tools/cjk_variants.c

   1 /* Copyright (C) 1999-2002, 2012, 2018 Free Software Foundation, Inc.
   2    This file is part of the GNU LIBICONV Tools.
   3
   4    This program is free software: you can redistribute it and/or modify
   5    it under the terms of the GNU General Public License as published by
   6    the Free Software Foundation; either version 3 of the License, or
   7    (at your option) any later version.
   8
   9    This program is distributed in the hope that it will be useful,
  10    but WITHOUT ANY WARRANTY; without even the implied warranty of
  11    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  12    GNU General Public License for more details.
  13
  14    You should have received a copy of the GNU General Public License
  15    along with this program; if not, see <https://www.gnu.org/licenses/>.  */
  16
  17 /*
  18  * Generates Unicode variants table from Koichi Yasuoka's UniVariants file.
  19  */
  20
  21 #include <stdio.h>
  22 #include <stdlib.h>
  23
  24 #define ENTRIES  8176  /* number of lines in UniVariants file */
  25 #define MAX_PER_ENTRY  10  /* max number of entries per line in file */
  26
  27 int main (int argc, char *argv[])
  28 {
  29   int variants[MAX_PER_ENTRY*ENTRIES];
  30   int uni2index[0x10000];
  31   int index;
  32
  33   if (argc != 1)
  34     exit(1);
  35
  36   printf("/*\n");
  37   printf(" * Copyright (C) 1999-2002 Free Software Foundation, Inc.\n");
  38   printf(" * This file is part of the GNU LIBICONV Library.\n");
  39   printf(" *\n");
  40   printf(" * The GNU LIBICONV Library is free software; you can redistribute it\n");
  41   printf(" * and/or modify it under the terms of the GNU Lesser General Public\n");
  42   printf(" * License as published by the Free Software Foundation; either version 2\n");
  43   printf(" * of the License, or (at your option) any later version.\n");
  44   printf(" *\n");
  45   printf(" * The GNU LIBICONV Library is distributed in the hope that it will be\n");
  46   printf(" * useful, but WITHOUT ANY WARRANTY; without even the implied warranty of\n");
  47   printf(" * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU\n");
  48   printf(" * Lesser General Public License for more details.\n");
  49   printf(" *\n");
  50   printf(" * You should have received a copy of the GNU Lesser General Public\n");
  51   printf(" * License along with the GNU LIBICONV Library; see the file COPYING.LIB.\n");
  52   printf(" * If not, see <https://www.gnu.org/licenses/>.\n");
  53   printf(" */\n");
  54   printf("\n");
  55   printf("/*\n");
  56   printf(" * CJK variants table\n");
  57   printf(" */\n");
  58   printf("\n");
  59   {
  60     int c;
  61     int j;
  62     for (j = 0; j < 0x10000; j++)
  63       uni2index[j] = -1;
  64     index = 0;
  65     for (;;) {
  66       c = getc(stdin);
  67       if (c == EOF)
  68         break;
  69       if (c == '#') {
  70         do { c = getc(stdin); } while (!(c == EOF || c == '\n'));
  71         continue;
  72       }
  73       ungetc(c,stdin);
  74       if (scanf("%x",&j) != 1)
  75         exit(1);
  76       c = getc(stdin);
  77       if (c != '\t')
  78         exit(1);
  79       uni2index[j] = index;
  80       for (;;) {
  81         int i;
  82         if (scanf("%x",&i) != 1)
  83           exit(1);
  84         if (!(i >= 0x3000 && i < 0x3000+0x8000))
  85           exit(1);
  86         variants[index++] = i-0x3000;
  87         c = getc(stdin);
  88         if (c != ' ')
  89           break;
  90       }
  91       variants[index-1] |= 0x8000; /* end of list marker */
  92       if (c != '\n')
  93         exit(1);
  94     }
  95   }
  96   printf("static const unsigned short cjk_variants[%d] = {",index);
  97   {
  98     int i;
  99     for (i = 0; i < index; i++) {
 100       if ((i % 8) == 0)
 101         printf("\n ");
 102       printf(" 0x%04x,",variants[i]);
 103     }
 104     printf("\n};\n");
 105   }
 106   printf("\n");
 107   printf("static const short cjk_variants_indx[0x5200] = {\n");
 108   {
 109     int j;
 110     for (j = 0x4e00; j < 0xa000; j++) {
 111       if ((j % 0x100) == 0)
 112         printf("  /* 0x%04x */\n", j);
 113       if ((j % 8) == 0)
 114         printf(" ");
 115       printf(" %5d,",uni2index[j]);
 116       if ((j % 8) == 7)
 117         printf("\n");
 118     }
 119     printf("};\n");
 120   }
 121   printf("\n");
 122
 123   return 0;
 124 }