Add comments to UTF-8 declarations.

Add US ASCII optimizations macros.
#ifdef out unused routines
Ready to hack getdn.c and others to support UTF-8

diff --git a/include/ldap_pvt.h b/include/ldap_pvt.h
index f1a0ab6aef69aca2da5b5f6fded57f2aefafff4a..8bb43326fe0152d3edbc6c7d51f053c80ce632c2 100644 (file)
--- a/include/ldap_pvt.h
+++ b/include/ldap_pvt.h
@@ -147,25 +147,54 @@ LIBLDAP_F (int) ldap_pvt_tls_start LDAP_P(( Sockbuf *sb, void *ctx_arg ));
  * UTF-8 (in utf-8.c)
  */
 
+/* returns the number of bytes in the UTF-8 string
+       (counting the NULL) */
 LIBLDAP_F (ber_len_t) ldap_utf8_bytes( const char * );
+/* returns the number of UTF-8 characters in the string */
 LIBLDAP_F (ber_len_t) ldap_utf8_chars( const char * );
+/* returns the length (in bytes) of a UTF-8 string */
 LIBLDAP_F (int) ldap_utf8_charlen( const char * );
+/* copies a UTF-8 character and returning number of bytes copied */
+LIBLDAP_F (int) ldap_utf8_copy( char *, const char *);
 
+/* returns pointer of next UTF-8 character in string */
 LIBLDAP_F (char*) ldap_utf8_next( const char * );
+/* returns pointer of previous UTF-8 character in string */
 LIBLDAP_F (char*) ldap_utf8_prev( const char * );
 
+/* primitive ctype routines -- not aware of non-ascii characters */
 LIBLDAP_F (int) ldap_utf8_isascii( const char * );
-LIBLDAP_F (int) ldap_utf8_isalpa( const char * );
+LIBLDAP_F (int) ldap_utf8_isalpha( const char * );
 LIBLDAP_F (int) ldap_utf8_isalnum( const char * );
 LIBLDAP_F (int) ldap_utf8_isdigit( const char * );
 LIBLDAP_F (int) ldap_utf8_isxdigit( const char * );
 LIBLDAP_F (int) ldap_utf8_isspace( const char * );
 
+/* span characters not in set, return bytes spanned */
 LIBLDAP_F (ber_len_t) ldap_utf8_strcspn( const char* str, const char *set);
+/* span characters in set, return bytes spanned */
 LIBLDAP_F (ber_len_t) ldap_utf8_strspn( const char* str, const char *set);
+/* return first character of set in string */
 LIBLDAP_F (char *) ldap_utf8_strpbrk( const char* str, const char *set);
+/* reentrant tokenizer */
 LIBLDAP_F (char*) ldap_utf8_strtok( char* sp, const char* sep, char **last);
 
+/* Optimizations */
+#define LDAP_UTF8_CHARLEN(p) (*(unsigned char *)(p) < 0x100 \
+       ? 1 : ldap_utf8_charlen((p)))
+
+#define LDAP_UTF8_COPY(p) (*(unsigned char *)(s) < 0x100 \
+       ? (*(d) = *(s), 1) : ldap_utf8_cpy((d),(s)))
+
+#define LDAP_UTF8_NEXT(p) (*(unsigned char *)(p) < 0x100 \
+       ? &(p)[1] : ldap_utf8_next((p)))
+
+#define LDAP_UTF8_INCR(p) (*(unsigned char *)(p) < 0x100 \
+       ? ++(p) : (p)=ldap_utf8_next((p)))
+
+/* For symmetry */
+#define LDAP_UTF8_PREV(p) (ldap_utf8_prev((p)))
+#define LDAP_UTF8_DECR(p) ((p)=ldap_utf8_prev((p)))
 
 LDAP_END_DECL
 

diff --git a/libraries/libldap/utf-8.c b/libraries/libldap/utf-8.c
index e14ebf84d842e79895719f88b3bd7aa1e4078a68..29e96a18a871f3519a36b2dfdb482798ac2d9c92 100644 (file)
--- a/libraries/libldap/utf-8.c
+++ b/libraries/libldap/utf-8.c
@@ -308,7 +308,6 @@ int ldap_utf8_isupper( const char * p )
 
        return ( c >= 'A' && c <= 'Z' );
 }
-#endif
 
 /*
  * get one UTF-8 character
@@ -352,6 +351,7 @@ char* ldap_utf8_fgetc( FILE *s, char *buf )
 
        return buf;
 }
+#endif
 
 
 /*
@@ -423,7 +423,7 @@ char *(ldap_utf8_strpbrk)( const char *str, const char *set )
 
                for( cset = set; ; cset += len ) {
                        if( ldap_utf8_to_ucs4( cstr ) == ldap_utf8_to_ucs4( cset ) ) {
-                               return cstr;
+                               return (char *) cstr;
                        } 
 
                        len = ldap_utf8_charlen(cset);