1 /*      $Id: man_hash.c,v 1.25 2011/07/24 18:15:14 kristaps 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 <stdlib.h>
  27 #include <string.h>
  28 
  29 #include "man.h"
  30 #include "mandoc.h"
  31 #include "libman.h"
  32 
  33 #define HASH_DEPTH       6
  34 
  35 #define HASH_ROW(x) do { \
  36                 if (isupper((unsigned char)(x))) \
  37                         (x) -= 65; \
  38                 else \
  39                         (x) -= 97; \
  40                 (x) *= HASH_DEPTH; \
  41         } while (/* CONSTCOND */ 0)
  42 
  43 /*
  44  * Lookup table is indexed first by lower-case first letter (plus one
  45  * for the period, which is stored in the last row), then by lower or
  46  * uppercase second letter.  Buckets correspond to the index of the
  47  * macro (the integer value of the enum stored as a char to save a bit
  48  * of space).
  49  */
  50 static  unsigned char    table[26 * HASH_DEPTH];
  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(/* LINTED */ 
  64                         MAN_MAX < UCHAR_MAX);
  65 
  66         for (i = 0; i < (int)MAN_MAX; i++) {
  67                 x = man_macronames[i][0];
  68 
  69                 assert(isalpha((unsigned char)x));
  70 
  71                 HASH_ROW(x);
  72 
  73                 for (j = 0; j < HASH_DEPTH; j++)
  74                         if (UCHAR_MAX == table[x + j]) {
  75                                 table[x + j] = (unsigned char)i;
  76                                 break;
  77                         }
  78 
  79                 assert(j < HASH_DEPTH);
  80         }
  81 }
  82 
  83 
  84 enum mant
  85 man_hash_find(const char *tmp)
  86 {
  87         int              x, y, i;
  88         enum mant        tok;
  89 
  90         if ('\0' == (x = tmp[0]))
  91                 return(MAN_MAX);
  92         if ( ! (isalpha((unsigned char)x)))
  93                 return(MAN_MAX);
  94 
  95         HASH_ROW(x);
  96 
  97         for (i = 0; i < HASH_DEPTH; i++) {
  98                 if (UCHAR_MAX == (y = table[x + i]))
  99                         return(MAN_MAX);
 100 
 101                 tok = (enum mant)y;
 102                 if (0 == strcmp(tmp, man_macronames[tok]))
 103                         return(tok);
 104         }
 105 
 106         return(MAN_MAX);
 107 }