src/unigen.c

   1 /* coded by Ketmar // Vampire Avalon (psyc://ketmar.no-ip.org/~Ketmar)
   2  * Understanding is not required. Only obedience.
   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 <http://www.gnu.org/licenses/>.
  16  */
  17 #include <ctype.h>
  18 #include <stdio.h>
  19 #include <stdlib.h>
  20 #include <string.h>
  21 #include <unistd.h>
  22
  23
  24 static char *trim (char *str) {
  25   char *s;
  26   if (str == NULL) return NULL;
  27   for (s = str; *s && isspace(*s); ++s) ;
  28   if (s > str) memmove(str, s, strlen(s)+1);
  29   for (s = str+strlen(str)-1; s >= str && isspace(*s); --s) ;
  30   s[1] = 0;
  31   return str;
  32 }
  33
  34
  35 static int hex2num (char *s) {
  36   char *end;
  37   int res;
  38   trim(s);
  39   if (!s[0]) return -1;
  40   res = strtol(s, &end, 16);
  41   if (end[0] || res < 0) return -1;
  42   return res;
  43 }
  44
  45
  46 typedef struct {
  47   int code;
  48   const char *name;
  49   const char *class;
  50   int upper;
  51   int lower;
  52 } UCInfo;
  53
  54
  55 static char *stx = NULL;
  56
  57 static char *tok (void) {
  58   char *s = stx;
  59   for (; *stx && *stx != ';'; ++stx) ;
  60   if (*stx) *stx++ = 0;
  61   return s;
  62 }
  63
  64
  65 // <0: eof; 0: bad record; >0: good record
  66 static int read_record (FILE *fi, UCInfo *ui) {
  67   static char str[8192];
  68   char *tk, *u, *l;
  69   int f;
  70   if (fgets(str, sizeof(str)-1, fi) == NULL) return -1;
  71   stx = trim(str);
  72   if ((tk = tok()) == NULL) return 0;
  73   if ((ui->code = hex2num(tk)) < 0) return 0;
  74   if (ui->code > 65535) return -1;
  75   if ((ui->name = trim(tok())) == NULL) return 0;
  76   if ((ui->class = trim(tok())) == NULL) return 0;
  77   // skip unused fields
  78   for (f = 9; f > 0; --f) if (tok() == NULL) { printf("%d\n", f); return 0; }
  79   if ((u = trim(tok())) == NULL) return 0;
  80   if ((l = trim(tok())) == NULL) return 0;
  81   if (!u[0]) ui->upper = ui->code; else ui->upper = hex2num(u);
  82   if (!l[0]) ui->lower = ui->code; else ui->lower = hex2num(l);
  83   if (ui->upper < 0 || ui->lower < 0) return 0;
  84   if (ui->upper > 65535 || ui->lower > 65535) abort();
  85   return 1;
  86 }
  87
  88
  89 typedef struct {
  90   int code, l, u;
  91 } mm;
  92
  93 static mm map[65535];
  94
  95
  96 static int mm_cmp (const void *p0, const void *p1) {
  97   const mm *i0 = (const mm *)p0;
  98   const mm *i1 = (const mm *)p1;
  99   return (i0->code < i1->code ? -1 : (i0->code > i1->code ? 1 : 0));
 100 }
 101
 102
 103 int main (int argc, char *argv[]) {
 104   UCInfo ui;
 105   int rc, f, totalmap = 0;
 106   FILE *fi, *fo;
 107   if (argc != 3) { fprintf(stderr, "usage: %s unitable.c unitable.txt\n", argv[0]); exit(1); }
 108   if ((fi = fopen(argv[2], "r")) == NULL) { fprintf(stderr, "FATAL: can't open input file: '%s'\n", argv[2]); exit(1); }
 109   for (;;) {
 110     rc = read_record(fi, &ui);
 111     if (rc < 0) break;
 112     if (rc == 0) continue;
 113     if (ui.code < 128) continue;
 114     if (strcmp(ui.class, "Lu") != 0 && strcmp(ui.class, "Ll") != 0) continue;
 115     if (ui.upper == ui.code && ui.lower == ui.code) continue;
 116     for (f = 0; f < totalmap; ++f) {
 117       if (map[f].code == ui.code) { fprintf(stderr, "FATAL: duplicate entries in the map!\n"); exit(1); }
 118       if (map[f].code > ui.code) { fprintf(stderr, "FATAL: invalid entry order in the map!\n"); exit(1); }
 119     }
 120     if (totalmap > 65535) { fprintf(stderr, "FATAL: too many entries in the map!\n"); exit(1); }
 121     map[totalmap].code = ui.code;
 122     map[totalmap].l = ui.lower;
 123     map[totalmap].u = ui.upper;
 124     ++totalmap;
 125   }
 126   fclose(fi);
 127   qsort(map, totalmap, sizeof(map[0]), mm_cmp);
 128   fo = fopen(argv[1], "w");
 129   if (fo == NULL) { fprintf(stderr, "FATAL: can't create output file: '%s'\n", argv[1]); exit(1); }
 130   fprintf(fo, "static const struct casemap unicode_case_mapping[%d] = {\n", totalmap);
 131   for (f = 0; f < totalmap; ++f) fprintf(fo, "{0x%04x,0x%04x,0x%04x},\n", map[f].code, map[f].l, map[f].u);
 132   fprintf(fo, "%s\n", "};");
 133   fclose(fo);
 134   return 0;
 135 }