desr: desr: text/Utf8Utils.h Source File

00001 /*
00002 **  Tanl C++ Library
00003 **  ixe/text/Utf8Utils.h: utilities for handling UTF-8
00004 **  ----------------------------------------------------------------------
00005 **  Copyright (c) 2002  Giuseppe Attardi (attardi@di.unipi.it).
00006 **  ----------------------------------------------------------------------
00007 **
00008 **  This file is part of DeSR.
00009 **
00010 **  DeSR is free software; you can redistribute it and/or modify it
00011 **  under the terms of the GNU General Public License, version 3,
00012 **  as published by the Free Software Foundation.
00013 **
00014 **  DeSR is distributed in the hope that it will be useful,
00015 **  but WITHOUT ANY WARRANTY; without even the implied warranty of
00016 **  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
00017 **  GNU General Public License for more details.
00018 **
00019 **  You should have received a copy of the GNU General Public License
00020 **  along with this program.  If not, see <http://www.gnu.org/licenses/>.
00021 **  ----------------------------------------------------------------------
00022 */
00023 
00024 #ifndef Tanl_text_Utf8Utils_h
00025 #define Tanl_text_Utf8Utils_h
00026 
00027 // Standard
00028 #include <sys/types.h>
00029 #include <string.h>
00030 
00031 #if defined(_WIN32)
00032 # include <tchar.h>
00033 #endif
00034 
00035 // local
00036 #include "text.h"
00037 
00038 namespace Tanl {
00039 namespace Text {
00040 
00044 namespace Unicode
00045 {
00046   typedef unsigned char byte;
00047 
00048   void incUtfPtr(const char*& ptr, const char* end);
00049   void decUtfPtr(const char*& ptr, const char* begin);
00050   size_t utfDiff(const char* end, const char* begin);
00051   bool isAscii(const char* begin, const char* end);
00052 
00053   UCS2 fetchChar(const char*& begin, const char* end);
00054   inline UCS2 toChar(const char* begin, const char* end)
00055   {
00056     const char* it = begin;
00057     return fetchChar(it, end);
00058   }
00059   int byteLength(const char* it);
00063   int utfLength(UCS2 ch);
00064   
00068   int utfLength(const UCS2* begin, const UCS2* end);
00072   int ucLength(const UCS2* ch);
00073   inline int uc2Length(const UCS2* ch) { return ucLength(ch); }
00077   int uc4Length(const UCS4* ch);
00087   int ToUtf8(byte*& it, byte* end, UCS2 ucc);
00095   int ToUtf8(byte* it, const size_t max, const UCS2* ucbegin, const size_t len);
00096 
00106   UCS2* ToUc(UCS2* buffer, const char* source, int length);
00117   UCS2* ToUc(UCS2* buffer, const UCS4* source, int length = -1);
00128   UCS2* ToUc(UCS2* buffer, const wchar_t* source, int length = -1);
00132   UCS2* utf8ToUc(UCS2* buffer, const char* source, int length = -1);
00133 
00134   inline size_t stringLength(const char* str) { return strlen(str); }
00135   inline size_t stringLength(const UCS2* str) { return ucLength(str); }
00136   inline size_t stringLength(const UCS4* str) { return uc4Length(str); }
00137   
00138   template <typename ToCharType, typename FromCharType>
00139   inline ToCharType* convertTo(ToCharType* buffer, const FromCharType* source, int length = -1)
00140   {
00141     if (length == -1)
00142       length = stringLength(source);
00143     ToCharType* tit = buffer;
00144     register const FromCharType* fit = source;
00145     register const FromCharType* end = source + length;
00146     for (; fit < end; ++tit, ++fit)
00147       *tit = *fit;
00148     return buffer;
00149   }
00150 
00151 } // namespace Unicode
00152 } // namespace text
00153 } // namespace Tanl
00154 
00155 
00156 #endif //Tanl_text_Utf8Utils_h