ref: 64afa666f159be2f11dc99a3f42a72c48fa38b6f
dir: /utf8.h/
#ifndef __UTF8_H_
#define __UTF8_H_
/* is c the start of a utf8 sequence? */
#define isutf(c) (((c)&0xC0) != 0x80)
int u8_iswprint(uint32_t c);
/* convert UTF-8 data to wide character */
size_t u8_toucs(uint32_t *dest, size_t sz, const char *src, size_t srcsz);
/* the opposite conversion */
size_t u8_toutf8(char *dest, size_t sz, const uint32_t *src, size_t srcsz);
/* single character to UTF-8, returns # bytes written */
size_t u8_wc_toutf8(char *dest, uint32_t ch);
/* byte offset to character number */
size_t u8_charnum(const char *s, size_t offset);
/* return next character, updating an index variable */
uint32_t u8_nextchar(const char *s, size_t *i);
/* next character without NUL character terminator */
uint32_t u8_nextmemchar(const char *s, size_t *i);
/* returns length of next utf-8 sequence */
size_t u8_seqlen(const char *s);
/* returns the # of bytes needed to encode a certain character */
size_t u8_charlen(uint32_t ch);
/* computes the # of bytes needed to encode a WC string as UTF-8 */
size_t u8_codingsize(uint32_t *wcstr, size_t n);
char read_escape_control_char(char c);
/* given a wide character, convert it to an ASCII escape sequence stored in
buf, where buf is "sz" bytes. returns the number of characters output.
sz must be at least 3. */
int u8_escape_wchar(char *buf, size_t sz, uint32_t ch);
/* convert UTF-8 "src" to escape sequences.
sz is buf size in bytes. must be at least 12.
if escape_quotes is nonzero, quote characters will be escaped.
if ascii is nonzero, the output is 7-bit ASCII, no UTF-8 survives.
starts at src[*pi], updates *pi to point to the first unprocessed
byte of the input.
end is one more than the last allowable value of *pi.
returns number of bytes placed in buf, including a NUL terminator.
*/
size_t u8_escape(char *buf, size_t sz, const char *src, size_t *pi, size_t end,
int escape_quotes, int ascii);
/* utility predicates used by the above */
int octal_digit(char c);
int hex_digit(char c);
/* same as the above, but searches a buffer of a given size instead of
a NUL-terminated string. */
char *u8_memchr(const char *s, uint32_t ch, size_t sz, size_t *charn);
/* number of columns occupied by a string */
size_t u8_strwidth(const char *s);
/* determine whether a sequence of bytes is valid UTF-8. length is in bytes */
int u8_isvalid(const char *str, int length);
/* reverse a UTF-8 string. len is length in bytes. dest and src must both
be allocated to at least len+1 bytes. returns 1 for error, 0 otherwise */
int u8_reverse(char *dest, char *src, size_t len);
#endif