duckdb::Utf8Proc Class Reference

Static Public Member Functions
static UnicodeType	Analyze (const char *s, size_t len, UnicodeInvalidReason *invalid_reason=nullptr, size_t *invalid_pos=nullptr)
	Distinguishes ASCII, Valid UTF8 and Invalid UTF8 strings.
static char *	Normalize (const char *s, size_t len)
	Performs UTF NFC normalization of string, return value needs to be free'd.
static bool	IsValid (const char *s, size_t len)
	Returns whether or not the UTF8 string is valid.
static void	MakeValid (char *s, size_t len, char special_flag='?')
	Makes Invalid Unicode valid by replacing invalid parts with a given character.
static std::string	RemoveInvalid (const char *s, size_t len)
	Creates a new string with invalid UTF-8 characters removed.
static size_t	NextGraphemeCluster (const char *s, size_t len, size_t pos)
	Returns the position (in bytes) of the next grapheme cluster.
static size_t	PreviousGraphemeCluster (const char *s, size_t len, size_t pos)
	Returns the position (in bytes) of the previous grapheme cluster.
static bool	CodepointToUtf8 (int cp, int &sz, char *c)
	Transform a codepoint to utf8 and writes it to "c", sets "sz" to the size of the codepoint.
static int	CodepointLength (int cp)
	Returns the codepoint length in bytes when encoded in UTF8.
static int32_t	UTF8ToCodepoint (const char *c, int &sz)
	Transform a UTF8 string to a codepoint; returns the codepoint and writes the length of the codepoint (in UTF8) to sz.
static size_t	RenderWidth (const char *s, size_t len, size_t pos)
	Returns the render width of a single character in a string.
static size_t	RenderWidth (const std::string &str)
static int32_t	CodepointToUpper (int32_t codepoint)
static int32_t	CodepointToLower (int32_t codepoint)
static GraphemeIterator	GraphemeClusters (const char *s, size_t len)
	Constructs a class that can be iterated over to fetch grapheme clusters in a string.
static size_t	GraphemeCount (const char *s, size_t len)
	Returns the number of grapheme clusters in a string.

---

The documentation for this class was generated from the following file:

• external/duckdb/duckdb.cpp