std::text_encoding
| Defined in header <text_encoding>
|
||
| struct text_encoding; |
(since C++26) | |
The class text_encoding provides a mechanism for identifying character encodings. It is used in determining the ordinary character literal encoding and the character encoding of the current environment.
Each text_encoding object encapsulates a character encoding scheme, uniquely identified by an enumerator in text_encoding::id and a corresponding name represented by a null-terminated string. These can be accessed through the mib() and name() member functions, respectively. The determination of whether an object represents a character encoding scheme implemented in the translation or execution environment is implementation-defined.
The class text_encoding is a TriviallyCopyable type. It guarantees that its value is allocated as part of the object storage. No dynamic memory allocation takes place. The stored name is limited to a maximum of max_name_length characters excluding the null character '\0'.
The class supports both registered and non-registered character encodings. Registered encodings are those found in the IANA Character Sets Registry excluding the following character encodings:
- NATS-DANO (33)
- NATS-DANO-ADD (34).
Non-registered encodings can be represented with an enumerator id::other or id::unknown and a custom name.
For registered character encodings, the class provides access to:
- Primary name: The official name specified in the registry.
- Aliases: An implementation-defined superset of aliases from the registry.
- MIBenum value: A unique identifier for use in identifying coded character encodings.
Contents |
[edit] Member types
| represents the MIBenum value of the character encoding (public member enum) | |
a view over aliases of the character encoding (public member class) |
[edit] Member constant
| Name | Value |
| constexpr std::size_t max_name_length [static] |
63 (public static member constant) |
[edit] Data members
| Member | Description |
std::text_encoding::id mib_ (private)
|
a MIBenum value with id::unknown as the default value (exposition-only member object*) |
char[max_name_length + 1] name_ (private)
|
a stored primary name (exposition-only member object*) |
[edit] Member functions
Creation | |
constructs new text_encoding object (public member function) | |
| [static] |
constructs a new text_encoding representing the ordinary character literal encoding (public static member function) |
| [static] |
constructs a new text_encoding representing the implementation-defined character encoding scheme of the environment (public static member function) |
Observers | |
| returns the MIBenum value of the current character encoding (public member function) | |
| returns the primary name of the current character encoding (public member function) | |
returns a view over aliases of the current character encoding (public member function) | |
| [static] |
checks the character encoding scheme of the environment with the specified MIB value (public static member function) |
Helpers | |
| [static](private) |
compares two alias names using Charset Alias Matching (exposition-only static member function*) |
[edit] Non-member functions
compares two text_encoding objects. (public member function) |
[edit] Helper classes
| hash support for std::text_encoding (class template specialization) |
[edit] Notes
When working with character encodings, it is important to note that the primary names or aliases of two distinct registered character encodings are not equivalent when compared using Charset Alias Matching as described by the Unicode Technical Standard.
For convenience, the enumerators of text_encoding::id are introduced as members of text_encoding and can be accessed directly. This means that text_encoding::ASCII and text_encoding::id::ASCII refer to the same entity.
It is recommended that the implementation should treat registered encodings as not interchangeable. Additionally, the primary name of a registered encoding should not be used to describe a similar but different non-registered encoding, unless there is a clear precedent for doing so.
| Feature-test macro | Value | Std | Feature |
|---|---|---|---|
__cpp_lib_text_encoding |
202306L | (C++26) | std::text_encoding
|
[edit] Example
#include <locale> #include <print> #include <text_encoding> int main() { // literal encoding is known at compile-time constexpr std::text_encoding literal_encoding = std::text_encoding::literal(); // check for literal encoding static_assert(literal_encoding.mib() != std::text_encoding::other && literal_encoding.mib() != std::text_encoding::unknown); // environment encoding is only known at runtime std::text_encoding env_encoding = std::text_encoding::environment(); // associated encoding of the default locale std::text_encoding locale_encoding = std::locale("").encoding(); std::println("The literal encoding is {}", literal_encoding.name()); std::println("The aliases of literal encoding:"); for (const char* alias_name : literal_encoding.aliases()) std::println(" -> {}", alias_name); if (env_encoding == locale_encoding) std::println("In this case, both environment and locale encodings are the same"); std::println("The environment encoding is {}", env_encoding.name()); std::println("The aliases of environment encoding:"); for (const char* alias_name : env_encoding.aliases()) std::println(" -> {}", alias_name); }
Output:
The literal encoding is UTF-8 The aliases of literal encoding: -> UTF-8 -> csUTF8 In this case, both environment and locale encodings are the same The environment encoding is ANSI_X3.4-1968 The aliases of environment encoding: -> US-ASCII -> iso-ir-6 -> ANSI_X3.4-1968 -> ANSI_X3.4-1986 -> ISO_646.irv:1991 -> ISO646-US -> us -> IBM367 -> cp367 -> csASCII -> ASCII
[edit] See also
| set of polymorphic facets that encapsulate cultural differences (class) |
