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, version 3 of the License ONLY.
   7  *
   8  * This program is distributed in the hope that it will be useful,
   9  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  10  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  11  * GNU General Public License for more details.
  12  *
  13  * You should have received a copy of the GNU General Public License
  14  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  15  */
  16 #include <ctype.h>
  17 #include <stdio.h>
  18 #include <stdlib.h>
  19 #include <string.h>
  20 #include <unistd.h>
  21
  22
  23 static char *trim (char *str) {
  24   char *s;
  25   if (str == NULL) return NULL;
  26   for (s = str; *s && isspace(*s); ++s) ;
  27   if (s > str) memmove(str, s, strlen(s)+1);
  28   for (s = str+strlen(str)-1; s >= str && isspace(*s); --s) ;
  29   s[1] = 0;
  30   return str;
  31 }
  32
  33
  34 static int hex2num (char *s) {
  35   char *end;
  36   int res;
  37   trim(s);
  38   if (!s[0]) return -1;
  39   res = strtol(s, &end, 16);
  40   if (end[0] || res < 0) return -1;
  41   return res;
  42 }
  43
  44
  45 typedef struct {
  46   int code;
  47   const char *name;
  48   const char *class;
  49   int upper;
  50   int lower;
  51 } UCInfo;
  52
  53
  54 static char *stx = NULL;
  55
  56 static char *tok (void) {
  57   char *s = stx;
  58   for (; *stx && *stx != ';'; ++stx) ;
  59   if (*stx) *stx++ = 0;
  60   return s;
  61 }
  62
  63
  64 // <0: eof; 0: bad record; >0: good record
  65 static int read_record (FILE *fi, UCInfo *ui) {
  66   static char str[8192];
  67   char *tk, *u, *l;
  68   int f;
  69   if (fgets(str, sizeof(str)-1, fi) == NULL) return -1;
  70   stx = trim(str);
  71   if ((tk = tok()) == NULL) return 0;
  72   if ((ui->code = hex2num(tk)) < 0) return 0;
  73   if (ui->code > 65535) return -1;
  74   if ((ui->name = trim(tok())) == NULL) return 0;
  75   if ((ui->class = trim(tok())) == NULL) return 0;
  76   // skip unused fields
  77   for (f = 9; f > 0; --f) if (tok() == NULL) { printf("%d\n", f); return 0; }
  78   if ((u = trim(tok())) == NULL) return 0;
  79   if ((l = trim(tok())) == NULL) return 0;
  80   if (!u[0]) ui->upper = ui->code; else ui->upper = hex2num(u);
  81   if (!l[0]) ui->lower = ui->code; else ui->lower = hex2num(l);
  82   if (ui->upper < 0 || ui->lower < 0) return 0;
  83   if (ui->upper > 65535 || ui->lower > 65535) abort();
  84   return 1;
  85 }
  86
  87
  88 typedef struct {
  89   int code, l, u;
  90 } mm;
  91
  92 static mm map[65535];
  93
  94
  95 static int mm_cmp (const void *p0, const void *p1) {
  96   const mm *i0 = (const mm *)p0;
  97   const mm *i1 = (const mm *)p1;
  98   return (i0->code < i1->code ? -1 : (i0->code > i1->code ? 1 : 0));
  99 }
 100
 101
 102 int main (int argc, char *argv[]) {
 103   UCInfo ui;
 104   int rc, f, totalmap = 0;
 105   FILE *fi, *fo;
 106   if (argc != 3) { fprintf(stderr, "usage: %s unitable.c unitable.txt\n", argv[0]); exit(1); }
 107   if ((fi = fopen(argv[2], "r")) == NULL) { fprintf(stderr, "FATAL: can't open input file: '%s'\n", argv[2]); exit(1); }
 108   for (;;) {
 109     rc = read_record(fi, &ui);
 110     if (rc < 0) break;
 111     if (rc == 0) continue;
 112     if (ui.code < 128) continue;
 113     if (strcmp(ui.class, "Lu") != 0 && strcmp(ui.class, "Ll") != 0) continue;
 114     if (ui.upper == ui.code && ui.lower == ui.code) continue;
 115     for (f = 0; f < totalmap; ++f) {
 116       if (map[f].code == ui.code) { fprintf(stderr, "FATAL: duplicate entries in the map!\n"); exit(1); }
 117       if (map[f].code > ui.code) { fprintf(stderr, "FATAL: invalid entry order in the map!\n"); exit(1); }
 118     }
 119     if (totalmap > 65535) { fprintf(stderr, "FATAL: too many entries in the map!\n"); exit(1); }
 120     map[totalmap].code = ui.code;
 121     map[totalmap].l = ui.lower;
 122     map[totalmap].u = ui.upper;
 123     ++totalmap;
 124   }
 125   fclose(fi);
 126   qsort(map, totalmap, sizeof(map[0]), mm_cmp);
 127   fo = fopen(argv[1], "w");
 128   if (fo == NULL) { fprintf(stderr, "FATAL: can't create output file: '%s'\n", argv[1]); exit(1); }
 129   fprintf(fo, "static const struct casemap unicode_case_mapping[%d] = {\n", totalmap);
 130   for (f = 0; f < totalmap; ++f) fprintf(fo, "{0x%04x,0x%04x,0x%04x},\n", map[f].code, map[f].l, map[f].u);
 131   fprintf(fo, "%s\n", "};");
 132   fclose(fo);
 133   return 0;
 134 }