chars.c

   1 /*      $Id: chars.c,v 1.56 2014/03/23 11:25:25 schwarze Exp $ */
   2 /*
   3  * Copyright (c) 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
   4  * Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org>
   5  *
   6  * Permission to use, copy, modify, and distribute this software for any
   7  * purpose with or without fee is hereby granted, provided that the above
   8  * copyright notice and this permission notice appear in all copies.
   9  *
  10  * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
  11  * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
  12  * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
  13  * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
  14  * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
  15  * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
  16  * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
  17  */
  18 #ifdef HAVE_CONFIG_H
  19 #include "config.h"
  20 #endif
  21
  22 #include <assert.h>
  23 #include <ctype.h>
  24 #include <stdlib.h>
  25 #include <string.h>
  26
  27 #include "mandoc.h"
  28 #include "mandoc_aux.h"
  29 #include "libmandoc.h"
  30
  31 #define PRINT_HI         126
  32 #define PRINT_LO         32
  33
  34 struct  ln {
  35         struct ln        *next;
  36         const char       *code;
  37         const char       *ascii;
  38         int               unicode;
  39 };
  40
  41 #define LINES_MAX         330
  42
  43 #define CHAR(in, ch, code) \
  44         { NULL, (in), (ch), (code) },
  45
  46 #define CHAR_TBL_START    static struct ln lines[LINES_MAX] = {
  47 #define CHAR_TBL_END      };
  48
  49 #include "chars.in"
  50
  51 struct  mchars {
  52         struct ln       **htab;
  53 };
  54
  55 static  const struct ln  *find(const struct mchars *,
  56                                 const char *, size_t);
  57
  58 void
  59 mchars_free(struct mchars *arg)
  60 {
  61
  62         free(arg->htab);
  63         free(arg);
  64 }
  65
  66 struct mchars *
  67 mchars_alloc(void)
  68 {
  69         struct mchars    *tab;
  70         struct ln       **htab;
  71         struct ln        *pp;
  72         int               i, hash;
  73
  74         /*
  75          * Constructs a very basic chaining hashtable.  The hash routine
  76          * is simply the integral value of the first character.
  77          * Subsequent entries are chained in the order they're processed.
  78          */
  79
  80         tab = mandoc_malloc(sizeof(struct mchars));
  81         htab = mandoc_calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln *));
  82
  83         for (i = 0; i < LINES_MAX; i++) {
  84                 hash = (int)lines[i].code[0] - PRINT_LO;
  85
  86                 if (NULL == (pp = htab[hash])) {
  87                         htab[hash] = &lines[i];
  88                         continue;
  89                 }
  90
  91                 for ( ; pp->next; pp = pp->next)
  92                         /* Scan ahead. */ ;
  93                 pp->next = &lines[i];
  94         }
  95
  96         tab->htab = htab;
  97         return(tab);
  98 }
  99
 100 int
 101 mchars_spec2cp(const struct mchars *arg, const char *p, size_t sz)
 102 {
 103         const struct ln *ln;
 104
 105         ln = find(arg, p, sz);
 106         if (NULL == ln)
 107                 return(-1);
 108         return(ln->unicode);
 109 }
 110
 111 char
 112 mchars_num2char(const char *p, size_t sz)
 113 {
 114         int               i;
 115
 116         if ((i = mandoc_strntoi(p, sz, 10)) < 0)
 117                 return('\0');
 118         return(i > 0 && i < 256 && isprint(i) ?
 119                         /* LINTED */ i : '\0');
 120 }
 121
 122 int
 123 mchars_num2uc(const char *p, size_t sz)
 124 {
 125         int               i;
 126
 127         if ((i = mandoc_strntoi(p, sz, 16)) < 0)
 128                 return('\0');
 129         /* FIXME: make sure we're not in a bogus range. */
 130         return(i > 0x80 && i <= 0x10FFFF ? i : '\0');
 131 }
 132
 133 const char *
 134 mchars_spec2str(const struct mchars *arg,
 135                 const char *p, size_t sz, size_t *rsz)
 136 {
 137         const struct ln *ln;
 138
 139         ln = find(arg, p, sz);
 140         if (NULL == ln) {
 141                 *rsz = 1;
 142                 return(NULL);
 143         }
 144
 145         *rsz = strlen(ln->ascii);
 146         return(ln->ascii);
 147 }
 148
 149 static const struct ln *
 150 find(const struct mchars *tab, const char *p, size_t sz)
 151 {
 152         const struct ln  *pp;
 153         int               hash;
 154
 155         assert(p);
 156
 157         if (0 == sz || p[0] < PRINT_LO || p[0] > PRINT_HI)
 158                 return(NULL);
 159
 160         hash = (int)p[0] - PRINT_LO;
 161
 162         for (pp = tab->htab[hash]; pp; pp = pp->next)
 163                 if (0 == strncmp(pp->code, p, sz) &&
 164                                 '\0' == pp->code[(int)sz])
 165                         return(pp);
 166
 167         return(NULL);
 168 }