Class template argument deduction (CTAD) (since C++17) (original) (raw)

In order to instantiate a class template, every template argument must be known, but not every template argument has to be specified. In the following contexts the compiler will deduce the template arguments from the type of the initializer:

• any declaration that specifies initialization of a variable and variable template, whose declared type is the class template (possibly cv-qualified):

std::pair p(2, 4.5); // deduces to std::pair<int, double> p(2, 4.5); std::tuple t(4, 3, 2.5); // same as auto t = std::make_tuple(4, 3, 2.5); std::less l; // same as std::less l;

• new-expressions:

template struct A { A(T, T); };   auto y = new A{1, 2}; // allocated type is A

• function-style cast expressions:

auto lck = std::lock_guard(mtx); // deduces to std::lock_guardstd::mutex std::copy_n(vi1, 3, std::back_insert_iterator(vi2)); // deduces to std::back_insert_iterator, // where T is the type of the container vi2 std::for_each(vi.begin(), vi.end(), Foo([&](int i) {...})); // deduces to Foo, // where T is the unique lambda type

Contents

• 1 Deduction for class templates
  • 1.1 Implicitly-generated deduction guides
  • 1.2 User-defined deduction guides
• 2 Deduction for alias templates
• 3 Notes
• 4 Defect reports

[edit] Deduction for class templates

[edit] Implicitly-generated deduction guides

When, in a function-style cast or in a variable's declaration, the type specifier consists solely of the name of a primary class template C (i.e., there is no accompanying template argument list), candidates for deduction are formed as follows:

• If C is defined, for each constructor (or constructor template) Ci declared in the named primary template, a fictional function template Fi, is constructed, such that all following conditions are satisfied:

• The template parameters of Fi are the template parameters of C followed (if Ci is a constructor template) by the template parameters of Ci (default template arguments are included too).

• The parameter list of Fi is the parameter list of Ci.

• The return type of Fi is C followed by the template parameters of the class template enclosed in <>.

• If C is not defined or does not declare any constructors, an additional fictional function template is added, derived as above from a hypothetical constructor C().

• In any case, an additional fictional function template derived as above from a hypothetical constructor C(C) is added, called the copy deduction candidate.

• For each user-defined deduction guide Gi, a fictional function or function template Fi, is constructed, such that all following conditions are satisfied:

• The parameter list of Fi is the parameter list of Gi.

• The return type of Fi is the simple template identifier of Gi.

• If Gi has template parameters (syntax (2)), Fi is a function template, and its template parameter list is the template parameter list of Gi. Otherwise, Fi is a function.

Template argument deduction and overload resolution is then performed for initialization of a fictional object of hypothetical class type, whose constructor signatures match the guides (except for return type) for the purpose of forming an overload set, and the initializer is provided by the context in which class template argument deduction was performed, except that the first phase of list-initialization (considering initializer-list constructors) is omitted if the initializer list consists of a single expression of type (possibly cv-qualified) U, where U is a specialization of C or a class derived from a specialization of C.

These fictional constructors are public members of the hypothetical class type. They are explicit if the guide was formed from an explicit constructor. If overload resolution fails, the program is ill-formed. Otherwise, the return type of the selected F template specialization becomes the deduced class template specialization.

template struct UniquePtr { UniquePtr(T* t); };   UniquePtr dp{new auto(2.0)};   // One declared constructor: // C1: UniquePtr(T*);   // Set of implicitly-generated deduction guides:   // F1: template // UniquePtr F(T* p);   // F2: template // UniquePtr F(UniquePtr); // copy deduction candidate   // imaginary class to initialize: // struct X // { // template // X(T* p); // from F1 //
// template // X(UniquePtr); // from F2 // };   // direct-initialization of an X object // with "new double(2.0)" as the initializer // selects the constructor that corresponds to the guide F1 with T = double // For F1 with T=double, the return type is UniquePtr   // result: // UniquePtr dp{new auto(2.0)}

Or, for a more complex example (note: "S::N" would not compile: scope resolution qualifiers are not something that can be deduced):

template struct S { template struct N { N(T); N(T, U);   template N(V, U); }; };   S::N x{2.0, 1};   // the implicitly-generated deduction guides are (note that T is already known to be int)   // F1: template // S::N F(int);   // F2: template // S::N F(int, U);   // F3: template<class U, class V> // S::N F(V, U);   // F4: template // S::N F(S::N); (copy deduction candidate)   // Overload resolution for direct-list-init with "{2.0, 1}" as the initializer // chooses F3 with U=int and V=double. // The return type is S::N   // result: // S::N x{2.0, 1};

[edit] User-defined deduction guides

The syntax of a user-defined deduction guide is the syntax of a function (template) declaration with a trailing return type, except that it uses the name of a class template as the function name:

| | | | | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | --- | | | explicit (optional) template-name ( parameter-list ) -> simple-template-id requires-clause (optional) ; | (1) | | | | | | | template <**template-parameter-list **> requires-clause (optional)explicit (optional) template-name ( parameter-list ) -> simple-template-id requires-clause (optional) ; | (2) | | | | | |

User-defined deduction guides must name a class template and must be introduced within the same semantic scope of the class template (which could be namespace or enclosing class) and, for a member class template, must have the same access, but deduction guides do not become members of that scope.

A deduction guide is not a function and does not have a body. Deduction guides are not found by name lookup and do not participate in overload resolution except for the overload resolution against other deduction guides when deducing class template arguments. Deduction guides cannot be redeclared in the same translation unit for the same class template.

// declaration of the template template struct container { container(T t) {}   template container(Iter beg, Iter end); };   // additional deduction guide template container(Iter b, Iter e) -> container<typename std::iterator_traits::value_type>;   // uses container c(7); // OK: deduces T=int using an implicitly-generated guide std::vector v = {/* ... */}; auto d = container(v.begin(), v.end()); // OK: deduces T=double container e{5, 6}; // Error: there is no std::iterator_traits::value_type

The fictional constructors for the purpose of overload resolution (described above) are explicit if they correspond to an implicitly-generated deduction guide formed from an explicit constructor or to a user-defined deduction guide that is declared explicit. As always, such constructors are ignored in copy-initialization context:

template struct A { explicit A(const T&, ...) noexcept; // #1 A(T&&, ...); // #2 };   int i; A a1 = {i, i}; // error: cannot deduce from rvalue reference in #2, // and #1 is explicit, and not considered in copy-initialization. A a2{i, i}; // OK, #1 deduces to A and also initializes A a3{0, i}; // OK, #2 deduces to A and also initializes A a4 = {0, i}; // OK, #2 deduces to A and also initializes   template A(const T&, const T&) -> A<T&>; // #3   template explicit A(T&&, T&&) -> A; // #4   A a5 = {0, 1}; // error: #3 deduces to A<int&> // and #1 & #2 result in same parameter constructors. A a6{0, 1}; // OK, #4 deduces to A and #2 initializes A a7 = {0, i}; // error: #3 deduces to A<int&> A a8{0, i}; // error: #3 deduces to A<int&>   // Note: check https://github.com/cplusplus/CWG/issues/647, claiming that // examples a7 and a8 are incorrect, to be possibly replaced as //A a7 = {0, i}; // error: #2 and #3 both match, overload resolution fails //A a8{i,i}; // error: #3 deduces to A<int&>, // // #1 and #2 declare same constructor

Using a member typedef or alias template in a constructor or constructor template's parameter list does not, by itself, render the corresponding parameter of the implicitly generated guide a non-deduced context.

template struct B { template using TA = T;   template B(U, TA); // #1 };   // Implicit deduction guide generated from #1 is the equivalent of // template<class T, class U> // B(U, T) -> B; // rather than // template<class T, class U> // B(U, typename B::template TA) -> B; // which would not have been deducible   B b{(int*)0, (char*)0}; // OK, deduces B<char*>

Deduction for alias templates When a function-style cast or declaration of a variable uses the name of an alias template A without an argument list as the type specifier, where A is defined as an alias of B, the scope of B is non-dependent, and B is either a class template or a similarly-defined alias template, deduction will proceed in the same way as for class templates, except that the guides are instead generated from the guides of B, as follows: For each guide f of B, deduce the template arguments of the return type of f from B using template argument deduction, except that deduction does not fail if some arguments are not deduced. If deduction fails for another reason, proceed with an empty set of deduced template arguments. Substitute the result of above deduction into f, if substitution fails, no guide is produced; otherwise, let g denote the result of substitution, a guide f' is formed, such that The parameter types and the return type of f' are the same as g If f is a template, f' is a function template whose template parameter list consists of all the template parameters of A (including their default template arguments) that appear in the above deductions or (recursively) in their default template arguments, followed by the template parameters of f that were not deduced (including their default template arguments); otherwise (f is not a template), f' is a function The associated constraints of f' are the conjunction of the associated constraints of g and a constraint that is satisfied if and only if the arguments of A are deducible from the result type template<class T> class unique_ptr { /* ... */ };   template<class T> class unique_ptr<T[]> { /* ... */ };   template<class T> unique_ptr(T*) -> unique_ptr<T>; // #1   template<class T> unique_ptr(T*) -> unique_ptr<T[]>; // #2   template<class T> concept NonArray = <T>;   template<NonArray A> using unique_ptr_nonarray = unique_ptr<A>;   template<class A> using unique_ptr_array = unique_ptr<A[]>;   // generated guide for unique_ptr_nonarray:   // from #1 (deduction of unique_ptr from unique_ptr yields T = A): // template // requires(argument_of_unique_ptr_nonarray_is_deducible_from<unique_ptr>) // auto F(A*) -> unique_ptr;   // from #2 (deduction of unique_ptr<T[]> from unique_ptr yields nothing): // template // requires(argument_of_unique_ptr_nonarray_is_deducible_from<unique_ptr<T[]>>) // auto F(T*) -> unique_ptr<T[]>;   // where argument_of_unique_ptr_nonarray_is_deducible_from can be defined as   // template // class AA;   // template // class AA<unique_ptr_nonarray> {};   // template // concept argument_of_unique_ptr_nonarray_is_deducible_from = // requires { sizeof(AA); };   // generated guide for unique_ptr_array:   // from #1 (deduction of unique_ptr from unique_ptr<A[]> yields T = A[]): // template // requires(argument_of_unique_ptr_array_is_deducible_from<unique_ptr<A[]>>) // auto F(A(*)[]) -> unique_ptr<A[]>;   // from #2 (deduction of unique_ptr<T[]> from unique_ptr<A[]> yields T = A): // template // requires(argument_of_unique_ptr_array_is_deducible_from<unique_ptr<A[]>>) // auto F(A*) -> unique_ptr<A[]>;   // where argument_of_unique_ptr_array_is_deducible_from can be defined as   // template // class BB;   // template // class BB<unique_ptr_array> {};   // template // concept argument_of_unique_ptr_array_is_deducible_from = // requires { sizeof(BB); };   // Use: unique_ptr_nonarray p(new int); // deduced to unique_ptr // deduction guide generated from #1 returns unique_ptr // deduction guide generated from #2 returns unique_ptr<int[]>, which is ignored because // argument_of_unique_ptr_nonarray_is_deducible_from<unique_ptr<int[]>> is unsatisfied   unique_ptr_array q(new int[42]); // deduced to unique_ptr<int[]> // deduction guide generated from #1 fails (cannot deduce A in A(*)[] from new int[42]) // deduction guide generated from #2 returns unique_ptr<int[]>

(since C++20)

[edit] Notes

Class template argument deduction is only performed if no template argument list is present. If a template argument list is specified, deduction does not take place.

std::tuple t1(1, 2, 3); // OK: deduction std::tuple<int, int, int> t2(1, 2, 3); // OK: all arguments are provided   std::tuple<> t3(1, 2, 3); // Error: no matching constructor in tuple<>. // No deduction performed. std::tuple t4(1, 2, 3); // Error

User-defined deduction guides do not have to be templates:

template struct S { S(T); }; S(char const*) -> S<std::string>;   S s{"hello"}; // deduced to Sstd::string

Within the scope of a class template, the name of the template without a parameter list is an injected class name, and can be used as a type. In that case, class argument deduction does not happen and template parameters must be supplied explicitly:

template struct X { X(T) {}   template X(Iter b, Iter e) {}   template auto foo(Iter b, Iter e) { return X(b, e); // no deduction: X is the current X }   template auto bar(Iter b, Iter e) { return X(b, e); // must specify what we want }   auto baz() { return ::X(0); // not the injected-class-name; deduced to be X } };

In overload resolution, partial ordering takes precedence over whether a function template is generated from a user-defined deduction guide: if the function template generated from the constructor is more specialized than the one generated from the user-defined deduction guide, the one generated from the constructor is chosen. Because the copy deduction candidate is typically more specialized than a wrapping constructor, this rule means that copying is generally preferred over wrapping.

template struct A { A(T, int*); // #1 A(A&, int*); // #2   enum { value }; };   template<class T, int N = T::value> A(T&&, int*) -> A; //#3   A a{1, 0}; // uses #1 to deduce A and initializes with #1 A b{a, 0}; // uses #2 (more specialized than #3) to deduce A and initializes with #2

When earlier tiebreakers, including partial ordering, failed to distinguish between two candidate function templates, the following rules apply:

• A function template generated from a user-defined deduction guide is preferred over one implicitly generated from a constructor or constructor template.
• The copy deduction candidate is preferred over all other function templates implicitly generated from a constructor or constructor template.
• A function template implicitly generated from a non-template constructor is preferred over a function template implicitly generated from a constructor template.

template struct A { using value_type = T;   A(value_type); // #1 A(const A&); // #2 A(T, T, int); // #3   template A(int, T, U); // #4 }; // #5, the copy deduction candidate A(A);   A x(1, 2, 3); // uses #3, generated from a non-template constructor   template A(T) -> A; // #6, less specialized than #5   A a(42); // uses #6 to deduce A and #1 to initialize A b = a; // uses #5 to deduce A and #2 to initialize   template A(A) -> A<A>; // #7, as specialized as #5   A b2 = a; // uses #7 to deduce A<A> and #1 to initialize

An rvalue reference to a cv-unqualified template parameter is not a forwarding reference if that parameter is a class template parameter:

template struct A { template A(T&&, U&&, int*); // #1: T&& is not a forwarding reference // U&& is a forwarding reference   A(T&&, int*); // #2: T&& is not a forwarding reference };   template A(T&&, int*) -> A; // #3: T&& is a forwarding reference   int i, *ip; A a{i, 0, ip}; // error, cannot deduce from #1 A a0{0, 0, ip}; // uses #1 to deduce A and #1 to initialize A a2{i, ip}; // uses #3 to deduce A<int&> and #2 to initialize

When initializing from a single argument of a type that is a specialization of the class template at issue, copying deduction is generally preferred over wrapping by default:

std::tuple t1{1}; //std::tuple std::tuple t2{t1}; //std::tuple, not std::tuple<std::tuple>   std::vector v1{1, 2}; // std::vector std::vector v2{v1}; // std::vector, not std::vector<std::vector> (P0702R1) std::vector v3{v1, v2}; // std::vector<std::vector>

Outside the special case for copying vs. wrapping, the strong preference for initializer-list constructors in list-initialization remains intact.

std::vector v1{1, 2}; // std::vector   std::vector v2(v1.begin(), v1.end()); // std::vector std::vector v3{v1.begin(), v1.end()}; // std::vector<std::vector::iterator>

Before class template argument deduction was introduced, a common approach to avoiding explicitly specifying arguments is to use a function template:

std::tuple p1{1, 1.0}; //std::tuple<int, double>, using deduction auto p2 = std::make_tuple(1, 1.0); //std::tuple<int, double>, pre-C++17

[edit] Defect reports

The following behavior-changing defect reports were applied retroactively to previously published C++ standards.