contrib/mdocml/man_hash.c

   1 /*      $Id: man_hash.c,v 1.27 2014/04/20 16:46:04 schwarze Exp $ */
   2 /*
   3  * Copyright (c) 2008, 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
   4  *
   5  * Permission to use, copy, modify, and distribute this software for any
   6  * purpose with or without fee is hereby granted, provided that the above
   7  * copyright notice and this permission notice appear in all copies.
   8  *
   9  * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
  10  * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
  11  * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
  12  * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
  13  * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
  14  * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
  15  * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
  16  */
  17 #ifdef HAVE_CONFIG_H
  18 #include "config.h"
  19 #endif
  20
  21 #include <sys/types.h>
  22
  23 #include <assert.h>
  24 #include <ctype.h>
  25 #include <limits.h>
  26 #include <string.h>
  27
  28 #include "man.h"
  29 #include "mandoc.h"
  30 #include "libman.h"
  31
  32 #define HASH_DEPTH       6
  33
  34 #define HASH_ROW(x) do { \
  35                 if (isupper((unsigned char)(x))) \
  36                         (x) -= 65; \
  37                 else \
  38                         (x) -= 97; \
  39                 (x) *= HASH_DEPTH; \
  40         } while (/* CONSTCOND */ 0)
  41
  42 /*
  43  * Lookup table is indexed first by lower-case first letter (plus one
  44  * for the period, which is stored in the last row), then by lower or
  45  * uppercase second letter.  Buckets correspond to the index of the
  46  * macro (the integer value of the enum stored as a char to save a bit
  47  * of space).
  48  */
  49 static  unsigned char    table[26 * HASH_DEPTH];
  50
  51
  52 /*
  53  * XXX - this hash has global scope, so if intended for use as a library
  54  * with multiple callers, it will need re-invocation protection.
  55  */
  56 void
  57 man_hash_init(void)
  58 {
  59         int              i, j, x;
  60
  61         memset(table, UCHAR_MAX, sizeof(table));
  62
  63         assert(MAN_MAX < UCHAR_MAX);
  64
  65         for (i = 0; i < (int)MAN_MAX; i++) {
  66                 x = man_macronames[i][0];
  67
  68                 assert(isalpha((unsigned char)x));
  69
  70                 HASH_ROW(x);
  71
  72                 for (j = 0; j < HASH_DEPTH; j++)
  73                         if (UCHAR_MAX == table[x + j]) {
  74                                 table[x + j] = (unsigned char)i;
  75                                 break;
  76                         }
  77
  78                 assert(j < HASH_DEPTH);
  79         }
  80 }
  81
  82 enum mant
  83 man_hash_find(const char *tmp)
  84 {
  85         int              x, y, i;
  86         enum mant        tok;
  87
  88         if ('\0' == (x = tmp[0]))
  89                 return(MAN_MAX);
  90         if ( ! (isalpha((unsigned char)x)))
  91                 return(MAN_MAX);
  92
  93         HASH_ROW(x);
  94
  95         for (i = 0; i < HASH_DEPTH; i++) {
  96                 if (UCHAR_MAX == (y = table[x + i]))
  97                         return(MAN_MAX);
  98
  99                 tok = (enum mant)y;
 100                 if (0 == strcmp(tmp, man_macronames[tok]))
 101                         return(tok);
 102         }
 103
 104         return(MAN_MAX);
 105 }