clang/AST/TypeLoc.h"
#include "clang/AST/TypeOrdering.h"
#include "clang/Sema/DeclSpec.h"
#include "clang/Sema/ParsedTemplate.h"
#include "clang/Basic/PartialDiagnostic.h"
#include "clang/Lex/Preprocessor.h"
#include "llvm/ADT/DenseSet.h"
#include "llvm/ADT/STLExtras.h"
#include <map>
#include <set>
using namespace clang;
//===----------------------------------------------------------------------===//
// CheckDefaultArgumentVisitor
//===----------------------------------------------------------------------===//
namespace {
/// CheckDefaultArgumentVisitor - C++ [dcl.fct.default] Traverses
/// the default argument of a parameter to determine whether it
/// contains any ill-formed subexpressions. For example, this will
/// diagnose the use of local variables or parameters within the
/// default argument expression.
class CheckDefaultArgumentVisitor
: public StmtVisitor<CheckDefaultArgumentVisitor, bool> {
Expr *DefaultArg;
Sema *S;
public:
CheckDefaultArgumentVisitor(Expr *defarg, Sema *s)
: DefaultArg(defarg), S(s) {}
bool VisitExpr(Expr *Node);
bool VisitDeclRefExpr(DeclRefExpr *DRE);
bool VisitCXXThisExpr(CXXThisExpr *ThisE);
};
/// VisitExpr - Visit all of the children of this expression.
bool CheckDefaultArgumentVisitor::VisitExpr(Expr *Node) {
bool IsInvalid = false;
for (Stmt::child_range I = Node->children(); I; ++I)
IsInvalid |= Visit(*I);
return IsInvalid;
}
/// VisitDeclRefExpr - Visit a reference to a declaration, to
/// determine whether this declaration can be used in the default
/// argument expression.
bool CheckDefaultArgumentVisitor::VisitDeclRefExpr(DeclRefExpr *DRE) {
NamedDecl *Decl = DRE->getDecl();
if (ParmVarDecl *Param = dyn_cast<ParmVarDecl>(Decl)) {
// C++ [dcl.fct.default]p9
// Default arguments are evaluated each time the function is
// called. The order of evaluation of function arguments is
// unspecified. Consequently, parameters of a function shall not
// be used in default argument expressions, even if they are not
// evaluated. Parameters of a function declared before a default
// argument expression are in scope and can hide namespace and
// class member names.
return S->Diag(DRE->getSourceRange().getBegin(),
diag::err_param_default_argument_references_param)
<< Param->getDeclName() << DefaultArg->getSourceRange();
} else if (VarDecl *VDecl = dyn_cast<VarDecl>(Decl)) {
// C++ [dcl.fct.default]p7
// Local variables shall not be used in default argument
// expressions.
if (VDecl->isLocalVarDecl())
return S->Diag(DRE->getSourceRange().getBegin(),
diag::err_param_default_argument_references_local)
<< VDecl->getDeclName() << DefaultArg->getSourceRange();
}
return false;
}
/// VisitCXXThisExpr - Visit a C++ "this" expression.
bool CheckDefaultArgumentVisitor::VisitCXXThisExpr(CXXThisExpr *ThisE) {
// C++ [dcl.fct.default]p8:
// The keyword this shall not be used in a default argument of a
// member function.
return S->Diag(ThisE->getSourceRange().getBegin(),
diag::err_param_default_argument_references_this)
<< ThisE->getSourceRange();
}
}
bool
Sema::SetParamDefaultArgument(ParmVarDecl *Param, Expr *Arg,
SourceLocation EqualLoc) {
if (RequireCompleteType(Param->getLocation(), Param->getType(),
diag::err_typecheck_decl_incomplete_type)) {
Param->setInvalidDecl();
return true;
}
// C++ [dcl.fct.default]p5
// A default argument expression is implicitly converted (clause
// 4) to the parameter type. The default argument expression has
// the same semantic constraints as the initializer expression in
// a declaration of a variable of the parameter type, using the
// copy-initialization semantics (8.5).
InitializedEntity Entity = InitializedEntity::InitializeParameter(Context,
Param);
InitializationKind Kind = InitializationKind::CreateCopy(Param->getLocation(),
EqualLoc);
InitializationSequence InitSeq(*this, Entity, Kind, &Arg, 1);
ExprResult Result = InitSeq.Perform(*this, Entity, Kind,
MultiExprArg(*this, &Arg, 1));
if (Result.isInvalid())
return true;
Arg = Result.takeAs<Expr>();
CheckImplicitConversions(Arg, EqualLoc);
Arg = MaybeCreateExprWithCleanups(Arg);
// Okay: add the default argument to the parameter
Param->setDefaultArg(Arg);
// We have already instantiated this parameter; provide each of the
// instantiations with the uninstantiated default argument.
UnparsedDefaultArgInstantiationsMap::iterator InstPos
= UnparsedDefaultArgInstantiations.find(Param);
if (InstPos != UnparsedDefaultArgInstantiations.end()) {
for (unsigned I = 0, N = InstPos->second.size(); I != N; ++I)
InstPos->second[I]->setUninstantiatedDefaultArg(Arg);
// We're done tracking this parameter's instantiations.
UnparsedDefaultArgInstantiations.erase(InstPos);
}
return false;
}
/// ActOnParamDefaultArgument - Check whether the default argument
/// provided for a function parameter is well-formed. If so, attach it
/// to the parameter declaration.
void
2010-08-21 17:40:31 +08:00
Sema::ActOnParamDefaultArgument(Decl *param, SourceLocation EqualLoc,
Expr *DefaultArg) {
if (!param || !DefaultArg)
return;
2010-08-21 17:40:31 +08:00
ParmVarDecl *Param = cast<ParmVarDecl>(param);
UnparsedDefaultArgLocs.erase(Param);
// Default arguments are only permitted in C++
if (!getLangOptions().CPlusPlus) {
Diag(EqualLoc, diag::err_param_default_argument)
<< DefaultArg->getSourceRange();
Param->setInvalidDecl();
return;
}
// Check for unexpanded parameter packs.
if (DiagnoseUnexpandedParameterPack(DefaultArg, UPPC_DefaultArgument)) {
Param->setInvalidDecl();
return;
}
// Check that the default argument is well-formed
CheckDefaultArgumentVisitor DefaultArgChecker(DefaultArg, this);
if (DefaultArgChecker.Visit(DefaultArg)) {
Param->setInvalidDecl();
return;
}
SetParamDefaultArgument(Param, DefaultArg, EqualLoc);
}
/// ActOnParamUnparsedDefaultArgument - We've seen a default
/// argument for a function parameter, but we can't parse it yet
/// because we're inside a class definition. Note that this default
/// argument will be parsed later.
2010-08-21 17:40:31 +08:00
void Sema::ActOnParamUnparsedDefaultArgument(Decl *param,
SourceLocation EqualLoc,
SourceLocation ArgLoc) {
if (!param)
return;
2010-08-21 17:40:31 +08:00
ParmVarDecl *Param = cast<ParmVarDecl>(param);
if (Param)
Param->setUnparsedDefaultArg();
UnparsedDefaultArgLocs[Param] = ArgLoc;
}
/// ActOnParamDefaultArgumentError - Parsing or semantic analysis of
/// the default argument for the parameter param failed.
2010-08-21 17:40:31 +08:00
void Sema::ActOnParamDefaultArgumentError(Decl *param) {
if (!param)
return;
2010-08-21 17:40:31 +08:00
ParmVarDecl *Param = cast<ParmVarDecl>(param);
Param->setInvalidDecl();
UnparsedDefaultArgLocs.erase(Param);
}
/// CheckExtraCXXDefaultArguments - Check for any extra default
/// arguments in the declarator, which is not a function declaration
/// or definition and therefore is not permitted to have default
/// arguments. This routine should be invoked for every declarator
/// that is not a function declaration or definition.
void Sema::CheckExtraCXXDefaultArguments(Declarator &D) {
// C++ [dcl.fct.default]p3
// A default argument expression shall be specified only in the
// parameter-declaration-clause of a function declaration or in a
// template-parameter (14.1). It shall not be specified for a
// parameter pack. If it is specified in a
// parameter-declaration-clause, it shall not occur within a
// declarator or abstract-declarator of a parameter-declaration.
for (unsigned i = 0, e = D.getNumTypeObjects(); i != e; ++i) {
DeclaratorChunk &chunk = D.getTypeObject(i);
if (chunk.Kind == DeclaratorChunk::Function) {
for (unsigned argIdx = 0, e = chunk.Fun.NumArgs; argIdx != e; ++argIdx) {
ParmVarDecl *Param =
2010-08-21 17:40:31 +08:00
cast<ParmVarDecl>(chunk.Fun.ArgInfo[argIdx].Param);
if (Param->hasUnparsedDefaultArg()) {
CachedTokens *Toks = chunk.Fun.ArgInfo[argIdx].DefaultArgTokens;
Diag(Param->getLocation(), diag::err_param_default_argument_nonfunc)
<< SourceRange((*Toks)[1].getLocation(), Toks->back().getLocation());
delete Toks;
chunk.Fun.ArgInfo[argIdx].DefaultArgTokens = 0;
} else if (Param->getDefaultArg()) {
Diag(Param->getLocation(), diag::err_param_default_argument_nonfunc)
<< Param->getDefaultArg()->getSourceRange();
Param->setDefaultArg(0);
}
}
}
}
}
// MergeCXXFunctionDecl - Merge two declarations of the same C++
// function, once we already know that they have the same
// type. Subroutine of MergeFunctionDecl. Returns true if there was an
// error, false otherwise.
bool Sema::MergeCXXFunctionDecl(FunctionDecl *New, FunctionDecl *Old) {
bool Invalid = false;
// C++ [dcl.fct.default]p4:
// For non-template functions, default arguments can be added in
// later declarations of a function in the same
// scope. Declarations in different scopes have completely
// distinct sets of default arguments. That is, declarations in
// inner scopes do not acquire default arguments from
// declarations in outer scopes, and vice versa. In a given
// function declaration, all parameters subsequent to a
// parameter with a default argument shall have default
// arguments supplied in this or previous declarations. A
// default argument shall not be redefined by a later
// declaration (not even to the same value).
//
// C++ [dcl.fct.default]p6:
// Except for member functions of class templates, the default arguments
// in a member function definition that appears outside of the class
// definition are added to the set of default arguments provided by the
// member function declaration in the class definition.
for (unsigned p = 0, NumParams = Old->getNumParams(); p < NumParams; ++p) {
ParmVarDecl *OldParam = Old->getParamDecl(p);
ParmVarDecl *NewParam = New->getParamDecl(p);
if (OldParam->hasDefaultArg() && NewParam->hasDefaultArg()) {
// FIXME: If we knew where the '=' was, we could easily provide a fix-it
// hint here. Alternatively, we could walk the type-source information
// for NewParam to find the last source location in the type... but it
// isn't worth the effort right now. This is the kind of test case that
// is hard to get right:
// int f(int);
// void g(int (*fp)(int) = f);
// void g(int (*fp)(int) = &f);
Diag(NewParam->getLocation(),
diag::err_param_default_argument_redefinition)
<< NewParam->getDefaultArgRange();
// Look for the function declaration where the default argument was
// actually written, which may be a declaration prior to Old.
for (FunctionDecl *Older = Old->getPreviousDeclaration();
Older; Older = Older->getPreviousDeclaration()) {
if (!Older->getParamDecl(p)->hasDefaultArg())
break;
OldParam = Older->getParamDecl(p);
}
Diag(OldParam->getLocation(), diag::note_previous_definition)
<< OldParam->getDefaultArgRange();
Invalid = true;
} else if (OldParam->hasDefaultArg()) {
// Merge the old default argument into the new parameter.
// It's important to use getInit() here; getDefaultArg()
// strips off any top-level ExprWithCleanups.
NewParam->setHasInheritedDefaultArg();
if (OldParam->hasUninstantiatedDefaultArg())
NewParam->setUninstantiatedDefaultArg(
OldParam->getUninstantiatedDefaultArg());
else
NewParam->setDefaultArg(OldParam->getInit());
} else if (NewParam->hasDefaultArg()) {
if (New->getDescribedFunctionTemplate()) {
// Paragraph 4, quoted above, only applies to non-template functions.
Diag(NewParam->getLocation(),
diag::err_param_default_argument_template_redecl)
<< NewParam->getDefaultArgRange();
Diag(Old->getLocation(), diag::note_template_prev_declaration)
<< false;
} else if (New->getTemplateSpecializationKind()
!= TSK_ImplicitInstantiation &&
New->getTemplateSpecializationKind() != TSK_Undeclared) {
// C++ [temp.expr.spec]p21:
// Default function arguments shall not be specified in a declaration
// or a definition for one of the following explicit specializations:
// - the explicit specialization of a function template;
// - the explicit specialization of a member function template;
// - the explicit specialization of a member function of a class
// template where the class template specialization to which the
// member function specialization belongs is implicitly
// instantiated.
Diag(NewParam->getLocation(), diag::err_template_spec_default_arg)
<< (New->getTemplateSpecializationKind() ==TSK_ExplicitSpecialization)
<< New->getDeclName()
<< NewParam->getDefaultArgRange();
} else if (New->getDeclContext()->isDependentContext()) {
// C++ [dcl.fct.default]p6 (DR217):
// Default arguments for a member function of a class template shall
// be specified on the initial declaration of the member function
// within the class template.
//
// Reading the tea leaves a bit in DR217 and its reference to DR205
// leads me to the conclusion that one cannot add default function
// arguments for an out-of-line definition of a member function of a
// dependent type.
int WhichKind = 2;
if (CXXRecordDecl *Record
= dyn_cast<CXXRecordDecl>(New->getDeclContext())) {
if (Record->getDescribedClassTemplate())
WhichKind = 0;
else if (isa<ClassTemplatePartialSpecializationDecl>(Record))
WhichKind = 1;
else
WhichKind = 2;
}
Diag(NewParam->getLocation(),
diag::err_param_default_argument_member_template_redecl)
<< WhichKind
<< NewParam->getDefaultArgRange();
}
}
}
if (CheckEquivalentExceptionSpec(Old, New))
Invalid = true;
return Invalid;
}
/// CheckCXXDefaultArguments - Verify that the default arguments for a
/// function declaration are well-formed according to C++
/// [dcl.fct.default].
void Sema::CheckCXXDefaultArguments(FunctionDecl *FD) {
unsigned NumParams = FD->getNumParams();
unsigned p;
// Find first parameter with a default argument
for (p = 0; p < NumParams; ++p) {
ParmVarDecl *Param = FD->getParamDecl(p);
if (Param->hasDefaultArg())
break;
}
// C++ [dcl.fct.default]p4:
// In a given function declaration, all parameters
// subsequent to a parameter with a default argument shall
// have default arguments supplied in this or previous
// declarations. A default argument shall not be redefined
// by a later declaration (not even to the same value).
unsigned LastMissingDefaultArg = 0;
for (; p < NumParams; ++p) {
ParmVarDecl *Param = FD->getParamDecl(p);
if (!Param->hasDefaultArg()) {
if (Param->isInvalidDecl())
/* We already complained about this parameter. */;
else if (Param->getIdentifier())
Diag(Param->getLocation(),
diag::err_param_default_argument_missing_name)
<< Param->getIdentifier();
else
Diag(Param->getLocation(),
diag::err_param_default_argument_missing);
LastMissingDefaultArg = p;
}
}
if (LastMissingDefaultArg > 0) {
// Some default arguments were missing. Clear out all of the
// default arguments up to (and including) the last missing
// default argument, so that we leave the function parameters
// in a semantically valid state.
for (p = 0; p <= LastMissingDefaultArg; ++p) {
ParmVarDecl *Param = FD->getParamDecl(p);
if (Param->hasDefaultArg()) {
Param->setDefaultArg(0);
}
}
}
}
/// isCurrentClassName - Determine whether the identifier II is the
/// name of the class type currently being defined. In the case of
/// nested classes, this will only return true if II is the name of
/// the innermost class.
bool Sema::isCurrentClassName(const IdentifierInfo &II, Scope *,
const CXXScopeSpec *SS) {
assert(getLangOptions().CPlusPlus && "No class names in C!");
CXXRecordDecl *CurDecl;
2009-03-19 08:18:19 +08:00
if (SS && SS->isSet() && !SS->isInvalid()) {
DeclContext *DC = computeDeclContext(*SS, true);
CurDecl = dyn_cast_or_null<CXXRecordDecl>(DC);
} else
CurDecl = dyn_cast_or_null<CXXRecordDecl>(CurContext);
if (CurDecl && CurDecl->getIdentifier())
return &II == CurDecl->getIdentifier();
else
return false;
}
/// \brief Check the validity of a C++ base class specifier.
///
/// \returns a new CXXBaseSpecifier if well-formed, emits diagnostics
/// and returns NULL otherwise.
CXXBaseSpecifier *
Sema::CheckBaseSpecifier(CXXRecordDecl *Class,
SourceRange SpecifierRange,
bool Virtual, AccessSpecifier Access,
TypeSourceInfo *TInfo,
SourceLocation EllipsisLoc) {
QualType BaseType = TInfo->getType();
// C++ [class.union]p1:
// A union shall not have base classes.
if (Class->isUnion()) {
Diag(Class->getLocation(), diag::err_base_clause_on_union)
<< SpecifierRange;
return 0;
}
if (EllipsisLoc.isValid() &&
!TInfo->getType()->containsUnexpandedParameterPack()) {
Diag(EllipsisLoc, diag::err_pack_expansion_without_parameter_packs)
<< TInfo->getTypeLoc().getSourceRange();
EllipsisLoc = SourceLocation();
}
if (BaseType->isDependentType())
return new (Context) CXXBaseSpecifier(SpecifierRange, Virtual,
Class->getTagKind() == TTK_Class,
Access, TInfo, EllipsisLoc);
SourceLocation BaseLoc = TInfo->getTypeLoc().getBeginLoc();
// Base specifiers must be record types.
if (!BaseType->isRecordType()) {
Diag(BaseLoc, diag::err_base_must_be_class) << SpecifierRange;
return 0;
}
// C++ [class.union]p1:
// A union shall not be used as a base class.
if (BaseType->isUnionType()) {
Diag(BaseLoc, diag::err_union_as_base_class) << SpecifierRange;
return 0;
}
// C++ [class.derived]p2:
// The class-name in a base-specifier shall not be an incompletely
// defined class.
if (RequireCompleteType(BaseLoc, BaseType,
PDiag(diag::err_incomplete_base_class)
<< SpecifierRange)) {
Class->setInvalidDecl();
return 0;
}
// If the base class is polymorphic or isn't empty, the new one is/isn't, too.
RecordDecl *BaseDecl = BaseType->getAs<RecordType>()->getDecl();
assert(BaseDecl && "Record type has no declaration");
BaseDecl = BaseDecl->getDefinition();
assert(BaseDecl && "Base type is not incomplete, but has no definition");
CXXRecordDecl * CXXBaseDecl = cast<CXXRecordDecl>(BaseDecl);
assert(CXXBaseDecl && "Base type is not a C++ type");
// C++ [class.derived]p2:
// If a class is marked with the class-virt-specifier final and it appears
// as a base-type-specifier in a base-clause (10 class.derived), the program
// is ill-formed.
if (CXXBaseDecl->hasAttr<FinalAttr>()) {
Diag(BaseLoc, diag::err_class_marked_final_used_as_base)
<< CXXBaseDecl->getDeclName();
Diag(CXXBaseDecl->getLocation(), diag::note_previous_decl)
<< CXXBaseDecl->getDeclName();
return 0;
}
if (BaseDecl->isInvalidDecl())
Class->setInvalidDecl();
// Create the base specifier.
return new (Context) CXXBaseSpecifier(SpecifierRange, Virtual,
Class->getTagKind() == TTK_Class,
Access, TInfo, EllipsisLoc);
}
/// ActOnBaseSpecifier - Parsed a base specifier. A base specifier is
/// one entry in the base class list of a class specifier, for
/// example:
/// class foo : public bar, virtual private baz {
/// 'public bar' and 'virtual private baz' are each base-specifiers.
BaseResult
2010-08-21 17:40:31 +08:00
Sema::ActOnBaseSpecifier(Decl *classdecl, SourceRange SpecifierRange,
bool Virtual, AccessSpecifier Access,
ParsedType basetype, SourceLocation BaseLoc,
SourceLocation EllipsisLoc) {
if (!classdecl)
return true;
AdjustDeclIfTemplate(classdecl);
2010-08-21 17:40:31 +08:00
CXXRecordDecl *Class = dyn_cast<CXXRecordDecl>(classdecl);
if (!Class)
return true;
TypeSourceInfo *TInfo = 0;
GetTypeFromParser(basetype, &TInfo);
if (EllipsisLoc.isInvalid() &&
DiagnoseUnexpandedParameterPack(SpecifierRange.getBegin(), TInfo,
UPPC_BaseType))
return true;
if (CXXBaseSpecifier *BaseSpec = CheckBaseSpecifier(Class, SpecifierRange,
Virtual, Access, TInfo,
EllipsisLoc))
return BaseSpec;
return true;
}
/// \brief Performs the actual work of attaching the given base class
/// specifiers to a C++ class.
bool Sema::AttachBaseSpecifiers(CXXRecordDecl *Class, CXXBaseSpecifier **Bases,
unsigned NumBases) {
if (NumBases == 0)
return false;
// Used to keep track of which base types we have already seen, so
// that we can properly diagnose redundant direct base types. Note
// that the key is always the unqualified canonical type of the base
// class.
std::map<QualType, CXXBaseSpecifier*, QualTypeOrdering> KnownBaseTypes;
// Copy non-redundant base specifiers into permanent storage.
unsigned NumGoodBases = 0;
bool Invalid = false;
for (unsigned idx = 0; idx < NumBases; ++idx) {
QualType NewBaseType
= Context.getCanonicalType(Bases[idx]->getType());
NewBaseType = NewBaseType.getLocalUnqualifiedType();
if (!Class->hasObjectMember()) {
if (const RecordType *FDTTy =
NewBaseType.getTypePtr()->getAs<RecordType>())
if (FDTTy->getDecl()->hasObjectMember())
Class->setHasObjectMember(true);
}
if (KnownBaseTypes[NewBaseType]) {
// C++ [class.mi]p3:
// A class shall not be specified as a direct base class of a
// derived class more than once.
Diag(Bases[idx]->getSourceRange().getBegin(),
diag::err_duplicate_base_class)
<< KnownBaseTypes[NewBaseType]->getType()
<< Bases[idx]->getSourceRange();
// Delete the duplicate base class specifier; we're going to
// overwrite its pointer later.
Context.Deallocate(Bases[idx]);
Invalid = true;
} else {
// Okay, add this new base class.
KnownBaseTypes[NewBaseType] = Bases[idx];
Bases[NumGoodBases++] = Bases[idx];
}
}
// Attach the remaining base class specifiers to the derived class.
Class->setBases(Bases, NumGoodBases);
// Delete the remaining (good) base class specifiers, since their
// data has been copied into the CXXRecordDecl.
for (unsigned idx = 0; idx < NumGoodBases; ++idx)
Context.Deallocate(Bases[idx]);
return Invalid;
}
/// ActOnBaseSpecifiers - Attach the given base specifiers to the
/// class, after checking whether there are any duplicate base
/// classes.
2010-08-21 17:40:31 +08:00
void Sema::ActOnBaseSpecifiers(Decl *ClassDecl, BaseTy **Bases,
unsigned NumBases) {
if (!ClassDecl || !Bases || !NumBases)
return;
AdjustDeclIfTemplate(ClassDecl);
2010-08-21 17:40:31 +08:00
AttachBaseSpecifiers(cast<CXXRecordDecl>(ClassDecl),
(CXXBaseSpecifier**)(Bases), NumBases);
}
static CXXRecordDecl *GetClassForType(QualType T) {
if (const RecordType *RT = T->getAs<RecordType>())
return cast<CXXRecordDecl>(RT->getDecl());
else if (const InjectedClassNameType *ICT = T->getAs<InjectedClassNameType>())
return ICT->getDecl();
else
return 0;
}
/// \brief Determine whether the type \p Derived is a C++ class that is
/// derived from the type \p Base.
bool Sema::IsDerivedFrom(QualType Derived, QualType Base) {
if (!getLangOptions().CPlusPlus)
return false;
CXXRecordDecl *DerivedRD = GetClassForType(Derived);
if (!DerivedRD)
return false;
CXXRecordDecl *BaseRD = GetClassForType(Base);
if (!BaseRD)
return false;
// FIXME: instantiate DerivedRD if necessary. We need a PoI for this.
return DerivedRD->hasDefinition() && DerivedRD->isDerivedFrom(BaseRD);
}
/// \brief Determine whether the type \p Derived is a C++ class that is
/// derived from the type \p Base.
bool Sema::IsDerivedFrom(QualType Derived, QualType Base, CXXBasePaths &Paths) {
if (!getLangOptions().CPlusPlus)
return false;
CXXRecordDecl *DerivedRD = GetClassForType(Derived);
if (!DerivedRD)
return false;
CXXRecordDecl *BaseRD = GetClassForType(Base);
if (!BaseRD)
return false;
return DerivedRD->isDerivedFrom(BaseRD, Paths);
}
void Sema::BuildBasePathArray(const CXXBasePaths &Paths,
CXXCastPath &BasePathArray) {
assert(BasePathArray.empty() && "Base path array must be empty!");
assert(Paths.isRecordingPaths() && "Must record paths!");
const CXXBasePath &Path = Paths.front();
// We first go backward and check if we have a virtual base.
// FIXME: It would be better if CXXBasePath had the base specifier for
// the nearest virtual base.
unsigned Start = 0;
for (unsigned I = Path.size(); I != 0; --I) {
if (Path[I - 1].Base->isVirtual()) {
Start = I - 1;
break;
}
}
// Now add all bases.
for (unsigned I = Start, E = Path.size(); I != E; ++I)
BasePathArray.push_back(const_cast<CXXBaseSpecifier*>(Path[I].Base));
}
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
/// \brief Determine whether the given base path includes a virtual
/// base class.
bool Sema::BasePathInvolvesVirtualBase(const CXXCastPath &BasePath) {
for (CXXCastPath::const_iterator B = BasePath.begin(),
BEnd = BasePath.end();
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
B != BEnd; ++B)
if ((*B)->isVirtual())
return true;
return false;
}
/// CheckDerivedToBaseConversion - Check whether the Derived-to-Base
/// conversion (where Derived and Base are class types) is
/// well-formed, meaning that the conversion is unambiguous (and
/// that all of the base classes are accessible). Returns true
/// and emits a diagnostic if the code is ill-formed, returns false
/// otherwise. Loc is the location where this routine should point to
/// if there is an error, and Range is the source range to highlight
/// if there is an error.
bool
Sema::CheckDerivedToBaseConversion(QualType Derived, QualType Base,
unsigned InaccessibleBaseID,
unsigned AmbigiousBaseConvID,
SourceLocation Loc, SourceRange Range,
DeclarationName Name,
CXXCastPath *BasePath) {
// First, determine whether the path from Derived to Base is
// ambiguous. This is slightly more expensive than checking whether
// the Derived to Base conversion exists, because here we need to
// explore multiple paths to determine if there is an ambiguity.
CXXBasePaths Paths(/*FindAmbiguities=*/true, /*RecordPaths=*/true,
/*DetectVirtual=*/false);
bool DerivationOkay = IsDerivedFrom(Derived, Base, Paths);
assert(DerivationOkay &&
"Can only be used with a derived-to-base conversion");
(void)DerivationOkay;
if (!Paths.isAmbiguous(Context.getCanonicalType(Base).getUnqualifiedType())) {
if (InaccessibleBaseID) {
// Check that the base class can be accessed.
switch (CheckBaseClassAccess(Loc, Base, Derived, Paths.front(),
InaccessibleBaseID)) {
case AR_inaccessible:
return true;
case AR_accessible:
case AR_dependent:
case AR_delayed:
break;
}
}
// Build a base path if necessary.
if (BasePath)
BuildBasePathArray(Paths, *BasePath);
return false;
}
// We know that the derived-to-base conversion is ambiguous, and
// we're going to produce a diagnostic. Perform the derived-to-base
// search just one more time to compute all of the possible paths so
// that we can print them out. This is more expensive than any of
// the previous derived-to-base checks we've done, but at this point
// performance isn't as much of an issue.
Paths.clear();
Paths.setRecordingPaths(true);
bool StillOkay = IsDerivedFrom(Derived, Base, Paths);
assert(StillOkay && "Can only be used with a derived-to-base conversion");
(void)StillOkay;
// Build up a textual representation of the ambiguous paths, e.g.,
// D -> B -> A, that will be used to illustrate the ambiguous
// conversions in the diagnostic. We only print one of the paths
// to each base class subobject.
std::string PathDisplayStr = getAmbiguousPathsDisplayString(Paths);
Diag(Loc, AmbigiousBaseConvID)
<< Derived << Base << PathDisplayStr << Range << Name;
return true;
}
bool
Sema::CheckDerivedToBaseConversion(QualType Derived, QualType Base,
SourceLocation Loc, SourceRange Range,
CXXCastPath *BasePath,
bool IgnoreAccess) {
return CheckDerivedToBaseConversion(Derived, Base,
IgnoreAccess ? 0
: diag::err_upcast_to_inaccessible_base,
diag::err_ambiguous_derived_to_base_conv,
Loc, Range, DeclarationName(),
BasePath);
}
/// @brief Builds a string representing ambiguous paths from a
/// specific derived class to different subobjects of the same base
/// class.
///
/// This function builds a string that can be used in error messages
/// to show the different paths that one can take through the
/// inheritance hierarchy to go from the derived class to different
/// subobjects of a base class. The result looks something like this:
/// @code
/// struct D -> struct B -> struct A
/// struct D -> struct C -> struct A
/// @endcode
std::string Sema::getAmbiguousPathsDisplayString(CXXBasePaths &Paths) {
std::string PathDisplayStr;
std::set<unsigned> DisplayedPaths;
for (CXXBasePaths::paths_iterator Path = Paths.begin();
Path != Paths.end(); ++Path) {
if (DisplayedPaths.insert(Path->back().SubobjectNumber).second) {
// We haven't displayed a path to this particular base
// class subobject yet.
PathDisplayStr += "\n ";
PathDisplayStr += Context.getTypeDeclType(Paths.getOrigin()).getAsString();
for (CXXBasePath::const_iterator Element = Path->begin();
Element != Path->end(); ++Element)
PathDisplayStr += " -> " + Element->Base->getType().getAsString();
}
}
return PathDisplayStr;
}
//===----------------------------------------------------------------------===//
// C++ class member Handling
//===----------------------------------------------------------------------===//
/// ActOnAccessSpecifier - Parsed an access specifier followed by a colon.
2010-08-21 17:40:31 +08:00
Decl *Sema::ActOnAccessSpecifier(AccessSpecifier Access,
SourceLocation ASLoc,
SourceLocation ColonLoc) {
assert(Access != AS_none && "Invalid kind for syntactic access specifier!");
2010-08-21 17:40:31 +08:00
AccessSpecDecl *ASDecl = AccessSpecDecl::Create(Context, Access, CurContext,
ASLoc, ColonLoc);
CurContext->addHiddenDecl(ASDecl);
2010-08-21 17:40:31 +08:00
return ASDecl;
}
/// CheckOverrideControl - Check C++0x override control semantics.
void Sema::CheckOverrideControl(const Decl *D) {
const CXXMethodDecl *MD = llvm::dyn_cast<CXXMethodDecl>(D);
if (!MD || !MD->isVirtual())
return;
if (MD->isDependentContext())
return;
// C++0x [class.virtual]p3:
// If a virtual function is marked with the virt-specifier override and does
// not override a member function of a base class,
// the program is ill-formed.
bool HasOverriddenMethods =
MD->begin_overridden_methods() != MD->end_overridden_methods();
if (MD->hasAttr<OverrideAttr>() && !HasOverriddenMethods) {
Diag(MD->getLocation(),
diag::err_function_marked_override_not_overriding)
<< MD->getDeclName();
return;
}
// C++0x [class.derived]p8:
// In a class definition marked with the class-virt-specifier explicit,
// if a virtual member function that is neither implicitly-declared nor a
// destructor overrides a member function of a base class and it is not
// marked with the virt-specifier override, the program is ill-formed.
if (MD->getParent()->hasAttr<ExplicitAttr>() && !isa<CXXDestructorDecl>(MD) &&
HasOverriddenMethods && !MD->hasAttr<OverrideAttr>()) {
llvm::SmallVector<const CXXMethodDecl*, 4>
OverriddenMethods(MD->begin_overridden_methods(),
MD->end_overridden_methods());
Diag(MD->getLocation(), diag::err_function_overriding_without_override)
<< MD->getDeclName()
<< (unsigned)OverriddenMethods.size();
for (unsigned I = 0; I != OverriddenMethods.size(); ++I)
Diag(OverriddenMethods[I]->getLocation(),
diag::note_overridden_virtual_function);
}
}
/// CheckIfOverriddenFunctionIsMarkedFinal - Checks whether a virtual member
/// function overrides a virtual member function marked 'final', according to
/// C++0x [class.virtual]p3.
bool Sema::CheckIfOverriddenFunctionIsMarkedFinal(const CXXMethodDecl *New,
const CXXMethodDecl *Old) {
if (!Old->hasAttr<FinalAttr>())
return false;
Diag(New->getLocation(), diag::err_final_function_overridden)
<< New->getDeclName();
Diag(Old->getLocation(), diag::note_overridden_virtual_function);
return true;
}
/// ActOnCXXMemberDeclarator - This is invoked when a C++ class member
/// declarator is parsed. 'AS' is the access specifier, 'BW' specifies the
/// bitfield width if there is one and 'InitExpr' specifies the initializer if
/// any.
2010-08-21 17:40:31 +08:00
Decl *
Sema::ActOnCXXMemberDeclarator(Scope *S, AccessSpecifier AS, Declarator &D,
MultiTemplateParamsArg TemplateParameterLists,
ExprTy *BW, const VirtSpecifiers &VS,
ExprTy *InitExpr, bool IsDefinition,
bool Deleted) {
const DeclSpec &DS = D.getDeclSpec();
DeclarationNameInfo NameInfo = GetNameForDeclarator(D);
DeclarationName Name = NameInfo.getName();
SourceLocation Loc = NameInfo.getLoc();
// For anonymous bitfields, the location should point to the type.
if (Loc.isInvalid())
Loc = D.getSourceRange().getBegin();
Expr *BitWidth = static_cast<Expr*>(BW);
Expr *Init = static_cast<Expr*>(InitExpr);
assert(isa<CXXRecordDecl>(CurContext));
assert(!DS.isFriendSpecified());
bool isFunc = false;
if (D.isFunctionDeclarator())
isFunc = true;
else if (D.getNumTypeObjects() == 0 &&
D.getDeclSpec().getTypeSpecType() == DeclSpec::TST_typename) {
QualType TDType = GetTypeFromParser(DS.getRepAsType());
isFunc = TDType->isFunctionType();
}
// C++ 9.2p6: A member shall not be declared to have automatic storage
// duration (auto, register) or with the extern storage-class-specifier.
// C++ 7.1.1p8: The mutable specifier can be applied only to names of class
// data members and cannot be applied to names declared const or static,
// and cannot be applied to reference members.
switch (DS.getStorageClassSpec()) {
case DeclSpec::SCS_unspecified:
case DeclSpec::SCS_typedef:
case DeclSpec::SCS_static:
// FALL THROUGH.
break;
case DeclSpec::SCS_mutable:
if (isFunc) {
if (DS.getStorageClassSpecLoc().isValid())
Diag(DS.getStorageClassSpecLoc(), diag::err_mutable_function);
else
Diag(DS.getThreadSpecLoc(), diag::err_mutable_function);
// FIXME: It would be nicer if the keyword was ignored only for this
// declarator. Otherwise we could get follow-up errors.
D.getMutableDeclSpec().ClearStorageClassSpecs();
}
break;
default:
if (DS.getStorageClassSpecLoc().isValid())
Diag(DS.getStorageClassSpecLoc(),
diag::err_storageclass_invalid_for_member);
else
Diag(DS.getThreadSpecLoc(), diag::err_storageclass_invalid_for_member);
D.getMutableDeclSpec().ClearStorageClassSpecs();
}
bool isInstField = ((DS.getStorageClassSpec() == DeclSpec::SCS_unspecified ||
DS.getStorageClassSpec() == DeclSpec::SCS_mutable) &&
!isFunc);
Decl *Member;
if (isInstField) {
CXXScopeSpec &SS = D.getCXXScopeSpec();
if (SS.isSet() && !SS.isInvalid()) {
// The user provided a superfluous scope specifier inside a class
// definition:
//
// class X {
// int X::member;
// };
DeclContext *DC = 0;
if ((DC = computeDeclContext(SS, false)) && DC->Equals(CurContext))
Diag(D.getIdentifierLoc(), diag::warn_member_extra_qualification)
<< Name << FixItHint::CreateRemoval(SS.getRange());
else
Diag(D.getIdentifierLoc(), diag::err_member_qualification)
<< Name << SS.getRange();
SS.clear();
}
// FIXME: Check for template parameters!
// FIXME: Check that the name is an identifier!
Member = HandleField(S, cast<CXXRecordDecl>(CurContext), Loc, D, BitWidth,
AS);
assert(Member && "HandleField never returns null");
} else {
2010-08-21 17:40:31 +08:00
Member = HandleDeclarator(S, D, move(TemplateParameterLists), IsDefinition);
if (!Member) {
2010-08-21 17:40:31 +08:00
return 0;
}
// Non-instance-fields can't have a bitfield.
if (BitWidth) {
if (Member->isInvalidDecl()) {
// don't emit another diagnostic.
} else if (isa<VarDecl>(Member)) {
// C++ 9.6p3: A bit-field shall not be a static member.
// "static member 'A' cannot be a bit-field"
Diag(Loc, diag::err_static_not_bitfield)
<< Name << BitWidth->getSourceRange();
} else if (isa<TypedefDecl>(Member)) {
// "typedef member 'x' cannot be a bit-field"
Diag(Loc, diag::err_typedef_not_bitfield)
<< Name << BitWidth->getSourceRange();
} else {
// A function typedef ("typedef int f(); f a;").
// C++ 9.6p3: A bit-field shall have integral or enumeration type.
Diag(Loc, diag::err_not_integral_type_bitfield)
<< Name << cast<ValueDecl>(Member)->getType()
<< BitWidth->getSourceRange();
}
BitWidth = 0;
Member->setInvalidDecl();
}
Member->setAccess(AS);
// If we have declared a member function template, set the access of the
// templated declaration as well.
if (FunctionTemplateDecl *FunTmpl = dyn_cast<FunctionTemplateDecl>(Member))
FunTmpl->getTemplatedDecl()->setAccess(AS);
}
if (VS.isOverrideSpecified()) {
CXXMethodDecl *MD = dyn_cast<CXXMethodDecl>(Member);
if (!MD || !MD->isVirtual()) {
Diag(Member->getLocStart(),
diag::override_keyword_only_allowed_on_virtual_member_functions)
<< "override" << FixItHint::CreateRemoval(VS.getOverrideLoc());
} else
MD->addAttr(new (Context) OverrideAttr(VS.getOverrideLoc(), Context));
}
if (VS.isFinalSpecified()) {
CXXMethodDecl *MD = dyn_cast<CXXMethodDecl>(Member);
if (!MD || !MD->isVirtual()) {
Diag(Member->getLocStart(),
diag::override_keyword_only_allowed_on_virtual_member_functions)
<< "final" << FixItHint::CreateRemoval(VS.getFinalLoc());
} else
MD->addAttr(new (Context) FinalAttr(VS.getFinalLoc(), Context));
}
CheckOverrideControl(Member);
assert((Name || isInstField) && "No identifier for non-field ?");
if (Init)
AddInitializerToDecl(Member, Init, false);
if (Deleted) // FIXME: Source location is not very good.
2010-08-21 17:40:31 +08:00
SetDeclDeleted(Member, D.getSourceRange().getBegin());
if (isInstField)
FieldCollector->Add(cast<FieldDecl>(Member));
2010-08-21 17:40:31 +08:00
return Member;
}
/// \brief Find the direct and/or virtual base specifiers that
/// correspond to the given base type, for use in base initialization
/// within a constructor.
static bool FindBaseInitializer(Sema &SemaRef,
CXXRecordDecl *ClassDecl,
QualType BaseType,
const CXXBaseSpecifier *&DirectBaseSpec,
const CXXBaseSpecifier *&VirtualBaseSpec) {
// First, check for a direct base class.
DirectBaseSpec = 0;
for (CXXRecordDecl::base_class_const_iterator Base
= ClassDecl->bases_begin();
Base != ClassDecl->bases_end(); ++Base) {
if (SemaRef.Context.hasSameUnqualifiedType(BaseType, Base->getType())) {
// We found a direct base of this type. That's what we're
// initializing.
DirectBaseSpec = &*Base;
break;
}
}
// Check for a virtual base class.
// FIXME: We might be able to short-circuit this if we know in advance that
// there are no virtual bases.
VirtualBaseSpec = 0;
if (!DirectBaseSpec || !DirectBaseSpec->isVirtual()) {
// We haven't found a base yet; search the class hierarchy for a
// virtual base class.
CXXBasePaths Paths(/*FindAmbiguities=*/true, /*RecordPaths=*/true,
/*DetectVirtual=*/false);
if (SemaRef.IsDerivedFrom(SemaRef.Context.getTypeDeclType(ClassDecl),
BaseType, Paths)) {
for (CXXBasePaths::paths_iterator Path = Paths.begin();
Path != Paths.end(); ++Path) {
if (Path->back().Base->isVirtual()) {
VirtualBaseSpec = Path->back().Base;
break;
}
}
}
}
return DirectBaseSpec || VirtualBaseSpec;
}
/// ActOnMemInitializer - Handle a C++ member initializer.
MemInitResult
2010-08-21 17:40:31 +08:00
Sema::ActOnMemInitializer(Decl *ConstructorD,
Scope *S,
CXXScopeSpec &SS,
IdentifierInfo *MemberOrBase,
ParsedType TemplateTypeTy,
SourceLocation IdLoc,
SourceLocation LParenLoc,
ExprTy **Args, unsigned NumArgs,
SourceLocation RParenLoc,
SourceLocation EllipsisLoc) {
if (!ConstructorD)
return true;
AdjustDeclIfTemplate(ConstructorD);
CXXConstructorDecl *Constructor
2010-08-21 17:40:31 +08:00
= dyn_cast<CXXConstructorDecl>(ConstructorD);
if (!Constructor) {
// The user wrote a constructor initializer on a function that is
// not a C++ constructor. Ignore the error for now, because we may
// have more member initializers coming; we'll diagnose it just
// once in ActOnMemInitializers.
return true;
}
CXXRecordDecl *ClassDecl = Constructor->getParent();
// C++ [class.base.init]p2:
// Names in a mem-initializer-id are looked up in the scope of the
// constructor's class and, if not found in that scope, are looked
// up in the scope containing the constructor's definition.
// [Note: if the constructor's class contains a member with the
// same name as a direct or virtual base class of the class, a
// mem-initializer-id naming the member or base class and composed
// of a single identifier refers to the class member. A
// mem-initializer-id for the hidden base class may be specified
// using a qualified name. ]
if (!SS.getScopeRep() && !TemplateTypeTy) {
// Look for a member, first.
FieldDecl *Member = 0;
DeclContext::lookup_result Result
= ClassDecl->lookup(MemberOrBase);
if (Result.first != Result.second) {
Member = dyn_cast<FieldDecl>(*Result.first);
if (Member) {
if (EllipsisLoc.isValid())
Diag(EllipsisLoc, diag::err_pack_expansion_member_init)
<< MemberOrBase << SourceRange(IdLoc, RParenLoc);
return BuildMemberInitializer(Member, (Expr**)Args, NumArgs, IdLoc,
LParenLoc, RParenLoc);
}
// Handle anonymous union case.
if (IndirectFieldDecl* IndirectField
= dyn_cast<IndirectFieldDecl>(*Result.first)) {
if (EllipsisLoc.isValid())
Diag(EllipsisLoc, diag::err_pack_expansion_member_init)
<< MemberOrBase << SourceRange(IdLoc, RParenLoc);
return BuildMemberInitializer(IndirectField, (Expr**)Args,
NumArgs, IdLoc,
LParenLoc, RParenLoc);
}
}
}
// It didn't name a member, so see if it names a class.
QualType BaseType;
TypeSourceInfo *TInfo = 0;
if (TemplateTypeTy) {
BaseType = GetTypeFromParser(TemplateTypeTy, &TInfo);
} else {
LookupResult R(*this, MemberOrBase, IdLoc, LookupOrdinaryName);
LookupParsedName(R, S, &SS);
TypeDecl *TyD = R.getAsSingle<TypeDecl>();
if (!TyD) {
if (R.isAmbiguous()) return true;
// We don't want access-control diagnostics here.
R.suppressDiagnostics();
if (SS.isSet() && isDependentScopeSpecifier(SS)) {
bool NotUnknownSpecialization = false;
DeclContext *DC = computeDeclContext(SS, false);
if (CXXRecordDecl *Record = dyn_cast_or_null<CXXRecordDecl>(DC))
NotUnknownSpecialization = !Record->hasAnyDependentBases();
if (!NotUnknownSpecialization) {
// When the scope specifier can refer to a member of an unknown
// specialization, we take it as a type name.
BaseType = CheckTypenameType(ETK_None,
(NestedNameSpecifier *)SS.getScopeRep(),
*MemberOrBase, SourceLocation(),
SS.getRange(), IdLoc);
if (BaseType.isNull())
return true;
R.clear();
R.setLookupName(MemberOrBase);
}
}
// If no results were found, try to correct typos.
if (R.empty() && BaseType.isNull() &&
CorrectTypo(R, S, &SS, ClassDecl, 0, CTC_NoKeywords) &&
R.isSingleResult()) {
if (FieldDecl *Member = R.getAsSingle<FieldDecl>()) {
if (Member->getDeclContext()->getRedeclContext()->Equals(ClassDecl)) {
// We have found a non-static data member with a similar
// name to what was typed; complain and initialize that
// member.
Diag(R.getNameLoc(), diag::err_mem_init_not_member_or_class_suggest)
<< MemberOrBase << true << R.getLookupName()
<< FixItHint::CreateReplacement(R.getNameLoc(),
R.getLookupName().getAsString());
Diag(Member->getLocation(), diag::note_previous_decl)
<< Member->getDeclName();
return BuildMemberInitializer(Member, (Expr**)Args, NumArgs, IdLoc,
LParenLoc, RParenLoc);
}
} else if (TypeDecl *Type = R.getAsSingle<TypeDecl>()) {
const CXXBaseSpecifier *DirectBaseSpec;
const CXXBaseSpecifier *VirtualBaseSpec;
if (FindBaseInitializer(*this, ClassDecl,
Context.getTypeDeclType(Type),
DirectBaseSpec, VirtualBaseSpec)) {
// We have found a direct or virtual base class with a
// similar name to what was typed; complain and initialize
// that base class.
Diag(R.getNameLoc(), diag::err_mem_init_not_member_or_class_suggest)
<< MemberOrBase << false << R.getLookupName()
<< FixItHint::CreateReplacement(R.getNameLoc(),
R.getLookupName().getAsString());
const CXXBaseSpecifier *BaseSpec = DirectBaseSpec? DirectBaseSpec
: VirtualBaseSpec;
Diag(BaseSpec->getSourceRange().getBegin(),
diag::note_base_class_specified_here)
<< BaseSpec->getType()
<< BaseSpec->getSourceRange();
TyD = Type;
}
}
}
if (!TyD && BaseType.isNull()) {
Diag(IdLoc, diag::err_mem_init_not_member_or_class)
<< MemberOrBase << SourceRange(IdLoc, RParenLoc);
return true;
}
}
if (BaseType.isNull()) {
BaseType = Context.getTypeDeclType(TyD);
if (SS.isSet()) {
NestedNameSpecifier *Qualifier =
static_cast<NestedNameSpecifier*>(SS.getScopeRep());
// FIXME: preserve source range information
BaseType = Context.getElaboratedType(ETK_None, Qualifier, BaseType);
}
}
}
if (!TInfo)
TInfo = Context.getTrivialTypeSourceInfo(BaseType, IdLoc);
return BuildBaseInitializer(BaseType, TInfo, (Expr **)Args, NumArgs,
LParenLoc, RParenLoc, ClassDecl, EllipsisLoc);
}
/// Checks an initializer expression for use of uninitialized fields, such as
/// containing the field that is being initialized. Returns true if there is an
/// uninitialized field was used an updates the SourceLocation parameter; false
/// otherwise.
static bool InitExprContainsUninitializedFields(const Stmt *S,
const ValueDecl *LhsField,
SourceLocation *L) {
assert(isa<FieldDecl>(LhsField) || isa<IndirectFieldDecl>(LhsField));
if (isa<CallExpr>(S)) {
// Do not descend into function calls or constructors, as the use
// of an uninitialized field may be valid. One would have to inspect
// the contents of the function/ctor to determine if it is safe or not.
// i.e. Pass-by-value is never safe, but pass-by-reference and pointers
// may be safe, depending on what the function/ctor does.
return false;
}
if (const MemberExpr *ME = dyn_cast<MemberExpr>(S)) {
const NamedDecl *RhsField = ME->getMemberDecl();
if (const VarDecl *VD = dyn_cast<VarDecl>(RhsField)) {
// The member expression points to a static data member.
assert(VD->isStaticDataMember() &&
"Member points to non-static data member!");
(void)VD;
return false;
}
if (isa<EnumConstantDecl>(RhsField)) {
// The member expression points to an enum.
return false;
}
if (RhsField == LhsField) {
// Initializing a field with itself. Throw a warning.
// But wait; there are exceptions!
// Exception #1: The field may not belong to this record.
// e.g. Foo(const Foo& rhs) : A(rhs.A) {}
const Expr *base = ME->getBase();
if (base != NULL && !isa<CXXThisExpr>(base->IgnoreParenCasts())) {
// Even though the field matches, it does not belong to this record.
return false;
}
// None of the exceptions triggered; return true to indicate an
// uninitialized field was used.
*L = ME->getMemberLoc();
return true;
}
} else if (isa<SizeOfAlignOfExpr>(S)) {
// sizeof/alignof doesn't reference contents, do not warn.
return false;
} else if (const UnaryOperator *UOE = dyn_cast<UnaryOperator>(S)) {
// address-of doesn't reference contents (the pointer may be dereferenced
// in the same expression but it would be rare; and weird).
if (UOE->getOpcode() == UO_AddrOf)
return false;
}
for (Stmt::const_child_range it = S->children(); it; ++it) {
if (!*it) {
// An expression such as 'member(arg ?: "")' may trigger this.
continue;
}
if (InitExprContainsUninitializedFields(*it, LhsField, L))
return true;
}
return false;
}
MemInitResult
Sema::BuildMemberInitializer(ValueDecl *Member, Expr **Args,
unsigned NumArgs, SourceLocation IdLoc,
SourceLocation LParenLoc,
SourceLocation RParenLoc) {
FieldDecl *DirectMember = dyn_cast<FieldDecl>(Member);
IndirectFieldDecl *IndirectMember = dyn_cast<IndirectFieldDecl>(Member);
assert((DirectMember || IndirectMember) &&
"Member must be a FieldDecl or IndirectFieldDecl");
if (Member->isInvalidDecl())
return true;
// Diagnose value-uses of fields to initialize themselves, e.g.
// foo(foo)
// where foo is not also a parameter to the constructor.
// TODO: implement -Wuninitialized and fold this into that framework.
for (unsigned i = 0; i < NumArgs; ++i) {
SourceLocation L;
if (InitExprContainsUninitializedFields(Args[i], Member, &L)) {
// FIXME: Return true in the case when other fields are used before being
// uninitialized. For example, let this field be the i'th field. When
// initializing the i'th field, throw a warning if any of the >= i'th
// fields are used, as they are not yet initialized.
// Right now we are only handling the case where the i'th field uses
// itself in its initializer.
Diag(L, diag::warn_field_is_uninit);
}
}
bool HasDependentArg = false;
for (unsigned i = 0; i < NumArgs; i++)
HasDependentArg |= Args[i]->isTypeDependent();
Expr *Init;
if (Member->getType()->isDependentType() || HasDependentArg) {
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
// Can't check initialization for a member of dependent type or when
// any of the arguments are type-dependent expressions.
Init = new (Context) ParenListExpr(Context, LParenLoc, Args, NumArgs,
RParenLoc);
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
// Erase any temporaries within this evaluation context; we're not
// going to track them in the AST, since we'll be rebuilding the
// ASTs during template instantiation.
ExprTemporaries.erase(
ExprTemporaries.begin() + ExprEvalContexts.back().NumTemporaries,
ExprTemporaries.end());
} else {
// Initialize the member.
InitializedEntity MemberEntity =
DirectMember ? InitializedEntity::InitializeMember(DirectMember, 0)
: InitializedEntity::InitializeMember(IndirectMember, 0);
InitializationKind Kind =
InitializationKind::CreateDirect(IdLoc, LParenLoc, RParenLoc);
InitializationSequence InitSeq(*this, MemberEntity, Kind, Args, NumArgs);
ExprResult MemberInit =
InitSeq.Perform(*this, MemberEntity, Kind,
MultiExprArg(*this, Args, NumArgs), 0);
if (MemberInit.isInvalid())
return true;
CheckImplicitConversions(MemberInit.get(), LParenLoc);
// C++0x [class.base.init]p7:
// The initialization of each base and member constitutes a
// full-expression.
MemberInit = MaybeCreateExprWithCleanups(MemberInit);
if (MemberInit.isInvalid())
return true;
// If we are in a dependent context, template instantiation will
// perform this type-checking again. Just save the arguments that we
// received in a ParenListExpr.
// FIXME: This isn't quite ideal, since our ASTs don't capture all
// of the information that we have about the member
// initializer. However, deconstructing the ASTs is a dicey process,
// and this approach is far more likely to get the corner cases right.
if (CurContext->isDependentContext())
Init = new (Context) ParenListExpr(Context, LParenLoc, Args, NumArgs,
RParenLoc);
else
Init = MemberInit.get();
}
if (DirectMember) {
return new (Context) CXXCtorInitializer(Context, DirectMember,
IdLoc, LParenLoc, Init,
RParenLoc);
} else {
return new (Context) CXXCtorInitializer(Context, IndirectMember,
IdLoc, LParenLoc, Init,
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
RParenLoc);
}
}
MemInitResult
Sema::BuildDelegatingInitializer(TypeSourceInfo *TInfo,
Expr **Args, unsigned NumArgs,
SourceLocation LParenLoc,
SourceLocation RParenLoc,
CXXRecordDecl *ClassDecl,
SourceLocation EllipsisLoc) {
SourceLocation Loc = TInfo->getTypeLoc().getLocalSourceRange().getBegin();
if (!LangOpts.CPlusPlus0x)
return Diag(Loc, diag::err_delegation_0x_only)
<< TInfo->getTypeLoc().getLocalSourceRange();
return Diag(Loc, diag::err_delegation_unimplemented)
<< TInfo->getTypeLoc().getLocalSourceRange();
}
MemInitResult
Sema::BuildBaseInitializer(QualType BaseType, TypeSourceInfo *BaseTInfo,
Expr **Args, unsigned NumArgs,
SourceLocation LParenLoc, SourceLocation RParenLoc,
CXXRecordDecl *ClassDecl,
SourceLocation EllipsisLoc) {
bool HasDependentArg = false;
for (unsigned i = 0; i < NumArgs; i++)
HasDependentArg |= Args[i]->isTypeDependent();
SourceLocation BaseLoc
= BaseTInfo->getTypeLoc().getLocalSourceRange().getBegin();
if (!BaseType->isDependentType() && !BaseType->isRecordType())
return Diag(BaseLoc, diag::err_base_init_does_not_name_class)
<< BaseType << BaseTInfo->getTypeLoc().getLocalSourceRange();
// C++ [class.base.init]p2:
// [...] Unless the mem-initializer-id names a nonstatic data
// member of the constructor's class or a direct or virtual base
// of that class, the mem-initializer is ill-formed. A
// mem-initializer-list can initialize a base class using any
// name that denotes that base class type.
bool Dependent = BaseType->isDependentType() || HasDependentArg;
if (EllipsisLoc.isValid()) {
// This is a pack expansion.
if (!BaseType->containsUnexpandedParameterPack()) {
Diag(EllipsisLoc, diag::err_pack_expansion_without_parameter_packs)
<< SourceRange(BaseLoc, RParenLoc);
EllipsisLoc = SourceLocation();
}
} else {
// Check for any unexpanded parameter packs.
if (DiagnoseUnexpandedParameterPack(BaseLoc, BaseTInfo, UPPC_Initializer))
return true;
for (unsigned I = 0; I != NumArgs; ++I)
if (DiagnoseUnexpandedParameterPack(Args[I]))
return true;
}
// Check for direct and virtual base classes.
const CXXBaseSpecifier *DirectBaseSpec = 0;
const CXXBaseSpecifier *VirtualBaseSpec = 0;
if (!Dependent) {
if (Context.hasSameUnqualifiedType(QualType(ClassDecl->getTypeForDecl(),0),
BaseType))
return BuildDelegatingInitializer(BaseTInfo, Args, NumArgs,
LParenLoc, RParenLoc, ClassDecl,
EllipsisLoc);
FindBaseInitializer(*this, ClassDecl, BaseType, DirectBaseSpec,
VirtualBaseSpec);
// C++ [base.class.init]p2:
// Unless the mem-initializer-id names a nonstatic data member of the
// constructor's class or a direct or virtual base of that class, the
// mem-initializer is ill-formed.
if (!DirectBaseSpec && !VirtualBaseSpec) {
// If the class has any dependent bases, then it's possible that
// one of those types will resolve to the same type as
// BaseType. Therefore, just treat this as a dependent base
// class initialization. FIXME: Should we try to check the
// initialization anyway? It seems odd.
if (ClassDecl->hasAnyDependentBases())
Dependent = true;
else
return Diag(BaseLoc, diag::err_not_direct_base_or_virtual)
<< BaseType << Context.getTypeDeclType(ClassDecl)
<< BaseTInfo->getTypeLoc().getLocalSourceRange();
}
}
if (Dependent) {
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
// Can't check initialization for a base of dependent type or when
// any of the arguments are type-dependent expressions.
ExprResult BaseInit
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
= Owned(new (Context) ParenListExpr(Context, LParenLoc, Args, NumArgs,
RParenLoc));
// Erase any temporaries within this evaluation context; we're not
// going to track them in the AST, since we'll be rebuilding the
// ASTs during template instantiation.
ExprTemporaries.erase(
ExprTemporaries.begin() + ExprEvalContexts.back().NumTemporaries,
ExprTemporaries.end());
return new (Context) CXXCtorInitializer(Context, BaseTInfo,
/*IsVirtual=*/false,
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
LParenLoc,
BaseInit.takeAs<Expr>(),
RParenLoc,
EllipsisLoc);
}
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
// C++ [base.class.init]p2:
// If a mem-initializer-id is ambiguous because it designates both
// a direct non-virtual base class and an inherited virtual base
// class, the mem-initializer is ill-formed.
if (DirectBaseSpec && VirtualBaseSpec)
return Diag(BaseLoc, diag::err_base_init_direct_and_virtual)
<< BaseType << BaseTInfo->getTypeLoc().getLocalSourceRange();
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
CXXBaseSpecifier *BaseSpec
= const_cast<CXXBaseSpecifier *>(DirectBaseSpec);
if (!BaseSpec)
BaseSpec = const_cast<CXXBaseSpecifier *>(VirtualBaseSpec);
// Initialize the base.
InitializedEntity BaseEntity =
InitializedEntity::InitializeBase(Context, BaseSpec, VirtualBaseSpec);
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
InitializationKind Kind =
InitializationKind::CreateDirect(BaseLoc, LParenLoc, RParenLoc);
InitializationSequence InitSeq(*this, BaseEntity, Kind, Args, NumArgs);
ExprResult BaseInit =
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
InitSeq.Perform(*this, BaseEntity, Kind,
MultiExprArg(*this, Args, NumArgs), 0);
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
if (BaseInit.isInvalid())
return true;
CheckImplicitConversions(BaseInit.get(), LParenLoc);
2009-11-14 03:21:49 +08:00
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
// C++0x [class.base.init]p7:
// The initialization of each base and member constitutes a
// full-expression.
BaseInit = MaybeCreateExprWithCleanups(BaseInit);
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
if (BaseInit.isInvalid())
return true;
// If we are in a dependent context, template instantiation will
// perform this type-checking again. Just save the arguments that we
// received in a ParenListExpr.
// FIXME: This isn't quite ideal, since our ASTs don't capture all
// of the information that we have about the base
// initializer. However, deconstructing the ASTs is a dicey process,
// and this approach is far more likely to get the corner cases right.
if (CurContext->isDependentContext()) {
ExprResult Init
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
= Owned(new (Context) ParenListExpr(Context, LParenLoc, Args, NumArgs,
RParenLoc));
return new (Context) CXXCtorInitializer(Context, BaseTInfo,
BaseSpec->isVirtual(),
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
LParenLoc,
Init.takeAs<Expr>(),
RParenLoc,
EllipsisLoc);
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
}
return new (Context) CXXCtorInitializer(Context, BaseTInfo,
BaseSpec->isVirtual(),
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
LParenLoc,
BaseInit.takeAs<Expr>(),
RParenLoc,
EllipsisLoc);
}
/// ImplicitInitializerKind - How an implicit base or member initializer should
/// initialize its base or member.
enum ImplicitInitializerKind {
IIK_Default,
IIK_Copy,
IIK_Move
};
static bool
BuildImplicitBaseInitializer(Sema &SemaRef, CXXConstructorDecl *Constructor,
ImplicitInitializerKind ImplicitInitKind,
CXXBaseSpecifier *BaseSpec,
bool IsInheritedVirtualBase,
CXXCtorInitializer *&CXXBaseInit) {
InitializedEntity InitEntity
= InitializedEntity::InitializeBase(SemaRef.Context, BaseSpec,
IsInheritedVirtualBase);
ExprResult BaseInit;
switch (ImplicitInitKind) {
case IIK_Default: {
InitializationKind InitKind
= InitializationKind::CreateDefault(Constructor->getLocation());
InitializationSequence InitSeq(SemaRef, InitEntity, InitKind, 0, 0);
BaseInit = InitSeq.Perform(SemaRef, InitEntity, InitKind,
MultiExprArg(SemaRef, 0, 0));
break;
}
case IIK_Copy: {
ParmVarDecl *Param = Constructor->getParamDecl(0);
QualType ParamType = Param->getType().getNonReferenceType();
Expr *CopyCtorArg =
DeclRefExpr::Create(SemaRef.Context, 0, SourceRange(), Param,
Constructor->getLocation(), ParamType,
VK_LValue, 0);
// Cast to the base class to avoid ambiguities.
QualType ArgTy =
SemaRef.Context.getQualifiedType(BaseSpec->getType().getUnqualifiedType(),
ParamType.getQualifiers());
CXXCastPath BasePath;
BasePath.push_back(BaseSpec);
SemaRef.ImpCastExprToType(CopyCtorArg, ArgTy,
CK_UncheckedDerivedToBase,
VK_LValue, &BasePath);
InitializationKind InitKind
= InitializationKind::CreateDirect(Constructor->getLocation(),
SourceLocation(), SourceLocation());
InitializationSequence InitSeq(SemaRef, InitEntity, InitKind,
&CopyCtorArg, 1);
BaseInit = InitSeq.Perform(SemaRef, InitEntity, InitKind,
MultiExprArg(&CopyCtorArg, 1));
break;
}
case IIK_Move:
assert(false && "Unhandled initializer kind!");
}
BaseInit = SemaRef.MaybeCreateExprWithCleanups(BaseInit);
if (BaseInit.isInvalid())
return true;
CXXBaseInit =
new (SemaRef.Context) CXXCtorInitializer(SemaRef.Context,
SemaRef.Context.getTrivialTypeSourceInfo(BaseSpec->getType(),
SourceLocation()),
BaseSpec->isVirtual(),
SourceLocation(),
BaseInit.takeAs<Expr>(),
SourceLocation(),
SourceLocation());
return false;
}
static bool
BuildImplicitMemberInitializer(Sema &SemaRef, CXXConstructorDecl *Constructor,
ImplicitInitializerKind ImplicitInitKind,
FieldDecl *Field,
CXXCtorInitializer *&CXXMemberInit) {
if (Field->isInvalidDecl())
return true;
SourceLocation Loc = Constructor->getLocation();
if (ImplicitInitKind == IIK_Copy) {
ParmVarDecl *Param = Constructor->getParamDecl(0);
QualType ParamType = Param->getType().getNonReferenceType();
Expr *MemberExprBase =
DeclRefExpr::Create(SemaRef.Context, 0, SourceRange(), Param,
Loc, ParamType, VK_LValue, 0);
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
// Build a reference to this field within the parameter.
CXXScopeSpec SS;
LookupResult MemberLookup(SemaRef, Field->getDeclName(), Loc,
Sema::LookupMemberName);
MemberLookup.addDecl(Field, AS_public);
MemberLookup.resolveKind();
ExprResult CopyCtorArg
= SemaRef.BuildMemberReferenceExpr(MemberExprBase,
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
ParamType, Loc,
/*IsArrow=*/false,
SS,
/*FirstQualifierInScope=*/0,
MemberLookup,
/*TemplateArgs=*/0);
if (CopyCtorArg.isInvalid())
return true;
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
// When the field we are copying is an array, create index variables for
// each dimension of the array. We use these index variables to subscript
// the source array, and other clients (e.g., CodeGen) will perform the
// necessary iteration with these index variables.
llvm::SmallVector<VarDecl *, 4> IndexVariables;
QualType BaseType = Field->getType();
QualType SizeType = SemaRef.Context.getSizeType();
while (const ConstantArrayType *Array
= SemaRef.Context.getAsConstantArrayType(BaseType)) {
// Create the iteration variable for this array index.
IdentifierInfo *IterationVarName = 0;
{
llvm::SmallString<8> Str;
llvm::raw_svector_ostream OS(Str);
OS << "__i" << IndexVariables.size();
IterationVarName = &SemaRef.Context.Idents.get(OS.str());
}
VarDecl *IterationVar
= VarDecl::Create(SemaRef.Context, SemaRef.CurContext, Loc,
IterationVarName, SizeType,
SemaRef.Context.getTrivialTypeSourceInfo(SizeType, Loc),
SC_None, SC_None);
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
IndexVariables.push_back(IterationVar);
// Create a reference to the iteration variable.
ExprResult IterationVarRef
= SemaRef.BuildDeclRefExpr(IterationVar, SizeType, VK_RValue, Loc);
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
assert(!IterationVarRef.isInvalid() &&
"Reference to invented variable cannot fail!");
// Subscript the array with this iteration variable.
CopyCtorArg = SemaRef.CreateBuiltinArraySubscriptExpr(CopyCtorArg.take(),
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
Loc,
IterationVarRef.take(),
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
Loc);
if (CopyCtorArg.isInvalid())
return true;
BaseType = Array->getElementType();
}
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
// Construct the entity that we will be initializing. For an array, this
// will be first element in the array, which may require several levels
// of array-subscript entities.
llvm::SmallVector<InitializedEntity, 4> Entities;
Entities.reserve(1 + IndexVariables.size());
Entities.push_back(InitializedEntity::InitializeMember(Field));
for (unsigned I = 0, N = IndexVariables.size(); I != N; ++I)
Entities.push_back(InitializedEntity::InitializeElement(SemaRef.Context,
0,
Entities.back()));
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
// Direct-initialize to use the copy constructor.
InitializationKind InitKind =
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
InitializationKind::CreateDirect(Loc, SourceLocation(), SourceLocation());
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
Expr *CopyCtorArgE = CopyCtorArg.takeAs<Expr>();
InitializationSequence InitSeq(SemaRef, Entities.back(), InitKind,
&CopyCtorArgE, 1);
ExprResult MemberInit
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
= InitSeq.Perform(SemaRef, Entities.back(), InitKind,
MultiExprArg(&CopyCtorArgE, 1));
MemberInit = SemaRef.MaybeCreateExprWithCleanups(MemberInit);
if (MemberInit.isInvalid())
return true;
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
CXXMemberInit
= CXXCtorInitializer::Create(SemaRef.Context, Field, Loc, Loc,
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
MemberInit.takeAs<Expr>(), Loc,
IndexVariables.data(),
IndexVariables.size());
return false;
}
assert(ImplicitInitKind == IIK_Default && "Unhandled implicit init kind!");
QualType FieldBaseElementType =
SemaRef.Context.getBaseElementType(Field->getType());
if (FieldBaseElementType->isRecordType()) {
InitializedEntity InitEntity = InitializedEntity::InitializeMember(Field);
InitializationKind InitKind =
InitializationKind::CreateDefault(Loc);
InitializationSequence InitSeq(SemaRef, InitEntity, InitKind, 0, 0);
ExprResult MemberInit =
InitSeq.Perform(SemaRef, InitEntity, InitKind, MultiExprArg());
MemberInit = SemaRef.MaybeCreateExprWithCleanups(MemberInit);
if (MemberInit.isInvalid())
return true;
CXXMemberInit =
new (SemaRef.Context) CXXCtorInitializer(SemaRef.Context,
Field, Loc, Loc,
MemberInit.get(),
Loc);
return false;
}
if (FieldBaseElementType->isReferenceType()) {
SemaRef.Diag(Constructor->getLocation(),
diag::err_uninitialized_member_in_ctor)
<< (int)Constructor->isImplicit()
<< SemaRef.Context.getTagDeclType(Constructor->getParent())
<< 0 << Field->getDeclName();
SemaRef.Diag(Field->getLocation(), diag::note_declared_at);
return true;
}
if (FieldBaseElementType.isConstQualified()) {
SemaRef.Diag(Constructor->getLocation(),
diag::err_uninitialized_member_in_ctor)
<< (int)Constructor->isImplicit()
<< SemaRef.Context.getTagDeclType(Constructor->getParent())
<< 1 << Field->getDeclName();
SemaRef.Diag(Field->getLocation(), diag::note_declared_at);
return true;
}
// Nothing to initialize.
CXXMemberInit = 0;
return false;
}
namespace {
struct BaseAndFieldInfo {
Sema &S;
CXXConstructorDecl *Ctor;
bool AnyErrorsInInits;
ImplicitInitializerKind IIK;
llvm::DenseMap<const void *, CXXCtorInitializer*> AllBaseFields;
llvm::SmallVector<CXXCtorInitializer*, 8> AllToInit;
BaseAndFieldInfo(Sema &S, CXXConstructorDecl *Ctor, bool ErrorsInInits)
: S(S), Ctor(Ctor), AnyErrorsInInits(ErrorsInInits) {
// FIXME: Handle implicit move constructors.
if (Ctor->isImplicit() && Ctor->isCopyConstructor())
IIK = IIK_Copy;
else
IIK = IIK_Default;
}
};
}
static bool CollectFieldInitializer(BaseAndFieldInfo &Info,
FieldDecl *Top, FieldDecl *Field) {
// Overwhelmingly common case: we have a direct initializer for this field.
if (CXXCtorInitializer *Init = Info.AllBaseFields.lookup(Field)) {
Info.AllToInit.push_back(Init);
return false;
}
if (Info.IIK == IIK_Default && Field->isAnonymousStructOrUnion()) {
const RecordType *FieldClassType = Field->getType()->getAs<RecordType>();
assert(FieldClassType && "anonymous struct/union without record type");
CXXRecordDecl *FieldClassDecl
= cast<CXXRecordDecl>(FieldClassType->getDecl());
// Even though union members never have non-trivial default
// constructions in C++03, we still build member initializers for aggregate
// record types which can be union members, and C++0x allows non-trivial
// default constructors for union members, so we ensure that only one
// member is initialized for these.
if (FieldClassDecl->isUnion()) {
// First check for an explicit initializer for one field.
for (RecordDecl::field_iterator FA = FieldClassDecl->field_begin(),
EA = FieldClassDecl->field_end(); FA != EA; FA++) {
if (CXXCtorInitializer *Init = Info.AllBaseFields.lookup(*FA)) {
Info.AllToInit.push_back(Init);
// Once we've initialized a field of an anonymous union, the union
// field in the class is also initialized, so exit immediately.
return false;
} else if ((*FA)->isAnonymousStructOrUnion()) {
if (CollectFieldInitializer(Info, Top, *FA))
return true;
}
}
// Fallthrough and construct a default initializer for the union as
// a whole, which can call its default constructor if such a thing exists
// (C++0x perhaps). FIXME: It's not clear that this is the correct
// behavior going forward with C++0x, when anonymous unions there are
// finalized, we should revisit this.
} else {
// For structs, we simply descend through to initialize all members where
// necessary.
for (RecordDecl::field_iterator FA = FieldClassDecl->field_begin(),
EA = FieldClassDecl->field_end(); FA != EA; FA++) {
if (CollectFieldInitializer(Info, Top, *FA))
return true;
}
}
}
// Don't try to build an implicit initializer if there were semantic
// errors in any of the initializers (and therefore we might be
// missing some that the user actually wrote).
if (Info.AnyErrorsInInits)
return false;
CXXCtorInitializer *Init = 0;
if (BuildImplicitMemberInitializer(Info.S, Info.Ctor, Info.IIK, Field, Init))
return true;
if (Init)
Info.AllToInit.push_back(Init);
return false;
}
bool
Sema::SetCtorInitializers(CXXConstructorDecl *Constructor,
CXXCtorInitializer **Initializers,
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
unsigned NumInitializers,
bool AnyErrors) {
if (Constructor->getDeclContext()->isDependentContext()) {
// Just store the initializers as written, they will be checked during
// instantiation.
if (NumInitializers > 0) {
Constructor->setNumCtorInitializers(NumInitializers);
CXXCtorInitializer **baseOrMemberInitializers =
new (Context) CXXCtorInitializer*[NumInitializers];
memcpy(baseOrMemberInitializers, Initializers,
NumInitializers * sizeof(CXXCtorInitializer*));
Constructor->setCtorInitializers(baseOrMemberInitializers);
}
return false;
}
BaseAndFieldInfo Info(*this, Constructor, AnyErrors);
// We need to build the initializer AST according to order of construction
// and not what user specified in the Initializers list.
CXXRecordDecl *ClassDecl = Constructor->getParent()->getDefinition();
if (!ClassDecl)
return true;
bool HadError = false;
for (unsigned i = 0; i < NumInitializers; i++) {
CXXCtorInitializer *Member = Initializers[i];
if (Member->isBaseInitializer())
Info.AllBaseFields[Member->getBaseClass()->getAs<RecordType>()] = Member;
else
Info.AllBaseFields[Member->getAnyMember()] = Member;
}
// Keep track of the direct virtual bases.
llvm::SmallPtrSet<CXXBaseSpecifier *, 16> DirectVBases;
for (CXXRecordDecl::base_class_iterator I = ClassDecl->bases_begin(),
E = ClassDecl->bases_end(); I != E; ++I) {
if (I->isVirtual())
DirectVBases.insert(I);
}
// Push virtual bases before others.
for (CXXRecordDecl::base_class_iterator VBase = ClassDecl->vbases_begin(),
E = ClassDecl->vbases_end(); VBase != E; ++VBase) {
if (CXXCtorInitializer *Value
= Info.AllBaseFields.lookup(VBase->getType()->getAs<RecordType>())) {
Info.AllToInit.push_back(Value);
} else if (!AnyErrors) {
bool IsInheritedVirtualBase = !DirectVBases.count(VBase);
CXXCtorInitializer *CXXBaseInit;
if (BuildImplicitBaseInitializer(*this, Constructor, Info.IIK,
VBase, IsInheritedVirtualBase,
CXXBaseInit)) {
HadError = true;
continue;
}
Info.AllToInit.push_back(CXXBaseInit);
}
}
// Non-virtual bases.
for (CXXRecordDecl::base_class_iterator Base = ClassDecl->bases_begin(),
E = ClassDecl->bases_end(); Base != E; ++Base) {
// Virtuals are in the virtual base list and already constructed.
if (Base->isVirtual())
continue;
if (CXXCtorInitializer *Value
= Info.AllBaseFields.lookup(Base->getType()->getAs<RecordType>())) {
Info.AllToInit.push_back(Value);
} else if (!AnyErrors) {
CXXCtorInitializer *CXXBaseInit;
if (BuildImplicitBaseInitializer(*this, Constructor, Info.IIK,
Base, /*IsInheritedVirtualBase=*/false,
CXXBaseInit)) {
HadError = true;
continue;
}
Info.AllToInit.push_back(CXXBaseInit);
}
}
// Fields.
for (CXXRecordDecl::field_iterator Field = ClassDecl->field_begin(),
E = ClassDecl->field_end(); Field != E; ++Field) {
if ((*Field)->getType()->isIncompleteArrayType()) {
assert(ClassDecl->hasFlexibleArrayMember() &&
"Incomplete array type is not valid");
continue;
}
if (CollectFieldInitializer(Info, *Field, *Field))
HadError = true;
}
NumInitializers = Info.AllToInit.size();
if (NumInitializers > 0) {
Constructor->setNumCtorInitializers(NumInitializers);
CXXCtorInitializer **baseOrMemberInitializers =
new (Context) CXXCtorInitializer*[NumInitializers];
memcpy(baseOrMemberInitializers, Info.AllToInit.data(),
NumInitializers * sizeof(CXXCtorInitializer*));
Constructor->setCtorInitializers(baseOrMemberInitializers);
// Constructors implicitly reference the base and member
// destructors.
MarkBaseAndMemberDestructorsReferenced(Constructor->getLocation(),
Constructor->getParent());
}
return HadError;
}
static void *GetKeyForTopLevelField(FieldDecl *Field) {
// For anonymous unions, use the class declaration as the key.
if (const RecordType *RT = Field->getType()->getAs<RecordType>()) {
if (RT->getDecl()->isAnonymousStructOrUnion())
return static_cast<void *>(RT->getDecl());
}
return static_cast<void *>(Field);
}
static void *GetKeyForBase(ASTContext &Context, QualType BaseType) {
return const_cast<Type*>(Context.getCanonicalType(BaseType).getTypePtr());
}
static void *GetKeyForMember(ASTContext &Context,
CXXCtorInitializer *Member) {
if (!Member->isAnyMemberInitializer())
return GetKeyForBase(Context, QualType(Member->getBaseClass(), 0));
2010-03-30 23:39:27 +08:00
// For fields injected into the class via declaration of an anonymous union,
// use its anonymous union class declaration as the unique key.
FieldDecl *Field = Member->getAnyMember();
// If the field is a member of an anonymous struct or union, our key
// is the anonymous record decl that's a direct child of the class.
RecordDecl *RD = Field->getParent();
if (RD->isAnonymousStructOrUnion()) {
while (true) {
RecordDecl *Parent = cast<RecordDecl>(RD->getDeclContext());
if (Parent->isAnonymousStructOrUnion())
RD = Parent;
else
break;
}
return static_cast<void *>(RD);
}
2010-03-30 23:39:27 +08:00
return static_cast<void *>(Field);
}
2010-04-02 11:37:03 +08:00
static void
DiagnoseBaseOrMemInitializerOrder(Sema &SemaRef,
2010-04-02 11:38:04 +08:00
const CXXConstructorDecl *Constructor,
CXXCtorInitializer **Inits,
unsigned NumInits) {
if (Constructor->getDeclContext()->isDependentContext())
return;
// Don't check initializers order unless the warning is enabled at the
// location of at least one initializer.
bool ShouldCheckOrder = false;
for (unsigned InitIndex = 0; InitIndex != NumInits; ++InitIndex) {
CXXCtorInitializer *Init = Inits[InitIndex];
if (SemaRef.Diags.getDiagnosticLevel(diag::warn_initializer_out_of_order,
Init->getSourceLocation())
!= Diagnostic::Ignored) {
ShouldCheckOrder = true;
break;
}
}
if (!ShouldCheckOrder)
return;
2010-04-02 11:37:03 +08:00
// Build the list of bases and members in the order that they'll
// actually be initialized. The explicit initializers should be in
// this same order but may be missing things.
llvm::SmallVector<const void*, 32> IdealInitKeys;
2010-04-02 11:38:04 +08:00
const CXXRecordDecl *ClassDecl = Constructor->getParent();
// 1. Virtual bases.
2010-04-02 11:38:04 +08:00
for (CXXRecordDecl::base_class_const_iterator VBase =
ClassDecl->vbases_begin(),
E = ClassDecl->vbases_end(); VBase != E; ++VBase)
IdealInitKeys.push_back(GetKeyForBase(SemaRef.Context, VBase->getType()));
// 2. Non-virtual bases.
2010-04-02 11:38:04 +08:00
for (CXXRecordDecl::base_class_const_iterator Base = ClassDecl->bases_begin(),
E = ClassDecl->bases_end(); Base != E; ++Base) {
if (Base->isVirtual())
continue;
IdealInitKeys.push_back(GetKeyForBase(SemaRef.Context, Base->getType()));
}
// 3. Direct fields.
for (CXXRecordDecl::field_iterator Field = ClassDecl->field_begin(),
E = ClassDecl->field_end(); Field != E; ++Field)
IdealInitKeys.push_back(GetKeyForTopLevelField(*Field));
unsigned NumIdealInits = IdealInitKeys.size();
unsigned IdealIndex = 0;
CXXCtorInitializer *PrevInit = 0;
for (unsigned InitIndex = 0; InitIndex != NumInits; ++InitIndex) {
CXXCtorInitializer *Init = Inits[InitIndex];
void *InitKey = GetKeyForMember(SemaRef.Context, Init);
// Scan forward to try to find this initializer in the idealized
// initializers list.
for (; IdealIndex != NumIdealInits; ++IdealIndex)
if (InitKey == IdealInitKeys[IdealIndex])
break;
// If we didn't find this initializer, it must be because we
// scanned past it on a previous iteration. That can only
// happen if we're out of order; emit a warning.
if (IdealIndex == NumIdealInits && PrevInit) {
Sema::SemaDiagnosticBuilder D =
SemaRef.Diag(PrevInit->getSourceLocation(),
diag::warn_initializer_out_of_order);
if (PrevInit->isAnyMemberInitializer())
D << 0 << PrevInit->getAnyMember()->getDeclName();
else
D << 1 << PrevInit->getBaseClassInfo()->getType();
if (Init->isAnyMemberInitializer())
D << 0 << Init->getAnyMember()->getDeclName();
else
D << 1 << Init->getBaseClassInfo()->getType();
// Move back to the initializer's location in the ideal list.
for (IdealIndex = 0; IdealIndex != NumIdealInits; ++IdealIndex)
if (InitKey == IdealInitKeys[IdealIndex])
break;
assert(IdealIndex != NumIdealInits &&
"initializer not found in initializer list");
}
PrevInit = Init;
}
}
namespace {
bool CheckRedundantInit(Sema &S,
CXXCtorInitializer *Init,
CXXCtorInitializer *&PrevInit) {
if (!PrevInit) {
PrevInit = Init;
return false;
}
if (FieldDecl *Field = Init->getMember())
S.Diag(Init->getSourceLocation(),
diag::err_multiple_mem_initialization)
<< Field->getDeclName()
<< Init->getSourceRange();
else {
const Type *BaseClass = Init->getBaseClass();
assert(BaseClass && "neither field nor base");
S.Diag(Init->getSourceLocation(),
diag::err_multiple_base_initialization)
<< QualType(BaseClass, 0)
<< Init->getSourceRange();
}
S.Diag(PrevInit->getSourceLocation(), diag::note_previous_initializer)
<< 0 << PrevInit->getSourceRange();
return true;
}
typedef std::pair<NamedDecl *, CXXCtorInitializer *> UnionEntry;
typedef llvm::DenseMap<RecordDecl*, UnionEntry> RedundantUnionMap;
bool CheckRedundantUnionInit(Sema &S,
CXXCtorInitializer *Init,
RedundantUnionMap &Unions) {
FieldDecl *Field = Init->getAnyMember();
RecordDecl *Parent = Field->getParent();
if (!Parent->isAnonymousStructOrUnion())
return false;
NamedDecl *Child = Field;
do {
if (Parent->isUnion()) {
UnionEntry &En = Unions[Parent];
if (En.first && En.first != Child) {
S.Diag(Init->getSourceLocation(),
diag::err_multiple_mem_union_initialization)
<< Field->getDeclName()
<< Init->getSourceRange();
S.Diag(En.second->getSourceLocation(), diag::note_previous_initializer)
<< 0 << En.second->getSourceRange();
return true;
} else if (!En.first) {
En.first = Child;
En.second = Init;
}
}
Child = Parent;
Parent = cast<RecordDecl>(Parent->getDeclContext());
} while (Parent->isAnonymousStructOrUnion());
return false;
}
}
2010-04-02 11:37:03 +08:00
/// ActOnMemInitializers - Handle the member initializers for a constructor.
2010-08-21 17:40:31 +08:00
void Sema::ActOnMemInitializers(Decl *ConstructorDecl,
2010-04-02 11:37:03 +08:00
SourceLocation ColonLoc,
MemInitTy **meminits, unsigned NumMemInits,
bool AnyErrors) {
if (!ConstructorDecl)
return;
AdjustDeclIfTemplate(ConstructorDecl);
CXXConstructorDecl *Constructor
2010-08-21 17:40:31 +08:00
= dyn_cast<CXXConstructorDecl>(ConstructorDecl);
2010-04-02 11:37:03 +08:00
if (!Constructor) {
Diag(ColonLoc, diag::err_only_constructors_take_base_inits);
return;
}
CXXCtorInitializer **MemInits =
reinterpret_cast<CXXCtorInitializer **>(meminits);
// Mapping for the duplicate initializers check.
// For member initializers, this is keyed with a FieldDecl*.
// For base initializers, this is keyed with a Type*.
llvm::DenseMap<void*, CXXCtorInitializer *> Members;
// Mapping for the inconsistent anonymous-union initializers check.
RedundantUnionMap MemberUnions;
bool HadError = false;
for (unsigned i = 0; i < NumMemInits; i++) {
CXXCtorInitializer *Init = MemInits[i];
2010-04-02 11:37:03 +08:00
// Set the source order index.
Init->setSourceOrder(i);
if (Init->isAnyMemberInitializer()) {
FieldDecl *Field = Init->getAnyMember();
if (CheckRedundantInit(*this, Init, Members[Field]) ||
CheckRedundantUnionInit(*this, Init, MemberUnions))
HadError = true;
} else {
void *Key = GetKeyForBase(Context, QualType(Init->getBaseClass(), 0));
if (CheckRedundantInit(*this, Init, Members[Key]))
HadError = true;
}
2010-04-02 11:37:03 +08:00
}
if (HadError)
return;
2010-04-02 11:37:03 +08:00
DiagnoseBaseOrMemInitializerOrder(*this, Constructor, MemInits, NumMemInits);
2010-04-02 11:43:34 +08:00
SetCtorInitializers(Constructor, MemInits, NumMemInits, AnyErrors);
2010-04-02 11:37:03 +08:00
}
void
Sema::MarkBaseAndMemberDestructorsReferenced(SourceLocation Location,
CXXRecordDecl *ClassDecl) {
// Ignore dependent contexts.
if (ClassDecl->isDependentContext())
return;
// FIXME: all the access-control diagnostics are positioned on the
// field/base declaration. That's probably good; that said, the
// user might reasonably want to know why the destructor is being
// emitted, and we currently don't say.
// Non-static data members.
for (CXXRecordDecl::field_iterator I = ClassDecl->field_begin(),
E = ClassDecl->field_end(); I != E; ++I) {
FieldDecl *Field = *I;
if (Field->isInvalidDecl())
continue;
QualType FieldType = Context.getBaseElementType(Field->getType());
const RecordType* RT = FieldType->getAs<RecordType>();
if (!RT)
continue;
CXXRecordDecl *FieldClassDecl = cast<CXXRecordDecl>(RT->getDecl());
if (FieldClassDecl->hasTrivialDestructor())
continue;
CXXDestructorDecl *Dtor = LookupDestructor(FieldClassDecl);
CheckDestructorAccess(Field->getLocation(), Dtor,
PDiag(diag::err_access_dtor_field)
<< Field->getDeclName()
<< FieldType);
MarkDeclarationReferenced(Location, const_cast<CXXDestructorDecl*>(Dtor));
}
llvm::SmallPtrSet<const RecordType *, 8> DirectVirtualBases;
// Bases.
for (CXXRecordDecl::base_class_iterator Base = ClassDecl->bases_begin(),
E = ClassDecl->bases_end(); Base != E; ++Base) {
// Bases are always records in a well-formed non-dependent class.
const RecordType *RT = Base->getType()->getAs<RecordType>();
// Remember direct virtual bases.
if (Base->isVirtual())
DirectVirtualBases.insert(RT);
// Ignore trivial destructors.
CXXRecordDecl *BaseClassDecl = cast<CXXRecordDecl>(RT->getDecl());
if (BaseClassDecl->hasTrivialDestructor())
continue;
CXXDestructorDecl *Dtor = LookupDestructor(BaseClassDecl);
// FIXME: caret should be on the start of the class name
CheckDestructorAccess(Base->getSourceRange().getBegin(), Dtor,
PDiag(diag::err_access_dtor_base)
<< Base->getType()
<< Base->getSourceRange());
MarkDeclarationReferenced(Location, const_cast<CXXDestructorDecl*>(Dtor));
}
// Virtual bases.
for (CXXRecordDecl::base_class_iterator VBase = ClassDecl->vbases_begin(),
E = ClassDecl->vbases_end(); VBase != E; ++VBase) {
// Bases are always records in a well-formed non-dependent class.
const RecordType *RT = VBase->getType()->getAs<RecordType>();
// Ignore direct virtual bases.
if (DirectVirtualBases.count(RT))
continue;
// Ignore trivial destructors.
CXXRecordDecl *BaseClassDecl = cast<CXXRecordDecl>(RT->getDecl());
if (BaseClassDecl->hasTrivialDestructor())
continue;
CXXDestructorDecl *Dtor = LookupDestructor(BaseClassDecl);
CheckDestructorAccess(ClassDecl->getLocation(), Dtor,
PDiag(diag::err_access_dtor_vbase)
<< VBase->getType());
MarkDeclarationReferenced(Location, const_cast<CXXDestructorDecl*>(Dtor));
}
}
2010-08-21 17:40:31 +08:00
void Sema::ActOnDefaultCtorInitializers(Decl *CDtorDecl) {
if (!CDtorDecl)
return;
if (CXXConstructorDecl *Constructor
2010-08-21 17:40:31 +08:00
= dyn_cast<CXXConstructorDecl>(CDtorDecl))
SetCtorInitializers(Constructor, 0, 0, /*AnyErrors=*/false);
}
bool Sema::RequireNonAbstractType(SourceLocation Loc, QualType T,
unsigned DiagID, AbstractDiagSelID SelID) {
if (SelID == -1)
return RequireNonAbstractType(Loc, T, PDiag(DiagID));
else
return RequireNonAbstractType(Loc, T, PDiag(DiagID) << SelID);
}
bool Sema::RequireNonAbstractType(SourceLocation Loc, QualType T,
const PartialDiagnostic &PD) {
if (!getLangOptions().CPlusPlus)
return false;
if (const ArrayType *AT = Context.getAsArrayType(T))
return RequireNonAbstractType(Loc, AT->getElementType(), PD);
if (const PointerType *PT = T->getAs<PointerType>()) {
// Find the innermost pointer type.
while (const PointerType *T = PT->getPointeeType()->getAs<PointerType>())
PT = T;
if (const ArrayType *AT = Context.getAsArrayType(PT->getPointeeType()))
return RequireNonAbstractType(Loc, AT->getElementType(), PD);
}
const RecordType *RT = T->getAs<RecordType>();
if (!RT)
return false;
const CXXRecordDecl *RD = cast<CXXRecordDecl>(RT->getDecl());
// We can't answer whether something is abstract until it has a
// definition. If it's currently being defined, we'll walk back
// over all the declarations when we have a full definition.
const CXXRecordDecl *Def = RD->getDefinition();
if (!Def || Def->isBeingDefined())
return false;
if (!RD->isAbstract())
return false;
Diag(Loc, PD) << RD->getDeclName();
DiagnoseAbstractType(RD);
return true;
}
void Sema::DiagnoseAbstractType(const CXXRecordDecl *RD) {
// Check if we've already emitted the list of pure virtual functions
// for this class.
if (PureVirtualClassDiagSet && PureVirtualClassDiagSet->count(RD))
return;
CXXFinalOverriderMap FinalOverriders;
RD->getFinalOverriders(FinalOverriders);
// Keep a set of seen pure methods so we won't diagnose the same method
// more than once.
llvm::SmallPtrSet<const CXXMethodDecl *, 8> SeenPureMethods;
for (CXXFinalOverriderMap::iterator M = FinalOverriders.begin(),
MEnd = FinalOverriders.end();
M != MEnd;
++M) {
for (OverridingMethods::iterator SO = M->second.begin(),
SOEnd = M->second.end();
SO != SOEnd; ++SO) {
// C++ [class.abstract]p4:
// A class is abstract if it contains or inherits at least one
// pure virtual function for which the final overrider is pure
// virtual.
//
if (SO->second.size() != 1)
continue;
if (!SO->second.front().Method->isPure())
continue;
if (!SeenPureMethods.insert(SO->second.front().Method))
continue;
Diag(SO->second.front().Method->getLocation(),
diag::note_pure_virtual_function)
<< SO->second.front().Method->getDeclName();
}
}
if (!PureVirtualClassDiagSet)
PureVirtualClassDiagSet.reset(new RecordDeclSetTy);
PureVirtualClassDiagSet->insert(RD);
}
namespace {
struct AbstractUsageInfo {
Sema &S;
CXXRecordDecl *Record;
CanQualType AbstractType;
bool Invalid;
AbstractUsageInfo(Sema &S, CXXRecordDecl *Record)
: S(S), Record(Record),
AbstractType(S.Context.getCanonicalType(
S.Context.getTypeDeclType(Record))),
Invalid(false) {}
void DiagnoseAbstractType() {
if (Invalid) return;
S.DiagnoseAbstractType(Record);
Invalid = true;
}
void CheckType(const NamedDecl *D, TypeLoc TL, Sema::AbstractDiagSelID Sel);
};
struct CheckAbstractUsage {
AbstractUsageInfo &Info;
const NamedDecl *Ctx;
CheckAbstractUsage(AbstractUsageInfo &Info, const NamedDecl *Ctx)
: Info(Info), Ctx(Ctx) {}
void Visit(TypeLoc TL, Sema::AbstractDiagSelID Sel) {
switch (TL.getTypeLocClass()) {
#define ABSTRACT_TYPELOC(CLASS, PARENT)
#define TYPELOC(CLASS, PARENT) \
case TypeLoc::CLASS: Check(cast<CLASS##TypeLoc>(TL), Sel); break;
#include "clang/AST/TypeLocNodes.def"
}
}
void Check(FunctionProtoTypeLoc TL, Sema::AbstractDiagSelID Sel) {
Visit(TL.getResultLoc(), Sema::AbstractReturnType);
for (unsigned I = 0, E = TL.getNumArgs(); I != E; ++I) {
TypeSourceInfo *TSI = TL.getArg(I)->getTypeSourceInfo();
if (TSI) Visit(TSI->getTypeLoc(), Sema::AbstractParamType);
}
}
void Check(ArrayTypeLoc TL, Sema::AbstractDiagSelID Sel) {
Visit(TL.getElementLoc(), Sema::AbstractArrayType);
}
void Check(TemplateSpecializationTypeLoc TL, Sema::AbstractDiagSelID Sel) {
// Visit the type parameters from a permissive context.
for (unsigned I = 0, E = TL.getNumArgs(); I != E; ++I) {
TemplateArgumentLoc TAL = TL.getArgLoc(I);
if (TAL.getArgument().getKind() == TemplateArgument::Type)
if (TypeSourceInfo *TSI = TAL.getTypeSourceInfo())
Visit(TSI->getTypeLoc(), Sema::AbstractNone);
// TODO: other template argument types?
}
}
// Visit pointee types from a permissive context.
#define CheckPolymorphic(Type) \
void Check(Type TL, Sema::AbstractDiagSelID Sel) { \
Visit(TL.getNextTypeLoc(), Sema::AbstractNone); \
}
CheckPolymorphic(PointerTypeLoc)
CheckPolymorphic(ReferenceTypeLoc)
CheckPolymorphic(MemberPointerTypeLoc)
CheckPolymorphic(BlockPointerTypeLoc)
/// Handle all the types we haven't given a more specific
/// implementation for above.
void Check(TypeLoc TL, Sema::AbstractDiagSelID Sel) {
// Every other kind of type that we haven't called out already
// that has an inner type is either (1) sugar or (2) contains that
// inner type in some way as a subobject.
if (TypeLoc Next = TL.getNextTypeLoc())
return Visit(Next, Sel);
// If there's no inner type and we're in a permissive context,
// don't diagnose.
if (Sel == Sema::AbstractNone) return;
// Check whether the type matches the abstract type.
QualType T = TL.getType();
if (T->isArrayType()) {
Sel = Sema::AbstractArrayType;
T = Info.S.Context.getBaseElementType(T);
}
CanQualType CT = T->getCanonicalTypeUnqualified().getUnqualifiedType();
if (CT != Info.AbstractType) return;
// It matched; do some magic.
if (Sel == Sema::AbstractArrayType) {
Info.S.Diag(Ctx->getLocation(), diag::err_array_of_abstract_type)
<< T << TL.getSourceRange();
} else {
Info.S.Diag(Ctx->getLocation(), diag::err_abstract_type_in_decl)
<< Sel << T << TL.getSourceRange();
}
Info.DiagnoseAbstractType();
}
};
void AbstractUsageInfo::CheckType(const NamedDecl *D, TypeLoc TL,
Sema::AbstractDiagSelID Sel) {
CheckAbstractUsage(*this, D).Visit(TL, Sel);
}
}
/// Check for invalid uses of an abstract type in a method declaration.
static void CheckAbstractClassUsage(AbstractUsageInfo &Info,
CXXMethodDecl *MD) {
// No need to do the check on definitions, which require that
// the return/param types be complete.
if (MD->isThisDeclarationADefinition())
return;
// For safety's sake, just ignore it if we don't have type source
// information. This should never happen for non-implicit methods,
// but...
if (TypeSourceInfo *TSI = MD->getTypeSourceInfo())
Info.CheckType(MD, TSI->getTypeLoc(), Sema::AbstractNone);
}
/// Check for invalid uses of an abstract type within a class definition.
static void CheckAbstractClassUsage(AbstractUsageInfo &Info,
CXXRecordDecl *RD) {
for (CXXRecordDecl::decl_iterator
I = RD->decls_begin(), E = RD->decls_end(); I != E; ++I) {
Decl *D = *I;
if (D->isImplicit()) continue;
// Methods and method templates.
if (isa<CXXMethodDecl>(D)) {
CheckAbstractClassUsage(Info, cast<CXXMethodDecl>(D));
} else if (isa<FunctionTemplateDecl>(D)) {
FunctionDecl *FD = cast<FunctionTemplateDecl>(D)->getTemplatedDecl();
CheckAbstractClassUsage(Info, cast<CXXMethodDecl>(FD));
// Fields and static variables.
} else if (isa<FieldDecl>(D)) {
FieldDecl *FD = cast<FieldDecl>(D);
if (TypeSourceInfo *TSI = FD->getTypeSourceInfo())
Info.CheckType(FD, TSI->getTypeLoc(), Sema::AbstractFieldType);
} else if (isa<VarDecl>(D)) {
VarDecl *VD = cast<VarDecl>(D);
if (TypeSourceInfo *TSI = VD->getTypeSourceInfo())
Info.CheckType(VD, TSI->getTypeLoc(), Sema::AbstractVariableType);
// Nested classes and class templates.
} else if (isa<CXXRecordDecl>(D)) {
CheckAbstractClassUsage(Info, cast<CXXRecordDecl>(D));
} else if (isa<ClassTemplateDecl>(D)) {
CheckAbstractClassUsage(Info,
cast<ClassTemplateDecl>(D)->getTemplatedDecl());
}
}
}
/// \brief Perform semantic checks on a class definition that has been
/// completing, introducing implicitly-declared members, checking for
/// abstract types, etc.
void Sema::CheckCompletedCXXClass(CXXRecordDecl *Record) {
if (!Record)
return;
if (Record->isAbstract() && !Record->isInvalidDecl()) {
AbstractUsageInfo Info(*this, Record);
CheckAbstractClassUsage(Info, Record);
}
// If this is not an aggregate type and has no user-declared constructor,
// complain about any non-static data members of reference or const scalar
// type, since they will never get initializers.
if (!Record->isInvalidDecl() && !Record->isDependentType() &&
!Record->isAggregate() && !Record->hasUserDeclaredConstructor()) {
bool Complained = false;
for (RecordDecl::field_iterator F = Record->field_begin(),
FEnd = Record->field_end();
F != FEnd; ++F) {
if (F->getType()->isReferenceType() ||
2010-04-17 01:43:15 +08:00
(F->getType().isConstQualified() && F->getType()->isScalarType())) {
if (!Complained) {
Diag(Record->getLocation(), diag::warn_no_constructor_for_refconst)
<< Record->getTagKind() << Record;
Complained = true;
}
Diag(F->getLocation(), diag::note_refconst_member_not_initialized)
<< F->getType()->isReferenceType()
<< F->getDeclName();
}
}
}
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
if (Record->isDynamicClass() && !Record->isDependentType())
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
DynamicClasses.push_back(Record);
if (Record->getIdentifier()) {
// C++ [class.mem]p13:
// If T is the name of a class, then each of the following shall have a
// name different from T:
// - every member of every anonymous union that is a member of class T.
//
// C++ [class.mem]p14:
// In addition, if class T has a user-declared constructor (12.1), every
// non-static data member of class T shall have a name different from T.
for (DeclContext::lookup_result R = Record->lookup(Record->getDeclName());
R.first != R.second; ++R.first) {
NamedDecl *D = *R.first;
if ((isa<FieldDecl>(D) && Record->hasUserDeclaredConstructor()) ||
isa<IndirectFieldDecl>(D)) {
Diag(D->getLocation(), diag::err_member_name_of_class)
<< D->getDeclName();
break;
}
}
}
// Warn if the class has virtual methods but non-virtual public destructor.
if (Record->isDynamicClass() && !Record->isDependentType()) {
CXXDestructorDecl *dtor = Record->getDestructor();
if (!dtor || (!dtor->isVirtual() && dtor->getAccess() == AS_public))
Diag(dtor ? dtor->getLocation() : Record->getLocation(),
diag::warn_non_virtual_dtor) << Context.getRecordType(Record);
}
// See if a method overloads virtual methods in a base
/// class without overriding any.
if (!Record->isDependentType()) {
for (CXXRecordDecl::method_iterator M = Record->method_begin(),
MEnd = Record->method_end();
M != MEnd; ++M) {
DiagnoseHiddenVirtualMethods(Record, *M);
}
}
// Declare inherited constructors. We do this eagerly here because:
// - The standard requires an eager diagnostic for conflicting inherited
// constructors from different classes.
// - The lazy declaration of the other implicit constructors is so as to not
// waste space and performance on classes that are not meant to be
// instantiated (e.g. meta-functions). This doesn't apply to classes that
// have inherited constructors.
DeclareInheritedConstructors(Record);
}
/// \brief Data used with FindHiddenVirtualMethod
struct FindHiddenVirtualMethodData {
Sema *S;
CXXMethodDecl *Method;
llvm::SmallPtrSet<const CXXMethodDecl *, 8> OverridenAndUsingBaseMethods;
llvm::SmallVector<CXXMethodDecl *, 8> OverloadedMethods;
};
/// \brief Member lookup function that determines whether a given C++
/// method overloads virtual methods in a base class without overriding any,
/// to be used with CXXRecordDecl::lookupInBases().
static bool FindHiddenVirtualMethod(const CXXBaseSpecifier *Specifier,
CXXBasePath &Path,
void *UserData) {
RecordDecl *BaseRecord = Specifier->getType()->getAs<RecordType>()->getDecl();
FindHiddenVirtualMethodData &Data
= *static_cast<FindHiddenVirtualMethodData*>(UserData);
DeclarationName Name = Data.Method->getDeclName();
assert(Name.getNameKind() == DeclarationName::Identifier);
bool foundSameNameMethod = false;
llvm::SmallVector<CXXMethodDecl *, 8> overloadedMethods;
for (Path.Decls = BaseRecord->lookup(Name);
Path.Decls.first != Path.Decls.second;
++Path.Decls.first) {
NamedDecl *D = *Path.Decls.first;
if (CXXMethodDecl *MD = dyn_cast<CXXMethodDecl>(D)) {
MD = MD->getCanonicalDecl();
foundSameNameMethod = true;
// Interested only in hidden virtual methods.
if (!MD->isVirtual())
continue;
// If the method we are checking overrides a method from its base
// don't warn about the other overloaded methods.
if (!Data.S->IsOverload(Data.Method, MD, false))
return true;
// Collect the overload only if its hidden.
if (!Data.OverridenAndUsingBaseMethods.count(MD))
overloadedMethods.push_back(MD);
}
}
if (foundSameNameMethod)
Data.OverloadedMethods.append(overloadedMethods.begin(),
overloadedMethods.end());
return foundSameNameMethod;
}
/// \brief See if a method overloads virtual methods in a base class without
/// overriding any.
void Sema::DiagnoseHiddenVirtualMethods(CXXRecordDecl *DC, CXXMethodDecl *MD) {
if (Diags.getDiagnosticLevel(diag::warn_overloaded_virtual,
MD->getLocation()) == Diagnostic::Ignored)
return;
if (MD->getDeclName().getNameKind() != DeclarationName::Identifier)
return;
CXXBasePaths Paths(/*FindAmbiguities=*/true, // true to look in all bases.
/*bool RecordPaths=*/false,
/*bool DetectVirtual=*/false);
FindHiddenVirtualMethodData Data;
Data.Method = MD;
Data.S = this;
// Keep the base methods that were overriden or introduced in the subclass
// by 'using' in a set. A base method not in this set is hidden.
for (DeclContext::lookup_result res = DC->lookup(MD->getDeclName());
res.first != res.second; ++res.first) {
if (CXXMethodDecl *MD = dyn_cast<CXXMethodDecl>(*res.first))
for (CXXMethodDecl::method_iterator I = MD->begin_overridden_methods(),
E = MD->end_overridden_methods();
I != E; ++I)
Data.OverridenAndUsingBaseMethods.insert((*I)->getCanonicalDecl());
if (UsingShadowDecl *shad = dyn_cast<UsingShadowDecl>(*res.first))
if (CXXMethodDecl *MD = dyn_cast<CXXMethodDecl>(shad->getTargetDecl()))
Data.OverridenAndUsingBaseMethods.insert(MD->getCanonicalDecl());
}
if (DC->lookupInBases(&FindHiddenVirtualMethod, &Data, Paths) &&
!Data.OverloadedMethods.empty()) {
Diag(MD->getLocation(), diag::warn_overloaded_virtual)
<< MD << (Data.OverloadedMethods.size() > 1);
for (unsigned i = 0, e = Data.OverloadedMethods.size(); i != e; ++i) {
CXXMethodDecl *overloadedMD = Data.OverloadedMethods[i];
Diag(overloadedMD->getLocation(),
diag::note_hidden_overloaded_virtual_declared_here) << overloadedMD;
}
}
}
void Sema::ActOnFinishCXXMemberSpecification(Scope* S, SourceLocation RLoc,
2010-08-21 17:40:31 +08:00
Decl *TagDecl,
SourceLocation LBrac,
SourceLocation RBrac,
AttributeList *AttrList) {
if (!TagDecl)
return;
AdjustDeclIfTemplate(TagDecl);
ActOnFields(S, RLoc, TagDecl,
2010-08-21 17:40:31 +08:00
// strict aliasing violation!
reinterpret_cast<Decl**>(FieldCollector->getCurFields()),
FieldCollector->getCurNumFields(), LBrac, RBrac, AttrList);
CheckCompletedCXXClass(
2010-08-21 17:40:31 +08:00
dyn_cast_or_null<CXXRecordDecl>(TagDecl));
}
namespace {
/// \brief Helper class that collects exception specifications for
/// implicitly-declared special member functions.
class ImplicitExceptionSpecification {
ASTContext &Context;
bool AllowsAllExceptions;
llvm::SmallPtrSet<CanQualType, 4> ExceptionsSeen;
llvm::SmallVector<QualType, 4> Exceptions;
public:
explicit ImplicitExceptionSpecification(ASTContext &Context)
: Context(Context), AllowsAllExceptions(false) { }
/// \brief Whether the special member function should have any
/// exception specification at all.
bool hasExceptionSpecification() const {
return !AllowsAllExceptions;
}
/// \brief Whether the special member function should have a
/// throw(...) exception specification (a Microsoft extension).
bool hasAnyExceptionSpecification() const {
return false;
}
/// \brief The number of exceptions in the exception specification.
unsigned size() const { return Exceptions.size(); }
/// \brief The set of exceptions in the exception specification.
const QualType *data() const { return Exceptions.data(); }
/// \brief Note that
void CalledDecl(CXXMethodDecl *Method) {
// If we already know that we allow all exceptions, do nothing.
if (AllowsAllExceptions || !Method)
return;
const FunctionProtoType *Proto
= Method->getType()->getAs<FunctionProtoType>();
// If this function can throw any exceptions, make a note of that.
if (!Proto->hasExceptionSpec() || Proto->hasAnyExceptionSpec()) {
AllowsAllExceptions = true;
ExceptionsSeen.clear();
Exceptions.clear();
return;
}
// Record the exceptions in this function's exception specification.
for (FunctionProtoType::exception_iterator E = Proto->exception_begin(),
EEnd = Proto->exception_end();
E != EEnd; ++E)
if (ExceptionsSeen.insert(Context.getCanonicalType(*E)))
Exceptions.push_back(*E);
}
};
}
/// AddImplicitlyDeclaredMembersToClass - Adds any implicitly-declared
/// special functions, such as the default constructor, copy
/// constructor, or destructor, to the given C++ class (C++
/// [special]p1). This routine can only be executed just before the
/// definition of the class is complete.
void Sema::AddImplicitlyDeclaredMembersToClass(CXXRecordDecl *ClassDecl) {
if (!ClassDecl->hasUserDeclaredConstructor())
++ASTContext::NumImplicitDefaultConstructors;
if (!ClassDecl->hasUserDeclaredCopyConstructor())
++ASTContext::NumImplicitCopyConstructors;
if (!ClassDecl->hasUserDeclaredCopyAssignment()) {
++ASTContext::NumImplicitCopyAssignmentOperators;
// If we have a dynamic class, then the copy assignment operator may be
// virtual, so we have to declare it immediately. This ensures that, e.g.,
// it shows up in the right place in the vtable and that we diagnose
// problems with the implicit exception specification.
if (ClassDecl->isDynamicClass())
DeclareImplicitCopyAssignment(ClassDecl);
}
if (!ClassDecl->hasUserDeclaredDestructor()) {
++ASTContext::NumImplicitDestructors;
// If we have a dynamic class, then the destructor may be virtual, so we
// have to declare the destructor immediately. This ensures that, e.g., it
// shows up in the right place in the vtable and that we diagnose problems
// with the implicit exception specification.
if (ClassDecl->isDynamicClass())
DeclareImplicitDestructor(ClassDecl);
}
}
2010-08-21 17:40:31 +08:00
void Sema::ActOnReenterTemplateScope(Scope *S, Decl *D) {
if (!D)
return;
TemplateParameterList *Params = 0;
if (TemplateDecl *Template = dyn_cast<TemplateDecl>(D))
Params = Template->getTemplateParameters();
else if (ClassTemplatePartialSpecializationDecl *PartialSpec
= dyn_cast<ClassTemplatePartialSpecializationDecl>(D))
Params = PartialSpec->getTemplateParameters();
else
return;
for (TemplateParameterList::iterator Param = Params->begin(),
ParamEnd = Params->end();
Param != ParamEnd; ++Param) {
NamedDecl *Named = cast<NamedDecl>(*Param);
if (Named->getDeclName()) {
2010-08-21 17:40:31 +08:00
S->AddDecl(Named);
IdResolver.AddDecl(Named);
}
}
}
2010-08-21 17:40:31 +08:00
void Sema::ActOnStartDelayedMemberDeclarations(Scope *S, Decl *RecordD) {
if (!RecordD) return;
AdjustDeclIfTemplate(RecordD);
2010-08-21 17:40:31 +08:00
CXXRecordDecl *Record = cast<CXXRecordDecl>(RecordD);
PushDeclContext(S, Record);
}
2010-08-21 17:40:31 +08:00
void Sema::ActOnFinishDelayedMemberDeclarations(Scope *S, Decl *RecordD) {
if (!RecordD) return;
PopDeclContext();
}
/// ActOnStartDelayedCXXMethodDeclaration - We have completed
/// parsing a top-level (non-nested) C++ class, and we are now
/// parsing those parts of the given Method declaration that could
/// not be parsed earlier (C++ [class.mem]p2), such as default
/// arguments. This action should enter the scope of the given
/// Method declaration as if we had just parsed the qualified method
/// name. However, it should not bring the parameters into scope;
/// that will be performed by ActOnDelayedCXXMethodParameter.
2010-08-21 17:40:31 +08:00
void Sema::ActOnStartDelayedCXXMethodDeclaration(Scope *S, Decl *MethodD) {
}
/// ActOnDelayedCXXMethodParameter - We've already started a delayed
/// C++ method declaration. We're (re-)introducing the given
/// function parameter into scope for use in parsing later parts of
/// the method declaration. For example, we could see an
/// ActOnParamDefaultArgument event for this parameter.
2010-08-21 17:40:31 +08:00
void Sema::ActOnDelayedCXXMethodParameter(Scope *S, Decl *ParamD) {
if (!ParamD)
return;
2010-08-21 17:40:31 +08:00
ParmVarDecl *Param = cast<ParmVarDecl>(ParamD);
// If this parameter has an unparsed default argument, clear it out
// to make way for the parsed default argument.
if (Param->hasUnparsedDefaultArg())
Param->setDefaultArg(0);
2010-08-21 17:40:31 +08:00
S->AddDecl(Param);
if (Param->getDeclName())
IdResolver.AddDecl(Param);
}
/// ActOnFinishDelayedCXXMethodDeclaration - We have finished
/// processing the delayed method declaration for Method. The method
/// declaration is now considered finished. There may be a separate
/// ActOnStartOfFunctionDef action later (not necessarily
/// immediately!) for this method, if it was also defined inside the
/// class body.
2010-08-21 17:40:31 +08:00
void Sema::ActOnFinishDelayedCXXMethodDeclaration(Scope *S, Decl *MethodD) {
if (!MethodD)
return;
AdjustDeclIfTemplate(MethodD);
2010-08-21 17:40:31 +08:00
FunctionDecl *Method = cast<FunctionDecl>(MethodD);
// Now that we have our default arguments, check the constructor
// again. It could produce additional diagnostics or affect whether
// the class has implicitly-declared destructors, among other
// things.
if (CXXConstructorDecl *Constructor = dyn_cast<CXXConstructorDecl>(Method))
CheckConstructor(Constructor);
// Check the default arguments, which we may have added.
if (!Method->isInvalidDecl())
CheckCXXDefaultArguments(Method);
}
/// CheckConstructorDeclarator - Called by ActOnDeclarator to check
/// the well-formedness of the constructor declarator @p D with type @p
/// R. If there are any errors in the declarator, this routine will
/// emit diagnostics and set the invalid bit to true. In any case, the type
/// will be updated to reflect a well-formed type for the constructor and
/// returned.
QualType Sema::CheckConstructorDeclarator(Declarator &D, QualType R,
StorageClass &SC) {
bool isVirtual = D.getDeclSpec().isVirtualSpecified();
// C++ [class.ctor]p3:
// A constructor shall not be virtual (10.3) or static (9.4). A
// constructor can be invoked for a const, volatile or const
// volatile object. A constructor shall not be declared const,
// volatile, or const volatile (9.3.2).
if (isVirtual) {
if (!D.isInvalidType())
Diag(D.getIdentifierLoc(), diag::err_constructor_cannot_be)
<< "virtual" << SourceRange(D.getDeclSpec().getVirtualSpecLoc())
<< SourceRange(D.getIdentifierLoc());
D.setInvalidType();
}
if (SC == SC_Static) {
if (!D.isInvalidType())
Diag(D.getIdentifierLoc(), diag::err_constructor_cannot_be)
<< "static" << SourceRange(D.getDeclSpec().getStorageClassSpecLoc())
<< SourceRange(D.getIdentifierLoc());
D.setInvalidType();
SC = SC_None;
}
DeclaratorChunk::FunctionTypeInfo &FTI = D.getFunctionTypeInfo();
if (FTI.TypeQuals != 0) {
if (FTI.TypeQuals & Qualifiers::Const)
Diag(D.getIdentifierLoc(), diag::err_invalid_qualified_constructor)
<< "const" << SourceRange(D.getIdentifierLoc());
if (FTI.TypeQuals & Qualifiers::Volatile)
Diag(D.getIdentifierLoc(), diag::err_invalid_qualified_constructor)
<< "volatile" << SourceRange(D.getIdentifierLoc());
if (FTI.TypeQuals & Qualifiers::Restrict)
Diag(D.getIdentifierLoc(), diag::err_invalid_qualified_constructor)
<< "restrict" << SourceRange(D.getIdentifierLoc());
D.setInvalidType();
}
// C++0x [class.ctor]p4:
// A constructor shall not be declared with a ref-qualifier.
if (FTI.hasRefQualifier()) {
Diag(FTI.getRefQualifierLoc(), diag::err_ref_qualifier_constructor)
<< FTI.RefQualifierIsLValueRef
<< FixItHint::CreateRemoval(FTI.getRefQualifierLoc());
D.setInvalidType();
}
// Rebuild the function type "R" without any type qualifiers (in
// case any of the errors above fired) and with "void" as the
// return type, since constructors don't have return types.
const FunctionProtoType *Proto = R->getAs<FunctionProtoType>();
if (Proto->getResultType() == Context.VoidTy && !D.isInvalidType())
return R;
FunctionProtoType::ExtProtoInfo EPI = Proto->getExtProtoInfo();
EPI.TypeQuals = 0;
EPI.RefQualifier = RQ_None;
return Context.getFunctionType(Context.VoidTy, Proto->arg_type_begin(),
Proto->getNumArgs(), EPI);
}
/// CheckConstructor - Checks a fully-formed constructor for
/// well-formedness, issuing any diagnostics required. Returns true if
/// the constructor declarator is invalid.
void Sema::CheckConstructor(CXXConstructorDecl *Constructor) {
CXXRecordDecl *ClassDecl
= dyn_cast<CXXRecordDecl>(Constructor->getDeclContext());
if (!ClassDecl)
return Constructor->setInvalidDecl();
// C++ [class.copy]p3:
// A declaration of a constructor for a class X is ill-formed if
// its first parameter is of type (optionally cv-qualified) X and
// either there are no other parameters or else all other
// parameters have default arguments.
if (!Constructor->isInvalidDecl() &&
((Constructor->getNumParams() == 1) ||
(Constructor->getNumParams() > 1 &&
Constructor->getParamDecl(1)->hasDefaultArg())) &&
Constructor->getTemplateSpecializationKind()
!= TSK_ImplicitInstantiation) {
QualType ParamType = Constructor->getParamDecl(0)->getType();
QualType ClassTy = Context.getTagDeclType(ClassDecl);
if (Context.getCanonicalType(ParamType).getUnqualifiedType() == ClassTy) {
SourceLocation ParamLoc = Constructor->getParamDecl(0)->getLocation();
const char *ConstRef
= Constructor->getParamDecl(0)->getIdentifier() ? "const &"
: " const &";
Diag(ParamLoc, diag::err_constructor_byvalue_arg)
<< FixItHint::CreateInsertion(ParamLoc, ConstRef);
// FIXME: Rather that making the constructor invalid, we should endeavor
// to fix the type.
Constructor->setInvalidDecl();
}
}
}
/// CheckDestructor - Checks a fully-formed destructor definition for
/// well-formedness, issuing any diagnostics required. Returns true
/// on error.
bool Sema::CheckDestructor(CXXDestructorDecl *Destructor) {
CXXRecordDecl *RD = Destructor->getParent();
if (Destructor->isVirtual()) {
SourceLocation Loc;
if (!Destructor->isImplicit())
Loc = Destructor->getLocation();
else
Loc = RD->getLocation();
// If we have a virtual destructor, look up the deallocation function
FunctionDecl *OperatorDelete = 0;
DeclarationName Name =
Context.DeclarationNames.getCXXOperatorName(OO_Delete);
if (FindDeallocationFunction(Loc, RD, Name, OperatorDelete))
return true;
MarkDeclarationReferenced(Loc, OperatorDelete);
Destructor->setOperatorDelete(OperatorDelete);
}
return false;
}
static inline bool
FTIHasSingleVoidArgument(DeclaratorChunk::FunctionTypeInfo &FTI) {
return (FTI.NumArgs == 1 && !FTI.isVariadic && FTI.ArgInfo[0].Ident == 0 &&
FTI.ArgInfo[0].Param &&
2010-08-21 17:40:31 +08:00
cast<ParmVarDecl>(FTI.ArgInfo[0].Param)->getType()->isVoidType());
}
/// CheckDestructorDeclarator - Called by ActOnDeclarator to check
/// the well-formednes of the destructor declarator @p D with type @p
/// R. If there are any errors in the declarator, this routine will
/// emit diagnostics and set the declarator to invalid. Even if this happens,
/// will be updated to reflect a well-formed type for the destructor and
/// returned.
QualType Sema::CheckDestructorDeclarator(Declarator &D, QualType R,
StorageClass& SC) {
// C++ [class.dtor]p1:
// [...] A typedef-name that names a class is a class-name
// (7.1.3); however, a typedef-name that names a class shall not
// be used as the identifier in the declarator for a destructor
// declaration.
QualType DeclaratorType = GetTypeFromParser(D.getName().DestructorName);
if (isa<TypedefType>(DeclaratorType))
Diag(D.getIdentifierLoc(), diag::err_destructor_typedef_name)
<< DeclaratorType;
// C++ [class.dtor]p2:
// A destructor is used to destroy objects of its class type. A
// destructor takes no parameters, and no return type can be
// specified for it (not even void). The address of a destructor
// shall not be taken. A destructor shall not be static. A
// destructor can be invoked for a const, volatile or const
// volatile object. A destructor shall not be declared const,
// volatile or const volatile (9.3.2).
if (SC == SC_Static) {
if (!D.isInvalidType())
Diag(D.getIdentifierLoc(), diag::err_destructor_cannot_be)
<< "static" << SourceRange(D.getDeclSpec().getStorageClassSpecLoc())
<< SourceRange(D.getIdentifierLoc())
<< FixItHint::CreateRemoval(D.getDeclSpec().getStorageClassSpecLoc());
SC = SC_None;
}
if (D.getDeclSpec().hasTypeSpecifier() && !D.isInvalidType()) {
// Destructors don't have return types, but the parser will
// happily parse something like:
//
// class X {
// float ~X();
// };
//
// The return type will be eliminated later.
Diag(D.getIdentifierLoc(), diag::err_destructor_return_type)
<< SourceRange(D.getDeclSpec().getTypeSpecTypeLoc())
<< SourceRange(D.getIdentifierLoc());
}
DeclaratorChunk::FunctionTypeInfo &FTI = D.getFunctionTypeInfo();
if (FTI.TypeQuals != 0 && !D.isInvalidType()) {
if (FTI.TypeQuals & Qualifiers::Const)
Diag(D.getIdentifierLoc(), diag::err_invalid_qualified_destructor)
<< "const" << SourceRange(D.getIdentifierLoc());
if (FTI.TypeQuals & Qualifiers::Volatile)
Diag(D.getIdentifierLoc(), diag::err_invalid_qualified_destructor)
<< "volatile" << SourceRange(D.getIdentifierLoc());
if (FTI.TypeQuals & Qualifiers::Restrict)
Diag(D.getIdentifierLoc(), diag::err_invalid_qualified_destructor)
<< "restrict" << SourceRange(D.getIdentifierLoc());
D.setInvalidType();
}
// C++0x [class.dtor]p2:
// A destructor shall not be declared with a ref-qualifier.
if (FTI.hasRefQualifier()) {
Diag(FTI.getRefQualifierLoc(), diag::err_ref_qualifier_destructor)
<< FTI.RefQualifierIsLValueRef
<< FixItHint::CreateRemoval(FTI.getRefQualifierLoc());
D.setInvalidType();
}
// Make sure we don't have any parameters.
if (FTI.NumArgs > 0 && !FTIHasSingleVoidArgument(FTI)) {
Diag(D.getIdentifierLoc(), diag::err_destructor_with_params);
// Delete the parameters.
FTI.freeArgs();
D.setInvalidType();
}
// Make sure the destructor isn't variadic.
if (FTI.isVariadic) {
Diag(D.getIdentifierLoc(), diag::err_destructor_variadic);
D.setInvalidType();
}
// Rebuild the function type "R" without any type qualifiers or
// parameters (in case any of the errors above fired) and with
// "void" as the return type, since destructors don't have return
// types.
if (!D.isInvalidType())
return R;
const FunctionProtoType *Proto = R->getAs<FunctionProtoType>();
FunctionProtoType::ExtProtoInfo EPI = Proto->getExtProtoInfo();
EPI.Variadic = false;
EPI.TypeQuals = 0;
EPI.RefQualifier = RQ_None;
return Context.getFunctionType(Context.VoidTy, 0, 0, EPI);
}
/// CheckConversionDeclarator - Called by ActOnDeclarator to check the
/// well-formednes of the conversion function declarator @p D with
/// type @p R. If there are any errors in the declarator, this routine
/// will emit diagnostics and return true. Otherwise, it will return
/// false. Either way, the type @p R will be updated to reflect a
/// well-formed type for the conversion operator.
void Sema::CheckConversionDeclarator(Declarator &D, QualType &R,
StorageClass& SC) {
// C++ [class.conv.fct]p1:
// Neither parameter types nor return type can be specified. The
// type of a conversion function (8.3.5) is "function taking no
// parameter returning conversion-type-id."
if (SC == SC_Static) {
if (!D.isInvalidType())
Diag(D.getIdentifierLoc(), diag::err_conv_function_not_member)
<< "static" << SourceRange(D.getDeclSpec().getStorageClassSpecLoc())
<< SourceRange(D.getIdentifierLoc());
D.setInvalidType();
SC = SC_None;
}
QualType ConvType = GetTypeFromParser(D.getName().ConversionFunctionId);
if (D.getDeclSpec().hasTypeSpecifier() && !D.isInvalidType()) {
// Conversion functions don't have return types, but the parser will
// happily parse something like:
//
// class X {
// float operator bool();
// };
//
// The return type will be changed later anyway.
Diag(D.getIdentifierLoc(), diag::err_conv_function_return_type)
<< SourceRange(D.getDeclSpec().getTypeSpecTypeLoc())
<< SourceRange(D.getIdentifierLoc());
D.setInvalidType();
}
const FunctionProtoType *Proto = R->getAs<FunctionProtoType>();
// Make sure we don't have any parameters.
if (Proto->getNumArgs() > 0) {
Diag(D.getIdentifierLoc(), diag::err_conv_function_with_params);
// Delete the parameters.
D.getFunctionTypeInfo().freeArgs();
D.setInvalidType();
} else if (Proto->isVariadic()) {
Diag(D.getIdentifierLoc(), diag::err_conv_function_variadic);
D.setInvalidType();
}
// Diagnose "&operator bool()" and other such nonsense. This
// is actually a gcc extension which we don't support.
if (Proto->getResultType() != ConvType) {
Diag(D.getIdentifierLoc(), diag::err_conv_function_with_complex_decl)
<< Proto->getResultType();
D.setInvalidType();
ConvType = Proto->getResultType();
}
// C++ [class.conv.fct]p4:
// The conversion-type-id shall not represent a function type nor
// an array type.
if (ConvType->isArrayType()) {
Diag(D.getIdentifierLoc(), diag::err_conv_function_to_array);
ConvType = Context.getPointerType(ConvType);
D.setInvalidType();
} else if (ConvType->isFunctionType()) {
Diag(D.getIdentifierLoc(), diag::err_conv_function_to_function);
ConvType = Context.getPointerType(ConvType);
D.setInvalidType();
}
// Rebuild the function type "R" without any parameters (in case any
// of the errors above fired) and with the conversion type as the
// return type.
if (D.isInvalidType())
R = Context.getFunctionType(ConvType, 0, 0, Proto->getExtProtoInfo());
// C++0x explicit conversion operators.
if (D.getDeclSpec().isExplicitSpecified() && !getLangOptions().CPlusPlus0x)
Diag(D.getDeclSpec().getExplicitSpecLoc(),
diag::warn_explicit_conversion_functions)
<< SourceRange(D.getDeclSpec().getExplicitSpecLoc());
}
/// ActOnConversionDeclarator - Called by ActOnDeclarator to complete
/// the declaration of the given C++ conversion function. This routine
/// is responsible for recording the conversion function in the C++
/// class, if possible.
2010-08-21 17:40:31 +08:00
Decl *Sema::ActOnConversionDeclarator(CXXConversionDecl *Conversion) {
assert(Conversion && "Expected to receive a conversion function declaration");
CXXRecordDecl *ClassDecl = cast<CXXRecordDecl>(Conversion->getDeclContext());
// Make sure we aren't redeclaring the conversion function.
QualType ConvType = Context.getCanonicalType(Conversion->getConversionType());
// C++ [class.conv.fct]p1:
// [...] A conversion function is never used to convert a
// (possibly cv-qualified) object to the (possibly cv-qualified)
// same object type (or a reference to it), to a (possibly
// cv-qualified) base class of that type (or a reference to it),
// or to (possibly cv-qualified) void.
2009-05-16 15:39:55 +08:00
// FIXME: Suppress this warning if the conversion function ends up being a
// virtual function that overrides a virtual function in a base class.
QualType ClassType
= Context.getCanonicalType(Context.getTypeDeclType(ClassDecl));
if (const ReferenceType *ConvTypeRef = ConvType->getAs<ReferenceType>())
ConvType = ConvTypeRef->getPointeeType();
if (Conversion->getTemplateSpecializationKind() != TSK_Undeclared &&
Conversion->getTemplateSpecializationKind() != TSK_ExplicitSpecialization)
/* Suppress diagnostics for instantiations. */;
else if (ConvType->isRecordType()) {
ConvType = Context.getCanonicalType(ConvType).getUnqualifiedType();
if (ConvType == ClassType)
Diag(Conversion->getLocation(), diag::warn_conv_to_self_not_used)
<< ClassType;
else if (IsDerivedFrom(ClassType, ConvType))
Diag(Conversion->getLocation(), diag::warn_conv_to_base_not_used)
<< ClassType << ConvType;
} else if (ConvType->isVoidType()) {
Diag(Conversion->getLocation(), diag::warn_conv_to_void_not_used)
<< ClassType << ConvType;
}
if (FunctionTemplateDecl *ConversionTemplate
= Conversion->getDescribedFunctionTemplate())
return ConversionTemplate;
2010-08-21 17:40:31 +08:00
return Conversion;
}
//===----------------------------------------------------------------------===//
// Namespace Handling
//===----------------------------------------------------------------------===//
/// ActOnStartNamespaceDef - This is called at the start of a namespace
/// definition.
2010-08-21 17:40:31 +08:00
Decl *Sema::ActOnStartNamespaceDef(Scope *NamespcScope,
SourceLocation InlineLoc,
SourceLocation IdentLoc,
IdentifierInfo *II,
SourceLocation LBrace,
AttributeList *AttrList) {
// anonymous namespace starts at its left brace
NamespaceDecl *Namespc = NamespaceDecl::Create(Context, CurContext,
(II ? IdentLoc : LBrace) , II);
Namespc->setLBracLoc(LBrace);
Namespc->setInline(InlineLoc.isValid());
Scope *DeclRegionScope = NamespcScope->getParent();
ProcessDeclAttributeList(DeclRegionScope, Namespc, AttrList);
if (const VisibilityAttr *Attr = Namespc->getAttr<VisibilityAttr>())
PushNamespaceVisibilityAttr(Attr);
if (II) {
// C++ [namespace.def]p2:
// The identifier in an original-namespace-definition shall not
// have been previously defined in the declarative region in
// which the original-namespace-definition appears. The
// identifier in an original-namespace-definition is the name of
// the namespace. Subsequently in that declarative region, it is
// treated as an original-namespace-name.
//
// Since namespace names are unique in their scope, and we don't
// look through using directives, just
DeclContext::lookup_result R = CurContext->getRedeclContext()->lookup(II);
NamedDecl *PrevDecl = R.first == R.second? 0 : *R.first;
if (NamespaceDecl *OrigNS = dyn_cast_or_null<NamespaceDecl>(PrevDecl)) {
// This is an extended namespace definition.
if (Namespc->isInline() != OrigNS->isInline()) {
// inline-ness must match
Diag(Namespc->getLocation(), diag::err_inline_namespace_mismatch)
<< Namespc->isInline();
Diag(OrigNS->getLocation(), diag::note_previous_definition);
Namespc->setInvalidDecl();
// Recover by ignoring the new namespace's inline status.
Namespc->setInline(OrigNS->isInline());
}
// Attach this namespace decl to the chain of extended namespace
// definitions.
OrigNS->setNextNamespace(Namespc);
Namespc->setOriginalNamespace(OrigNS->getOriginalNamespace());
// Remove the previous declaration from the scope.
2010-08-21 17:40:31 +08:00
if (DeclRegionScope->isDeclScope(OrigNS)) {
IdResolver.RemoveDecl(OrigNS);
2010-08-21 17:40:31 +08:00
DeclRegionScope->RemoveDecl(OrigNS);
}
} else if (PrevDecl) {
// This is an invalid name redefinition.
Diag(Namespc->getLocation(), diag::err_redefinition_different_kind)
<< Namespc->getDeclName();
Diag(PrevDecl->getLocation(), diag::note_previous_definition);
Namespc->setInvalidDecl();
// Continue on to push Namespc as current DeclContext and return it.
} else if (II->isStr("std") &&
CurContext->getRedeclContext()->isTranslationUnit()) {
// This is the first "real" definition of the namespace "std", so update
// our cache of the "std" namespace to point at this definition.
if (NamespaceDecl *StdNS = getStdNamespace()) {
// We had already defined a dummy namespace "std". Link this new
// namespace definition to the dummy namespace "std".
StdNS->setNextNamespace(Namespc);
StdNS->setLocation(IdentLoc);
Namespc->setOriginalNamespace(StdNS->getOriginalNamespace());
}
// Make our StdNamespace cache point at the first real definition of the
// "std" namespace.
StdNamespace = Namespc;
}
PushOnScopeChains(Namespc, DeclRegionScope);
} else {
// Anonymous namespaces.
assert(Namespc->isAnonymousNamespace());
// Link the anonymous namespace into its parent.
NamespaceDecl *PrevDecl;
DeclContext *Parent = CurContext->getRedeclContext();
if (TranslationUnitDecl *TU = dyn_cast<TranslationUnitDecl>(Parent)) {
PrevDecl = TU->getAnonymousNamespace();
TU->setAnonymousNamespace(Namespc);
} else {
NamespaceDecl *ND = cast<NamespaceDecl>(Parent);
PrevDecl = ND->getAnonymousNamespace();
ND->setAnonymousNamespace(Namespc);
}
// Link the anonymous namespace with its previous declaration.
if (PrevDecl) {
assert(PrevDecl->isAnonymousNamespace());
assert(!PrevDecl->getNextNamespace());
Namespc->setOriginalNamespace(PrevDecl->getOriginalNamespace());
PrevDecl->setNextNamespace(Namespc);
if (Namespc->isInline() != PrevDecl->isInline()) {
// inline-ness must match
Diag(Namespc->getLocation(), diag::err_inline_namespace_mismatch)
<< Namespc->isInline();
Diag(PrevDecl->getLocation(), diag::note_previous_definition);
Namespc->setInvalidDecl();
// Recover by ignoring the new namespace's inline status.
Namespc->setInline(PrevDecl->isInline());
}
}
CurContext->addDecl(Namespc);
// C++ [namespace.unnamed]p1. An unnamed-namespace-definition
// behaves as if it were replaced by
// namespace unique { /* empty body */ }
// using namespace unique;
// namespace unique { namespace-body }
// where all occurrences of 'unique' in a translation unit are
// replaced by the same identifier and this identifier differs
// from all other identifiers in the entire program.
// We just create the namespace with an empty name and then add an
// implicit using declaration, just like the standard suggests.
//
// CodeGen enforces the "universally unique" aspect by giving all
// declarations semantically contained within an anonymous
// namespace internal linkage.
if (!PrevDecl) {
UsingDirectiveDecl* UD
= UsingDirectiveDecl::Create(Context, CurContext,
/* 'using' */ LBrace,
/* 'namespace' */ SourceLocation(),
/* qualifier */ SourceRange(),
/* NNS */ NULL,
/* identifier */ SourceLocation(),
Namespc,
/* Ancestor */ CurContext);
UD->setImplicit();
CurContext->addDecl(UD);
}
}
// Although we could have an invalid decl (i.e. the namespace name is a
// redefinition), push it as current DeclContext and try to continue parsing.
2009-05-16 15:39:55 +08:00
// FIXME: We should be able to push Namespc here, so that the each DeclContext
// for the namespace has the declarations that showed up in that particular
// namespace definition.
PushDeclContext(NamespcScope, Namespc);
2010-08-21 17:40:31 +08:00
return Namespc;
}
/// getNamespaceDecl - Returns the namespace a decl represents. If the decl
/// is a namespace alias, returns the namespace it points to.
static inline NamespaceDecl *getNamespaceDecl(NamedDecl *D) {
if (NamespaceAliasDecl *AD = dyn_cast_or_null<NamespaceAliasDecl>(D))
return AD->getNamespace();
return dyn_cast_or_null<NamespaceDecl>(D);
}
/// ActOnFinishNamespaceDef - This callback is called after a namespace is
/// exited. Decl is the DeclTy returned by ActOnStartNamespaceDef.
2010-08-21 17:40:31 +08:00
void Sema::ActOnFinishNamespaceDef(Decl *Dcl, SourceLocation RBrace) {
NamespaceDecl *Namespc = dyn_cast_or_null<NamespaceDecl>(Dcl);
assert(Namespc && "Invalid parameter, expected NamespaceDecl");
Namespc->setRBracLoc(RBrace);
PopDeclContext();
if (Namespc->hasAttr<VisibilityAttr>())
PopPragmaVisibility();
}
CXXRecordDecl *Sema::getStdBadAlloc() const {
return cast_or_null<CXXRecordDecl>(
StdBadAlloc.get(Context.getExternalSource()));
}
NamespaceDecl *Sema::getStdNamespace() const {
return cast_or_null<NamespaceDecl>(
StdNamespace.get(Context.getExternalSource()));
}
/// \brief Retrieve the special "std" namespace, which may require us to
/// implicitly define the namespace.
NamespaceDecl *Sema::getOrCreateStdNamespace() {
if (!StdNamespace) {
// The "std" namespace has not yet been defined, so build one implicitly.
StdNamespace = NamespaceDecl::Create(Context,
Context.getTranslationUnitDecl(),
SourceLocation(),
&PP.getIdentifierTable().get("std"));
getStdNamespace()->setImplicit(true);
}
return getStdNamespace();
}
2010-08-21 17:40:31 +08:00
Decl *Sema::ActOnUsingDirective(Scope *S,
SourceLocation UsingLoc,
SourceLocation NamespcLoc,
CXXScopeSpec &SS,
SourceLocation IdentLoc,
IdentifierInfo *NamespcName,
AttributeList *AttrList) {
assert(!SS.isInvalid() && "Invalid CXXScopeSpec.");
assert(NamespcName && "Invalid NamespcName.");
assert(IdentLoc.isValid() && "Invalid NamespceName location.");
// This can only happen along a recovery path.
while (S->getFlags() & Scope::TemplateParamScope)
S = S->getParent();
assert(S->getFlags() & Scope::DeclScope && "Invalid Scope.");
UsingDirectiveDecl *UDir = 0;
NestedNameSpecifier *Qualifier = 0;
if (SS.isSet())
Qualifier = static_cast<NestedNameSpecifier *>(SS.getScopeRep());
// Lookup namespace name.
LookupResult R(*this, NamespcName, IdentLoc, LookupNamespaceName);
LookupParsedName(R, S, &SS);
if (R.isAmbiguous())
2010-08-21 17:40:31 +08:00
return 0;
if (R.empty()) {
// Allow "using namespace std;" or "using namespace ::std;" even if
// "std" hasn't been defined yet, for GCC compatibility.
if ((!Qualifier || Qualifier->getKind() == NestedNameSpecifier::Global) &&
NamespcName->isStr("std")) {
Diag(IdentLoc, diag::ext_using_undefined_std);
R.addDecl(getOrCreateStdNamespace());
R.resolveKind();
}
// Otherwise, attempt typo correction.
else if (DeclarationName Corrected = CorrectTypo(R, S, &SS, 0, false,
CTC_NoKeywords, 0)) {
if (R.getAsSingle<NamespaceDecl>() ||
R.getAsSingle<NamespaceAliasDecl>()) {
if (DeclContext *DC = computeDeclContext(SS, false))
Diag(IdentLoc, diag::err_using_directive_member_suggest)
<< NamespcName << DC << Corrected << SS.getRange()
<< FixItHint::CreateReplacement(IdentLoc, Corrected.getAsString());
else
Diag(IdentLoc, diag::err_using_directive_suggest)
<< NamespcName << Corrected
<< FixItHint::CreateReplacement(IdentLoc, Corrected.getAsString());
Diag(R.getFoundDecl()->getLocation(), diag::note_namespace_defined_here)
<< Corrected;
NamespcName = Corrected.getAsIdentifierInfo();
} else {
R.clear();
R.setLookupName(NamespcName);
}
}
}
if (!R.empty()) {
NamedDecl *Named = R.getFoundDecl();
assert((isa<NamespaceDecl>(Named) || isa<NamespaceAliasDecl>(Named))
&& "expected namespace decl");
// C++ [namespace.udir]p1:
// A using-directive specifies that the names in the nominated
// namespace can be used in the scope in which the
// using-directive appears after the using-directive. During
// unqualified name lookup (3.4.1), the names appear as if they
// were declared in the nearest enclosing namespace which
// contains both the using-directive and the nominated
// namespace. [Note: in this context, "contains" means "contains
// directly or indirectly". ]
// Find enclosing context containing both using-directive and
// nominated namespace.
NamespaceDecl *NS = getNamespaceDecl(Named);
DeclContext *CommonAncestor = cast<DeclContext>(NS);
while (CommonAncestor && !CommonAncestor->Encloses(CurContext))
CommonAncestor = CommonAncestor->getParent();
UDir = UsingDirectiveDecl::Create(Context, CurContext, UsingLoc, NamespcLoc,
SS.getRange(),
(NestedNameSpecifier *)SS.getScopeRep(),
IdentLoc, Named, CommonAncestor);
PushUsingDirective(S, UDir);
} else {
Diag(IdentLoc, diag::err_expected_namespace_name) << SS.getRange();
}
// FIXME: We ignore attributes for now.
2010-08-21 17:40:31 +08:00
return UDir;
}
void Sema::PushUsingDirective(Scope *S, UsingDirectiveDecl *UDir) {
// If scope has associated entity, then using directive is at namespace
// or translation unit scope. We add UsingDirectiveDecls, into
// it's lookup structure.
if (DeclContext *Ctx = static_cast<DeclContext*>(S->getEntity()))
Ctx->addDecl(UDir);
else
// Otherwise it is block-sope. using-directives will affect lookup
// only to the end of scope.
2010-08-21 17:40:31 +08:00
S->PushUsingDirective(UDir);
}
2010-08-21 17:40:31 +08:00
Decl *Sema::ActOnUsingDeclaration(Scope *S,
AccessSpecifier AS,
bool HasUsingKeyword,
SourceLocation UsingLoc,
CXXScopeSpec &SS,
UnqualifiedId &Name,
AttributeList *AttrList,
bool IsTypeName,
SourceLocation TypenameLoc) {
assert(S->getFlags() & Scope::DeclScope && "Invalid Scope.");
switch (Name.getKind()) {
case UnqualifiedId::IK_Identifier:
case UnqualifiedId::IK_OperatorFunctionId:
case UnqualifiedId::IK_LiteralOperatorId:
case UnqualifiedId::IK_ConversionFunctionId:
break;
case UnqualifiedId::IK_ConstructorName:
case UnqualifiedId::IK_ConstructorTemplateId:
// C++0x inherited constructors.
if (getLangOptions().CPlusPlus0x) break;
Diag(Name.getSourceRange().getBegin(), diag::err_using_decl_constructor)
<< SS.getRange();
2010-08-21 17:40:31 +08:00
return 0;
case UnqualifiedId::IK_DestructorName:
Diag(Name.getSourceRange().getBegin(), diag::err_using_decl_destructor)
<< SS.getRange();
2010-08-21 17:40:31 +08:00
return 0;
case UnqualifiedId::IK_TemplateId:
Diag(Name.getSourceRange().getBegin(), diag::err_using_decl_template_id)
<< SourceRange(Name.TemplateId->LAngleLoc, Name.TemplateId->RAngleLoc);
2010-08-21 17:40:31 +08:00
return 0;
}
DeclarationNameInfo TargetNameInfo = GetNameFromUnqualifiedId(Name);
DeclarationName TargetName = TargetNameInfo.getName();
if (!TargetName)
2010-08-21 17:40:31 +08:00
return 0;
// Warn about using declarations.
// TODO: store that the declaration was written without 'using' and
// talk about access decls instead of using decls in the
// diagnostics.
if (!HasUsingKeyword) {
UsingLoc = Name.getSourceRange().getBegin();
Diag(UsingLoc, diag::warn_access_decl_deprecated)
<< FixItHint::CreateInsertion(SS.getRange().getBegin(), "using ");
}
if (DiagnoseUnexpandedParameterPack(SS, UPPC_UsingDeclaration) ||
DiagnoseUnexpandedParameterPack(TargetNameInfo, UPPC_UsingDeclaration))
return 0;
NamedDecl *UD = BuildUsingDeclaration(S, AS, UsingLoc, SS,
TargetNameInfo, AttrList,
/* IsInstantiation */ false,
IsTypeName, TypenameLoc);
if (UD)
PushOnScopeChains(UD, S, /*AddToContext*/ false);
2010-08-21 17:40:31 +08:00
return UD;
}
/// \brief Determine whether a using declaration considers the given
/// declarations as "equivalent", e.g., if they are redeclarations of
/// the same entity or are both typedefs of the same type.
static bool
IsEquivalentForUsingDecl(ASTContext &Context, NamedDecl *D1, NamedDecl *D2,
bool &SuppressRedeclaration) {
if (D1->getCanonicalDecl() == D2->getCanonicalDecl()) {
SuppressRedeclaration = false;
return true;
}
if (TypedefDecl *TD1 = dyn_cast<TypedefDecl>(D1))
if (TypedefDecl *TD2 = dyn_cast<TypedefDecl>(D2)) {
SuppressRedeclaration = true;
return Context.hasSameType(TD1->getUnderlyingType(),
TD2->getUnderlyingType());
}
return false;
}
/// Determines whether to create a using shadow decl for a particular
/// decl, given the set of decls existing prior to this using lookup.
bool Sema::CheckUsingShadowDecl(UsingDecl *Using, NamedDecl *Orig,
const LookupResult &Previous) {
// Diagnose finding a decl which is not from a base class of the
// current class. We do this now because there are cases where this
// function will silently decide not to build a shadow decl, which
// will pre-empt further diagnostics.
//
// We don't need to do this in C++0x because we do the check once on
// the qualifier.
//
// FIXME: diagnose the following if we care enough:
// struct A { int foo; };
// struct B : A { using A::foo; };
// template <class T> struct C : A {};
// template <class T> struct D : C<T> { using B::foo; } // <---
// This is invalid (during instantiation) in C++03 because B::foo
// resolves to the using decl in B, which is not a base class of D<T>.
// We can't diagnose it immediately because C<T> is an unknown
// specialization. The UsingShadowDecl in D<T> then points directly
// to A::foo, which will look well-formed when we instantiate.
// The right solution is to not collapse the shadow-decl chain.
if (!getLangOptions().CPlusPlus0x && CurContext->isRecord()) {
DeclContext *OrigDC = Orig->getDeclContext();
// Handle enums and anonymous structs.
if (isa<EnumDecl>(OrigDC)) OrigDC = OrigDC->getParent();
CXXRecordDecl *OrigRec = cast<CXXRecordDecl>(OrigDC);
while (OrigRec->isAnonymousStructOrUnion())
OrigRec = cast<CXXRecordDecl>(OrigRec->getDeclContext());
if (cast<CXXRecordDecl>(CurContext)->isProvablyNotDerivedFrom(OrigRec)) {
if (OrigDC == CurContext) {
Diag(Using->getLocation(),
diag::err_using_decl_nested_name_specifier_is_current_class)
<< Using->getNestedNameRange();
Diag(Orig->getLocation(), diag::note_using_decl_target);
return true;
}
Diag(Using->getNestedNameRange().getBegin(),
diag::err_using_decl_nested_name_specifier_is_not_base_class)
<< Using->getTargetNestedNameDecl()
<< cast<CXXRecordDecl>(CurContext)
<< Using->getNestedNameRange();
Diag(Orig->getLocation(), diag::note_using_decl_target);
return true;
}
}
if (Previous.empty()) return false;
NamedDecl *Target = Orig;
if (isa<UsingShadowDecl>(Target))
Target = cast<UsingShadowDecl>(Target)->getTargetDecl();
// If the target happens to be one of the previous declarations, we
// don't have a conflict.
//
// FIXME: but we might be increasing its access, in which case we
// should redeclare it.
NamedDecl *NonTag = 0, *Tag = 0;
for (LookupResult::iterator I = Previous.begin(), E = Previous.end();
I != E; ++I) {
NamedDecl *D = (*I)->getUnderlyingDecl();
bool Result;
if (IsEquivalentForUsingDecl(Context, D, Target, Result))
return Result;
(isa<TagDecl>(D) ? Tag : NonTag) = D;
}
if (Target->isFunctionOrFunctionTemplate()) {
FunctionDecl *FD;
if (isa<FunctionTemplateDecl>(Target))
FD = cast<FunctionTemplateDecl>(Target)->getTemplatedDecl();
else
FD = cast<FunctionDecl>(Target);
NamedDecl *OldDecl = 0;
switch (CheckOverload(0, FD, Previous, OldDecl, /*IsForUsingDecl*/ true)) {
case Ovl_Overload:
return false;
case Ovl_NonFunction:
Diag(Using->getLocation(), diag::err_using_decl_conflict);
break;
// We found a decl with the exact signature.
case Ovl_Match:
// If we're in a record, we want to hide the target, so we
// return true (without a diagnostic) to tell the caller not to
// build a shadow decl.
if (CurContext->isRecord())
return true;
// If we're not in a record, this is an error.
Diag(Using->getLocation(), diag::err_using_decl_conflict);
break;
}
Diag(Target->getLocation(), diag::note_using_decl_target);
Diag(OldDecl->getLocation(), diag::note_using_decl_conflict);
return true;
}
// Target is not a function.
if (isa<TagDecl>(Target)) {
// No conflict between a tag and a non-tag.
if (!Tag) return false;
Diag(Using->getLocation(), diag::err_using_decl_conflict);
Diag(Target->getLocation(), diag::note_using_decl_target);
Diag(Tag->getLocation(), diag::note_using_decl_conflict);
return true;
}
// No conflict between a tag and a non-tag.
if (!NonTag) return false;
Diag(Using->getLocation(), diag::err_using_decl_conflict);
Diag(Target->getLocation(), diag::note_using_decl_target);
Diag(NonTag->getLocation(), diag::note_using_decl_conflict);
return true;
}
/// Builds a shadow declaration corresponding to a 'using' declaration.
UsingShadowDecl *Sema::BuildUsingShadowDecl(Scope *S,
UsingDecl *UD,
NamedDecl *Orig) {
// If we resolved to another shadow declaration, just coalesce them.
NamedDecl *Target = Orig;
if (isa<UsingShadowDecl>(Target)) {
Target = cast<UsingShadowDecl>(Target)->getTargetDecl();
assert(!isa<UsingShadowDecl>(Target) && "nested shadow declaration");
}
UsingShadowDecl *Shadow
= UsingShadowDecl::Create(Context, CurContext,
UD->getLocation(), UD, Target);
UD->addShadowDecl(Shadow);
Shadow->setAccess(UD->getAccess());
if (Orig->isInvalidDecl() || UD->isInvalidDecl())
Shadow->setInvalidDecl();
if (S)
PushOnScopeChains(Shadow, S);
else
CurContext->addDecl(Shadow);
return Shadow;
}
/// Hides a using shadow declaration. This is required by the current
/// using-decl implementation when a resolvable using declaration in a
/// class is followed by a declaration which would hide or override
/// one or more of the using decl's targets; for example:
///
/// struct Base { void foo(int); };
/// struct Derived : Base {
/// using Base::foo;
/// void foo(int);
/// };
///
/// The governing language is C++03 [namespace.udecl]p12:
///
/// When a using-declaration brings names from a base class into a
/// derived class scope, member functions in the derived class
/// override and/or hide member functions with the same name and
/// parameter types in a base class (rather than conflicting).
///
/// There are two ways to implement this:
/// (1) optimistically create shadow decls when they're not hidden
/// by existing declarations, or
/// (2) don't create any shadow decls (or at least don't make them
/// visible) until we've fully parsed/instantiated the class.
/// The problem with (1) is that we might have to retroactively remove
/// a shadow decl, which requires several O(n) operations because the
/// decl structures are (very reasonably) not designed for removal.
/// (2) avoids this but is very fiddly and phase-dependent.
void Sema::HideUsingShadowDecl(Scope *S, UsingShadowDecl *Shadow) {
if (Shadow->getDeclName().getNameKind() ==
DeclarationName::CXXConversionFunctionName)
cast<CXXRecordDecl>(Shadow->getDeclContext())->removeConversion(Shadow);
// Remove it from the DeclContext...
Shadow->getDeclContext()->removeDecl(Shadow);
// ...and the scope, if applicable...
if (S) {
2010-08-21 17:40:31 +08:00
S->RemoveDecl(Shadow);
IdResolver.RemoveDecl(Shadow);
}
// ...and the using decl.
Shadow->getUsingDecl()->removeShadowDecl(Shadow);
// TODO: complain somehow if Shadow was used. It shouldn't
// be possible for this to happen, because...?
}
/// Builds a using declaration.
///
/// \param IsInstantiation - Whether this call arises from an
/// instantiation of an unresolved using declaration. We treat
/// the lookup differently for these declarations.
NamedDecl *Sema::BuildUsingDeclaration(Scope *S, AccessSpecifier AS,
SourceLocation UsingLoc,
CXXScopeSpec &SS,
const DeclarationNameInfo &NameInfo,
AttributeList *AttrList,
bool IsInstantiation,
bool IsTypeName,
SourceLocation TypenameLoc) {
assert(!SS.isInvalid() && "Invalid CXXScopeSpec.");
SourceLocation IdentLoc = NameInfo.getLoc();
assert(IdentLoc.isValid() && "Invalid TargetName location.");
// FIXME: We ignore attributes for now.
if (SS.isEmpty()) {
Diag(IdentLoc, diag::err_using_requires_qualname);
return 0;
}
// Do the redeclaration lookup in the current scope.
LookupResult Previous(*this, NameInfo, LookupUsingDeclName,
ForRedeclaration);
Previous.setHideTags(false);
if (S) {
LookupName(Previous, S);
// It is really dumb that we have to do this.
LookupResult::Filter F = Previous.makeFilter();
while (F.hasNext()) {
NamedDecl *D = F.next();
if (!isDeclInScope(D, CurContext, S))
F.erase();
}
F.done();
} else {
assert(IsInstantiation && "no scope in non-instantiation");
assert(CurContext->isRecord() && "scope not record in instantiation");
LookupQualifiedName(Previous, CurContext);
}
NestedNameSpecifier *NNS = SS.getScopeRep();
// Check for invalid redeclarations.
if (CheckUsingDeclRedeclaration(UsingLoc, IsTypeName, SS, IdentLoc, Previous))
return 0;
// Check for bad qualifiers.
if (CheckUsingDeclQualifier(UsingLoc, SS, IdentLoc))
return 0;
DeclContext *LookupContext = computeDeclContext(SS);
NamedDecl *D;
if (!LookupContext) {
if (IsTypeName) {
// FIXME: not all declaration name kinds are legal here
D = UnresolvedUsingTypenameDecl::Create(Context, CurContext,
UsingLoc, TypenameLoc,
SS.getRange(), NNS,
IdentLoc, NameInfo.getName());
} else {
D = UnresolvedUsingValueDecl::Create(Context, CurContext,
UsingLoc, SS.getRange(),
NNS, NameInfo);
}
} else {
D = UsingDecl::Create(Context, CurContext,
SS.getRange(), UsingLoc, NNS, NameInfo,
IsTypeName);
}
D->setAccess(AS);
CurContext->addDecl(D);
if (!LookupContext) return D;
UsingDecl *UD = cast<UsingDecl>(D);
if (RequireCompleteDeclContext(SS, LookupContext)) {
UD->setInvalidDecl();
return UD;
}
// Constructor inheriting using decls get special treatment.
if (NameInfo.getName().getNameKind() == DeclarationName::CXXConstructorName) {
if (CheckInheritedConstructorUsingDecl(UD))
UD->setInvalidDecl();
return UD;
}
// Otherwise, look up the target name.
LookupResult R(*this, NameInfo, LookupOrdinaryName);
// Unlike most lookups, we don't always want to hide tag
// declarations: tag names are visible through the using declaration
// even if hidden by ordinary names, *except* in a dependent context
// where it's important for the sanity of two-phase lookup.
if (!IsInstantiation)
R.setHideTags(false);
LookupQualifiedName(R, LookupContext);
if (R.empty()) {
Diag(IdentLoc, diag::err_no_member)
<< NameInfo.getName() << LookupContext << SS.getRange();
UD->setInvalidDecl();
return UD;
}
if (R.isAmbiguous()) {
UD->setInvalidDecl();
return UD;
}
if (IsTypeName) {
// If we asked for a typename and got a non-type decl, error out.
if (!R.getAsSingle<TypeDecl>()) {
Diag(IdentLoc, diag::err_using_typename_non_type);
for (LookupResult::iterator I = R.begin(), E = R.end(); I != E; ++I)
Diag((*I)->getUnderlyingDecl()->getLocation(),
diag::note_using_decl_target);
UD->setInvalidDecl();
return UD;
}
} else {
// If we asked for a non-typename and we got a type, error out,
// but only if this is an instantiation of an unresolved using
// decl. Otherwise just silently find the type name.
if (IsInstantiation && R.getAsSingle<TypeDecl>()) {
Diag(IdentLoc, diag::err_using_dependent_value_is_type);
Diag(R.getFoundDecl()->getLocation(), diag::note_using_decl_target);
UD->setInvalidDecl();
return UD;
}
}
// C++0x N2914 [namespace.udecl]p6:
// A using-declaration shall not name a namespace.
if (R.getAsSingle<NamespaceDecl>()) {
Diag(IdentLoc, diag::err_using_decl_can_not_refer_to_namespace)
<< SS.getRange();
UD->setInvalidDecl();
return UD;
}
for (LookupResult::iterator I = R.begin(), E = R.end(); I != E; ++I) {
if (!CheckUsingShadowDecl(UD, *I, Previous))
BuildUsingShadowDecl(S, UD, *I);
}
return UD;
}
/// Additional checks for a using declaration referring to a constructor name.
bool Sema::CheckInheritedConstructorUsingDecl(UsingDecl *UD) {
if (UD->isTypeName()) {
// FIXME: Cannot specify typename when specifying constructor
return true;
}
const Type *SourceType = UD->getTargetNestedNameDecl()->getAsType();
assert(SourceType &&
"Using decl naming constructor doesn't have type in scope spec.");
CXXRecordDecl *TargetClass = cast<CXXRecordDecl>(CurContext);
// Check whether the named type is a direct base class.
CanQualType CanonicalSourceType = SourceType->getCanonicalTypeUnqualified();
CXXRecordDecl::base_class_iterator BaseIt, BaseE;
for (BaseIt = TargetClass->bases_begin(), BaseE = TargetClass->bases_end();
BaseIt != BaseE; ++BaseIt) {
CanQualType BaseType = BaseIt->getType()->getCanonicalTypeUnqualified();
if (CanonicalSourceType == BaseType)
break;
}
if (BaseIt == BaseE) {
// Did not find SourceType in the bases.
Diag(UD->getUsingLocation(),
diag::err_using_decl_constructor_not_in_direct_base)
<< UD->getNameInfo().getSourceRange()
<< QualType(SourceType, 0) << TargetClass;
return true;
}
BaseIt->setInheritConstructors();
return false;
}
/// Checks that the given using declaration is not an invalid
/// redeclaration. Note that this is checking only for the using decl
/// itself, not for any ill-formedness among the UsingShadowDecls.
bool Sema::CheckUsingDeclRedeclaration(SourceLocation UsingLoc,
bool isTypeName,
const CXXScopeSpec &SS,
SourceLocation NameLoc,
const LookupResult &Prev) {
// C++03 [namespace.udecl]p8:
// C++0x [namespace.udecl]p10:
// A using-declaration is a declaration and can therefore be used
// repeatedly where (and only where) multiple declarations are
// allowed.
//
// That's in non-member contexts.
if (!CurContext->getRedeclContext()->isRecord())
return false;
NestedNameSpecifier *Qual
= static_cast<NestedNameSpecifier*>(SS.getScopeRep());
for (LookupResult::iterator I = Prev.begin(), E = Prev.end(); I != E; ++I) {
NamedDecl *D = *I;
bool DTypename;
NestedNameSpecifier *DQual;
if (UsingDecl *UD = dyn_cast<UsingDecl>(D)) {
DTypename = UD->isTypeName();
DQual = UD->getTargetNestedNameDecl();
} else if (UnresolvedUsingValueDecl *UD
= dyn_cast<UnresolvedUsingValueDecl>(D)) {
DTypename = false;
DQual = UD->getTargetNestedNameSpecifier();
} else if (UnresolvedUsingTypenameDecl *UD
= dyn_cast<UnresolvedUsingTypenameDecl>(D)) {
DTypename = true;
DQual = UD->getTargetNestedNameSpecifier();
} else continue;
// using decls differ if one says 'typename' and the other doesn't.
// FIXME: non-dependent using decls?
if (isTypeName != DTypename) continue;
// using decls differ if they name different scopes (but note that
// template instantiation can cause this check to trigger when it
// didn't before instantiation).
if (Context.getCanonicalNestedNameSpecifier(Qual) !=
Context.getCanonicalNestedNameSpecifier(DQual))
continue;
Diag(NameLoc, diag::err_using_decl_redeclaration) << SS.getRange();
Diag(D->getLocation(), diag::note_using_decl) << 1;
return true;
}
return false;
}
/// Checks that the given nested-name qualifier used in a using decl
/// in the current context is appropriately related to the current
/// scope. If an error is found, diagnoses it and returns true.
bool Sema::CheckUsingDeclQualifier(SourceLocation UsingLoc,
const CXXScopeSpec &SS,
SourceLocation NameLoc) {
DeclContext *NamedContext = computeDeclContext(SS);
if (!CurContext->isRecord()) {
// C++03 [namespace.udecl]p3:
// C++0x [namespace.udecl]p8:
// A using-declaration for a class member shall be a member-declaration.
// If we weren't able to compute a valid scope, it must be a
// dependent class scope.
if (!NamedContext || NamedContext->isRecord()) {
Diag(NameLoc, diag::err_using_decl_can_not_refer_to_class_member)
<< SS.getRange();
return true;
}
// Otherwise, everything is known to be fine.
return false;
}
// The current scope is a record.
// If the named context is dependent, we can't decide much.
if (!NamedContext) {
// FIXME: in C++0x, we can diagnose if we can prove that the
// nested-name-specifier does not refer to a base class, which is
// still possible in some cases.
// Otherwise we have to conservatively report that things might be
// okay.
return false;
}
if (!NamedContext->isRecord()) {
// Ideally this would point at the last name in the specifier,
// but we don't have that level of source info.
Diag(SS.getRange().getBegin(),
diag::err_using_decl_nested_name_specifier_is_not_class)
<< (NestedNameSpecifier*) SS.getScopeRep() << SS.getRange();
return true;
}
if (!NamedContext->isDependentContext() &&
RequireCompleteDeclContext(const_cast<CXXScopeSpec&>(SS), NamedContext))
return true;
if (getLangOptions().CPlusPlus0x) {
// C++0x [namespace.udecl]p3:
// In a using-declaration used as a member-declaration, the
// nested-name-specifier shall name a base class of the class
// being defined.
if (cast<CXXRecordDecl>(CurContext)->isProvablyNotDerivedFrom(
cast<CXXRecordDecl>(NamedContext))) {
if (CurContext == NamedContext) {
Diag(NameLoc,
diag::err_using_decl_nested_name_specifier_is_current_class)
<< SS.getRange();
return true;
}
Diag(SS.getRange().getBegin(),
diag::err_using_decl_nested_name_specifier_is_not_base_class)
<< (NestedNameSpecifier*) SS.getScopeRep()
<< cast<CXXRecordDecl>(CurContext)
<< SS.getRange();
return true;
}
return false;
}
// C++03 [namespace.udecl]p4:
// A using-declaration used as a member-declaration shall refer
// to a member of a base class of the class being defined [etc.].
// Salient point: SS doesn't have to name a base class as long as
// lookup only finds members from base classes. Therefore we can
// diagnose here only if we can prove that that can't happen,
// i.e. if the class hierarchies provably don't intersect.
// TODO: it would be nice if "definitely valid" results were cached
// in the UsingDecl and UsingShadowDecl so that these checks didn't
// need to be repeated.
struct UserData {
llvm::DenseSet<const CXXRecordDecl*> Bases;
static bool collect(const CXXRecordDecl *Base, void *OpaqueData) {
UserData *Data = reinterpret_cast<UserData*>(OpaqueData);
Data->Bases.insert(Base);
return true;
}
bool hasDependentBases(const CXXRecordDecl *Class) {
return !Class->forallBases(collect, this);
}
/// Returns true if the base is dependent or is one of the
/// accumulated base classes.
static bool doesNotContain(const CXXRecordDecl *Base, void *OpaqueData) {
UserData *Data = reinterpret_cast<UserData*>(OpaqueData);
return !Data->Bases.count(Base);
}
bool mightShareBases(const CXXRecordDecl *Class) {
return Bases.count(Class) || !Class->forallBases(doesNotContain, this);
}
};
UserData Data;
// Returns false if we find a dependent base.
if (Data.hasDependentBases(cast<CXXRecordDecl>(CurContext)))
return false;
// Returns false if the class has a dependent base or if it or one
// of its bases is present in the base set of the current context.
if (Data.mightShareBases(cast<CXXRecordDecl>(NamedContext)))
return false;
Diag(SS.getRange().getBegin(),
diag::err_using_decl_nested_name_specifier_is_not_base_class)
<< (NestedNameSpecifier*) SS.getScopeRep()
<< cast<CXXRecordDecl>(CurContext)
<< SS.getRange();
return true;
}
2010-08-21 17:40:31 +08:00
Decl *Sema::ActOnNamespaceAliasDef(Scope *S,
SourceLocation NamespaceLoc,
SourceLocation AliasLoc,
IdentifierInfo *Alias,
CXXScopeSpec &SS,
SourceLocation IdentLoc,
IdentifierInfo *Ident) {
// Lookup the namespace name.
LookupResult R(*this, Ident, IdentLoc, LookupNamespaceName);
LookupParsedName(R, S, &SS);
// Check if we have a previous declaration with the same name.
NamedDecl *PrevDecl
= LookupSingleName(S, Alias, AliasLoc, LookupOrdinaryName,
ForRedeclaration);
if (PrevDecl && !isDeclInScope(PrevDecl, CurContext, S))
PrevDecl = 0;
if (PrevDecl) {
if (NamespaceAliasDecl *AD = dyn_cast<NamespaceAliasDecl>(PrevDecl)) {
// We already have an alias with the same name that points to the same
// namespace, so don't create a new one.
// FIXME: At some point, we'll want to create the (redundant)
// declaration to maintain better source information.
if (!R.isAmbiguous() && !R.empty() &&
AD->getNamespace()->Equals(getNamespaceDecl(R.getFoundDecl())))
2010-08-21 17:40:31 +08:00
return 0;
}
unsigned DiagID = isa<NamespaceDecl>(PrevDecl) ? diag::err_redefinition :
diag::err_redefinition_different_kind;
Diag(AliasLoc, DiagID) << Alias;
Diag(PrevDecl->getLocation(), diag::note_previous_definition);
2010-08-21 17:40:31 +08:00
return 0;
}
if (R.isAmbiguous())
2010-08-21 17:40:31 +08:00
return 0;
if (R.empty()) {
if (DeclarationName Corrected = CorrectTypo(R, S, &SS, 0, false,
CTC_NoKeywords, 0)) {
if (R.getAsSingle<NamespaceDecl>() ||
R.getAsSingle<NamespaceAliasDecl>()) {
if (DeclContext *DC = computeDeclContext(SS, false))
Diag(IdentLoc, diag::err_using_directive_member_suggest)
<< Ident << DC << Corrected << SS.getRange()
<< FixItHint::CreateReplacement(IdentLoc, Corrected.getAsString());
else
Diag(IdentLoc, diag::err_using_directive_suggest)
<< Ident << Corrected
<< FixItHint::CreateReplacement(IdentLoc, Corrected.getAsString());
Diag(R.getFoundDecl()->getLocation(), diag::note_namespace_defined_here)
<< Corrected;
Ident = Corrected.getAsIdentifierInfo();
} else {
R.clear();
R.setLookupName(Ident);
}
}
if (R.empty()) {
Diag(NamespaceLoc, diag::err_expected_namespace_name) << SS.getRange();
2010-08-21 17:40:31 +08:00
return 0;
}
}
NamespaceAliasDecl *AliasDecl =
NamespaceAliasDecl::Create(Context, CurContext, NamespaceLoc, AliasLoc,
Alias, SS.getRange(),
(NestedNameSpecifier *)SS.getScopeRep(),
IdentLoc, R.getFoundDecl());
PushOnScopeChains(AliasDecl, S);
2010-08-21 17:40:31 +08:00
return AliasDecl;
}
namespace {
/// \brief Scoped object used to handle the state changes required in Sema
/// to implicitly define the body of a C++ member function;
class ImplicitlyDefinedFunctionScope {
Sema &S;
Sema::ContextRAII SavedContext;
public:
ImplicitlyDefinedFunctionScope(Sema &S, CXXMethodDecl *Method)
: S(S), SavedContext(S, Method)
{
S.PushFunctionScope();
S.PushExpressionEvaluationContext(Sema::PotentiallyEvaluated);
}
~ImplicitlyDefinedFunctionScope() {
S.PopExpressionEvaluationContext();
S.PopFunctionOrBlockScope();
}
};
}
static CXXConstructorDecl *getDefaultConstructorUnsafe(Sema &Self,
CXXRecordDecl *D) {
ASTContext &Context = Self.Context;
QualType ClassType = Context.getTypeDeclType(D);
DeclarationName ConstructorName
= Context.DeclarationNames.getCXXConstructorName(
Context.getCanonicalType(ClassType.getUnqualifiedType()));
DeclContext::lookup_const_iterator Con, ConEnd;
for (llvm::tie(Con, ConEnd) = D->lookup(ConstructorName);
Con != ConEnd; ++Con) {
// FIXME: In C++0x, a constructor template can be a default constructor.
if (isa<FunctionTemplateDecl>(*Con))
continue;
CXXConstructorDecl *Constructor = cast<CXXConstructorDecl>(*Con);
if (Constructor->isDefaultConstructor())
return Constructor;
}
return 0;
}
CXXConstructorDecl *Sema::DeclareImplicitDefaultConstructor(
CXXRecordDecl *ClassDecl) {
// C++ [class.ctor]p5:
// A default constructor for a class X is a constructor of class X
// that can be called without an argument. If there is no
// user-declared constructor for class X, a default constructor is
// implicitly declared. An implicitly-declared default constructor
// is an inline public member of its class.
assert(!ClassDecl->hasUserDeclaredConstructor() &&
"Should not build implicit default constructor!");
// C++ [except.spec]p14:
// An implicitly declared special member function (Clause 12) shall have an
// exception-specification. [...]
ImplicitExceptionSpecification ExceptSpec(Context);
// Direct base-class destructors.
for (CXXRecordDecl::base_class_iterator B = ClassDecl->bases_begin(),
BEnd = ClassDecl->bases_end();
B != BEnd; ++B) {
if (B->isVirtual()) // Handled below.
continue;
if (const RecordType *BaseType = B->getType()->getAs<RecordType>()) {
CXXRecordDecl *BaseClassDecl = cast<CXXRecordDecl>(BaseType->getDecl());
if (!BaseClassDecl->hasDeclaredDefaultConstructor())
ExceptSpec.CalledDecl(DeclareImplicitDefaultConstructor(BaseClassDecl));
else if (CXXConstructorDecl *Constructor
= getDefaultConstructorUnsafe(*this, BaseClassDecl))
ExceptSpec.CalledDecl(Constructor);
}
}
// Virtual base-class destructors.
for (CXXRecordDecl::base_class_iterator B = ClassDecl->vbases_begin(),
BEnd = ClassDecl->vbases_end();
B != BEnd; ++B) {
if (const RecordType *BaseType = B->getType()->getAs<RecordType>()) {
CXXRecordDecl *BaseClassDecl = cast<CXXRecordDecl>(BaseType->getDecl());
if (!BaseClassDecl->hasDeclaredDefaultConstructor())
ExceptSpec.CalledDecl(DeclareImplicitDefaultConstructor(BaseClassDecl));
else if (CXXConstructorDecl *Constructor
= getDefaultConstructorUnsafe(*this, BaseClassDecl))
ExceptSpec.CalledDecl(Constructor);
}
}
// Field destructors.
for (RecordDecl::field_iterator F = ClassDecl->field_begin(),
FEnd = ClassDecl->field_end();
F != FEnd; ++F) {
if (const RecordType *RecordTy
= Context.getBaseElementType(F->getType())->getAs<RecordType>()) {
CXXRecordDecl *FieldClassDecl = cast<CXXRecordDecl>(RecordTy->getDecl());
if (!FieldClassDecl->hasDeclaredDefaultConstructor())
ExceptSpec.CalledDecl(
DeclareImplicitDefaultConstructor(FieldClassDecl));
else if (CXXConstructorDecl *Constructor
= getDefaultConstructorUnsafe(*this, FieldClassDecl))
ExceptSpec.CalledDecl(Constructor);
}
}
FunctionProtoType::ExtProtoInfo EPI;
EPI.HasExceptionSpec = ExceptSpec.hasExceptionSpecification();
EPI.HasAnyExceptionSpec = ExceptSpec.hasAnyExceptionSpecification();
EPI.NumExceptions = ExceptSpec.size();
EPI.Exceptions = ExceptSpec.data();
// Create the actual constructor declaration.
CanQualType ClassType
= Context.getCanonicalType(Context.getTypeDeclType(ClassDecl));
DeclarationName Name
= Context.DeclarationNames.getCXXConstructorName(ClassType);
DeclarationNameInfo NameInfo(Name, ClassDecl->getLocation());
CXXConstructorDecl *DefaultCon
= CXXConstructorDecl::Create(Context, ClassDecl, NameInfo,
Context.getFunctionType(Context.VoidTy,
0, 0, EPI),
/*TInfo=*/0,
/*isExplicit=*/false,
/*isInline=*/true,
/*isImplicitlyDeclared=*/true);
DefaultCon->setAccess(AS_public);
DefaultCon->setImplicit();
DefaultCon->setTrivial(ClassDecl->hasTrivialConstructor());
// Note that we have declared this constructor.
++ASTContext::NumImplicitDefaultConstructorsDeclared;
if (Scope *S = getScopeForContext(ClassDecl))
PushOnScopeChains(DefaultCon, S, false);
ClassDecl->addDecl(DefaultCon);
return DefaultCon;
}
void Sema::DefineImplicitDefaultConstructor(SourceLocation CurrentLocation,
CXXConstructorDecl *Constructor) {
assert((Constructor->isImplicit() && Constructor->isDefaultConstructor() &&
!Constructor->isUsed(false)) &&
"DefineImplicitDefaultConstructor - call it for implicit default ctor");
CXXRecordDecl *ClassDecl = Constructor->getParent();
assert(ClassDecl && "DefineImplicitDefaultConstructor - invalid constructor");
ImplicitlyDefinedFunctionScope Scope(*this, Constructor);
DiagnosticErrorTrap Trap(Diags);
if (SetCtorInitializers(Constructor, 0, 0, /*AnyErrors=*/false) ||
Trap.hasErrorOccurred()) {
Diag(CurrentLocation, diag::note_member_synthesized_at)
<< CXXConstructor << Context.getTagDeclType(ClassDecl);
Constructor->setInvalidDecl();
return;
}
SourceLocation Loc = Constructor->getLocation();
Constructor->setBody(new (Context) CompoundStmt(Context, 0, 0, Loc, Loc));
Constructor->setUsed();
MarkVTableUsed(CurrentLocation, ClassDecl);
}
void Sema::DeclareInheritedConstructors(CXXRecordDecl *ClassDecl) {
// We start with an initial pass over the base classes to collect those that
// inherit constructors from. If there are none, we can forgo all further
// processing.
typedef llvm::SmallVector<const RecordType *, 4> BasesVector;
BasesVector BasesToInheritFrom;
for (CXXRecordDecl::base_class_iterator BaseIt = ClassDecl->bases_begin(),
BaseE = ClassDecl->bases_end();
BaseIt != BaseE; ++BaseIt) {
if (BaseIt->getInheritConstructors()) {
QualType Base = BaseIt->getType();
if (Base->isDependentType()) {
// If we inherit constructors from anything that is dependent, just
// abort processing altogether. We'll get another chance for the
// instantiations.
return;
}
BasesToInheritFrom.push_back(Base->castAs<RecordType>());
}
}
if (BasesToInheritFrom.empty())
return;
// Now collect the constructors that we already have in the current class.
// Those take precedence over inherited constructors.
// C++0x [class.inhctor]p3: [...] a constructor is implicitly declared [...]
// unless there is a user-declared constructor with the same signature in
// the class where the using-declaration appears.
llvm::SmallSet<const Type *, 8> ExistingConstructors;
for (CXXRecordDecl::ctor_iterator CtorIt = ClassDecl->ctor_begin(),
CtorE = ClassDecl->ctor_end();
CtorIt != CtorE; ++CtorIt) {
ExistingConstructors.insert(
Context.getCanonicalType(CtorIt->getType()).getTypePtr());
}
Scope *S = getScopeForContext(ClassDecl);
DeclarationName CreatedCtorName =
Context.DeclarationNames.getCXXConstructorName(
ClassDecl->getTypeForDecl()->getCanonicalTypeUnqualified());
// Now comes the true work.
// First, we keep a map from constructor types to the base that introduced
// them. Needed for finding conflicting constructors. We also keep the
// actually inserted declarations in there, for pretty diagnostics.
typedef std::pair<CanQualType, CXXConstructorDecl *> ConstructorInfo;
typedef llvm::DenseMap<const Type *, ConstructorInfo> ConstructorToSourceMap;
ConstructorToSourceMap InheritedConstructors;
for (BasesVector::iterator BaseIt = BasesToInheritFrom.begin(),
BaseE = BasesToInheritFrom.end();
BaseIt != BaseE; ++BaseIt) {
const RecordType *Base = *BaseIt;
CanQualType CanonicalBase = Base->getCanonicalTypeUnqualified();
CXXRecordDecl *BaseDecl = cast<CXXRecordDecl>(Base->getDecl());
for (CXXRecordDecl::ctor_iterator CtorIt = BaseDecl->ctor_begin(),
CtorE = BaseDecl->ctor_end();
CtorIt != CtorE; ++CtorIt) {
// Find the using declaration for inheriting this base's constructors.
DeclarationName Name =
Context.DeclarationNames.getCXXConstructorName(CanonicalBase);
UsingDecl *UD = dyn_cast_or_null<UsingDecl>(
LookupSingleName(S, Name,SourceLocation(), LookupUsingDeclName));
SourceLocation UsingLoc = UD ? UD->getLocation() :
ClassDecl->getLocation();
// C++0x [class.inhctor]p1: The candidate set of inherited constructors
// from the class X named in the using-declaration consists of actual
// constructors and notional constructors that result from the
// transformation of defaulted parameters as follows:
// - all non-template default constructors of X, and
// - for each non-template constructor of X that has at least one
// parameter with a default argument, the set of constructors that
// results from omitting any ellipsis parameter specification and
// successively omitting parameters with a default argument from the
// end of the parameter-type-list.
CXXConstructorDecl *BaseCtor = *CtorIt;
bool CanBeCopyOrMove = BaseCtor->isCopyOrMoveConstructor();
const FunctionProtoType *BaseCtorType =
BaseCtor->getType()->getAs<FunctionProtoType>();
for (unsigned params = BaseCtor->getMinRequiredArguments(),
maxParams = BaseCtor->getNumParams();
params <= maxParams; ++params) {
// Skip default constructors. They're never inherited.
if (params == 0)
continue;
// Skip copy and move constructors for the same reason.
if (CanBeCopyOrMove && params == 1)
continue;
// Build up a function type for this particular constructor.
// FIXME: The working paper does not consider that the exception spec
// for the inheriting constructor might be larger than that of the
// source. This code doesn't yet, either.
const Type *NewCtorType;
if (params == maxParams)
NewCtorType = BaseCtorType;
else {
llvm::SmallVector<QualType, 16> Args;
for (unsigned i = 0; i < params; ++i) {
Args.push_back(BaseCtorType->getArgType(i));
}
FunctionProtoType::ExtProtoInfo ExtInfo =
BaseCtorType->getExtProtoInfo();
ExtInfo.Variadic = false;
NewCtorType = Context.getFunctionType(BaseCtorType->getResultType(),
Args.data(), params, ExtInfo)
.getTypePtr();
}
const Type *CanonicalNewCtorType =
Context.getCanonicalType(NewCtorType);
// Now that we have the type, first check if the class already has a
// constructor with this signature.
if (ExistingConstructors.count(CanonicalNewCtorType))
continue;
// Then we check if we have already declared an inherited constructor
// with this signature.
std::pair<ConstructorToSourceMap::iterator, bool> result =
InheritedConstructors.insert(std::make_pair(
CanonicalNewCtorType,
std::make_pair(CanonicalBase, (CXXConstructorDecl*)0)));
if (!result.second) {
// Already in the map. If it came from a different class, that's an
// error. Not if it's from the same.
CanQualType PreviousBase = result.first->second.first;
if (CanonicalBase != PreviousBase) {
const CXXConstructorDecl *PrevCtor = result.first->second.second;
const CXXConstructorDecl *PrevBaseCtor =
PrevCtor->getInheritedConstructor();
assert(PrevBaseCtor && "Conflicting constructor was not inherited");
Diag(UsingLoc, diag::err_using_decl_constructor_conflict);
Diag(BaseCtor->getLocation(),
diag::note_using_decl_constructor_conflict_current_ctor);
Diag(PrevBaseCtor->getLocation(),
diag::note_using_decl_constructor_conflict_previous_ctor);
Diag(PrevCtor->getLocation(),
diag::note_using_decl_constructor_conflict_previous_using);
}
continue;
}
// OK, we're there, now add the constructor.
// C++0x [class.inhctor]p8: [...] that would be performed by a
// user-writtern inline constructor [...]
DeclarationNameInfo DNI(CreatedCtorName, UsingLoc);
CXXConstructorDecl *NewCtor = CXXConstructorDecl::Create(
Context, ClassDecl, DNI, QualType(NewCtorType, 0), /*TInfo=*/0,
BaseCtor->isExplicit(), /*Inline=*/true,
/*ImplicitlyDeclared=*/true);
NewCtor->setAccess(BaseCtor->getAccess());
// Build up the parameter decls and add them.
llvm::SmallVector<ParmVarDecl *, 16> ParamDecls;
for (unsigned i = 0; i < params; ++i) {
ParamDecls.push_back(ParmVarDecl::Create(Context, NewCtor, UsingLoc,
/*IdentifierInfo=*/0,
BaseCtorType->getArgType(i),
/*TInfo=*/0, SC_None,
SC_None, /*DefaultArg=*/0));
}
NewCtor->setParams(ParamDecls.data(), ParamDecls.size());
NewCtor->setInheritedConstructor(BaseCtor);
PushOnScopeChains(NewCtor, S, false);
ClassDecl->addDecl(NewCtor);
result.first->second.second = NewCtor;
}
}
}
}
CXXDestructorDecl *Sema::DeclareImplicitDestructor(CXXRecordDecl *ClassDecl) {
// C++ [class.dtor]p2:
// If a class has no user-declared destructor, a destructor is
// declared implicitly. An implicitly-declared destructor is an
// inline public member of its class.
// C++ [except.spec]p14:
// An implicitly declared special member function (Clause 12) shall have
// an exception-specification.
ImplicitExceptionSpecification ExceptSpec(Context);
// Direct base-class destructors.
for (CXXRecordDecl::base_class_iterator B = ClassDecl->bases_begin(),
BEnd = ClassDecl->bases_end();
B != BEnd; ++B) {
if (B->isVirtual()) // Handled below.
continue;
if (const RecordType *BaseType = B->getType()->getAs<RecordType>())
ExceptSpec.CalledDecl(
LookupDestructor(cast<CXXRecordDecl>(BaseType->getDecl())));
}
// Virtual base-class destructors.
for (CXXRecordDecl::base_class_iterator B = ClassDecl->vbases_begin(),
BEnd = ClassDecl->vbases_end();
B != BEnd; ++B) {
if (const RecordType *BaseType = B->getType()->getAs<RecordType>())
ExceptSpec.CalledDecl(
LookupDestructor(cast<CXXRecordDecl>(BaseType->getDecl())));
}
// Field destructors.
for (RecordDecl::field_iterator F = ClassDecl->field_begin(),
FEnd = ClassDecl->field_end();
F != FEnd; ++F) {
if (const RecordType *RecordTy
= Context.getBaseElementType(F->getType())->getAs<RecordType>())
ExceptSpec.CalledDecl(
LookupDestructor(cast<CXXRecordDecl>(RecordTy->getDecl())));
}
// Create the actual destructor declaration.
FunctionProtoType::ExtProtoInfo EPI;
EPI.HasExceptionSpec = ExceptSpec.hasExceptionSpecification();
EPI.HasAnyExceptionSpec = ExceptSpec.hasAnyExceptionSpecification();
EPI.NumExceptions = ExceptSpec.size();
EPI.Exceptions = ExceptSpec.data();
QualType Ty = Context.getFunctionType(Context.VoidTy, 0, 0, EPI);
CanQualType ClassType
= Context.getCanonicalType(Context.getTypeDeclType(ClassDecl));
DeclarationName Name
= Context.DeclarationNames.getCXXDestructorName(ClassType);
DeclarationNameInfo NameInfo(Name, ClassDecl->getLocation());
CXXDestructorDecl *Destructor
= CXXDestructorDecl::Create(Context, ClassDecl, NameInfo, Ty, 0,
/*isInline=*/true,
/*isImplicitlyDeclared=*/true);
Destructor->setAccess(AS_public);
Destructor->setImplicit();
Destructor->setTrivial(ClassDecl->hasTrivialDestructor());
// Note that we have declared this destructor.
++ASTContext::NumImplicitDestructorsDeclared;
// Introduce this destructor into its scope.
if (Scope *S = getScopeForContext(ClassDecl))
PushOnScopeChains(Destructor, S, false);
ClassDecl->addDecl(Destructor);
// This could be uniqued if it ever proves significant.
Destructor->setTypeSourceInfo(Context.getTrivialTypeSourceInfo(Ty));
AddOverriddenMethods(ClassDecl, Destructor);
return Destructor;
}
void Sema::DefineImplicitDestructor(SourceLocation CurrentLocation,
CXXDestructorDecl *Destructor) {
assert((Destructor->isImplicit() && !Destructor->isUsed(false)) &&
"DefineImplicitDestructor - call it for implicit default dtor");
CXXRecordDecl *ClassDecl = Destructor->getParent();
assert(ClassDecl && "DefineImplicitDestructor - invalid destructor");
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
if (Destructor->isInvalidDecl())
return;
ImplicitlyDefinedFunctionScope Scope(*this, Destructor);
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
DiagnosticErrorTrap Trap(Diags);
MarkBaseAndMemberDestructorsReferenced(Destructor->getLocation(),
Destructor->getParent());
if (CheckDestructor(Destructor) || Trap.hasErrorOccurred()) {
Diag(CurrentLocation, diag::note_member_synthesized_at)
<< CXXDestructor << Context.getTagDeclType(ClassDecl);
Destructor->setInvalidDecl();
return;
}
SourceLocation Loc = Destructor->getLocation();
Destructor->setBody(new (Context) CompoundStmt(Context, 0, 0, Loc, Loc));
Destructor->setUsed();
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
MarkVTableUsed(CurrentLocation, ClassDecl);
}
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
/// \brief Builds a statement that copies the given entity from \p From to
/// \c To.
///
/// This routine is used to copy the members of a class with an
/// implicitly-declared copy assignment operator. When the entities being
/// copied are arrays, this routine builds for loops to copy them.
///
/// \param S The Sema object used for type-checking.
///
/// \param Loc The location where the implicit copy is being generated.
///
/// \param T The type of the expressions being copied. Both expressions must
/// have this type.
///
/// \param To The expression we are copying to.
///
/// \param From The expression we are copying from.
///
/// \param CopyingBaseSubobject Whether we're copying a base subobject.
/// Otherwise, it's a non-static member subobject.
///
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
/// \param Depth Internal parameter recording the depth of the recursion.
///
/// \returns A statement or a loop that copies the expressions.
static StmtResult
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
BuildSingleCopyAssign(Sema &S, SourceLocation Loc, QualType T,
Expr *To, Expr *From,
bool CopyingBaseSubobject, unsigned Depth = 0) {
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// C++0x [class.copy]p30:
// Each subobject is assigned in the manner appropriate to its type:
//
// - if the subobject is of class type, the copy assignment operator
// for the class is used (as if by explicit qualification; that is,
// ignoring any possible virtual overriding functions in more derived
// classes);
if (const RecordType *RecordTy = T->getAs<RecordType>()) {
CXXRecordDecl *ClassDecl = cast<CXXRecordDecl>(RecordTy->getDecl());
// Look for operator=.
DeclarationName Name
= S.Context.DeclarationNames.getCXXOperatorName(OO_Equal);
LookupResult OpLookup(S, Name, Loc, Sema::LookupOrdinaryName);
S.LookupQualifiedName(OpLookup, ClassDecl, false);
// Filter out any result that isn't a copy-assignment operator.
LookupResult::Filter F = OpLookup.makeFilter();
while (F.hasNext()) {
NamedDecl *D = F.next();
if (CXXMethodDecl *Method = dyn_cast<CXXMethodDecl>(D))
if (Method->isCopyAssignmentOperator())
continue;
F.erase();
}
F.done();
// Suppress the protected check (C++ [class.protected]) for each of the
// assignment operators we found. This strange dance is required when
// we're assigning via a base classes's copy-assignment operator. To
// ensure that we're getting the right base class subobject (without
// ambiguities), we need to cast "this" to that subobject type; to
// ensure that we don't go through the virtual call mechanism, we need
// to qualify the operator= name with the base class (see below). However,
// this means that if the base class has a protected copy assignment
// operator, the protected member access check will fail. So, we
// rewrite "protected" access to "public" access in this case, since we
// know by construction that we're calling from a derived class.
if (CopyingBaseSubobject) {
for (LookupResult::iterator L = OpLookup.begin(), LEnd = OpLookup.end();
L != LEnd; ++L) {
if (L.getAccess() == AS_protected)
L.setAccess(AS_public);
}
}
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Create the nested-name-specifier that will be used to qualify the
// reference to operator=; this is required to suppress the virtual
// call mechanism.
CXXScopeSpec SS;
SS.setRange(Loc);
SS.setScopeRep(NestedNameSpecifier::Create(S.Context, 0, false,
T.getTypePtr()));
// Create the reference to operator=.
ExprResult OpEqualRef
= S.BuildMemberReferenceExpr(To, T, Loc, /*isArrow=*/false, SS,
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
/*FirstQualifierInScope=*/0, OpLookup,
/*TemplateArgs=*/0,
/*SuppressQualifierCheck=*/true);
if (OpEqualRef.isInvalid())
return StmtError();
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Build the call to the assignment operator.
ExprResult Call = S.BuildCallToMemberFunction(/*Scope=*/0,
OpEqualRef.takeAs<Expr>(),
Loc, &From, 1, Loc);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
if (Call.isInvalid())
return StmtError();
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
return S.Owned(Call.takeAs<Stmt>());
}
// - if the subobject is of scalar type, the built-in assignment
// operator is used.
const ConstantArrayType *ArrayTy = S.Context.getAsConstantArrayType(T);
if (!ArrayTy) {
ExprResult Assignment = S.CreateBuiltinBinOp(Loc, BO_Assign, To, From);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
if (Assignment.isInvalid())
return StmtError();
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
return S.Owned(Assignment.takeAs<Stmt>());
}
// - if the subobject is an array, each element is assigned, in the
// manner appropriate to the element type;
// Construct a loop over the array bounds, e.g.,
//
// for (__SIZE_TYPE__ i0 = 0; i0 != array-size; ++i0)
//
// that will copy each of the array elements.
QualType SizeType = S.Context.getSizeType();
// Create the iteration variable.
IdentifierInfo *IterationVarName = 0;
{
llvm::SmallString<8> Str;
llvm::raw_svector_ostream OS(Str);
OS << "__i" << Depth;
IterationVarName = &S.Context.Idents.get(OS.str());
}
VarDecl *IterationVar = VarDecl::Create(S.Context, S.CurContext, Loc,
IterationVarName, SizeType,
S.Context.getTrivialTypeSourceInfo(SizeType, Loc),
SC_None, SC_None);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Initialize the iteration variable to zero.
llvm::APInt Zero(S.Context.getTypeSize(SizeType), 0);
IterationVar->setInit(IntegerLiteral::Create(S.Context, Zero, SizeType, Loc));
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Create a reference to the iteration variable; we'll use this several
// times throughout.
Expr *IterationVarRef
= S.BuildDeclRefExpr(IterationVar, SizeType, VK_RValue, Loc).take();
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
assert(IterationVarRef && "Reference to invented variable cannot fail!");
// Create the DeclStmt that holds the iteration variable.
Stmt *InitStmt = new (S.Context) DeclStmt(DeclGroupRef(IterationVar),Loc,Loc);
// Create the comparison against the array bound.
llvm::APInt Upper
= ArrayTy->getSize().zextOrTrunc(S.Context.getTypeSize(SizeType));
Expr *Comparison
= new (S.Context) BinaryOperator(IterationVarRef,
IntegerLiteral::Create(S.Context, Upper, SizeType, Loc),
BO_NE, S.Context.BoolTy,
VK_RValue, OK_Ordinary, Loc);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Create the pre-increment of the iteration variable.
Expr *Increment
= new (S.Context) UnaryOperator(IterationVarRef, UO_PreInc, SizeType,
VK_LValue, OK_Ordinary, Loc);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Subscript the "from" and "to" expressions with the iteration variable.
From = AssertSuccess(S.CreateBuiltinArraySubscriptExpr(From, Loc,
IterationVarRef, Loc));
To = AssertSuccess(S.CreateBuiltinArraySubscriptExpr(To, Loc,
IterationVarRef, Loc));
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Build the copy for an individual element of the array.
StmtResult Copy = BuildSingleCopyAssign(S, Loc, ArrayTy->getElementType(),
To, From, CopyingBaseSubobject,
Depth + 1);
if (Copy.isInvalid())
return StmtError();
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Construct the loop that copies all elements of this array.
return S.ActOnForStmt(Loc, Loc, InitStmt,
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
S.MakeFullExpr(Comparison),
2010-08-21 17:40:31 +08:00
0, S.MakeFullExpr(Increment),
Loc, Copy.take());
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
}
/// \brief Determine whether the given class has a copy assignment operator
/// that accepts a const-qualified argument.
static bool hasConstCopyAssignment(Sema &S, const CXXRecordDecl *CClass) {
CXXRecordDecl *Class = const_cast<CXXRecordDecl *>(CClass);
if (!Class->hasDeclaredCopyAssignment())
S.DeclareImplicitCopyAssignment(Class);
QualType ClassType = S.Context.getCanonicalType(S.Context.getTypeDeclType(Class));
DeclarationName OpName
= S.Context.DeclarationNames.getCXXOperatorName(OO_Equal);
DeclContext::lookup_const_iterator Op, OpEnd;
for (llvm::tie(Op, OpEnd) = Class->lookup(OpName); Op != OpEnd; ++Op) {
// C++ [class.copy]p9:
// A user-declared copy assignment operator is a non-static non-template
// member function of class X with exactly one parameter of type X, X&,
// const X&, volatile X& or const volatile X&.
const CXXMethodDecl* Method = dyn_cast<CXXMethodDecl>(*Op);
if (!Method)
continue;
if (Method->isStatic())
continue;
if (Method->getPrimaryTemplate())
continue;
const FunctionProtoType *FnType =
Method->getType()->getAs<FunctionProtoType>();
assert(FnType && "Overloaded operator has no prototype.");
// Don't assert on this; an invalid decl might have been left in the AST.
if (FnType->getNumArgs() != 1 || FnType->isVariadic())
continue;
bool AcceptsConst = true;
QualType ArgType = FnType->getArgType(0);
if (const LValueReferenceType *Ref = ArgType->getAs<LValueReferenceType>()){
ArgType = Ref->getPointeeType();
// Is it a non-const lvalue reference?
if (!ArgType.isConstQualified())
AcceptsConst = false;
}
if (!S.Context.hasSameUnqualifiedType(ArgType, ClassType))
continue;
// We have a single argument of type cv X or cv X&, i.e. we've found the
// copy assignment operator. Return whether it accepts const arguments.
return AcceptsConst;
}
assert(Class->isInvalidDecl() &&
"No copy assignment operator declared in valid code.");
return false;
}
CXXMethodDecl *Sema::DeclareImplicitCopyAssignment(CXXRecordDecl *ClassDecl) {
// Note: The following rules are largely analoguous to the copy
// constructor rules. Note that virtual bases are not taken into account
// for determining the argument type of the operator. Note also that
// operators taking an object instead of a reference are allowed.
// C++ [class.copy]p10:
// If the class definition does not explicitly declare a copy
// assignment operator, one is declared implicitly.
// The implicitly-defined copy assignment operator for a class X
// will have the form
//
// X& X::operator=(const X&)
//
// if
bool HasConstCopyAssignment = true;
// -- each direct base class B of X has a copy assignment operator
// whose parameter is of type const B&, const volatile B& or B,
// and
for (CXXRecordDecl::base_class_iterator Base = ClassDecl->bases_begin(),
BaseEnd = ClassDecl->bases_end();
HasConstCopyAssignment && Base != BaseEnd; ++Base) {
assert(!Base->getType()->isDependentType() &&
"Cannot generate implicit members for class with dependent bases.");
const CXXRecordDecl *BaseClassDecl
= cast<CXXRecordDecl>(Base->getType()->getAs<RecordType>()->getDecl());
HasConstCopyAssignment = hasConstCopyAssignment(*this, BaseClassDecl);
}
// -- for all the nonstatic data members of X that are of a class
// type M (or array thereof), each such class type has a copy
// assignment operator whose parameter is of type const M&,
// const volatile M& or M.
for (CXXRecordDecl::field_iterator Field = ClassDecl->field_begin(),
FieldEnd = ClassDecl->field_end();
HasConstCopyAssignment && Field != FieldEnd;
++Field) {
QualType FieldType = Context.getBaseElementType((*Field)->getType());
if (const RecordType *FieldClassType = FieldType->getAs<RecordType>()) {
const CXXRecordDecl *FieldClassDecl
= cast<CXXRecordDecl>(FieldClassType->getDecl());
HasConstCopyAssignment = hasConstCopyAssignment(*this, FieldClassDecl);
}
}
// Otherwise, the implicitly declared copy assignment operator will
// have the form
//
// X& X::operator=(X&)
QualType ArgType = Context.getTypeDeclType(ClassDecl);
QualType RetType = Context.getLValueReferenceType(ArgType);
if (HasConstCopyAssignment)
ArgType = ArgType.withConst();
ArgType = Context.getLValueReferenceType(ArgType);
// C++ [except.spec]p14:
// An implicitly declared special member function (Clause 12) shall have an
// exception-specification. [...]
ImplicitExceptionSpecification ExceptSpec(Context);
for (CXXRecordDecl::base_class_iterator Base = ClassDecl->bases_begin(),
BaseEnd = ClassDecl->bases_end();
Base != BaseEnd; ++Base) {
CXXRecordDecl *BaseClassDecl
= cast<CXXRecordDecl>(Base->getType()->getAs<RecordType>()->getDecl());
if (!BaseClassDecl->hasDeclaredCopyAssignment())
DeclareImplicitCopyAssignment(BaseClassDecl);
if (CXXMethodDecl *CopyAssign
= BaseClassDecl->getCopyAssignmentOperator(HasConstCopyAssignment))
ExceptSpec.CalledDecl(CopyAssign);
}
for (CXXRecordDecl::field_iterator Field = ClassDecl->field_begin(),
FieldEnd = ClassDecl->field_end();
Field != FieldEnd;
++Field) {
QualType FieldType = Context.getBaseElementType((*Field)->getType());
if (const RecordType *FieldClassType = FieldType->getAs<RecordType>()) {
CXXRecordDecl *FieldClassDecl
= cast<CXXRecordDecl>(FieldClassType->getDecl());
if (!FieldClassDecl->hasDeclaredCopyAssignment())
DeclareImplicitCopyAssignment(FieldClassDecl);
if (CXXMethodDecl *CopyAssign
= FieldClassDecl->getCopyAssignmentOperator(HasConstCopyAssignment))
ExceptSpec.CalledDecl(CopyAssign);
}
}
// An implicitly-declared copy assignment operator is an inline public
// member of its class.
FunctionProtoType::ExtProtoInfo EPI;
EPI.HasExceptionSpec = ExceptSpec.hasExceptionSpecification();
EPI.HasAnyExceptionSpec = ExceptSpec.hasAnyExceptionSpecification();
EPI.NumExceptions = ExceptSpec.size();
EPI.Exceptions = ExceptSpec.data();
DeclarationName Name = Context.DeclarationNames.getCXXOperatorName(OO_Equal);
DeclarationNameInfo NameInfo(Name, ClassDecl->getLocation());
CXXMethodDecl *CopyAssignment
= CXXMethodDecl::Create(Context, ClassDecl, NameInfo,
Context.getFunctionType(RetType, &ArgType, 1, EPI),
/*TInfo=*/0, /*isStatic=*/false,
/*StorageClassAsWritten=*/SC_None,
/*isInline=*/true);
CopyAssignment->setAccess(AS_public);
CopyAssignment->setImplicit();
CopyAssignment->setTrivial(ClassDecl->hasTrivialCopyAssignment());
// Add the parameter to the operator.
ParmVarDecl *FromParam = ParmVarDecl::Create(Context, CopyAssignment,
ClassDecl->getLocation(),
/*Id=*/0,
ArgType, /*TInfo=*/0,
SC_None,
SC_None, 0);
CopyAssignment->setParams(&FromParam, 1);
// Note that we have added this copy-assignment operator.
++ASTContext::NumImplicitCopyAssignmentOperatorsDeclared;
if (Scope *S = getScopeForContext(ClassDecl))
PushOnScopeChains(CopyAssignment, S, false);
ClassDecl->addDecl(CopyAssignment);
AddOverriddenMethods(ClassDecl, CopyAssignment);
return CopyAssignment;
}
void Sema::DefineImplicitCopyAssignment(SourceLocation CurrentLocation,
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
CXXMethodDecl *CopyAssignOperator) {
assert((CopyAssignOperator->isImplicit() &&
CopyAssignOperator->isOverloadedOperator() &&
CopyAssignOperator->getOverloadedOperator() == OO_Equal &&
!CopyAssignOperator->isUsed(false)) &&
"DefineImplicitCopyAssignment called for wrong function");
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
CXXRecordDecl *ClassDecl = CopyAssignOperator->getParent();
if (ClassDecl->isInvalidDecl() || CopyAssignOperator->isInvalidDecl()) {
CopyAssignOperator->setInvalidDecl();
return;
}
CopyAssignOperator->setUsed();
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
ImplicitlyDefinedFunctionScope Scope(*this, CopyAssignOperator);
DiagnosticErrorTrap Trap(Diags);
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// C++0x [class.copy]p30:
// The implicitly-defined or explicitly-defaulted copy assignment operator
// for a non-union class X performs memberwise copy assignment of its
// subobjects. The direct base classes of X are assigned first, in the
// order of their declaration in the base-specifier-list, and then the
// immediate non-static data members of X are assigned, in the order in
// which they were declared in the class definition.
// The statements that form the synthesized function body.
ASTOwningVector<Stmt*> Statements(*this);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// The parameter for the "other" object, which we are copying from.
ParmVarDecl *Other = CopyAssignOperator->getParamDecl(0);
Qualifiers OtherQuals = Other->getType().getQualifiers();
QualType OtherRefType = Other->getType();
if (const LValueReferenceType *OtherRef
= OtherRefType->getAs<LValueReferenceType>()) {
OtherRefType = OtherRef->getPointeeType();
OtherQuals = OtherRefType.getQualifiers();
}
// Our location for everything implicitly-generated.
SourceLocation Loc = CopyAssignOperator->getLocation();
// Construct a reference to the "other" object. We'll be using this
// throughout the generated ASTs.
Expr *OtherRef = BuildDeclRefExpr(Other, OtherRefType, VK_LValue, Loc).take();
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
assert(OtherRef && "Reference to parameter cannot fail!");
// Construct the "this" pointer. We'll be using this throughout the generated
// ASTs.
Expr *This = ActOnCXXThis(Loc).takeAs<Expr>();
assert(This && "Reference to this cannot fail!");
// Assign base classes.
bool Invalid = false;
for (CXXRecordDecl::base_class_iterator Base = ClassDecl->bases_begin(),
E = ClassDecl->bases_end(); Base != E; ++Base) {
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Form the assignment:
// static_cast<Base*>(this)->Base::operator=(static_cast<Base&>(other));
QualType BaseType = Base->getType().getUnqualifiedType();
if (!BaseType->isRecordType()) {
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
Invalid = true;
continue;
}
CXXCastPath BasePath;
BasePath.push_back(Base);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Construct the "from" expression, which is an implicit cast to the
// appropriately-qualified base type.
Expr *From = OtherRef;
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
ImpCastExprToType(From, Context.getQualifiedType(BaseType, OtherQuals),
CK_UncheckedDerivedToBase,
VK_LValue, &BasePath);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Dereference "this".
ExprResult To = CreateBuiltinUnaryOp(Loc, UO_Deref, This);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Implicitly cast "this" to the appropriately-qualified base type.
Expr *ToE = To.takeAs<Expr>();
ImpCastExprToType(ToE,
Context.getCVRQualifiedType(BaseType,
CopyAssignOperator->getTypeQualifiers()),
CK_UncheckedDerivedToBase,
VK_LValue, &BasePath);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
To = Owned(ToE);
// Build the copy.
StmtResult Copy = BuildSingleCopyAssign(*this, Loc, BaseType,
To.get(), From,
/*CopyingBaseSubobject=*/true);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
if (Copy.isInvalid()) {
Diag(CurrentLocation, diag::note_member_synthesized_at)
<< CXXCopyAssignment << Context.getTagDeclType(ClassDecl);
CopyAssignOperator->setInvalidDecl();
return;
}
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Success! Record the copy.
Statements.push_back(Copy.takeAs<Expr>());
}
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// \brief Reference to the __builtin_memcpy function.
Expr *BuiltinMemCpyRef = 0;
// \brief Reference to the __builtin_objc_memmove_collectable function.
Expr *CollectableMemCpyRef = 0;
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Assign non-static members.
for (CXXRecordDecl::field_iterator Field = ClassDecl->field_begin(),
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
FieldEnd = ClassDecl->field_end();
Field != FieldEnd; ++Field) {
// Check for members of reference type; we can't copy those.
if (Field->getType()->isReferenceType()) {
Diag(ClassDecl->getLocation(), diag::err_uninitialized_member_for_assign)
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
<< Context.getTagDeclType(ClassDecl) << 0 << Field->getDeclName();
Diag(Field->getLocation(), diag::note_declared_at);
Diag(CurrentLocation, diag::note_member_synthesized_at)
<< CXXCopyAssignment << Context.getTagDeclType(ClassDecl);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
Invalid = true;
continue;
}
// Check for members of const-qualified, non-class type.
QualType BaseType = Context.getBaseElementType(Field->getType());
if (!BaseType->getAs<RecordType>() && BaseType.isConstQualified()) {
Diag(ClassDecl->getLocation(), diag::err_uninitialized_member_for_assign)
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
<< Context.getTagDeclType(ClassDecl) << 1 << Field->getDeclName();
Diag(Field->getLocation(), diag::note_declared_at);
Diag(CurrentLocation, diag::note_member_synthesized_at)
<< CXXCopyAssignment << Context.getTagDeclType(ClassDecl);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
Invalid = true;
continue;
}
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
QualType FieldType = Field->getType().getNonReferenceType();
if (FieldType->isIncompleteArrayType()) {
assert(ClassDecl->hasFlexibleArrayMember() &&
"Incomplete array type is not valid");
continue;
}
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Build references to the field in the object we're copying from and to.
CXXScopeSpec SS; // Intentionally empty
LookupResult MemberLookup(*this, Field->getDeclName(), Loc,
LookupMemberName);
MemberLookup.addDecl(*Field);
MemberLookup.resolveKind();
ExprResult From = BuildMemberReferenceExpr(OtherRef, OtherRefType,
Loc, /*IsArrow=*/false,
SS, 0, MemberLookup, 0);
ExprResult To = BuildMemberReferenceExpr(This, This->getType(),
Loc, /*IsArrow=*/true,
SS, 0, MemberLookup, 0);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
assert(!From.isInvalid() && "Implicit field reference cannot fail");
assert(!To.isInvalid() && "Implicit field reference cannot fail");
// If the field should be copied with __builtin_memcpy rather than via
// explicit assignments, do so. This optimization only applies for arrays
// of scalars and arrays of class type with trivial copy-assignment
// operators.
if (FieldType->isArrayType() &&
(!BaseType->isRecordType() ||
cast<CXXRecordDecl>(BaseType->getAs<RecordType>()->getDecl())
->hasTrivialCopyAssignment())) {
// Compute the size of the memory buffer to be copied.
QualType SizeType = Context.getSizeType();
llvm::APInt Size(Context.getTypeSize(SizeType),
Context.getTypeSizeInChars(BaseType).getQuantity());
for (const ConstantArrayType *Array
= Context.getAsConstantArrayType(FieldType);
Array;
Array = Context.getAsConstantArrayType(Array->getElementType())) {
llvm::APInt ArraySize
= Array->getSize().zextOrTrunc(Size.getBitWidth());
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
Size *= ArraySize;
}
// Take the address of the field references for "from" and "to".
From = CreateBuiltinUnaryOp(Loc, UO_AddrOf, From.get());
To = CreateBuiltinUnaryOp(Loc, UO_AddrOf, To.get());
bool NeedsCollectableMemCpy =
(BaseType->isRecordType() &&
BaseType->getAs<RecordType>()->getDecl()->hasObjectMember());
if (NeedsCollectableMemCpy) {
if (!CollectableMemCpyRef) {
// Create a reference to the __builtin_objc_memmove_collectable function.
LookupResult R(*this,
&Context.Idents.get("__builtin_objc_memmove_collectable"),
Loc, LookupOrdinaryName);
LookupName(R, TUScope, true);
FunctionDecl *CollectableMemCpy = R.getAsSingle<FunctionDecl>();
if (!CollectableMemCpy) {
// Something went horribly wrong earlier, and we will have
// complained about it.
Invalid = true;
continue;
}
CollectableMemCpyRef = BuildDeclRefExpr(CollectableMemCpy,
CollectableMemCpy->getType(),
VK_LValue, Loc, 0).take();
assert(CollectableMemCpyRef && "Builtin reference cannot fail");
}
}
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
// Create a reference to the __builtin_memcpy builtin function.
else if (!BuiltinMemCpyRef) {
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
LookupResult R(*this, &Context.Idents.get("__builtin_memcpy"), Loc,
LookupOrdinaryName);
LookupName(R, TUScope, true);
FunctionDecl *BuiltinMemCpy = R.getAsSingle<FunctionDecl>();
if (!BuiltinMemCpy) {
// Something went horribly wrong earlier, and we will have complained
// about it.
Invalid = true;
continue;
}
BuiltinMemCpyRef = BuildDeclRefExpr(BuiltinMemCpy,
BuiltinMemCpy->getType(),
VK_LValue, Loc, 0).take();
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
assert(BuiltinMemCpyRef && "Builtin reference cannot fail");
}
ASTOwningVector<Expr*> CallArgs(*this);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
CallArgs.push_back(To.takeAs<Expr>());
CallArgs.push_back(From.takeAs<Expr>());
CallArgs.push_back(IntegerLiteral::Create(Context, Size, SizeType, Loc));
ExprResult Call = ExprError();
if (NeedsCollectableMemCpy)
Call = ActOnCallExpr(/*Scope=*/0,
CollectableMemCpyRef,
Loc, move_arg(CallArgs),
Loc);
else
Call = ActOnCallExpr(/*Scope=*/0,
BuiltinMemCpyRef,
Loc, move_arg(CallArgs),
Loc);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
assert(!Call.isInvalid() && "Call to __builtin_memcpy cannot fail!");
Statements.push_back(Call.takeAs<Expr>());
continue;
}
// Build the copy of this field.
StmtResult Copy = BuildSingleCopyAssign(*this, Loc, FieldType,
To.get(), From.get(),
/*CopyingBaseSubobject=*/false);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
if (Copy.isInvalid()) {
Diag(CurrentLocation, diag::note_member_synthesized_at)
<< CXXCopyAssignment << Context.getTagDeclType(ClassDecl);
CopyAssignOperator->setInvalidDecl();
return;
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
}
// Success! Record the copy.
Statements.push_back(Copy.takeAs<Stmt>());
}
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
if (!Invalid) {
// Add a "return *this;"
ExprResult ThisObj = CreateBuiltinUnaryOp(Loc, UO_Deref, This);
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
StmtResult Return = ActOnReturnStmt(Loc, ThisObj.get());
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
if (Return.isInvalid())
Invalid = true;
else {
Statements.push_back(Return.takeAs<Stmt>());
if (Trap.hasErrorOccurred()) {
Diag(CurrentLocation, diag::note_member_synthesized_at)
<< CXXCopyAssignment << Context.getTagDeclType(ClassDecl);
Invalid = true;
}
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
}
}
if (Invalid) {
CopyAssignOperator->setInvalidDecl();
return;
}
StmtResult Body = ActOnCompoundStmt(Loc, Loc, move_arg(Statements),
Complete reimplementation of the synthesis for implicitly-defined copy assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. llvm-svn: 102853
2010-05-02 04:49:11 +08:00
/*isStmtExpr=*/false);
assert(!Body.isInvalid() && "Compound statement creation cannot fail");
CopyAssignOperator->setBody(Body.takeAs<Stmt>());
}
CXXConstructorDecl *Sema::DeclareImplicitCopyConstructor(
CXXRecordDecl *ClassDecl) {
// C++ [class.copy]p4:
// If the class definition does not explicitly declare a copy
// constructor, one is declared implicitly.
// C++ [class.copy]p5:
// The implicitly-declared copy constructor for a class X will
// have the form
//
// X::X(const X&)
//
// if
bool HasConstCopyConstructor = true;
// -- each direct or virtual base class B of X has a copy
// constructor whose first parameter is of type const B& or
// const volatile B&, and
for (CXXRecordDecl::base_class_iterator Base = ClassDecl->bases_begin(),
BaseEnd = ClassDecl->bases_end();
HasConstCopyConstructor && Base != BaseEnd;
++Base) {
// Virtual bases are handled below.
if (Base->isVirtual())
continue;
CXXRecordDecl *BaseClassDecl
= cast<CXXRecordDecl>(Base->getType()->getAs<RecordType>()->getDecl());
if (!BaseClassDecl->hasDeclaredCopyConstructor())
DeclareImplicitCopyConstructor(BaseClassDecl);
HasConstCopyConstructor
= BaseClassDecl->hasConstCopyConstructor(Context);
}
for (CXXRecordDecl::base_class_iterator Base = ClassDecl->vbases_begin(),
BaseEnd = ClassDecl->vbases_end();
HasConstCopyConstructor && Base != BaseEnd;
++Base) {
CXXRecordDecl *BaseClassDecl
= cast<CXXRecordDecl>(Base->getType()->getAs<RecordType>()->getDecl());
if (!BaseClassDecl->hasDeclaredCopyConstructor())
DeclareImplicitCopyConstructor(BaseClassDecl);
HasConstCopyConstructor
= BaseClassDecl->hasConstCopyConstructor(Context);
}
// -- for all the nonstatic data members of X that are of a
// class type M (or array thereof), each such class type
// has a copy constructor whose first parameter is of type
// const M& or const volatile M&.
for (CXXRecordDecl::field_iterator Field = ClassDecl->field_begin(),
FieldEnd = ClassDecl->field_end();
HasConstCopyConstructor && Field != FieldEnd;
++Field) {
QualType FieldType = Context.getBaseElementType((*Field)->getType());
if (const RecordType *FieldClassType = FieldType->getAs<RecordType>()) {
CXXRecordDecl *FieldClassDecl
= cast<CXXRecordDecl>(FieldClassType->getDecl());
if (!FieldClassDecl->hasDeclaredCopyConstructor())
DeclareImplicitCopyConstructor(FieldClassDecl);
HasConstCopyConstructor
= FieldClassDecl->hasConstCopyConstructor(Context);
}
}
// Otherwise, the implicitly declared copy constructor will have
// the form
//
// X::X(X&)
QualType ClassType = Context.getTypeDeclType(ClassDecl);
QualType ArgType = ClassType;
if (HasConstCopyConstructor)
ArgType = ArgType.withConst();
ArgType = Context.getLValueReferenceType(ArgType);
// C++ [except.spec]p14:
// An implicitly declared special member function (Clause 12) shall have an
// exception-specification. [...]
ImplicitExceptionSpecification ExceptSpec(Context);
unsigned Quals = HasConstCopyConstructor? Qualifiers::Const : 0;
for (CXXRecordDecl::base_class_iterator Base = ClassDecl->bases_begin(),
BaseEnd = ClassDecl->bases_end();
Base != BaseEnd;
++Base) {
// Virtual bases are handled below.
if (Base->isVirtual())
continue;
CXXRecordDecl *BaseClassDecl
= cast<CXXRecordDecl>(Base->getType()->getAs<RecordType>()->getDecl());
if (!BaseClassDecl->hasDeclaredCopyConstructor())
DeclareImplicitCopyConstructor(BaseClassDecl);
if (CXXConstructorDecl *CopyConstructor
= BaseClassDecl->getCopyConstructor(Context, Quals))
ExceptSpec.CalledDecl(CopyConstructor);
}
for (CXXRecordDecl::base_class_iterator Base = ClassDecl->vbases_begin(),
BaseEnd = ClassDecl->vbases_end();
Base != BaseEnd;
++Base) {
CXXRecordDecl *BaseClassDecl
= cast<CXXRecordDecl>(Base->getType()->getAs<RecordType>()->getDecl());
if (!BaseClassDecl->hasDeclaredCopyConstructor())
DeclareImplicitCopyConstructor(BaseClassDecl);
if (CXXConstructorDecl *CopyConstructor
= BaseClassDecl->getCopyConstructor(Context, Quals))
ExceptSpec.CalledDecl(CopyConstructor);
}
for (CXXRecordDecl::field_iterator Field = ClassDecl->field_begin(),
FieldEnd = ClassDecl->field_end();
Field != FieldEnd;
++Field) {
QualType FieldType = Context.getBaseElementType((*Field)->getType());
if (const RecordType *FieldClassType = FieldType->getAs<RecordType>()) {
CXXRecordDecl *FieldClassDecl
= cast<CXXRecordDecl>(FieldClassType->getDecl());
if (!FieldClassDecl->hasDeclaredCopyConstructor())
DeclareImplicitCopyConstructor(FieldClassDecl);
if (CXXConstructorDecl *CopyConstructor
= FieldClassDecl->getCopyConstructor(Context, Quals))
ExceptSpec.CalledDecl(CopyConstructor);
}
}
// An implicitly-declared copy constructor is an inline public
// member of its class.
FunctionProtoType::ExtProtoInfo EPI;
EPI.HasExceptionSpec = ExceptSpec.hasExceptionSpecification();
EPI.HasAnyExceptionSpec = ExceptSpec.hasAnyExceptionSpecification();
EPI.NumExceptions = ExceptSpec.size();
EPI.Exceptions = ExceptSpec.data();
DeclarationName Name
= Context.DeclarationNames.getCXXConstructorName(
Context.getCanonicalType(ClassType));
DeclarationNameInfo NameInfo(Name, ClassDecl->getLocation());
CXXConstructorDecl *CopyConstructor
= CXXConstructorDecl::Create(Context, ClassDecl, NameInfo,
Context.getFunctionType(Context.VoidTy,
&ArgType, 1, EPI),
/*TInfo=*/0,
/*isExplicit=*/false,
/*isInline=*/true,
/*isImplicitlyDeclared=*/true);
CopyConstructor->setAccess(AS_public);
CopyConstructor->setTrivial(ClassDecl->hasTrivialCopyConstructor());
// Note that we have declared this constructor.
++ASTContext::NumImplicitCopyConstructorsDeclared;
// Add the parameter to the constructor.
ParmVarDecl *FromParam = ParmVarDecl::Create(Context, CopyConstructor,
ClassDecl->getLocation(),
/*IdentifierInfo=*/0,
ArgType, /*TInfo=*/0,
SC_None,
SC_None, 0);
CopyConstructor->setParams(&FromParam, 1);
if (Scope *S = getScopeForContext(ClassDecl))
PushOnScopeChains(CopyConstructor, S, false);
ClassDecl->addDecl(CopyConstructor);
return CopyConstructor;
}
void Sema::DefineImplicitCopyConstructor(SourceLocation CurrentLocation,
CXXConstructorDecl *CopyConstructor,
unsigned TypeQuals) {
assert((CopyConstructor->isImplicit() &&
CopyConstructor->isCopyConstructor(TypeQuals) &&
!CopyConstructor->isUsed(false)) &&
"DefineImplicitCopyConstructor - call it for implicit copy ctor");
2010-04-24 00:24:12 +08:00
CXXRecordDecl *ClassDecl = CopyConstructor->getParent();
assert(ClassDecl && "DefineImplicitCopyConstructor - invalid constructor");
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
ImplicitlyDefinedFunctionScope Scope(*this, CopyConstructor);
DiagnosticErrorTrap Trap(Diags);
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
if (SetCtorInitializers(CopyConstructor, 0, 0, /*AnyErrors=*/false) ||
Trap.hasErrorOccurred()) {
Diag(CurrentLocation, diag::note_member_synthesized_at)
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
<< CXXCopyConstructor << Context.getTagDeclType(ClassDecl);
CopyConstructor->setInvalidDecl();
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
} else {
CopyConstructor->setBody(ActOnCompoundStmt(CopyConstructor->getLocation(),
CopyConstructor->getLocation(),
MultiStmtArg(*this, 0, 0),
/*isStmtExpr=*/false)
.takeAs<Stmt>());
}
Reimplement code generation for copying fields in the implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. llvm-svn: 103079
2010-05-05 13:51:00 +08:00
CopyConstructor->setUsed();
}
ExprResult
Sema::BuildCXXConstructExpr(SourceLocation ConstructLoc, QualType DeclInitType,
CXXConstructorDecl *Constructor,
MultiExprArg ExprArgs,
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
bool RequiresZeroInit,
unsigned ConstructKind,
SourceRange ParenRange) {
bool Elidable = false;
// C++0x [class.copy]p34:
// When certain criteria are met, an implementation is allowed to
// omit the copy/move construction of a class object, even if the
// copy/move constructor and/or destructor for the object have
// side effects. [...]
// - when a temporary class object that has not been bound to a
// reference (12.2) would be copied/moved to a class object
// with the same cv-unqualified type, the copy/move operation
// can be omitted by constructing the temporary object
// directly into the target of the omitted copy/move
if (ConstructKind == CXXConstructExpr::CK_Complete &&
Constructor->isCopyOrMoveConstructor() && ExprArgs.size() >= 1) {
Expr *SubExpr = ((Expr **)ExprArgs.get())[0];
Elidable = SubExpr->isTemporaryObject(Context, Constructor->getParent());
}
return BuildCXXConstructExpr(ConstructLoc, DeclInitType, Constructor,
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
Elidable, move(ExprArgs), RequiresZeroInit,
ConstructKind, ParenRange);
}
/// BuildCXXConstructExpr - Creates a complete call to a constructor,
/// including handling of its default argument expressions.
ExprResult
Sema::BuildCXXConstructExpr(SourceLocation ConstructLoc, QualType DeclInitType,
CXXConstructorDecl *Constructor, bool Elidable,
MultiExprArg ExprArgs,
Rework base and member initialization in constructors, with several (necessarily simultaneous) changes: - CXXBaseOrMemberInitializer now contains only a single initializer rather than a set of initialiation arguments + a constructor. The single initializer covers all aspects of initialization, including constructor calls as necessary but also cleanup of temporaries created by the initializer (which we never handled before!). - Rework + simplify code generation for CXXBaseOrMemberInitializers, since we can now just emit the initializer as an initializer. - Switched base and member initialization over to the new initialization code (InitializationSequence), so that it - Improved diagnostics for the new initialization code when initializing bases and members, to match the diagnostics produced by the previous (special-purpose) code. - Simplify the representation of type-checked constructor initializers in templates; instead of keeping the fully-type-checked AST, which is rather hard to undo at template instantiation time, throw away the type-checked AST and store the raw expressions in the AST. This simplifies instantiation, but loses a little but of information in the AST. - When type-checking implicit base or member initializers within a dependent context, don't add the generated initializers into the AST, because they'll look like they were explicit. - Record in CXXConstructExpr when the constructor call is to initialize a base class, so that CodeGen does not have to infer it from context. This ensures that we call the right kind of constructor. There are also a few "opportunity" fixes here that were needed to not regress, for example: - Diagnose default-initialization of a const-qualified class that does not have a user-declared default constructor. We had this diagnostic specifically for bases and members, but missed it for variables. That's fixed now. - When defining the implicit constructors, destructor, and copy-assignment operator, set the CurContext to that constructor when we're defining the body. llvm-svn: 94952
2010-01-31 17:12:51 +08:00
bool RequiresZeroInit,
unsigned ConstructKind,
SourceRange ParenRange) {
unsigned NumExprs = ExprArgs.size();
Expr **Exprs = (Expr **)ExprArgs.release();
MarkDeclarationReferenced(ConstructLoc, Constructor);
return Owned(CXXConstructExpr::Create(Context, DeclInitType, ConstructLoc,
Constructor, Elidable, Exprs, NumExprs,
RequiresZeroInit,
static_cast<CXXConstructExpr::ConstructionKind>(ConstructKind),
ParenRange));
}
bool Sema::InitializeVarWithConstructor(VarDecl *VD,
CXXConstructorDecl *Constructor,
MultiExprArg Exprs) {
// FIXME: Provide the correct paren SourceRange when available.
ExprResult TempResult =
BuildCXXConstructExpr(VD->getLocation(), VD->getType(), Constructor,
move(Exprs), false, CXXConstructExpr::CK_Complete,
SourceRange());
if (TempResult.isInvalid())
return true;
Expr *Temp = TempResult.takeAs<Expr>();
CheckImplicitConversions(Temp, VD->getLocation());
MarkDeclarationReferenced(VD->getLocation(), Constructor);
Temp = MaybeCreateExprWithCleanups(Temp);
VD->setInit(Temp);
return false;
}
void Sema::FinalizeVarWithDestructor(VarDecl *VD, const RecordType *Record) {
CXXRecordDecl *ClassDecl = cast<CXXRecordDecl>(Record->getDecl());
if (!ClassDecl->isInvalidDecl() && !VD->isInvalidDecl() &&
!ClassDecl->hasTrivialDestructor() && !ClassDecl->isDependentContext()) {
CXXDestructorDecl *Destructor = LookupDestructor(ClassDecl);
MarkDeclarationReferenced(VD->getLocation(), Destructor);
CheckDestructorAccess(VD->getLocation(), Destructor,
PDiag(diag::err_access_dtor_var)
<< VD->getDeclName()
<< VD->getType());
// TODO: this should be re-enabled for static locals by !CXAAtExit
if (!VD->isInvalidDecl() && VD->hasGlobalStorage() && !VD->isStaticLocal())
Diag(VD->getLocation(), diag::warn_global_destructor);
}
}
/// AddCXXDirectInitializerToDecl - This action is called immediately after
/// ActOnDeclarator, when a C++ direct initializer is present.
/// e.g: "int x(1);"
2010-08-21 17:40:31 +08:00
void Sema::AddCXXDirectInitializerToDecl(Decl *RealDecl,
SourceLocation LParenLoc,
MultiExprArg Exprs,
SourceLocation RParenLoc) {
2009-12-25 03:19:26 +08:00
assert(Exprs.size() != 0 && Exprs.get() && "missing expressions");
// If there is no declaration, there was an error parsing it. Just ignore
// the initializer.
if (RealDecl == 0)
return;
VarDecl *VDecl = dyn_cast<VarDecl>(RealDecl);
if (!VDecl) {
Diag(RealDecl->getLocation(), diag::err_illegal_initializer);
RealDecl->setInvalidDecl();
return;
}
// We will represent direct-initialization similarly to copy-initialization:
// int x(1); -as-> int x = 1;
// ClassType x(a,b,c); -as-> ClassType x = ClassType(a,b,c);
//
// Clients that want to distinguish between the two forms, can check for
// direct initializer using VarDecl::hasCXXDirectInitializer().
// A major benefit is that clients that don't particularly care about which
// exactly form was it (like the CodeGen) can handle both cases without
// special case code.
// C++ 8.5p11:
// The form of initialization (using parentheses or '=') is generally
// insignificant, but does matter when the entity being initialized has a
// class type.
if (!VDecl->getType()->isDependentType() &&
RequireCompleteType(VDecl->getLocation(), VDecl->getType(),
diag::err_typecheck_decl_incomplete_type)) {
VDecl->setInvalidDecl();
return;
}
// The variable can not have an abstract class type.
if (RequireNonAbstractType(VDecl->getLocation(), VDecl->getType(),
diag::err_abstract_type_in_decl,
AbstractVariableType))
VDecl->setInvalidDecl();
const VarDecl *Def;
if ((Def = VDecl->getDefinition()) && Def != VDecl) {
Diag(VDecl->getLocation(), diag::err_redefinition)
<< VDecl->getDeclName();
Diag(Def->getLocation(), diag::note_previous_definition);
VDecl->setInvalidDecl();
return;
}
// C++ [class.static.data]p4
// If a static data member is of const integral or const
// enumeration type, its declaration in the class definition can
// specify a constant-initializer which shall be an integral
// constant expression (5.19). In that case, the member can appear
// in integral constant expressions. The member shall still be
// defined in a namespace scope if it is used in the program and the
// namespace scope definition shall not contain an initializer.
//
// We already performed a redefinition check above, but for static
// data members we also need to check whether there was an in-class
// declaration with an initializer.
const VarDecl* PrevInit = 0;
if (VDecl->isStaticDataMember() && VDecl->getAnyInitializer(PrevInit)) {
Diag(VDecl->getLocation(), diag::err_redefinition) << VDecl->getDeclName();
Diag(PrevInit->getLocation(), diag::note_previous_definition);
return;
}
bool IsDependent = false;
for (unsigned I = 0, N = Exprs.size(); I != N; ++I) {
if (DiagnoseUnexpandedParameterPack(Exprs.get()[I], UPPC_Expression)) {
VDecl->setInvalidDecl();
return;
}
if (Exprs.get()[I]->isTypeDependent())
IsDependent = true;
}
// If either the declaration has a dependent type or if any of the
// expressions is type-dependent, we represent the initialization
// via a ParenListExpr for later use during template instantiation.
if (VDecl->getType()->isDependentType() || IsDependent) {
// Let clients know that initialization was done with a direct initializer.
VDecl->setCXXDirectInitializer(true);
// Store the initialization expressions as a ParenListExpr.
unsigned NumExprs = Exprs.size();
VDecl->setInit(new (Context) ParenListExpr(Context, LParenLoc,
(Expr **)Exprs.release(),
NumExprs, RParenLoc));
return;
}
// Capture the variable that is being initialized and the style of
// initialization.
InitializedEntity Entity = InitializedEntity::InitializeVariable(VDecl);
// FIXME: Poor source location information.
InitializationKind Kind
= InitializationKind::CreateDirect(VDecl->getLocation(),
LParenLoc, RParenLoc);
InitializationSequence InitSeq(*this, Entity, Kind,
Exprs.get(), Exprs.size());
ExprResult Result = InitSeq.Perform(*this, Entity, Kind, move(Exprs));
if (Result.isInvalid()) {
VDecl->setInvalidDecl();
return;
}
CheckImplicitConversions(Result.get(), LParenLoc);
Result = MaybeCreateExprWithCleanups(Result);
VDecl->setInit(Result.takeAs<Expr>());
VDecl->setCXXDirectInitializer(true);
CheckCompleteVariableDeclaration(VDecl);
}
/// \brief Given a constructor and the set of arguments provided for the
/// constructor, convert the arguments and add any required default arguments
/// to form a proper call to this constructor.
///
/// \returns true if an error occurred, false otherwise.
bool
Sema::CompleteConstructorCall(CXXConstructorDecl *Constructor,
MultiExprArg ArgsPtr,
SourceLocation Loc,
ASTOwningVector<Expr*> &ConvertedArgs) {
// FIXME: This duplicates a lot of code from Sema::ConvertArgumentsForCall.
unsigned NumArgs = ArgsPtr.size();
Expr **Args = (Expr **)ArgsPtr.get();
const FunctionProtoType *Proto
= Constructor->getType()->getAs<FunctionProtoType>();
assert(Proto && "Constructor without a prototype?");
unsigned NumArgsInProto = Proto->getNumArgs();
// If too few arguments are available, we'll fill in the rest with defaults.
if (NumArgs < NumArgsInProto)
ConvertedArgs.reserve(NumArgsInProto);
else
ConvertedArgs.reserve(NumArgs);
VariadicCallType CallType =
Proto->isVariadic() ? VariadicConstructor : VariadicDoesNotApply;
llvm::SmallVector<Expr *, 8> AllArgs;
bool Invalid = GatherArgumentsForCall(Loc, Constructor,
Proto, 0, Args, NumArgs, AllArgs,
CallType);
for (unsigned i =0, size = AllArgs.size(); i < size; i++)
ConvertedArgs.push_back(AllArgs[i]);
return Invalid;
}
static inline bool
CheckOperatorNewDeleteDeclarationScope(Sema &SemaRef,
const FunctionDecl *FnDecl) {
const DeclContext *DC = FnDecl->getDeclContext()->getRedeclContext();
if (isa<NamespaceDecl>(DC)) {
return SemaRef.Diag(FnDecl->getLocation(),
diag::err_operator_new_delete_declared_in_namespace)
<< FnDecl->getDeclName();
}
if (isa<TranslationUnitDecl>(DC) &&
FnDecl->getStorageClass() == SC_Static) {
return SemaRef.Diag(FnDecl->getLocation(),
diag::err_operator_new_delete_declared_static)
<< FnDecl->getDeclName();
}
2009-12-12 10:43:16 +08:00
return false;
}
static inline bool
CheckOperatorNewDeleteTypes(Sema &SemaRef, const FunctionDecl *FnDecl,
CanQualType ExpectedResultType,
CanQualType ExpectedFirstParamType,
unsigned DependentParamTypeDiag,
unsigned InvalidParamTypeDiag) {
QualType ResultType =
FnDecl->getType()->getAs<FunctionType>()->getResultType();
// Check that the result type is not dependent.
if (ResultType->isDependentType())
return SemaRef.Diag(FnDecl->getLocation(),
diag::err_operator_new_delete_dependent_result_type)
<< FnDecl->getDeclName() << ExpectedResultType;
// Check that the result type is what we expect.
if (SemaRef.Context.getCanonicalType(ResultType) != ExpectedResultType)
return SemaRef.Diag(FnDecl->getLocation(),
diag::err_operator_new_delete_invalid_result_type)
<< FnDecl->getDeclName() << ExpectedResultType;
// A function template must have at least 2 parameters.
if (FnDecl->getDescribedFunctionTemplate() && FnDecl->getNumParams() < 2)
return SemaRef.Diag(FnDecl->getLocation(),
diag::err_operator_new_delete_template_too_few_parameters)
<< FnDecl->getDeclName();
// The function decl must have at least 1 parameter.
if (FnDecl->getNumParams() == 0)
return SemaRef.Diag(FnDecl->getLocation(),
diag::err_operator_new_delete_too_few_parameters)
<< FnDecl->getDeclName();
// Check the the first parameter type is not dependent.
QualType FirstParamType = FnDecl->getParamDecl(0)->getType();
if (FirstParamType->isDependentType())
return SemaRef.Diag(FnDecl->getLocation(), DependentParamTypeDiag)
<< FnDecl->getDeclName() << ExpectedFirstParamType;
// Check that the first parameter type is what we expect.
if (SemaRef.Context.getCanonicalType(FirstParamType).getUnqualifiedType() !=
ExpectedFirstParamType)
return SemaRef.Diag(FnDecl->getLocation(), InvalidParamTypeDiag)
<< FnDecl->getDeclName() << ExpectedFirstParamType;
return false;
}
static bool
CheckOperatorNewDeclaration(Sema &SemaRef, const FunctionDecl *FnDecl) {
// C++ [basic.stc.dynamic.allocation]p1:
// A program is ill-formed if an allocation function is declared in a
// namespace scope other than global scope or declared static in global
// scope.
if (CheckOperatorNewDeleteDeclarationScope(SemaRef, FnDecl))
return true;
CanQualType SizeTy =
SemaRef.Context.getCanonicalType(SemaRef.Context.getSizeType());
// C++ [basic.stc.dynamic.allocation]p1:
// The return type shall be void*. The first parameter shall have type
// std::size_t.
if (CheckOperatorNewDeleteTypes(SemaRef, FnDecl, SemaRef.Context.VoidPtrTy,
SizeTy,
diag::err_operator_new_dependent_param_type,
diag::err_operator_new_param_type))
return true;
// C++ [basic.stc.dynamic.allocation]p1:
// The first parameter shall not have an associated default argument.
if (FnDecl->getParamDecl(0)->hasDefaultArg())
return SemaRef.Diag(FnDecl->getLocation(),
diag::err_operator_new_default_arg)
<< FnDecl->getDeclName() << FnDecl->getParamDecl(0)->getDefaultArgRange();
return false;
}
static bool
CheckOperatorDeleteDeclaration(Sema &SemaRef, const FunctionDecl *FnDecl) {
// C++ [basic.stc.dynamic.deallocation]p1:
// A program is ill-formed if deallocation functions are declared in a
// namespace scope other than global scope or declared static in global
// scope.
if (CheckOperatorNewDeleteDeclarationScope(SemaRef, FnDecl))
return true;
// C++ [basic.stc.dynamic.deallocation]p2:
// Each deallocation function shall return void and its first parameter
// shall be void*.
if (CheckOperatorNewDeleteTypes(SemaRef, FnDecl, SemaRef.Context.VoidTy,
SemaRef.Context.VoidPtrTy,
diag::err_operator_delete_dependent_param_type,
diag::err_operator_delete_param_type))
return true;
return false;
}
/// CheckOverloadedOperatorDeclaration - Check whether the declaration
/// of this overloaded operator is well-formed. If so, returns false;
/// otherwise, emits appropriate diagnostics and returns true.
bool Sema::CheckOverloadedOperatorDeclaration(FunctionDecl *FnDecl) {
assert(FnDecl && FnDecl->isOverloadedOperator() &&
"Expected an overloaded operator declaration");
OverloadedOperatorKind Op = FnDecl->getOverloadedOperator();
// C++ [over.oper]p5:
// The allocation and deallocation functions, operator new,
// operator new[], operator delete and operator delete[], are
// described completely in 3.7.3. The attributes and restrictions
// found in the rest of this subclause do not apply to them unless
// explicitly stated in 3.7.3.
2009-12-12 07:31:21 +08:00
if (Op == OO_Delete || Op == OO_Array_Delete)
return CheckOperatorDeleteDeclaration(*this, FnDecl);
if (Op == OO_New || Op == OO_Array_New)
return CheckOperatorNewDeclaration(*this, FnDecl);
// C++ [over.oper]p6:
// An operator function shall either be a non-static member
// function or be a non-member function and have at least one
// parameter whose type is a class, a reference to a class, an
// enumeration, or a reference to an enumeration.
if (CXXMethodDecl *MethodDecl = dyn_cast<CXXMethodDecl>(FnDecl)) {
if (MethodDecl->isStatic())
return Diag(FnDecl->getLocation(),
diag::err_operator_overload_static) << FnDecl->getDeclName();
} else {
bool ClassOrEnumParam = false;
for (FunctionDecl::param_iterator Param = FnDecl->param_begin(),
ParamEnd = FnDecl->param_end();
Param != ParamEnd; ++Param) {
QualType ParamType = (*Param)->getType().getNonReferenceType();
if (ParamType->isDependentType() || ParamType->isRecordType() ||
ParamType->isEnumeralType()) {
ClassOrEnumParam = true;
break;
}
}
if (!ClassOrEnumParam)
return Diag(FnDecl->getLocation(),
diag::err_operator_overload_needs_class_or_enum)
<< FnDecl->getDeclName();
}
// C++ [over.oper]p8:
// An operator function cannot have default arguments (8.3.6),
// except where explicitly stated below.
//
// Only the function-call operator allows default arguments
// (C++ [over.call]p1).
if (Op != OO_Call) {
for (FunctionDecl::param_iterator Param = FnDecl->param_begin();
Param != FnDecl->param_end(); ++Param) {
if ((*Param)->hasDefaultArg())
return Diag((*Param)->getLocation(),
diag::err_operator_overload_default_arg)
<< FnDecl->getDeclName() << (*Param)->getDefaultArgRange();
}
}
static const bool OperatorUses[NUM_OVERLOADED_OPERATORS][3] = {
{ false, false, false }
#define OVERLOADED_OPERATOR(Name,Spelling,Token,Unary,Binary,MemberOnly) \
, { Unary, Binary, MemberOnly }
#include "clang/Basic/OperatorKinds.def"
};
bool CanBeUnaryOperator = OperatorUses[Op][0];
bool CanBeBinaryOperator = OperatorUses[Op][1];
bool MustBeMemberOperator = OperatorUses[Op][2];
// C++ [over.oper]p8:
// [...] Operator functions cannot have more or fewer parameters
// than the number required for the corresponding operator, as
// described in the rest of this subclause.
unsigned NumParams = FnDecl->getNumParams()
+ (isa<CXXMethodDecl>(FnDecl)? 1 : 0);
if (Op != OO_Call &&
((NumParams == 1 && !CanBeUnaryOperator) ||
(NumParams == 2 && !CanBeBinaryOperator) ||
(NumParams < 1) || (NumParams > 2))) {
// We have the wrong number of parameters.
unsigned ErrorKind;
if (CanBeUnaryOperator && CanBeBinaryOperator) {
ErrorKind = 2; // 2 -> unary or binary.
} else if (CanBeUnaryOperator) {
ErrorKind = 0; // 0 -> unary
} else {
assert(CanBeBinaryOperator &&
"All non-call overloaded operators are unary or binary!");
ErrorKind = 1; // 1 -> binary
}
return Diag(FnDecl->getLocation(), diag::err_operator_overload_must_be)
<< FnDecl->getDeclName() << NumParams << ErrorKind;
}
// Overloaded operators other than operator() cannot be variadic.
if (Op != OO_Call &&
FnDecl->getType()->getAs<FunctionProtoType>()->isVariadic()) {
return Diag(FnDecl->getLocation(), diag::err_operator_overload_variadic)
<< FnDecl->getDeclName();
}
// Some operators must be non-static member functions.
if (MustBeMemberOperator && !isa<CXXMethodDecl>(FnDecl)) {
return Diag(FnDecl->getLocation(),
diag::err_operator_overload_must_be_member)
<< FnDecl->getDeclName();
}
// C++ [over.inc]p1:
// The user-defined function called operator++ implements the
// prefix and postfix ++ operator. If this function is a member
// function with no parameters, or a non-member function with one
// parameter of class or enumeration type, it defines the prefix
// increment operator ++ for objects of that type. If the function
// is a member function with one parameter (which shall be of type
// int) or a non-member function with two parameters (the second
// of which shall be of type int), it defines the postfix
// increment operator ++ for objects of that type.
if ((Op == OO_PlusPlus || Op == OO_MinusMinus) && NumParams == 2) {
ParmVarDecl *LastParam = FnDecl->getParamDecl(FnDecl->getNumParams() - 1);
bool ParamIsInt = false;
if (const BuiltinType *BT = LastParam->getType()->getAs<BuiltinType>())
ParamIsInt = BT->getKind() == BuiltinType::Int;
if (!ParamIsInt)
return Diag(LastParam->getLocation(),
diag::err_operator_overload_post_incdec_must_be_int)
<< LastParam->getType() << (Op == OO_MinusMinus);
}
return false;
}
/// CheckLiteralOperatorDeclaration - Check whether the declaration
/// of this literal operator function is well-formed. If so, returns
/// false; otherwise, emits appropriate diagnostics and returns true.
bool Sema::CheckLiteralOperatorDeclaration(FunctionDecl *FnDecl) {
DeclContext *DC = FnDecl->getDeclContext();
Decl::Kind Kind = DC->getDeclKind();
if (Kind != Decl::TranslationUnit && Kind != Decl::Namespace &&
Kind != Decl::LinkageSpec) {
Diag(FnDecl->getLocation(), diag::err_literal_operator_outside_namespace)
<< FnDecl->getDeclName();
return true;
}
bool Valid = false;
// template <char...> type operator "" name() is the only valid template
// signature, and the only valid signature with no parameters.
if (FnDecl->param_size() == 0) {
if (FunctionTemplateDecl *TpDecl = FnDecl->getDescribedFunctionTemplate()) {
// Must have only one template parameter
TemplateParameterList *Params = TpDecl->getTemplateParameters();
if (Params->size() == 1) {
NonTypeTemplateParmDecl *PmDecl =
cast<NonTypeTemplateParmDecl>(Params->getParam(0));
// The template parameter must be a char parameter pack.
if (PmDecl && PmDecl->isTemplateParameterPack() &&
Context.hasSameType(PmDecl->getType(), Context.CharTy))
Valid = true;
}
}
} else {
// Check the first parameter
FunctionDecl::param_iterator Param = FnDecl->param_begin();
QualType T = (*Param)->getType();
// unsigned long long int, long double, and any character type are allowed
// as the only parameters.
if (Context.hasSameType(T, Context.UnsignedLongLongTy) ||
Context.hasSameType(T, Context.LongDoubleTy) ||
Context.hasSameType(T, Context.CharTy) ||
Context.hasSameType(T, Context.WCharTy) ||
Context.hasSameType(T, Context.Char16Ty) ||
Context.hasSameType(T, Context.Char32Ty)) {
if (++Param == FnDecl->param_end())
Valid = true;
goto FinishedParams;
}
// Otherwise it must be a pointer to const; let's strip those qualifiers.
const PointerType *PT = T->getAs<PointerType>();
if (!PT)
goto FinishedParams;
T = PT->getPointeeType();
if (!T.isConstQualified())
goto FinishedParams;
T = T.getUnqualifiedType();
// Move on to the second parameter;
++Param;
// If there is no second parameter, the first must be a const char *
if (Param == FnDecl->param_end()) {
if (Context.hasSameType(T, Context.CharTy))
Valid = true;
goto FinishedParams;
}
// const char *, const wchar_t*, const char16_t*, and const char32_t*
// are allowed as the first parameter to a two-parameter function
if (!(Context.hasSameType(T, Context.CharTy) ||
Context.hasSameType(T, Context.WCharTy) ||
Context.hasSameType(T, Context.Char16Ty) ||
Context.hasSameType(T, Context.Char32Ty)))
goto FinishedParams;
// The second and final parameter must be an std::size_t
T = (*Param)->getType().getUnqualifiedType();
if (Context.hasSameType(T, Context.getSizeType()) &&
++Param == FnDecl->param_end())
Valid = true;
}
// FIXME: This diagnostic is absolutely terrible.
FinishedParams:
if (!Valid) {
Diag(FnDecl->getLocation(), diag::err_literal_operator_params)
<< FnDecl->getDeclName();
return true;
}
return false;
}
/// ActOnStartLinkageSpecification - Parsed the beginning of a C++
/// linkage specification, including the language and (if present)
/// the '{'. ExternLoc is the location of the 'extern', LangLoc is
/// the location of the language string literal, which is provided
/// by Lang/StrSize. LBraceLoc, if valid, provides the location of
/// the '{' brace. Otherwise, this linkage specification does not
/// have any braces.
2010-11-10 04:15:55 +08:00
Decl *Sema::ActOnStartLinkageSpecification(Scope *S, SourceLocation ExternLoc,
SourceLocation LangLoc,
llvm::StringRef Lang,
SourceLocation LBraceLoc) {
LinkageSpecDecl::LanguageIDs Language;
if (Lang == "\"C\"")
Language = LinkageSpecDecl::lang_c;
else if (Lang == "\"C++\"")
Language = LinkageSpecDecl::lang_cxx;
else {
Diag(LangLoc, diag::err_bad_language);
2010-08-21 17:40:31 +08:00
return 0;
}
// FIXME: Add all the various semantics of linkage specifications
LinkageSpecDecl *D = LinkageSpecDecl::Create(Context, CurContext,
LangLoc, Language,
LBraceLoc.isValid());
CurContext->addDecl(D);
PushDeclContext(S, D);
2010-08-21 17:40:31 +08:00
return D;
}
/// ActOnFinishLinkageSpecification - Complete the definition of
/// the C++ linkage specification LinkageSpec. If RBraceLoc is
/// valid, it's the position of the closing '}' brace in a linkage
/// specification that uses braces.
2010-08-21 17:40:31 +08:00
Decl *Sema::ActOnFinishLinkageSpecification(Scope *S,
Decl *LinkageSpec,
SourceLocation RBraceLoc) {
if (LinkageSpec)
PopDeclContext();
return LinkageSpec;
}
/// \brief Perform semantic analysis for the variable declaration that
/// occurs within a C++ catch clause, returning the newly-created
/// variable.
VarDecl *Sema::BuildExceptionDeclaration(Scope *S,
TypeSourceInfo *TInfo,
IdentifierInfo *Name,
SourceLocation Loc) {
bool Invalid = false;
QualType ExDeclType = TInfo->getType();
// Arrays and functions decay.
if (ExDeclType->isArrayType())
ExDeclType = Context.getArrayDecayedType(ExDeclType);
else if (ExDeclType->isFunctionType())
ExDeclType = Context.getPointerType(ExDeclType);
// C++ 15.3p1: The exception-declaration shall not denote an incomplete type.
// The exception-declaration shall not denote a pointer or reference to an
// incomplete type, other than [cv] void*.
// N2844 forbids rvalue references.
if (!ExDeclType->isDependentType() && ExDeclType->isRValueReferenceType()) {
Diag(Loc, diag::err_catch_rvalue_ref);
Invalid = true;
}
// GCC allows catching pointers and references to incomplete types
// as an extension; so do we, but we warn by default.
QualType BaseType = ExDeclType;
int Mode = 0; // 0 for direct type, 1 for pointer, 2 for reference
unsigned DK = diag::err_catch_incomplete;
bool IncompleteCatchIsInvalid = true;
if (const PointerType *Ptr = BaseType->getAs<PointerType>()) {
BaseType = Ptr->getPointeeType();
Mode = 1;
DK = diag::ext_catch_incomplete_ptr;
IncompleteCatchIsInvalid = false;
} else if (const ReferenceType *Ref = BaseType->getAs<ReferenceType>()) {
// For the purpose of error recovery, we treat rvalue refs like lvalue refs.
BaseType = Ref->getPointeeType();
Mode = 2;
DK = diag::ext_catch_incomplete_ref;
IncompleteCatchIsInvalid = false;
}
if (!Invalid && (Mode == 0 || !BaseType->isVoidType()) &&
!BaseType->isDependentType() && RequireCompleteType(Loc, BaseType, DK) &&
IncompleteCatchIsInvalid)
Invalid = true;
if (!Invalid && !ExDeclType->isDependentType() &&
RequireNonAbstractType(Loc, ExDeclType,
diag::err_abstract_type_in_decl,
AbstractVariableType))
Invalid = true;
// Only the non-fragile NeXT runtime currently supports C++ catches
// of ObjC types, and no runtime supports catching ObjC types by value.
if (!Invalid && getLangOptions().ObjC1) {
QualType T = ExDeclType;
if (const ReferenceType *RT = T->getAs<ReferenceType>())
T = RT->getPointeeType();
if (T->isObjCObjectType()) {
Diag(Loc, diag::err_objc_object_catch);
Invalid = true;
} else if (T->isObjCObjectPointerType()) {
if (!getLangOptions().NeXTRuntime) {
Diag(Loc, diag::err_objc_pointer_cxx_catch_gnu);
Invalid = true;
} else if (!getLangOptions().ObjCNonFragileABI) {
Diag(Loc, diag::err_objc_pointer_cxx_catch_fragile);
Invalid = true;
}
}
}
VarDecl *ExDecl = VarDecl::Create(Context, CurContext, Loc,
Name, ExDeclType, TInfo, SC_None,
SC_None);
ExDecl->setExceptionVariable(true);
if (!Invalid) {
if (const RecordType *recordType = ExDeclType->getAs<RecordType>()) {
// C++ [except.handle]p16:
// The object declared in an exception-declaration or, if the
// exception-declaration does not specify a name, a temporary (12.2) is
// copy-initialized (8.5) from the exception object. [...]
// The object is destroyed when the handler exits, after the destruction
// of any automatic objects initialized within the handler.
//
// We just pretend to initialize the object with itself, then make sure
// it can be destroyed later.
QualType initType = ExDeclType;
InitializedEntity entity =
InitializedEntity::InitializeVariable(ExDecl);
InitializationKind initKind =
InitializationKind::CreateCopy(Loc, SourceLocation());
Expr *opaqueValue =
new (Context) OpaqueValueExpr(Loc, initType, VK_LValue, OK_Ordinary);
InitializationSequence sequence(*this, entity, initKind, &opaqueValue, 1);
ExprResult result = sequence.Perform(*this, entity, initKind,
MultiExprArg(&opaqueValue, 1));
if (result.isInvalid())
Invalid = true;
else {
// If the constructor used was non-trivial, set this as the
// "initializer".
CXXConstructExpr *construct = cast<CXXConstructExpr>(result.take());
if (!construct->getConstructor()->isTrivial()) {
Expr *init = MaybeCreateExprWithCleanups(construct);
ExDecl->setInit(init);
}
// And make sure it's destructable.
FinalizeVarWithDestructor(ExDecl, recordType);
}
}
}
if (Invalid)
ExDecl->setInvalidDecl();
return ExDecl;
}
/// ActOnExceptionDeclarator - Parsed the exception-declarator in a C++ catch
/// handler.
2010-08-21 17:40:31 +08:00
Decl *Sema::ActOnExceptionDeclarator(Scope *S, Declarator &D) {
TypeSourceInfo *TInfo = GetTypeForDeclarator(D, S);
bool Invalid = D.isInvalidType();
// Check for unexpanded parameter packs.
if (TInfo && DiagnoseUnexpandedParameterPack(D.getIdentifierLoc(), TInfo,
UPPC_ExceptionType)) {
TInfo = Context.getTrivialTypeSourceInfo(Context.IntTy,
D.getIdentifierLoc());
Invalid = true;
}
IdentifierInfo *II = D.getIdentifier();
if (NamedDecl *PrevDecl = LookupSingleName(S, II, D.getIdentifierLoc(),
LookupOrdinaryName,
ForRedeclaration)) {
// The scope should be freshly made just for us. There is just no way
// it contains any previous declaration.
2010-08-21 17:40:31 +08:00
assert(!S->isDeclScope(PrevDecl));
if (PrevDecl->isTemplateParameter()) {
// Maybe we will complain about the shadowed template parameter.
DiagnoseTemplateParameterShadow(D.getIdentifierLoc(), PrevDecl);
}
}
if (D.getCXXScopeSpec().isSet() && !Invalid) {
Diag(D.getIdentifierLoc(), diag::err_qualified_catch_declarator)
<< D.getCXXScopeSpec().getRange();
Invalid = true;
}
VarDecl *ExDecl = BuildExceptionDeclaration(S, TInfo,
D.getIdentifier(),
D.getIdentifierLoc());
if (Invalid)
ExDecl->setInvalidDecl();
// Add the exception declaration into this scope.
if (II)
PushOnScopeChains(ExDecl, S);
else
CurContext->addDecl(ExDecl);
ProcessDeclAttributes(S, ExDecl, D);
2010-08-21 17:40:31 +08:00
return ExDecl;
}
2010-08-21 17:40:31 +08:00
Decl *Sema::ActOnStaticAssertDeclaration(SourceLocation AssertLoc,
Expr *AssertExpr,
Expr *AssertMessageExpr_) {
StringLiteral *AssertMessage = cast<StringLiteral>(AssertMessageExpr_);
if (!AssertExpr->isTypeDependent() && !AssertExpr->isValueDependent()) {
llvm::APSInt Value(32);
if (!AssertExpr->isIntegerConstantExpr(Value, Context)) {
Diag(AssertLoc, diag::err_static_assert_expression_is_not_constant) <<
AssertExpr->getSourceRange();
2010-08-21 17:40:31 +08:00
return 0;
}
if (Value == 0) {
Diag(AssertLoc, diag::err_static_assert_failed)
<< AssertMessage->getString() << AssertExpr->getSourceRange();
}
}
if (DiagnoseUnexpandedParameterPack(AssertExpr, UPPC_StaticAssertExpression))
return 0;
Decl *Decl = StaticAssertDecl::Create(Context, CurContext, AssertLoc,
AssertExpr, AssertMessage);
CurContext->addDecl(Decl);
2010-08-21 17:40:31 +08:00
return Decl;
}
/// \brief Perform semantic analysis of the given friend type declaration.
///
/// \returns A friend declaration that.
FriendDecl *Sema::CheckFriendTypeDecl(SourceLocation FriendLoc,
TypeSourceInfo *TSInfo) {
assert(TSInfo && "NULL TypeSourceInfo for friend type declaration");
QualType T = TSInfo->getType();
SourceRange TypeRange = TSInfo->getTypeLoc().getLocalSourceRange();
if (!getLangOptions().CPlusPlus0x) {
// C++03 [class.friend]p2:
// An elaborated-type-specifier shall be used in a friend declaration
// for a class.*
//
// * The class-key of the elaborated-type-specifier is required.
if (!ActiveTemplateInstantiations.empty()) {
// Do not complain about the form of friend template types during
// template instantiation; we will already have complained when the
// template was declared.
} else if (!T->isElaboratedTypeSpecifier()) {
// If we evaluated the type to a record type, suggest putting
// a tag in front.
if (const RecordType *RT = T->getAs<RecordType>()) {
RecordDecl *RD = RT->getDecl();
std::string InsertionText = std::string(" ") + RD->getKindName();
Diag(TypeRange.getBegin(), diag::ext_unelaborated_friend_type)
<< (unsigned) RD->getTagKind()
<< T
<< FixItHint::CreateInsertion(PP.getLocForEndOfToken(FriendLoc),
InsertionText);
} else {
Diag(FriendLoc, diag::ext_nonclass_type_friend)
<< T
<< SourceRange(FriendLoc, TypeRange.getEnd());
}
} else if (T->getAs<EnumType>()) {
Diag(FriendLoc, diag::ext_enum_friend)
<< T
<< SourceRange(FriendLoc, TypeRange.getEnd());
}
}
// C++0x [class.friend]p3:
// If the type specifier in a friend declaration designates a (possibly
// cv-qualified) class type, that class is declared as a friend; otherwise,
// the friend declaration is ignored.
// FIXME: C++0x has some syntactic restrictions on friend type declarations
// in [class.friend]p3 that we do not implement.
return FriendDecl::Create(Context, CurContext, FriendLoc, TSInfo, FriendLoc);
}
/// Handle a friend tag declaration where the scope specifier was
/// templated.
Decl *Sema::ActOnTemplatedFriendTag(Scope *S, SourceLocation FriendLoc,
unsigned TagSpec, SourceLocation TagLoc,
CXXScopeSpec &SS,
IdentifierInfo *Name, SourceLocation NameLoc,
AttributeList *Attr,
MultiTemplateParamsArg TempParamLists) {
TagTypeKind Kind = TypeWithKeyword::getTagTypeKindForTypeSpec(TagSpec);
bool isExplicitSpecialization = false;
unsigned NumMatchedTemplateParamLists = TempParamLists.size();
bool Invalid = false;
if (TemplateParameterList *TemplateParams
= MatchTemplateParametersToScopeSpecifier(TagLoc, SS,
TempParamLists.get(),
TempParamLists.size(),
/*friend*/ true,
isExplicitSpecialization,
Invalid)) {
--NumMatchedTemplateParamLists;
if (TemplateParams->size() > 0) {
// This is a declaration of a class template.
if (Invalid)
return 0;
return CheckClassTemplate(S, TagSpec, TUK_Friend, TagLoc,
SS, Name, NameLoc, Attr,
TemplateParams, AS_public).take();
} else {
// The "template<>" header is extraneous.
Diag(TemplateParams->getTemplateLoc(), diag::err_template_tag_noparams)
<< TypeWithKeyword::getTagTypeKindName(Kind) << Name;
isExplicitSpecialization = true;
}
}
if (Invalid) return 0;
assert(SS.isNotEmpty() && "valid templated tag with no SS and no direct?");
bool isAllExplicitSpecializations = true;
for (unsigned I = 0; I != NumMatchedTemplateParamLists; ++I) {
if (TempParamLists.get()[I]->size()) {
isAllExplicitSpecializations = false;
break;
}
}
// FIXME: don't ignore attributes.
// If it's explicit specializations all the way down, just forget
// about the template header and build an appropriate non-templated
// friend. TODO: for source fidelity, remember the headers.
if (isAllExplicitSpecializations) {
ElaboratedTypeKeyword Keyword
= TypeWithKeyword::getKeywordForTagTypeKind(Kind);
QualType T = CheckTypenameType(Keyword, SS.getScopeRep(), *Name,
TagLoc, SS.getRange(), NameLoc);
if (T.isNull())
return 0;
TypeSourceInfo *TSI = Context.CreateTypeSourceInfo(T);
if (isa<DependentNameType>(T)) {
DependentNameTypeLoc TL = cast<DependentNameTypeLoc>(TSI->getTypeLoc());
TL.setKeywordLoc(TagLoc);
TL.setQualifierRange(SS.getRange());
TL.setNameLoc(NameLoc);
} else {
ElaboratedTypeLoc TL = cast<ElaboratedTypeLoc>(TSI->getTypeLoc());
TL.setKeywordLoc(TagLoc);
TL.setQualifierRange(SS.getRange());
cast<TypeSpecTypeLoc>(TL.getNamedTypeLoc()).setNameLoc(NameLoc);
}
FriendDecl *Friend = FriendDecl::Create(Context, CurContext, NameLoc,
TSI, FriendLoc);
Friend->setAccess(AS_public);
CurContext->addDecl(Friend);
return Friend;
}
// Handle the case of a templated-scope friend class. e.g.
// template <class T> class A<T>::B;
// FIXME: we don't support these right now.
ElaboratedTypeKeyword ETK = TypeWithKeyword::getKeywordForTagTypeKind(Kind);
QualType T = Context.getDependentNameType(ETK, SS.getScopeRep(), Name);
TypeSourceInfo *TSI = Context.CreateTypeSourceInfo(T);
DependentNameTypeLoc TL = cast<DependentNameTypeLoc>(TSI->getTypeLoc());
TL.setKeywordLoc(TagLoc);
TL.setQualifierRange(SS.getRange());
TL.setNameLoc(NameLoc);
FriendDecl *Friend = FriendDecl::Create(Context, CurContext, NameLoc,
TSI, FriendLoc);
Friend->setAccess(AS_public);
Friend->setUnsupportedFriend(true);
CurContext->addDecl(Friend);
return Friend;
}
/// Handle a friend type declaration. This works in tandem with
/// ActOnTag.
///
/// Notes on friend class templates:
///
/// We generally treat friend class declarations as if they were
/// declaring a class. So, for example, the elaborated type specifier
/// in a friend declaration is required to obey the restrictions of a
/// class-head (i.e. no typedefs in the scope chain), template
/// parameters are required to match up with simple template-ids, &c.
/// However, unlike when declaring a template specialization, it's
/// okay to refer to a template specialization without an empty
/// template parameter declaration, e.g.
/// friend class A<T>::B<unsigned>;
/// We permit this as a special case; if there are any template
/// parameters present at all, require proper matching, i.e.
/// template <> template <class T> friend class A<int>::B;
2010-08-21 17:40:31 +08:00
Decl *Sema::ActOnFriendTypeDecl(Scope *S, const DeclSpec &DS,
2010-10-16 15:23:36 +08:00
MultiTemplateParamsArg TempParams) {
SourceLocation Loc = DS.getSourceRange().getBegin();
assert(DS.isFriendSpecified());
assert(DS.getStorageClassSpec() == DeclSpec::SCS_unspecified);
// Try to convert the decl specifier to a type. This works for
// friend templates because ActOnTag never produces a ClassTemplateDecl
// for a TUK_Friend.
Declarator TheDeclarator(DS, Declarator::MemberContext);
TypeSourceInfo *TSI = GetTypeForDeclarator(TheDeclarator, S);
QualType T = TSI->getType();
if (TheDeclarator.isInvalidType())
2010-08-21 17:40:31 +08:00
return 0;
if (DiagnoseUnexpandedParameterPack(Loc, TSI, UPPC_FriendDeclaration))
return 0;
// This is definitely an error in C++98. It's probably meant to
// be forbidden in C++0x, too, but the specification is just
// poorly written.
//
// The problem is with declarations like the following:
// template <T> friend A<T>::foo;
// where deciding whether a class C is a friend or not now hinges
// on whether there exists an instantiation of A that causes
// 'foo' to equal C. There are restrictions on class-heads
// (which we declare (by fiat) elaborated friend declarations to
// be) that makes this tractable.
//
// FIXME: handle "template <> friend class A<T>;", which
// is possibly well-formed? Who even knows?
if (TempParams.size() && !T->isElaboratedTypeSpecifier()) {
Diag(Loc, diag::err_tagless_friend_type_template)
<< DS.getSourceRange();
2010-08-21 17:40:31 +08:00
return 0;
}
// C++98 [class.friend]p1: A friend of a class is a function
// or class that is not a member of the class . . .
// This is fixed in DR77, which just barely didn't make the C++03
// deadline. It's also a very silly restriction that seriously
// affects inner classes and which nobody else seems to implement;
// thus we never diagnose it, not even in -pedantic.
//
// But note that we could warn about it: it's always useless to
// friend one of your own members (it's not, however, worthless to
// friend a member of an arbitrary specialization of your template).
Decl *D;
if (unsigned NumTempParamLists = TempParams.size())
D = FriendTemplateDecl::Create(Context, CurContext, Loc,
NumTempParamLists,
2010-10-16 15:23:36 +08:00
TempParams.release(),
TSI,
DS.getFriendSpecLoc());
else
D = CheckFriendTypeDecl(DS.getFriendSpecLoc(), TSI);
if (!D)
2010-08-21 17:40:31 +08:00
return 0;
D->setAccess(AS_public);
CurContext->addDecl(D);
2010-08-21 17:40:31 +08:00
return D;
}
Decl *Sema::ActOnFriendFunctionDecl(Scope *S, Declarator &D, bool IsDefinition,
MultiTemplateParamsArg TemplateParams) {
const DeclSpec &DS = D.getDeclSpec();
assert(DS.isFriendSpecified());
assert(DS.getStorageClassSpec() == DeclSpec::SCS_unspecified);
SourceLocation Loc = D.getIdentifierLoc();
TypeSourceInfo *TInfo = GetTypeForDeclarator(D, S);
QualType T = TInfo->getType();
// C++ [class.friend]p1
// A friend of a class is a function or class....
// Note that this sees through typedefs, which is intended.
// It *doesn't* see through dependent types, which is correct
// according to [temp.arg.type]p3:
// If a declaration acquires a function type through a
// type dependent on a template-parameter and this causes
// a declaration that does not use the syntactic form of a
// function declarator to have a function type, the program
// is ill-formed.
if (!T->isFunctionType()) {
Diag(Loc, diag::err_unexpected_friend);
// It might be worthwhile to try to recover by creating an
// appropriate declaration.
2010-08-21 17:40:31 +08:00
return 0;
}
// C++ [namespace.memdef]p3
// - If a friend declaration in a non-local class first declares a
// class or function, the friend class or function is a member
// of the innermost enclosing namespace.
// - The name of the friend is not found by simple name lookup
// until a matching declaration is provided in that namespace
// scope (either before or after the class declaration granting
// friendship).
// - If a friend function is called, its name may be found by the
// name lookup that considers functions from namespaces and
// classes associated with the types of the function arguments.
// - When looking for a prior declaration of a class or a function
// declared as a friend, scopes outside the innermost enclosing
// namespace scope are not considered.
CXXScopeSpec &SS = D.getCXXScopeSpec();
DeclarationNameInfo NameInfo = GetNameForDeclarator(D);
DeclarationName Name = NameInfo.getName();
assert(Name);
// Check for unexpanded parameter packs.
if (DiagnoseUnexpandedParameterPack(Loc, TInfo, UPPC_FriendDeclaration) ||
DiagnoseUnexpandedParameterPack(NameInfo, UPPC_FriendDeclaration) ||
DiagnoseUnexpandedParameterPack(SS, UPPC_FriendDeclaration))
return 0;
// The context we found the declaration in, or in which we should
// create the declaration.
DeclContext *DC;
Scope *DCScope = S;
LookupResult Previous(*this, NameInfo, LookupOrdinaryName,
ForRedeclaration);
// FIXME: there are different rules in local classes
// There are four cases here.
// - There's no scope specifier, in which case we just go to the
// appropriate scope and look for a function or function template
// there as appropriate.
// Recover from invalid scope qualifiers as if they just weren't there.
if (SS.isInvalid() || !SS.isSet()) {
// C++0x [namespace.memdef]p3:
// If the name in a friend declaration is neither qualified nor
// a template-id and the declaration is a function or an
// elaborated-type-specifier, the lookup to determine whether
// the entity has been previously declared shall not consider
// any scopes outside the innermost enclosing namespace.
// C++0x [class.friend]p11:
// If a friend declaration appears in a local class and the name
// specified is an unqualified name, a prior declaration is
// looked up without considering scopes that are outside the
// innermost enclosing non-class scope. For a friend function
// declaration, if there is no prior declaration, the program is
// ill-formed.
bool isLocal = cast<CXXRecordDecl>(CurContext)->isLocalClass();
bool isTemplateId = D.getName().getKind() == UnqualifiedId::IK_TemplateId;
// Find the appropriate context according to the above.
DC = CurContext;
while (true) {
// Skip class contexts. If someone can cite chapter and verse
// for this behavior, that would be nice --- it's what GCC and
// EDG do, and it seems like a reasonable intent, but the spec
// really only says that checks for unqualified existing
// declarations should stop at the nearest enclosing namespace,
// not that they should only consider the nearest enclosing
// namespace.
while (DC->isRecord())
DC = DC->getParent();
LookupQualifiedName(Previous, DC);
// TODO: decide what we think about using declarations.
if (isLocal || !Previous.empty())
break;
if (isTemplateId) {
if (isa<TranslationUnitDecl>(DC)) break;
} else {
if (DC->isFileContext()) break;
}
DC = DC->getParent();
}
// C++ [class.friend]p1: A friend of a class is a function or
// class that is not a member of the class . . .
// C++0x changes this for both friend types and functions.
// Most C++ 98 compilers do seem to give an error here, so
// we do, too.
if (!Previous.empty() && DC->Equals(CurContext)
&& !getLangOptions().CPlusPlus0x)
Diag(DS.getFriendSpecLoc(), diag::err_friend_is_member);
DCScope = getScopeForDeclContext(S, DC);
// - There's a non-dependent scope specifier, in which case we
// compute it and do a previous lookup there for a function
// or function template.
} else if (!SS.getScopeRep()->isDependent()) {
DC = computeDeclContext(SS);
if (!DC) return 0;
if (RequireCompleteDeclContext(SS, DC)) return 0;
LookupQualifiedName(Previous, DC);
// Ignore things found implicitly in the wrong scope.
// TODO: better diagnostics for this case. Suggesting the right
// qualified scope would be nice...
LookupResult::Filter F = Previous.makeFilter();
while (F.hasNext()) {
NamedDecl *D = F.next();
if (!DC->InEnclosingNamespaceSetOf(
D->getDeclContext()->getRedeclContext()))
F.erase();
}
F.done();
if (Previous.empty()) {
D.setInvalidType();
Diag(Loc, diag::err_qualified_friend_not_found) << Name << T;
return 0;
}
// C++ [class.friend]p1: A friend of a class is a function or
// class that is not a member of the class . . .
if (DC->Equals(CurContext))
Diag(DS.getFriendSpecLoc(), diag::err_friend_is_member);
// - There's a scope specifier that does not match any template
// parameter lists, in which case we use some arbitrary context,
// create a method or method template, and wait for instantiation.
// - There's a scope specifier that does match some template
// parameter lists, which we don't handle right now.
} else {
DC = CurContext;
assert(isa<CXXRecordDecl>(DC) && "friend declaration not in class?");
}
if (!DC->isRecord()) {
// This implies that it has to be an operator or function.
if (D.getName().getKind() == UnqualifiedId::IK_ConstructorName ||
D.getName().getKind() == UnqualifiedId::IK_DestructorName ||
D.getName().getKind() == UnqualifiedId::IK_ConversionFunctionId) {
Diag(Loc, diag::err_introducing_special_friend) <<
(D.getName().getKind() == UnqualifiedId::IK_ConstructorName ? 0 :
D.getName().getKind() == UnqualifiedId::IK_DestructorName ? 1 : 2);
2010-08-21 17:40:31 +08:00
return 0;
}
}
bool Redeclaration = false;
NamedDecl *ND = ActOnFunctionDeclarator(DCScope, D, DC, T, TInfo, Previous,
move(TemplateParams),
IsDefinition,
Redeclaration);
2010-08-21 17:40:31 +08:00
if (!ND) return 0;
assert(ND->getDeclContext() == DC);
assert(ND->getLexicalDeclContext() == CurContext);
// Add the function declaration to the appropriate lookup tables,
// adjusting the redeclarations list as necessary. We don't
// want to do this yet if the friending class is dependent.
//
// Also update the scope-based lookup if the target context's
// lookup context is in lexical scope.
if (!CurContext->isDependentContext()) {
DC = DC->getRedeclContext();
DC->makeDeclVisibleInContext(ND, /* Recoverable=*/ false);
if (Scope *EnclosingScope = getScopeForDeclContext(S, DC))
PushOnScopeChains(ND, EnclosingScope, /*AddToContext=*/ false);
}
FriendDecl *FrD = FriendDecl::Create(Context, CurContext,
D.getIdentifierLoc(), ND,
DS.getFriendSpecLoc());
FrD->setAccess(AS_public);
CurContext->addDecl(FrD);
if (ND->isInvalidDecl())
FrD->setInvalidDecl();
else {
FunctionDecl *FD;
if (FunctionTemplateDecl *FTD = dyn_cast<FunctionTemplateDecl>(ND))
FD = FTD->getTemplatedDecl();
else
FD = cast<FunctionDecl>(ND);
// Mark templated-scope function declarations as unsupported.
if (FD->getNumTemplateParameterLists())
FrD->setUnsupportedFriend(true);
}
2010-08-21 17:40:31 +08:00
return ND;
}
2010-08-21 17:40:31 +08:00
void Sema::SetDeclDeleted(Decl *Dcl, SourceLocation DelLoc) {
AdjustDeclIfTemplate(Dcl);
FunctionDecl *Fn = dyn_cast<FunctionDecl>(Dcl);
if (!Fn) {
Diag(DelLoc, diag::err_deleted_non_function);
return;
}
if (const FunctionDecl *Prev = Fn->getPreviousDeclaration()) {
Diag(DelLoc, diag::err_deleted_decl_not_first);
Diag(Prev->getLocation(), diag::note_previous_declaration);
// If the declaration wasn't the first, we delete the function anyway for
// recovery.
}
Fn->setDeleted();
}
static void SearchForReturnInStmt(Sema &Self, Stmt *S) {
for (Stmt::child_range CI = S->children(); CI; ++CI) {
Stmt *SubStmt = *CI;
if (!SubStmt)
continue;
if (isa<ReturnStmt>(SubStmt))
Self.Diag(SubStmt->getSourceRange().getBegin(),
diag::err_return_in_constructor_handler);
if (!isa<Expr>(SubStmt))
SearchForReturnInStmt(Self, SubStmt);
}
}
void Sema::DiagnoseReturnInConstructorExceptionHandler(CXXTryStmt *TryBlock) {
for (unsigned I = 0, E = TryBlock->getNumHandlers(); I != E; ++I) {
CXXCatchStmt *Handler = TryBlock->getHandler(I);
SearchForReturnInStmt(*this, Handler);
}
}
bool Sema::CheckOverridingFunctionReturnType(const CXXMethodDecl *New,
const CXXMethodDecl *Old) {
QualType NewTy = New->getType()->getAs<FunctionType>()->getResultType();
QualType OldTy = Old->getType()->getAs<FunctionType>()->getResultType();
if (Context.hasSameType(NewTy, OldTy) ||
NewTy->isDependentType() || OldTy->isDependentType())
return false;
// Check if the return types are covariant
QualType NewClassTy, OldClassTy;
/// Both types must be pointers or references to classes.
if (const PointerType *NewPT = NewTy->getAs<PointerType>()) {
if (const PointerType *OldPT = OldTy->getAs<PointerType>()) {
NewClassTy = NewPT->getPointeeType();
OldClassTy = OldPT->getPointeeType();
}
} else if (const ReferenceType *NewRT = NewTy->getAs<ReferenceType>()) {
if (const ReferenceType *OldRT = OldTy->getAs<ReferenceType>()) {
if (NewRT->getTypeClass() == OldRT->getTypeClass()) {
NewClassTy = NewRT->getPointeeType();
OldClassTy = OldRT->getPointeeType();
}
}
}
// The return types aren't either both pointers or references to a class type.
if (NewClassTy.isNull()) {
Diag(New->getLocation(),
diag::err_different_return_type_for_overriding_virtual_function)
<< New->getDeclName() << NewTy << OldTy;
Diag(Old->getLocation(), diag::note_overridden_virtual_function);
return true;
}
// C++ [class.virtual]p6:
// If the return type of D::f differs from the return type of B::f, the
// class type in the return type of D::f shall be complete at the point of
// declaration of D::f or shall be the class type D.
if (const RecordType *RT = NewClassTy->getAs<RecordType>()) {
if (!RT->isBeingDefined() &&
RequireCompleteType(New->getLocation(), NewClassTy,
PDiag(diag::err_covariant_return_incomplete)
<< New->getDeclName()))
return true;
}
if (!Context.hasSameUnqualifiedType(NewClassTy, OldClassTy)) {
// Check if the new class derives from the old class.
if (!IsDerivedFrom(NewClassTy, OldClassTy)) {
Diag(New->getLocation(),
diag::err_covariant_return_not_derived)
<< New->getDeclName() << NewTy << OldTy;
Diag(Old->getLocation(), diag::note_overridden_virtual_function);
return true;
}
// Check if we the conversion from derived to base is valid.
if (CheckDerivedToBaseConversion(NewClassTy, OldClassTy,
diag::err_covariant_return_inaccessible_base,
diag::err_covariant_return_ambiguous_derived_to_base_conv,
// FIXME: Should this point to the return type?
New->getLocation(), SourceRange(), New->getDeclName(), 0)) {
// FIXME: this note won't trigger for delayed access control
// diagnostics, and it's impossible to get an undelayed error
// here from access control during the original parse because
// the ParsingDeclSpec/ParsingDeclarator are still in scope.
Diag(Old->getLocation(), diag::note_overridden_virtual_function);
return true;
}
}
// The qualifiers of the return types must be the same.
if (NewTy.getLocalCVRQualifiers() != OldTy.getLocalCVRQualifiers()) {
Diag(New->getLocation(),
diag::err_covariant_return_type_different_qualifications)
<< New->getDeclName() << NewTy << OldTy;
Diag(Old->getLocation(), diag::note_overridden_virtual_function);
return true;
};
// The new class type must have the same or less qualifiers as the old type.
if (NewClassTy.isMoreQualifiedThan(OldClassTy)) {
Diag(New->getLocation(),
diag::err_covariant_return_type_class_type_more_qualified)
<< New->getDeclName() << NewTy << OldTy;
Diag(Old->getLocation(), diag::note_overridden_virtual_function);
return true;
};
return false;
}
/// \brief Mark the given method pure.
///
/// \param Method the method to be marked pure.
///
/// \param InitRange the source range that covers the "0" initializer.
bool Sema::CheckPureMethod(CXXMethodDecl *Method, SourceRange InitRange) {
if (Method->isVirtual() || Method->getParent()->isDependentContext()) {
Method->setPure();
return false;
}
if (!Method->isInvalidDecl())
Diag(Method->getLocation(), diag::err_non_virtual_pure)
<< Method->getDeclName() << InitRange;
return true;
}
/// ActOnCXXEnterDeclInitializer - Invoked when we are about to parse
/// an initializer for the out-of-line declaration 'Dcl'. The scope
/// is a fresh scope pushed for just this purpose.
///
/// After this method is called, according to [C++ 3.4.1p13], if 'Dcl' is a
/// static data member of class X, names should be looked up in the scope of
/// class X.
2010-08-21 17:40:31 +08:00
void Sema::ActOnCXXEnterDeclInitializer(Scope *S, Decl *D) {
// If there is no declaration, there was an error parsing it.
if (D == 0) return;
// We should only get called for declarations with scope specifiers, like:
// int foo::bar;
assert(D->isOutOfLine());
EnterDeclaratorContext(S, D->getDeclContext());
}
/// ActOnCXXExitDeclInitializer - Invoked after we are finished parsing an
2010-08-21 17:40:31 +08:00
/// initializer for the out-of-line declaration 'D'.
void Sema::ActOnCXXExitDeclInitializer(Scope *S, Decl *D) {
// If there is no declaration, there was an error parsing it.
if (D == 0) return;
assert(D->isOutOfLine());
ExitDeclaratorContext(S);
}
/// ActOnCXXConditionDeclarationExpr - Parsed a condition declaration of a
/// C++ if/switch/while/for statement.
/// e.g: "if (int x = f()) {...}"
2010-08-21 17:40:31 +08:00
DeclResult Sema::ActOnCXXConditionDeclaration(Scope *S, Declarator &D) {
// C++ 6.4p2:
// The declarator shall not specify a function or an array.
// The type-specifier-seq shall not contain typedef and shall not declare a
// new class or enumeration.
assert(D.getDeclSpec().getStorageClassSpec() != DeclSpec::SCS_typedef &&
"Parser allowed 'typedef' as storage class of condition decl.");
TagDecl *OwnedTag = 0;
TypeSourceInfo *TInfo = GetTypeForDeclarator(D, S, &OwnedTag);
QualType Ty = TInfo->getType();
if (Ty->isFunctionType()) { // The declarator shall not specify a function...
// We exit without creating a CXXConditionDeclExpr because a FunctionDecl
// would be created and CXXConditionDeclExpr wants a VarDecl.
Diag(D.getIdentifierLoc(), diag::err_invalid_use_of_function_type)
<< D.getSourceRange();
return DeclResult();
} else if (OwnedTag && OwnedTag->isDefinition()) {
// The type-specifier-seq shall not declare a new class or enumeration.
Diag(OwnedTag->getLocation(), diag::err_type_defined_in_condition);
}
2010-08-21 17:40:31 +08:00
Decl *Dcl = ActOnDeclarator(S, D);
if (!Dcl)
return DeclResult();
return Dcl;
}
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
void Sema::MarkVTableUsed(SourceLocation Loc, CXXRecordDecl *Class,
bool DefinitionRequired) {
// Ignore any vtable uses in unevaluated operands or for classes that do
// not have a vtable.
if (!Class->isDynamicClass() || Class->isDependentContext() ||
CurContext->isDependentContext() ||
ExprEvalContexts.back().Context == Unevaluated)
return;
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
// Try to insert this class into the map.
Class = cast<CXXRecordDecl>(Class->getCanonicalDecl());
std::pair<llvm::DenseMap<CXXRecordDecl *, bool>::iterator, bool>
Pos = VTablesUsed.insert(std::make_pair(Class, DefinitionRequired));
if (!Pos.second) {
// If we already had an entry, check to see if we are promoting this vtable
// to required a definition. If so, we need to reappend to the VTableUses
// list, since we may have already processed the first entry.
if (DefinitionRequired && !Pos.first->second) {
Pos.first->second = true;
} else {
// Otherwise, we can early exit.
return;
}
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
}
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
// Local classes need to have their virtual members marked
// immediately. For all other classes, we mark their virtual members
// at the end of the translation unit.
if (Class->isLocalClass())
MarkVirtualMembersReferenced(Loc, Class);
else
VTableUses.push_back(std::make_pair(Class, Loc));
}
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
bool Sema::DefineUsedVTables() {
if (VTableUses.empty())
return false;
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
// Note: The VTableUses vector could grow as a result of marking
// the members of a class as "used", so we check the size each
// time through the loop and prefer indices (with are stable) to
// iterators (which are not).
for (unsigned I = 0; I != VTableUses.size(); ++I) {
2010-05-25 08:32:58 +08:00
CXXRecordDecl *Class = VTableUses[I].first->getDefinition();
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
if (!Class)
continue;
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
SourceLocation Loc = VTableUses[I].second;
// If this class has a key function, but that key function is
// defined in another translation unit, we don't need to emit the
// vtable even though we're using it.
const CXXMethodDecl *KeyFunction = Context.getKeyFunction(Class);
if (KeyFunction && !KeyFunction->hasBody()) {
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
switch (KeyFunction->getTemplateSpecializationKind()) {
case TSK_Undeclared:
case TSK_ExplicitSpecialization:
case TSK_ExplicitInstantiationDeclaration:
// The key function is in another translation unit.
continue;
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
case TSK_ExplicitInstantiationDefinition:
case TSK_ImplicitInstantiation:
// We will be instantiating the key function.
break;
}
} else if (!KeyFunction) {
// If we have a class with no key function that is the subject
// of an explicit instantiation declaration, suppress the
// vtable; it will live with the explicit instantiation
// definition.
bool IsExplicitInstantiationDeclaration
= Class->getTemplateSpecializationKind()
== TSK_ExplicitInstantiationDeclaration;
for (TagDecl::redecl_iterator R = Class->redecls_begin(),
REnd = Class->redecls_end();
R != REnd; ++R) {
TemplateSpecializationKind TSK
= cast<CXXRecordDecl>(*R)->getTemplateSpecializationKind();
if (TSK == TSK_ExplicitInstantiationDeclaration)
IsExplicitInstantiationDeclaration = true;
else if (TSK == TSK_ExplicitInstantiationDefinition) {
IsExplicitInstantiationDeclaration = false;
break;
}
}
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
if (IsExplicitInstantiationDeclaration)
continue;
}
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
// Mark all of the virtual members of this class as referenced, so
// that we can build a vtable. Then, tell the AST consumer that a
// vtable for this class is required.
MarkVirtualMembersReferenced(Loc, Class);
CXXRecordDecl *Canonical = cast<CXXRecordDecl>(Class->getCanonicalDecl());
Consumer.HandleVTable(Class, VTablesUsed[Canonical]);
// Optionally warn if we're emitting a weak vtable.
if (Class->getLinkage() == ExternalLinkage &&
Class->getTemplateSpecializationKind() != TSK_ImplicitInstantiation) {
if (!KeyFunction || (KeyFunction->hasBody() && KeyFunction->isInlined()))
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
Diag(Class->getLocation(), diag::warn_weak_vtable) << Class;
}
}
Rework when and how vtables are emitted, by tracking where vtables are "used" (e.g., we will refer to the vtable in the generated code) and when they are defined (i.e., because we've seen the key function definition). Previously, we were effectively tracking "potential definitions" rather than uses, so we were a bit too eager about emitting vtables for classes without key functions. The new scheme: - For every use of a vtable, Sema calls MarkVTableUsed() to indicate the use. For example, this occurs when calling a virtual member function of the class, defining a constructor of that class type, dynamic_cast'ing from that type to a derived class, casting to/through a virtual base class, etc. - For every definition of a vtable, Sema calls MarkVTableUsed() to indicate the definition. This happens at the end of the translation unit for classes whose key function has been defined (so we can delay computation of the key function; see PR6564), and will also occur with explicit template instantiation definitions. - For every vtable defined/used, we mark all of the virtual member functions of that vtable as defined/used, unless we know that the key function is in another translation unit. This instantiates virtual member functions when needed. - At the end of the translation unit, Sema tells CodeGen (via the ASTConsumer) which vtables must be defined (CodeGen will define them) and which may be used (for which CodeGen will define the vtables lazily). From a language perspective, both the old and the new schemes are permissible: we're allowed to instantiate virtual member functions whenever we want per the standard. However, all other C++ compilers were more lazy than we were, and our eagerness was both a performance issue (we instantiated too much) and a portability problem (we broke Boost test cases, which now pass). Notes: (1) There's a ton of churn in the tests, because the order in which vtables get emitted to IR has changed. I've tried to isolate some of the larger tests from these issues. (2) Some diagnostics related to implicitly-instantiated/implicitly-defined virtual member functions have moved to the point of first use/definition. It's better this way. (3) I could use a review of the places where we MarkVTableUsed, to see if I missed any place where the language effectively requires a vtable. Fixes PR7114 and PR6564. llvm-svn: 103718
2010-05-14 00:44:06 +08:00
VTableUses.clear();
return true;
}
void Sema::MarkVirtualMembersReferenced(SourceLocation Loc,
const CXXRecordDecl *RD) {
for (CXXRecordDecl::method_iterator i = RD->method_begin(),
e = RD->method_end(); i != e; ++i) {
CXXMethodDecl *MD = *i;
// C++ [basic.def.odr]p2:
// [...] A virtual member function is used if it is not pure. [...]
if (MD->isVirtual() && !MD->isPure())
MarkDeclarationReferenced(Loc, MD);
}
// Only classes that have virtual bases need a VTT.
if (RD->getNumVBases() == 0)
return;
for (CXXRecordDecl::base_class_const_iterator i = RD->bases_begin(),
e = RD->bases_end(); i != e; ++i) {
const CXXRecordDecl *Base =
cast<CXXRecordDecl>(i->getType()->getAs<RecordType>()->getDecl());
if (Base->getNumVBases() == 0)
continue;
MarkVirtualMembersReferenced(Loc, Base);
}
}
/// SetIvarInitializers - This routine builds initialization ASTs for the
/// Objective-C implementation whose ivars need be initialized.
void Sema::SetIvarInitializers(ObjCImplementationDecl *ObjCImplementation) {
if (!getLangOptions().CPlusPlus)
return;
if (ObjCInterfaceDecl *OID = ObjCImplementation->getClassInterface()) {
llvm::SmallVector<ObjCIvarDecl*, 8> ivars;
CollectIvarsToConstructOrDestruct(OID, ivars);
if (ivars.empty())
return;
llvm::SmallVector<CXXCtorInitializer*, 32> AllToInit;
for (unsigned i = 0; i < ivars.size(); i++) {
FieldDecl *Field = ivars[i];
if (Field->isInvalidDecl())
continue;
CXXCtorInitializer *Member;
InitializedEntity InitEntity = InitializedEntity::InitializeMember(Field);
InitializationKind InitKind =
InitializationKind::CreateDefault(ObjCImplementation->getLocation());
InitializationSequence InitSeq(*this, InitEntity, InitKind, 0, 0);
ExprResult MemberInit =
InitSeq.Perform(*this, InitEntity, InitKind, MultiExprArg());
MemberInit = MaybeCreateExprWithCleanups(MemberInit);
// Note, MemberInit could actually come back empty if no initialization
// is required (e.g., because it would call a trivial default constructor)
if (!MemberInit.get() || MemberInit.isInvalid())
continue;
Member =
new (Context) CXXCtorInitializer(Context, Field, SourceLocation(),
SourceLocation(),
MemberInit.takeAs<Expr>(),
SourceLocation());
AllToInit.push_back(Member);
// Be sure that the destructor is accessible and is marked as referenced.
if (const RecordType *RecordTy
= Context.getBaseElementType(Field->getType())
->getAs<RecordType>()) {
CXXRecordDecl *RD = cast<CXXRecordDecl>(RecordTy->getDecl());
if (CXXDestructorDecl *Destructor = LookupDestructor(RD)) {
MarkDeclarationReferenced(Field->getLocation(), Destructor);
CheckDestructorAccess(Field->getLocation(), Destructor,
PDiag(diag::err_access_dtor_ivar)
<< Context.getBaseElementType(Field->getType()));
}
}
}
ObjCImplementation->setIvarInitializers(Context,
AllToInit.data(), AllToInit.size());
}
}