3 * Copyright 1998-1999 The OpenLDAP Foundation, All Rights Reserved.
4 * COPYING RESTRICTIONS APPLY, see COPYRIGHT file
10 * These routines are not optimized.
17 #include <ac/stdlib.h>
19 #include <ac/socket.h>
20 #include <ac/string.h>
24 #include "ldap_defaults.h"
26 #define UTF8_ISASCII(u) ( !((u) & ~0x7f) )
27 #define UCS4_INVALID 0x80000000U
30 * return the number of bytes required to hold the
31 * NULL-terminated UTF-8 string INCLUDING the
34 ber_len_t ldap_utf8_bytes( const char * p )
38 if( p == NULL ) return bytes;
47 ber_len_t ldap_utf8_chars( const char * p )
49 /* could be optimized and could check for invalid sequences */
52 for( chars=0; *p ; chars++ ) {
53 int charlen = ldap_utf8_charlen( p );
55 if( !charlen ) return chars;
63 int ldap_utf8_charlen( const char * p )
65 unsigned c = * (const unsigned char *) p;
67 if ((c & 0xfe ) == 0xfc) {
71 if ((c & 0xfc ) == 0xf8) {
75 if ((c & 0xf8 ) == 0xf0) {
79 if ((c & 0xf0 ) == 0xe0) {
83 if ((c & 0xe0 ) == 0xc0) {
87 if ((c & 0x80 ) == 0x80) {
95 ber_int_t ldap_utf8_to_ucs4( const char * p )
98 ber_int_t c = * (const unsigned char *) p;
100 if ((c & 0xFE ) == 0xFC) {
104 } else if ((c & 0xFC ) == 0xF8) {
108 } else if ((c & 0xF8 ) == 0xF0) {
112 } else if ((c & 0xF0 ) == 0xE0) {
116 } else if ((c & 0xE0 ) == 0xC0) {
120 } else if ((c & 0x80 ) == 0x80) {
127 for(i=1; i < len; i++) {
128 ber_int_t ch = ((const unsigned char *) p)[i];
130 if ((ch & 0xc0) != 0x80) {
141 int ldap_ucs4_to_utf8( ber_int_t c, char *buf )
144 unsigned char* p = buf;
145 if(buf == NULL) return 0;
148 /* not a valid Unicode character */
150 } else if( c < 0x80 ) {
153 } else if( c < 0x800 ) {
154 p[len++] = 0xc0 | ( c >> 6 );
155 p[len++] = 0x80 | ( c & 0x3F );
157 } else if( c < 0x10000 ) {
158 p[len++] = 0xe0 | ( c >> 12 );
159 p[len++] = 0x80 | ( (c >> 6) & 0x3F );
160 p[len++] = 0x80 | ( c & 0x3F );
162 } else if( c < 0x200000 ) {
163 p[len++] = 0xf0 | ( c >> 18 );
164 p[len++] = 0x80 | ( (c >> 12) & 0x3F );
165 p[len++] = 0x80 | ( (c >> 6) & 0x3F );
166 p[len++] = 0x80 | ( c & 0x3F );
168 } else if( c < 0x400000 ) {
169 p[len++] = 0xf8 | ( c >> 24 );
170 p[len++] = 0x80 | ( (c >> 18) & 0x3F );
171 p[len++] = 0x80 | ( (c >> 12) & 0x3F );
172 p[len++] = 0x80 | ( (c >> 6) & 0x3F );
173 p[len++] = 0x80 | ( c & 0x3F );
175 } else /* if( c < 0x80000000 ) */ {
176 p[len++] = 0xfc | ( c >> 30 );
177 p[len++] = 0x80 | ( (c >> 24) & 0x3F );
178 p[len++] = 0x80 | ( (c >> 18) & 0x3F );
179 p[len++] = 0x80 | ( (c >> 12) & 0x3F );
180 p[len++] = 0x80 | ( (c >> 6) & 0x3F );
181 p[len++] = 0x80 | ( c & 0x3F );
188 char* ldap_utf8_next( const char * p )
190 int len = ldap_utf8_charlen( p );
192 return len ? (char *) &p[len] : NULL;
195 char* ldap_utf8_prev( const char * p )
198 const unsigned char *u = p;
200 for( i = -1; i >= -6 ; i-- ) {
201 if ( u[i] & 0xC0 != 0x80 ) return (char *) &p[i];
207 int ldap_utf8_isascii( const char * p )
209 unsigned c = * (const unsigned char *) p;
210 return UTF8_ISASCII(c);
213 int ldap_utf8_isdigit( const char * p )
215 unsigned c = * (const unsigned char *) p;
217 if(!UTF8_ISASCII(c)) return 0;
219 return c >= '0' && c <= '9';
222 int ldap_utf8_isxdigit( const char * p )
224 unsigned c = * (const unsigned char *) p;
226 if(!UTF8_ISASCII(c)) return 0;
228 return ( c >= '0' && c <= '9' )
229 || ( c >= 'A' && c <= 'F' )
230 || ( c >= 'a' && c <= 'f' );
233 int ldap_utf8_isalpha( const char * p )
235 unsigned c = * (const unsigned char *) p;
237 if(!UTF8_ISASCII(c)) return 0;
239 return ( c >= 'A' && c <= 'Z' )
240 || ( c >= 'a' && c <= 'z' );
243 int ldap_utf8_isalnum( const char * p )
245 unsigned c = * (const unsigned char *) p;
247 if(!UTF8_ISASCII(c)) return 0;
249 return ( c >= '0' && c <= '9' )
250 || ( c >= 'A' && c <= 'Z' )
251 || ( c >= 'a' && c <= 'z' );
254 int ldap_utf8_islower( const char * p )
256 unsigned c = * (const unsigned char *) p;
258 if(!UTF8_ISASCII(c)) return 0;
260 return ( c >= 'a' && c <= 'z' );
263 int ldap_utf8_isupper( const char * p )
265 unsigned c = * (const unsigned char *) p;
267 if(!UTF8_ISASCII(c)) return 0;
269 return ( c >= 'A' && c <= 'Z' );
272 int ldap_utf8_isspace( const char * p )
274 unsigned c = * (const unsigned char *) p;
276 if(!UTF8_ISASCII(c)) return 0;
291 char* ldap_utf8_fgetc( FILE *s, char *buf )
298 if( s == NULL ) return NULL;
310 len = ldap_utf8_charlen( buf );
312 if( len < 1 ) return NULL;
314 for( i = 1; i < len; i++ ) {
315 unsigned int c = fgetc( s );
320 if( c & 0xC0 != 0x80 ) {
331 ber_len_t (ldap_utf8_strcspn)( const char *str, const char *set )
336 for( cstr = str; *cstr != '\0'; cstr += len ) {
339 for( cset = set; ; cset += len ) {
340 if( ldap_utf8_to_ucs4( cstr ) == ldap_utf8_to_ucs4( cset ) ) {
344 len = ldap_utf8_charlen(cset);
348 len = ldap_utf8_charlen(cstr);
355 ber_len_t (ldap_utf8_strspn)( const char *str, const char *set )
360 for( cstr = str; *cstr != '\0'; cstr += len ) {
363 for( cset = set; ; cset += len ) {
364 if( *cset == '\0' ) {
368 if( ldap_utf8_to_ucs4( cstr ) == ldap_utf8_to_ucs4( cset ) ) {
372 len = ldap_utf8_charlen(cset);
376 len = ldap_utf8_charlen(cstr);
383 char *(ldap_utf8_strpbrk)( const char *str, const char *set )
388 for( cstr = str; *cstr != '\0'; cstr += len ) {
391 for( cset = set; ; cset += len ) {
392 if( ldap_utf8_to_ucs4( cstr ) == ldap_utf8_to_ucs4( cset ) ) {
396 len = ldap_utf8_charlen(cset);
400 len = ldap_utf8_charlen(cstr);
407 char *(ldap_utf8_strtok)(char *str, const char *sep, char **last)
412 if( last == NULL ) return NULL;
414 begin = str ? str : *last;
416 begin += ldap_utf8_strspn( begin, sep );
418 if( *begin == '\0' ) {
423 end = &begin[ ldap_utf8_strcpn( begin, sep ) ];
426 int len = ldap_utf8_charlen( end );