/*	Public domain	*/

/*
 * Compatibility string and UTF-8 conversion routines.
 */

__BEGIN_DECLS
size_t  CGI_Strlcpy(char *, const char *, size_t)
                    BOUNDED_ATTRIBUTE(__string__, 1, 3);
size_t  CGI_Strlcat(char *, const char *, size_t)
                    BOUNDED_ATTRIBUTE(__string__, 1, 3);
char   *CGI_Strsep(char **, const char *);
char   *CGI_Strdup(const char *);
char   *CGI_TryStrdup(const char *);

/*
 * Parse the first byte of a possible UTF-8 sequence and return the length
 * of the sequence in bytes (or 1 if there is none).
 */
static __inline__ int
CGI_CharLengthUTF8(unsigned char ch)
{
	if ((ch >> 7) == 0) {
		return (1);
	} else if (((ch & 0xe0) >> 5) == 0x6) {
		return (2);
	} else if (((ch & 0xf0) >> 4) == 0xe) {
		return (3);
	} else if (((ch & 0xf8) >> 3) == 0x1e) {
		return (4);
	} else if (((ch & 0xfc) >> 2) == 0x3e) {
		return (5);
	} else if (((ch & 0xfe) >> 1) == 0x7e) {
		return (6);
	} else {
		return (-1);
	}
}

/*
 * Return the number of characters in the given UTF-8 string, not counting
 * the terminating NUL.
 */
static __inline__ size_t
CGI_LengthUTF8(const char *s)
{
	const char *c = &s[0];
	size_t rv = 0;
	int i, cLen;

	if (s[0] == '\0') {
		return (0);
	}
	for (;;) {
		if ((cLen = CGI_CharLengthUTF8((unsigned char)*c)) == -1) {
			break;
		}
		for (i = 0; i < cLen; i++) {
			if (c[i] == '\0')
				return (rv);
		}
		rv++;
		c += cLen;
	}
	return (rv);
}
__END_DECLS

#if defined(_PERCGI_INTERNAL) || defined(_USE_PERCGI_STD)
#define Strlcat CGI_Strlcat
#define Strlcpy CGI_Strlcpy
#define Strsep CGI_Strsep
#define Strdup CGI_Strdup
#define TryStrdup CGI_TryStrdup
#define LengthUTF8 CGI_LengthUTF8
#define CharLengthUTF8 CGI_CharLengthUTF8
#endif