contrib/mdocml/chars.c

   1 /*      $Id: chars.c,v 1.58 2014/07/23 15:00:08 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
  59 void
  60 mchars_free(struct mchars *arg)
  61 {
  62
  63         free(arg->htab);
  64         free(arg);
  65 }
  66
  67 struct mchars *
  68 mchars_alloc(void)
  69 {
  70         struct mchars    *tab;
  71         struct ln       **htab;
  72         struct ln        *pp;
  73         int               i, hash;
  74
  75         /*
  76          * Constructs a very basic chaining hashtable.  The hash routine
  77          * is simply the integral value of the first character.
  78          * Subsequent entries are chained in the order they're processed.
  79          */
  80
  81         tab = mandoc_malloc(sizeof(struct mchars));
  82         htab = mandoc_calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln *));
  83
  84         for (i = 0; i < LINES_MAX; i++) {
  85                 hash = (int)lines[i].code[0] - PRINT_LO;
  86
  87                 if (NULL == (pp = htab[hash])) {
  88                         htab[hash] = &lines[i];
  89                         continue;
  90                 }
  91
  92                 for ( ; pp->next; pp = pp->next)
  93                         /* Scan ahead. */ ;
  94                 pp->next = &lines[i];
  95         }
  96
  97         tab->htab = htab;
  98         return(tab);
  99 }
 100
 101 int
 102 mchars_spec2cp(const struct mchars *arg, const char *p, size_t sz)
 103 {
 104         const struct ln *ln;
 105
 106         ln = find(arg, p, sz);
 107         if (NULL == ln)
 108                 return(-1);
 109         return(ln->unicode);
 110 }
 111
 112 char
 113 mchars_num2char(const char *p, size_t sz)
 114 {
 115         int       i;
 116
 117         if ((i = mandoc_strntoi(p, sz, 10)) < 0)
 118                 return('\0');
 119
 120         return(i > 0 && i < 256 && isprint(i) ? i : '\0');
 121 }
 122
 123 int
 124 mchars_num2uc(const char *p, size_t sz)
 125 {
 126         int      i;
 127
 128         if ((i = mandoc_strntoi(p, sz, 16)) < 0)
 129                 return('\0');
 130
 131         /*
 132          * Security warning:
 133          * Never extend the range of accepted characters
 134          * to overlap with the ASCII range, 0x00-0x7F
 135          * without re-auditing the callers of this function.
 136          * Some callers might relay on the fact that we never
 137          * return ASCII characters for their escaping decisions.
 138          *
 139          * XXX Code is missing here to exclude bogus ranges.
 140          */
 141
 142         return(i > 0x80 && i <= 0x10FFFF ? i : '\0');
 143 }
 144
 145 const char *
 146 mchars_spec2str(const struct mchars *arg,
 147                 const char *p, size_t sz, size_t *rsz)
 148 {
 149         const struct ln *ln;
 150
 151         ln = find(arg, p, sz);
 152         if (NULL == ln) {
 153                 *rsz = 1;
 154                 return(NULL);
 155         }
 156
 157         *rsz = strlen(ln->ascii);
 158         return(ln->ascii);
 159 }
 160
 161 static const struct ln *
 162 find(const struct mchars *tab, const char *p, size_t sz)
 163 {
 164         const struct ln  *pp;
 165         int               hash;
 166
 167         assert(p);
 168
 169         if (0 == sz || p[0] < PRINT_LO || p[0] > PRINT_HI)
 170                 return(NULL);
 171
 172         hash = (int)p[0] - PRINT_LO;
 173
 174         for (pp = tab->htab[hash]; pp; pp = pp->next)
 175                 if (0 == strncmp(pp->code, p, sz) &&
 176                     '\0' == pp->code[(int)sz])
 177                         return(pp);
 178
 179         return(NULL);
 180 }