CVS: tvision/classes codepage.cc,2.24.2.14,2.24.2.15
Brought to you by:
set
Menu
▾
▴
CVS: tvision/classes codepage.cc,2.24.2.14,2.24.2.15
|
From: Salvador E. T. <se...@us...> - 2017-08-01 14:10:43
|
Update of /cvsroot/tvision/tvision/classes In directory sfp-cvs-1.v30.ch3.sourceforge.com:/tmp/cvs-serv2537/classes Modified Files: Tag: r2_0_1u codepage.cc Log Message: * Added: [TVCodePage] Functions to convert the app cp to and from UTF8 (convertStrUTF8_2_CP and convertStrCP_2_UTF8). * Added: [X11] Support for copy/paste using UTF8. Index: codepage.cc =================================================================== RCS file: /cvsroot/tvision/tvision/classes/codepage.cc,v retrieving revision 2.24.2.14 retrieving revision 2.24.2.15 diff -C2 -d -r2.24.2.14 -r2.24.2.15 *** codepage.cc 10 Jul 2008 16:10:05 -0000 2.24.2.14 --- codepage.cc 1 Aug 2017 14:10:40 -0000 2.24.2.15 *************** *** 1,5 **** /**[txh]******************************************************************** ! Copyright 1996-2003 by Salvador Eduardo Tropea (SET) This file is covered by the GPL license. --- 1,5 ---- /**[txh]******************************************************************** ! Copyright 1996-2017 by Salvador Eduardo Tropea (SET) This file is covered by the GPL license. *************** *** 2962,2965 **** --- 2962,3160 ---- Description: + Converts a string containing UTF8 to Application Code Page. The len + indicates how many chars from orig we must process. The dest buffer must + be long enough to hold the result. If dest is NULL this routine just + meassure the length of the result. + + Return: -1 if the UTF8 sequence is wrong, the length of the converted + text otherwise. + + ***************************************************************************/ + + int TVCodePage::convertStrUTF8_2_CP(char *dest, const char *orig, + unsigned len) + { + const uchar *utf8=(const uchar *)orig; + uchar *d=(uchar *)dest; + unsigned unicode; + int retLen=0; + + uchar c=*(utf8++); + while (len--) + { + if ((c&0x80)==0) + {// 1 byte + unicode=c; + } + else if ((c&0xF8)==0xF0) + {// 4 bytes + if (len<3) return -1; + len-=3; + unicode =(c&0x07)<<18; + c=*(utf8++); + if ((c&0xC0)!=0x80) return -1; + unicode|=(c&0x3F)<<12; + c=*(utf8++); + if ((c&0xC0)!=0x80) return -1; + unicode|=(c&0x3F)<<6; + c=*(utf8++); + if ((c&0xC0)!=0x80) return -1; + unicode|=(c&0x3F); + } + else if ((c&0xE0)==0xE0) + {// 3 bytes + if (len<2) return -1; + len-=2; + unicode =(c&0x0F)<<12; + c=*(utf8++); + if ((c&0xC0)!=0x80) return -1; + unicode|=(c&0x3F)<<6; + c=*(utf8++); + if ((c&0xC0)!=0x80) return -1; + unicode|=(c&0x3F); + } + else + {// 2 bytes + if (!len) return -1; + len-=1; + unicode =(c&0x1F)<<6; + c=*(utf8++); + if ((c&0xC0)!=0x80) return -1; + unicode|=(c&0x3F); + } + c=*(utf8++); + // Convert the unicode value to the app code page + if (dest) + { + if (unicode>0xFFFF) + d+=sprintf((char *)d,"&#x%08X;",unicode); // � 12 + else + { + uint16 cp=unicodeToApp->search(unicode); + if (cp==0xFFFF) + d+=sprintf((char *)d,"&#x%04X;",unicode); // � 8 + else + *(d++)=(uchar)cp; + } + } + else + { + if (unicode>0xFFFF) + retLen+=12; + else + { + uint16 cp=unicodeToApp->search(unicode); + retLen+=cp==0xFFFF ? 8 : 1; + } + } + } + if (dest) + { + *d=0; + retLen=d-(uchar *)dest; + } + return retLen; + } + + /**[txh]******************************************************************** + + Description: + Converts a string containing Application Code Page to UTF8. Only len bytes + are processed. If dest is NULL it just meassures the destination length. + + Return: The destination length, or -1 if the string can't be represented + using UTF8. + + ***************************************************************************/ + + int TVCodePage::convertStrCP_2_UTF8(char *dest, const char *orig, + unsigned len) + { + const uchar *o=(const uchar *)orig; + uchar *d=(uchar *)dest; + unsigned unicode; + int retLen=0; + + while (len--) + { + if (*o=='&') + { + unicode='&'; + // Is it a 12 bytes escaped U32? + if (len>=11 && o[1]=='#' && o[2]=='x' && o[11]==';') + {// Seems to be, try + char *end; + unsigned v=strtoul((const char *)(o+3),&end,16); + if (end-(char *)o==11) + { + unicode=v; + len-=11; + o+=11; + } + } + // Is it an 8 bytes escaped U16? + else if (len>=7 && o[1]=='#' && o[2]=='x' && o[7]==';') + {// Seems to be, try + char *end; + unsigned v=strtoul((const char *)(o+3),&end,16); + if (end-(char *)o==7) + { + unicode=v; + len-=7; + o+=7; + } + } + } + else + unicode=appToUnicode[*o]; + o++; + // Now encode the unicode value using UTF8 + if (unicode<=0x7F) + { + retLen++; + if (d) + *(d++)=unicode; + } + else if (unicode<=0x7FF) + { + retLen+=2; + if (d) + { + *(d++)=0xC0 | (unicode>>6); + *(d++)=0x80 | (unicode&0x3F); + } + } + else if (unicode<=0xFFFF) + { + retLen+=3; + if (d) + { + *(d++)=0xE0 | (unicode>>12); + *(d++)=0x80 | ((unicode>>6)&0x3F); + *(d++)=0x80 | (unicode&0x3F); + } + } + else if (unicode<=0x1FFFFF) + { + retLen+=4; + if (d) + { + *(d++)=0xF0 | (unicode>>18); + *(d++)=0x80 | ((unicode>>12)&0x3F); + *(d++)=0x80 | ((unicode>> 6)&0x3F); + *(d++)=0x80 | (unicode&0x3F); + } + } + else // UTF8 doesn't support bigger values + return -1; + } + if (d) + *d=0; + return retLen; + } + + /**[txh]******************************************************************** + + Description: Converts a string containing Application Code Page to Unicode. The destination must be (len+1)*2 bytes for the EOS. *************** *** 2973,2977 **** { uint16 *d=(uint16 *)dest; ! uchar *o=(uchar *)orig; while (len--) *(d++)=appToUnicode[*(o++)]; --- 3168,3172 ---- { uint16 *d=(uint16 *)dest; ! const uchar *o=(const uchar *)orig; while (len--) *(d++)=appToUnicode[*(o++)]; |
