Identifiers
An identifier is an arbitrarily long sequence of digits, underscores, lowercase and uppercase Latin letters, and most Unicode characters.
The first character of a valid identifier must be one of the following:
- uppercase Latin letters A-Z
- lowercase Latin letters a-z
- underscore
- any Unicode character with the Unicode property XID_Start
Any other character of a valid identifier must be one of the following:
- digits 0-9
- uppercase Latin letters A-Z
- lowercase Latin letters a-z
- underscore
- any Unicode character with the Unicode property XID_Continue
The lists of characters with properties XID_Start and XID_Continue can be found in DerivedCoreProperties.txt.
Identifiers are case-sensitive (lowercase and uppercase letters are distinct), and every character is significant. Every identifier must conform to Normalization Form C.
Note: Support of Unicode identifiers is limited in most implementations, e.g. gcc (until 10).
In declarations
An identifier can be used to name objects, references, functions, enumerators, types, class members, namespaces, templates, template specializations, parameter packs(since C++11) goto labels, and other entities, with the following exceptions:
- The identifiers that are keywords cannot be used for other purposes.
|
(since C++11) |
- The identifiers that are alternative representations for certain operators and punctuators cannot be used for other purposes.
|
(since C++11) |
- Identifiers that appear as a token or preprocessing token (i.e., not in user-defined-string-literal like operator ""id)(since C++11) of one of the following forms are reserved:
- in the global namespace, identifiers that begin with an underscore
- identifiers that contain a double underscore or begin with an underscore followed by an uppercase letter, except the following identifiers:
- predefined macros (including language feature test macros)(since C++20)
(since C++11) |
|
(since C++11) |
(since C++20) |
“Reserved” here means that the standard library headers #define or declare such identifiers for their internal needs, the compiler may predefine non-standard identifiers of that kind, and that name mangling algorithm may assume that some of these identifiers are not in use. If the programmer uses such identifiers, the program is ill-formed, no diagnostic required.
In addition, it is undefined behavior to #define or #undef certain names in a translation unit, see reserved macro names for more details.
Zombie identifiers
As of C++14, some identifiers are removed from the C++ standard library. They are listed in the list of zombie names.
However, these identifiers are still reserved for previous standardization in a certain context. Removed member function names may not be used as a name for function-like macros, and other removed member names may not be used as a name for object-like macros in portable code.
In expressions
An identifier that names a variable, a function, specialization of a concept,(since C++20) or an enumerator can be used as an expression. The result of an expression consisting of just the identifier is the entity named by the identifier. The value category of the expression is lvalue if the identifier names a function, a variable, a template parameter object(since C++20), or a data member, and rvalue(until C++11)prvalue(since C++11) otherwise (e.g. an enumerator is an rvalue(until C++11)a prvalue(since C++11) expression, a specialization of a concept is a bool prvalue(since C++20)).
Type
The type of an identifier expression is the same as the type of the entity it names.
The following exceptions exist:
void f() { float x, &r = x; [=] { decltype(x) y1; // y1 has type float decltype((x)) y2 = y1; // y2 has type float const& because this lambda // is not mutable and x is an lvalue decltype(r) r1 = y1; // r1 has type float& decltype((r)) r2 = y2; // r2 has type float const& }; }
|
(since C++11) |
Unqualified identifiers
Besides suitably declared identifiers, the following can be used in expressions in the same role:
- an overloaded operator name in function notation, such as operator+ or operator new;
- a user-defined conversion function name, such as operator bool;
|
(since C++11) |
- a template name followed by its argument list, such as MyTemplate<int>;
- the character ~ followed by a class name, such as ~MyClass;
|
(since C++11) |
|
(since C++26) |
Together with identifiers they are known as unqualified identifier expressions.
Qualified identifiers
A qualified identifier expression is an unqualified identifier expression prepended by a scope resolution operator ::, and optionally, a sequence of any of the following separated by scope resolution operators:
- a namespace name;
- a class name;
|
(since C++11) |
|
(since C++26) |
For example, the expression std::string::npos is an expression that names the static member npos in the class string in namespace std. The expression ::tolower names the function tolower in the global namespace. The expression ::std::cout names the global variable cout in namespace std, which is a top-level namespace. The expression boost::signals2::connection names the type connection declared in namespace signals2, which is declared in namespace boost.
The keyword template may appear in qualified identifiers as necessary to disambiguate dependent template names.
See qualified lookup for the details of the name lookup for qualified identifiers.
Implicit member access transformation
If an identifier expression E denotes a non-static non-type member of some class C
and all following conditions are satisfied, E is transformed into the class member access expression this->E:
- E is not the right operand of a member access operator.
- If E is a qualified identifier expression, E not the un-parenthesized operand of an address-of operator.
- Any of the following conditions is satisfied:
- E is potentially evaluated.
-
C
is the innermost enclosing class at E. -
C
is a base class of the innermost enclosing class at E.
This transformation does not apply in the template definition context (see dependent names).
struct X { int x; }; struct B { int b; }; struct D : B { X d; void func() { d; // OK, will be transformed into this->d b; // OK, will be transformed into this->b x; // Error: this->x is ill-formed d.x; // OK, will be transformed into this->d.x // instead of d.this->x or this->d.this->x } };
Names
A name is the use of one of the following to refer to an entity:
- an identifier
- an overloaded operator name in function notation (operator+, operator new)
- a user-defined conversion function name (operator bool)
|
(since C++11) |
- a template name followed by its argument list (MyTemplate<int>)
Every name is introduced into the program by a declaration. A name used in more than one translation unit may refer to the same or different entities, depending on linkage.
When the compiler encounters an unknown name in a program, it associates it with the declaration that introduced the name by means of name lookup, except for the dependent names in template declarations and definitions (for those names, the compiler determines whether they name a type, a template, or some other entity, which may require explicit disambiguation).
Defect reports
The following behavior-changing defect reports were applied retroactively to previously published C++ standards.
DR | Applied to | Behavior as published | Correct behavior |
---|---|---|---|
CWG 1440 | C++11 | decltype expressions preceding :: could denote any type
|
can only denote class or enumeration types |
CWG 1963 | C++11 | implementation-defined characters other than digits, non-digits and universal character names could be used in an identifier |
prohibited |
CWG 2521 | C++11 | the identifier in user-defined-string-literal of a literal operator was reserved as usual |
the rules are different |
CWG 2771 | C++98 | &a was not transformed into &this->a in class contexts | it is transformed |
CWG 2777 | C++20 | the type of an identifier expression was unclear if it names a template parameter object |
made clear |
CWG 2818 | C++98 | predefined macro names are reserved | they are not reserved |
See also
C documentation for Identifiers
|