a11::utf8 Namespace Reference

Functions
bool	IsContinuation (char byte)
	Whether byte is a continuation byte – the middle of a character.
size_t	SequenceWidth (char lead)
	Bytes in the sequence lead starts, from the lead byte alone.
size_t	Utf16Units (char lead)
	UTF-16 code units the character at lead occupies: 2 above the BMP, else.
bool	IsValid (std::string_view text)
	Whether text is well-formed UTF-8.

Function Documentation

IsContinuation()

bool a11::utf8::IsContinuation ( char  byte )

inline

Whether byte is a continuation byte – the middle of a character.

IsValid()

bool a11::utf8::IsValid ( std::string_view  text )

inline

Whether text is well-formed UTF-8.

Strict: overlong encodings, surrogate halves and anything above U+10FFFF are rejected, which is the definition Python's bytes.decode("utf-8"), Kotlin's String(bytes) and JavaScript's TextDecoder with fatal all enforce. See a11::IsValidUtf8 in a11/json_codec.h for why A11 refuses such bytes on the way out rather than letting a peer fail on them.

SequenceWidth()

size_t a11::utf8::SequenceWidth ( char  lead )

inline

Bytes in the sequence lead starts, from the lead byte alone.

1 for ASCII and for a malformed lead, so that walking a document that is not quite UTF-8 advances rather than stalling or running off the end. A caller that must reject malformed input wants IsValid instead.

Utf16Units()

size_t a11::utf8::Utf16Units ( char  lead )

inline

UTF-16 code units the character at lead occupies: 2 above the BMP, else.