libguile/hash.c

   1 /*      Copyright (C) 1995,1996,1997, 2000, 2001 Free Software Foundation, Inc.
   2  *
   3  * This program is free software; you can redistribute it and/or modify
   4  * it under the terms of the GNU General Public License as published by
   5  * the Free Software Foundation; either version 2, or (at your option)
   6  * any later version.
   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 software; see the file COPYING.  If not, write to
  15  * the Free Software Foundation, Inc., 59 Temple Place, Suite 330,
  16  * Boston, MA 02111-1307 USA
  17  *
  18  * As a special exception, the Free Software Foundation gives permission
  19  * for additional uses of the text contained in its release of GUILE.
  20  *
  21  * The exception is that, if you link the GUILE library with other files
  22  * to produce an executable, this does not by itself cause the
  23  * resulting executable to be covered by the GNU General Public License.
  24  * Your use of that executable is in no way restricted on account of
  25  * linking the GUILE library code into it.
  26  *
  27  * This exception does not however invalidate any other reasons why
  28  * the executable file might be covered by the GNU General Public License.
  29  *
  30  * This exception applies only to the code released by the
  31  * Free Software Foundation under the name GUILE.  If you copy
  32  * code from other Free Software Foundation releases into a copy of
  33  * GUILE, as the General Public License permits, the exception does
  34  * not apply to the code that you add in this way.  To avoid misleading
  35  * anyone as to the status of such modified files, you must delete
  36  * this exception notice from them.
  37  *
  38  * If you write modifications of your own for GUILE, it is your choice
  39  * whether to permit this exception to apply to your modifications.
  40  * If you do not wish that, delete this exception notice.  */
  41
  42
  43 \f
  44
  45 #include "libguile/_scm.h"
  46 #include "libguile/chars.h"
  47 #include "libguile/ports.h"
  48 #include "libguile/strings.h"
  49 #include "libguile/symbols.h"
  50 #include "libguile/vectors.h"
  51
  52 #include "libguile/validate.h"
  53 #include "libguile/hash.h"
  54 \f
  55
  56 #ifndef floor
  57 extern double floor();
  58 #endif
  59
  60
  61 unsigned long
  62 scm_string_hash (const unsigned char *str, size_t len)
  63 {
  64   if (len > 5)
  65     {
  66       size_t i = 5;
  67       unsigned long h = 264;
  68       while (i--)
  69         h = (h << 8) + (unsigned) str[h % len];
  70       return h;
  71     }
  72   else
  73     {
  74       size_t i = len;
  75       unsigned long h = 0;
  76       while (i)
  77         h = (h << 8) + (unsigned) str[--i];
  78       return h;
  79     }
  80 }
  81
  82
  83 /* Dirk:FIXME:: why downcase for characters? (2x: scm_hasher, scm_ihashv) */
  84 /* Dirk:FIXME:: scm_hasher could be made static. */
  85
  86
  87 unsigned long
  88 scm_hasher(SCM obj, unsigned long n, size_t d)
  89 {
  90   switch (SCM_ITAG3 (obj)) {
  91   case scm_tc3_int_1:
  92   case scm_tc3_int_2:
  93     return SCM_INUM(obj) % n;   /* SCM_INUMP(obj) */
  94   case scm_tc3_imm24:
  95     if (SCM_CHARP(obj))
  96       return (unsigned)(scm_downcase(SCM_CHAR(obj))) % n;
  97     switch (SCM_UNPACK (obj)) {
  98 #ifndef SICP
  99     case SCM_UNPACK(SCM_EOL):
 100       d = 256;
 101       break;
 102 #endif
 103     case SCM_UNPACK(SCM_BOOL_T):
 104       d = 257;
 105       break;
 106     case SCM_UNPACK(SCM_BOOL_F):
 107       d = 258;
 108       break;
 109     case SCM_UNPACK(SCM_EOF_VAL):
 110       d = 259;
 111       break;
 112     default:
 113       d = 263;          /* perhaps should be error */
 114     }
 115     return d % n;
 116   default:
 117     return 263 % n;     /* perhaps should be error */
 118   case scm_tc3_cons:
 119     switch SCM_TYP7(obj) {
 120     default:
 121       return 263 % n;
 122     case scm_tc7_smob:
 123       switch SCM_TYP16 (obj) {
 124       case scm_tc16_big:
 125         return SCM_INUM (scm_modulo (obj, SCM_MAKINUM (n)));
 126       default:
 127         return 263 % n;
 128       case scm_tc16_real:
 129         {
 130           double r = SCM_REAL_VALUE (obj);
 131           if (floor (r) == r) {
 132             obj = scm_inexact_to_exact (obj);
 133             if SCM_IMP (obj) return SCM_INUM (obj) % n;
 134             return SCM_INUM (scm_modulo (obj, SCM_MAKINUM (n)));
 135           }
 136         }
 137       case scm_tc16_complex:
 138         obj = scm_number_to_string (obj, SCM_MAKINUM (10));
 139       }
 140     case scm_tc7_string:
 141       return scm_string_hash (SCM_STRING_UCHARS (obj), SCM_STRING_LENGTH (obj)) % n;
 142     case scm_tc7_symbol:
 143       return SCM_SYMBOL_HASH (obj) % n;
 144     case scm_tc7_wvect:
 145     case scm_tc7_vector:
 146       {
 147         size_t len = SCM_VECTOR_LENGTH(obj);
 148         SCM const *data = SCM_VELTS(obj);
 149         if (len > 5)
 150           {
 151             size_t i = d/2;
 152             unsigned long h = 1;
 153             while (i--) h = ((h << 8) + (scm_hasher (data[h % len], n, 2))) % n;
 154             return h;
 155           }
 156         else
 157           {
 158             size_t i = len;
 159             unsigned long h = (n)-1;
 160             while (i--) h = ((h << 8) + (scm_hasher (data[i], n, d/len))) % n;
 161             return h;
 162           }
 163       }
 164     case scm_tcs_cons_imcar:
 165     case scm_tcs_cons_nimcar:
 166       if (d) return (scm_hasher (SCM_CAR (obj), n, d/2)
 167                      + scm_hasher (SCM_CDR (obj), n, d/2)) % n;
 168       else return 1;
 169     case scm_tc7_port:
 170       return ((SCM_RDNG & SCM_CELL_WORD_0 (obj)) ? 260 : 261) % n;
 171     case scm_tcs_closures:
 172     case scm_tcs_subrs:
 173       return 262 % n;
 174     }
 175   }
 176 }
 177
 178
 179 \f
 180
 181
 182 unsigned long
 183 scm_ihashq (SCM obj, unsigned long n)
 184 {
 185   return (SCM_UNPACK (obj) >> 1) % n;
 186 }
 187
 188
 189 SCM_DEFINE (scm_hashq, "hashq", 2, 0, 0,
 190            (SCM key, SCM size),
 191             "Determine a hash value for @var{key} that is suitable for\n"
 192             "lookups in a hashtable of size @var{size}, where @code{eq?} is\n"
 193             "used as the equality predicate.  The function returns an\n"
 194             "integer in the range 0 to @var{size} - 1.  Note that\n"
 195             "@code{hashq} may use internal addresses.  Thus two calls to\n"
 196             "hashq where the keys are @code{eq?} are not guaranteed to\n"
 197             "deliver the same value if the key object gets garbage collected\n"
 198             "in between.  This can happen, for example with symbols:\n"
 199             "@code{(hashq 'foo n) (gc) (hashq 'foo n)} may produce two\n"
 200             "different values, since @code{foo} will be garbage collected.")
 201 #define FUNC_NAME s_scm_hashq
 202 {
 203   SCM_VALIDATE_INUM_MIN (2, size, 0);
 204   return SCM_MAKINUM (scm_ihashq (key, SCM_INUM (size)));
 205 }
 206 #undef FUNC_NAME
 207
 208
 209 \f
 210
 211
 212 unsigned long
 213 scm_ihashv (SCM obj, unsigned long n)
 214 {
 215   if (SCM_CHARP(obj))
 216     return ((unsigned long) (scm_downcase (SCM_CHAR (obj)))) % n; /* downcase!?!! */
 217
 218   if (SCM_NUMP(obj))
 219     return (unsigned long) scm_hasher(obj, n, 10);
 220   else
 221     return SCM_UNPACK (obj) % n;
 222 }
 223
 224
 225 SCM_DEFINE (scm_hashv, "hashv", 2, 0, 0,
 226            (SCM key, SCM size),
 227             "Determine a hash value for @var{key} that is suitable for\n"
 228             "lookups in a hashtable of size @var{size}, where @code{eqv?} is\n"
 229             "used as the equality predicate.  The function returns an\n"
 230             "integer in the range 0 to @var{size} - 1.  Note that\n"
 231             "@code{(hashv key)} may use internal addresses.  Thus two calls\n"
 232             "to hashv where the keys are @code{eqv?} are not guaranteed to\n"
 233             "deliver the same value if the key object gets garbage collected\n"
 234             "in between.  This can happen, for example with symbols:\n"
 235             "@code{(hashv 'foo n) (gc) (hashv 'foo n)} may produce two\n"
 236             "different values, since @code{foo} will be garbage collected.")
 237 #define FUNC_NAME s_scm_hashv
 238 {
 239   SCM_VALIDATE_INUM_MIN (2, size, 0);
 240   return SCM_MAKINUM (scm_ihashv (key, SCM_INUM (size)));
 241 }
 242 #undef FUNC_NAME
 243
 244
 245 \f
 246
 247
 248 unsigned long
 249 scm_ihash (SCM obj, unsigned long n)
 250 {
 251   return (unsigned long) scm_hasher (obj, n, 10);
 252 }
 253
 254 SCM_DEFINE (scm_hash, "hash", 2, 0, 0,
 255            (SCM key, SCM size),
 256             "Determine a hash value for @var{key} that is suitable for\n"
 257             "lookups in a hashtable of size @var{size}, where @code{equal?}\n"
 258             "is used as the equality predicate.  The function returns an\n"
 259             "integer in the range 0 to @var{size} - 1.")
 260 #define FUNC_NAME s_scm_hash
 261 {
 262   SCM_VALIDATE_INUM_MIN (2, size, 0);
 263   return SCM_MAKINUM (scm_ihash (key, SCM_INUM (size)));
 264 }
 265 #undef FUNC_NAME
 266
 267
 268 \f
 269
 270
 271 void
 272 scm_init_hash ()
 273 {
 274 #include "libguile/hash.x"
 275 }
 276
 277
 278 /*
 279   Local Variables:
 280   c-file-style: "gnu"
 281   End:
 282 */