X-Git-Url: https://git.sur5r.net/?a=blobdiff_plain;f=include%2Fldap_pvt_uc.h;h=a9b2515757c7012bbad8650da24d8d19f80ddd1f;hb=0f4a5f74453365236874d4223e1a177de0a2970d;hp=5292207630f912e22b390009a5c4c33f825bcde1;hpb=c38027902a17d6bd354074136ac1af4ecff38e4e;p=openldap diff --git a/include/ldap_pvt_uc.h b/include/ldap_pvt_uc.h index 5292207630..a9b2515757 100644 --- a/include/ldap_pvt_uc.h +++ b/include/ldap_pvt_uc.h @@ -1,14 +1,18 @@ /* $OpenLDAP$ */ -/* - * Copyright 1998-2002 The OpenLDAP Foundation, Redwood City, California, USA +/* This work is part of OpenLDAP Software . + * + * Copyright 1998-2005 The OpenLDAP Foundation. * All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted only as authorized by the OpenLDAP - * Public License. A copy of this license is available at - * http://www.OpenLDAP.org/license.html or in file LICENSE in the - * top-level directory of the distribution. + * Public License. + * + * A copy of this license is available in file LICENSE in the + * top-level directory of the distribution or, alternatively, at + * . */ + /* * ldap_pvt_uc.h - Header for Unicode functions. * These are meant to be used by the OpenLDAP distribution only. @@ -20,11 +24,8 @@ #include /* get ber_slen_t */ -#ifdef _MSC_VER +#include #include "../libraries/liblunicode/ucdata/ucdata.h" -#else -#include "../libraries/liblunicode/ucdata.h" -#endif LDAP_BEGIN_DECL @@ -39,6 +40,11 @@ typedef unsigned long ldap_unicode_t; #define ldap_utf8_to_unicode( p ) ldap_utf8_to_ucs4((p)) #define ldap_unicode_to_utf8( c, buf ) ldap_ucs4_to_ucs4((c),(buf)) +/* Convert a string with csize octets per character to UTF-8 */ +LDAP_F( int ) ldap_ucs_to_utf8s LDAP_P(( + struct berval *ucs, int csize, struct berval *utf8s )); + + /* returns the number of bytes in the UTF-8 string */ LDAP_F (ber_len_t) ldap_utf8_bytes( const char * ); /* returns the number of UTF-8 characters in the string */ @@ -47,6 +53,12 @@ LDAP_F (ber_len_t) ldap_utf8_chars( const char * ); LDAP_F (int) ldap_utf8_offset( const char * ); /* returns the length (in bytes) indicated by the UTF-8 character */ LDAP_F (int) ldap_utf8_charlen( const char * ); + +/* returns the length (in bytes) indicated by the UTF-8 character + * also checks that shortest possible encoding was used + */ +LDAP_F (int) ldap_utf8_charlen2( const char * ); + /* copies a UTF-8 character and returning number of bytes copied */ LDAP_F (int) ldap_utf8_copy( char *, const char *); @@ -76,10 +88,20 @@ LDAP_F (char*) ldap_utf8_strtok( char* sp, const char* sep, char **last); /* Optimizations */ LDAP_V (const char) ldap_utf8_lentab[128]; +LDAP_V (const char) ldap_utf8_mintab[32]; -#define LDAP_UTF8_ISASCII(p) ( *(unsigned char *)(p) ^ 0x80 ) +#define LDAP_UTF8_ISASCII(p) ( !(*(const unsigned char *)(p) & 0x80 ) ) #define LDAP_UTF8_CHARLEN(p) ( LDAP_UTF8_ISASCII(p) \ - ? 1 : ldap_utf8_lentab[*(unsigned char *)(p) ^ 0x80] ) + ? 1 : ldap_utf8_lentab[*(const unsigned char *)(p) ^ 0x80] ) + +/* This is like CHARLEN but additionally validates to make sure + * the char used the shortest possible encoding. + * 'l' is used to temporarily hold the result of CHARLEN. + */ +#define LDAP_UTF8_CHARLEN2(p, l) ( ( ( l = LDAP_UTF8_CHARLEN( p )) < 3 || \ + ( ldap_utf8_mintab[*(const unsigned char *)(p) & 0x1f] & (p)[1] ) ) ? \ + l : 0 ) + #define LDAP_UTF8_OFFSET(p) ( LDAP_UTF8_ISASCII(p) \ ? 1 : ldap_utf8_offset((p)) ) @@ -121,17 +143,23 @@ LDAP_LUNICODE_F(void) ucstr2upper( ldap_unicode_t *, ber_len_t ); -#define LDAP_UTF8_CASEFOLD 0x1U #define LDAP_UTF8_NOCASEFOLD 0x0U +#define LDAP_UTF8_CASEFOLD 0x1U +#define LDAP_UTF8_ARG1NFC 0x2U +#define LDAP_UTF8_ARG2NFC 0x4U +#define LDAP_UTF8_APPROX 0x8U -LDAP_LUNICODE_F(char *) UTF8normalize( +LDAP_LUNICODE_F(struct berval *) UTF8bvnormalize( + struct berval *, struct berval *, - unsigned ); + unsigned, + void *memctx ); -LDAP_LUNICODE_F(int) UTF8normcmp( - const char *, - const char *, - unsigned ); +LDAP_LUNICODE_F(int) UTF8bvnormcmp( + struct berval *, + struct berval *, + unsigned, + void *memctx ); LDAP_END_DECL