DCHookTest/StringLib.cpp

   1 //---------------------------------------------------------------------------
   2 #include <vcl.h>
   3 #pragma hdrstop
   4
   5 #include "StringLib.h"
   6
   7 //---------------------------------------------------------------------------
   8 #pragma package(smart_init)
   9
  10 // pos\82©\82ç\92P\8cê\82ð\90Ø\82è\8fo\82·
  11 // pos\82Ístr\82©\82ç\82Ì\83I\83t\83Z\83b\83g
  12 // start,end\82Ístr\82©\82ç\82Ì\83I\83t\83Z\83b\83g\82ð\95Ô\82·
  13 // prevstart\82Í\81Astart\82Ì\82P\82Â\91O\82Ì\92P\8cê\81A\96³\82¢\8fê\8d\87\82Ístart\82Æ\93¯\82¶\92l
  14 // fLongest\82ªTRUE\82Ì\8fê\8d\87\82Í\81A\8dÅ\8d\8210\8cê\82Ü\82Åget\82·\82é(.)\82ð\8fI\82í\82è\82Æ\82Ý\82È\82·
  15 // about   : \82¢\82¢\89Á\8c¸\82È\82Æ\82±\82ë\82ð\83N\83\8a\83b\83N\82µ\82Ä\82à\92P\8cê\82³\82¦\82 \82ê\82Î\95K\82¸get\82·\82é
  16 // alnum   : \89p\90\94\8e\9a\88È\8aO\82Í\96³\8e\8b
  17
  18 bool GetWord( const tchar *str, int pos, int &start, int &end, int &prevstart, bool fLongest, int wordcount, bool about, bool alnum )
  19 {
  20         if ( !about )
  21                 if ( !IsWordChar( *(str+pos) ) )
  22                         return false;
  23 //      while ( *(str+pos) == ' ' || *(str+pos) == '\t' ){
  24 //              pos++;
  25 //      }
  26
  27         // \89ü\8ds\82Ì\8fê\8d\87\82Í\96³\82µ
  28         if ( str[pos] == '\r' || str[pos] == '\n' ){
  29                 return false;
  30         }
  31
  32         // pos\82ª\95¶\8e\9a\97ñ\82Ì\8fI\92[\81A\82 \82é\82¢\82Í\95¶\8e\9a\97ñ\82Ì\8fI\92[\82©\82ç\82P\82Â\91O\82Ì\89ü\8ds\82Å\82 \82é\8fê\8d\87\82Í\82¾\82ß //
  33         if ( !str[pos] || (!str[pos+1] && ((tuchar)str[pos] < ' ')) )
  34                 return false;
  35
  36         const tchar *orgp = str;
  37         const tchar *p = orgp;
  38 rescan:
  39         // \92P\8cê\82Ì\90æ\93ª\82ð\92T\82· //
  40         const tchar *wordtop = NULL;
  41         while ( *p ){
  42                 if ( alnum ){
  43                         // \89p\90\94\8e\9a\82Ì\82Ý
  44                         if ( isalphanum(*p) ){
  45                                 // \92P\8cê\82Ì\8en\82Ü\82è
  46                                 wordtop = p;
  47                                 break;
  48                         }
  49                         p = NEXT_CHAR(p);
  50                 } else {
  51                         // \95¶\8e\9a\8eí\82ð\8bæ\95Ê\82µ\82È\82¢
  52                         if ( IsWordChar( *p ) )
  53                         {
  54                                 wordtop = p;
  55                                 break;
  56                         }
  57                         p++;
  58                 }
  59         }
  60         if ( !wordtop )
  61                 wordtop = p;
  62         const tchar *wordtail = NULL;
  63         const tchar *wordprev = NULL;
  64         bool fSpc = false;
  65         // pos\82É\82 \82é\92P\8cê\82Ìstart\82Æend\82ð\8b\81\82ß\82é
  66         while ( *p ){
  67                 if ( !IsWordChar( *p ) || (alnum && !isalphanum(*p)) ){
  68                         // \94ñ\92P\8cê\95¶\8e\9a
  69                         if ( orgp + pos < p ){
  70                                 // \92P\8cê\82Ì\8fI\82í\82è
  71                                 wordtail = p;
  72                                 break;
  73                         } else {
  74                                 if ( alnum ){
  75                                         if ( _ismbblead(*p) ){
  76                                                 // \93ú\96{\8cê\82Ì\8fê\8d\87\82Í\81A\92P\8f\83\82È\94ñ\92P\8cê\95¶\8e\9a\82Æ\82Í\82Ý\82È\82³\82¸\81A
  77                                                 // \8dÅ\8f\89\82©\82çscan\82ð\82â\82è\92¼\82·
  78                                                 goto rescan;
  79                                         }
  80                                 }
  81                         }
  82                         fSpc = true;
  83                 } else {
  84                         // \92P\8cê\95¶\8e\9a
  85                         if ( fSpc ){
  86                                 wordprev = wordtop;
  87                                 wordtop = p;
  88                                 fSpc = FALSE;
  89                         }
  90                 }
  91                 if ( alnum ){
  92                         p = NEXT_CHAR( p );
  93                 } else {
  94                         p++;
  95                 }
  96         }
  97
  98         // wordtop\82ªpos\82æ\82è\8cã\82ë
  99         if ( (unsigned)wordtop - (about ? 3 : 0) > (unsigned)(orgp + pos) ){
 100                 // \92P\8cê\82Ì\8bæ\90Ø\82è\82¾\82Á\82½\81A\92P\8cê\82Ì\8dÅ\8f\89\82ªpos\82æ\82è\8cã\82ë\82É\82 \82Á\82½
 101                 // about = true\82Ì\8fê\8d\87\82Í\81A\82R\95¶\8e\9a\95ª\82Ù\82Ç\82¢\82¢\82©\82°\82ñ\8d·\82ð\91«\82µ\82Ä\82¨\82
 102                 return false;
 103         }
 104
 105         if ( fLongest ){
 106 //              int wordcount = 10;     // 10\8cê\82Ü\82Å
 107                 while ( *p && *p != '.' ){
 108                         if ( !IsWordChar( *p ) || (alnum && !isalphanum(*p)) ){
 109                                 if ( !fSpc ){
 110                                         if ( --wordcount == 0 )
 111                                                 break;
 112                                         fSpc = true;
 113                                 }
 114                         } else {
 115                                 fSpc = FALSE;
 116                         }
 117                         if ( alnum ){
 118                                 p = NEXT_CHAR( p );
 119                         } else {
 120                                 p++;
 121                         }
 122                 }
 123                 wordtail = p;
 124         }
 125
 126         if ( !wordtail )
 127                 wordtail = p;
 128         start = STR_DIFF( wordtop, orgp );
 129         end = STR_DIFF( wordtail, orgp );
 130         if ( start == end )
 131                 return false;
 132         if ( wordprev ){
 133                 prevstart = STR_DIFF( wordprev, orgp );
 134         } else {
 135                 prevstart = start;
 136         }
 137         return true;
 138 }
 139 #if 0
 140 bool mbGetWord( const tchar *str, int pos, int &start, int &end, bool fLongest, int wordcount )
 141 {
 142         ushort c;
 143         const tchar *sp = str + pos;
 144         LD_CHAR( c, sp );
 145         if ( !mbIsWordChar( c ) )
 146                 return false;
 147
 148         const tchar *orgp = str;
 149         const tchar *p = orgp;
 150         const tchar *wordtop = p;
 151         const tchar *wordtail = NULL;
 152         bool fSpc = false;
 153         while ( 1 ){
 154                 sp = p;
 155                 LD_CHAR( c, p );
 156                 if ( (tuchar)c == 0x00 )
 157                         break;
 158                 if ( !mbIsWordChar( c ) ){
 159                         if ( orgp + pos < sp ){
 160                                 wordtail = sp;
 161                                 break;
 162                         }
 163                         fSpc = true;
 164                 } else {
 165                         if ( fSpc ){
 166                                 wordtop = sp;
 167                                 fSpc = false;
 168                         }
 169                 }
 170         }
 171         p = sp;
 172         if ( wordtop > orgp + pos ){
 173                 // \92P\8cê\82Ì\8bæ\90Ø\82è\82¾\82Á\82½
 174                 return false;
 175         }
 176
 177         if ( fLongest ){
 178 //              int wordcount = 10;     // 10\8cê\82Ü\82Å
 179                 while ( 1 ){
 180                         sp = p;
 181                         LD_CHAR( c, p );
 182                         if ( (tuchar)c == 0x00 )
 183                                 break;
 184                         if ( c == '.' )
 185                                 break;
 186                         if ( !mbIsWordChar( c ) ){
 187                                 if ( !fSpc ){
 188                                         if ( --wordcount == 0 )
 189                                                 break;
 190                                         fSpc = true;
 191                                 }
 192                         } else {
 193                                 fSpc = false;
 194                         }
 195                 }
 196                 wordtail = sp;
 197         }
 198
 199         if ( !wordtail )
 200                 wordtail = sp;
 201         start = STR_DIFF( wordtop, orgp );
 202         end = STR_DIFF( wordtail, orgp );
 203         return true;
 204 }
 205 bool mbIsWordChar( unsigned short c )
 206 {
 207         if ( c < 0x100 ){
 208                 // \94¼\8ap\95¶\8e\9a
 209                 if ( isalphanum( c ) || c == '-' || c == '\'' ||
 210                          (c >= 0x86 && c <= 0xBF ) )
 211                         return true;
 212                 else
 213                         return false;
 214         }
 215         // \91S\8ap\95¶\8e\9a
 216         if (
 217                 // \89p\90\94\8e\9a\81A\83J\83^\83J\83i\81A\82Ð\82ç\82ª\82È\81A\83M\83\8a\83V\83\83\81A\83\8d\83V\83A
 218                 (c >= 0x824f && c <= 0x8491) ||
 219                 // \8a¿\8e\9a
 220                 (c >= 0x889F) ||
 221                 // \83A\83|\83X\83g\83\8d\83t\83B
 222                 (c == CODE_APOSTROPHE)
 223                 )
 224                 return true;
 225         else
 226                 return false;
 227 }
 228 #endif
 229 //\89p\90\94\8e\9a\82Ì\94»\95Ê
 230 int isalphanum( tuchar c )
 231 {
 232         if ( c >= _t('A') && c <= _t('Z') ){
 233                 return 1;
 234         }
 235         if ( c >= _t('a') && c <= _t('z') ){
 236                 return 1;
 237         }
 238         if ( c >= _t('0') && c <= _t('9') ){
 239                 return 1;
 240         }
 241         return 0;
 242 }
 243