libguile/symbols.c

   1 /* Copyright (C) 1995,1996,1997,1998,2000,2001, 2003, 2004 Free Software Foundation, Inc.
   2  *
   3  * This library is free software; you can redistribute it and/or
   4  * modify it under the terms of the GNU Lesser General Public
   5  * License as published by the Free Software Foundation; either
   6  * version 2.1 of the License, or (at your option) any later version.
   7  *
   8  * This library 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 GNU
  11  * Lesser General Public License for more details.
  12  *
  13  * You should have received a copy of the GNU Lesser General Public
  14  * License along with this library; if not, write to the Free Software
  15  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
  16  */
  17
  18
  19 \f
  20 #if HAVE_CONFIG_H
  21 #  include <config.h>
  22 #endif
  23
  24 #include "libguile/_scm.h"
  25 #include "libguile/chars.h"
  26 #include "libguile/eval.h"
  27 #include "libguile/hash.h"
  28 #include "libguile/smob.h"
  29 #include "libguile/variable.h"
  30 #include "libguile/alist.h"
  31 #include "libguile/fluids.h"
  32 #include "libguile/strings.h"
  33 #include "libguile/vectors.h"
  34 #include "libguile/hashtab.h"
  35 #include "libguile/weaks.h"
  36 #include "libguile/modules.h"
  37 #include "libguile/read.h"
  38 #include "libguile/srfi-13.h"
  39
  40 #include "libguile/validate.h"
  41 #include "libguile/symbols.h"
  42
  43 #ifdef HAVE_STRING_H
  44 #include <string.h>
  45 #endif
  46
  47 \f
  48
  49 static SCM symbols;
  50
  51 #ifdef GUILE_DEBUG
  52 SCM_DEFINE (scm_sys_symbols, "%symbols", 0, 0, 0,
  53             (),
  54             "Return the system symbol obarray.")
  55 #define FUNC_NAME s_scm_sys_symbols
  56 {
  57   return symbols;
  58 }
  59 #undef FUNC_NAME
  60 #endif
  61
  62 \f
  63
  64 /* {Symbols}
  65  */
  66
  67 /* In order to optimize reading speed, this function breaks part of
  68  * the hashtable abstraction.  The optimizations are:
  69  *
  70  * 1. The argument string can be compared directly to symbol objects
  71  *    without first creating an SCM string object.  (This would have
  72  *    been necessary if we had used the hashtable API in hashtab.h.)
  73  *
  74  * 2. We can use the raw hash value stored in scm_i_symbol_hash (sym)
  75  *    to speed up lookup.
  76  *
  77  * Both optimizations might be possible without breaking the
  78  * abstraction if the API in hashtab.c is improved.
  79  */
  80
  81 unsigned long
  82 scm_i_hash_symbol (SCM obj, unsigned long n, void *closure)
  83 {
  84   return scm_i_symbol_hash (obj) % n;
  85 }
  86
  87 static SCM
  88 scm_i_mem2symbol (SCM str)
  89 {
  90   const char *name = scm_i_string_chars (str);
  91   size_t len = scm_i_string_length (str);
  92
  93   size_t raw_hash = scm_string_hash ((const unsigned char *) name, len) / 2;
  94   size_t hash = raw_hash % SCM_HASHTABLE_N_BUCKETS (symbols);
  95
  96   {
  97     /* Try to find the symbol in the symbols table */
  98
  99     SCM l;
 100
 101     for (l = SCM_HASHTABLE_BUCKETS (symbols) [hash];
 102          !SCM_NULLP (l);
 103          l = SCM_CDR (l))
 104       {
 105         SCM sym = SCM_CAAR (l);
 106         if (scm_i_symbol_hash (sym) == raw_hash
 107             && scm_i_symbol_length (sym) == len)
 108           {
 109             const char *chrs = scm_i_symbol_chars (sym);
 110             size_t i = len;
 111
 112             while (i != 0)
 113               {
 114                 --i;
 115                 if (name[i] != chrs[i])
 116                   goto next_symbol;
 117               }
 118
 119             return sym;
 120           }
 121       next_symbol:
 122         ;
 123       }
 124   }
 125
 126   {
 127     /* The symbol was not found - create it. */
 128     SCM symbol = scm_i_make_symbol (str, raw_hash,
 129                                     scm_cons (SCM_BOOL_F, SCM_EOL));
 130
 131     SCM slot = SCM_HASHTABLE_BUCKETS (symbols) [hash];
 132     SCM cell = scm_cons (symbol, SCM_UNDEFINED);
 133     SCM_SET_HASHTABLE_BUCKET (symbols, hash, scm_cons (cell, slot));
 134     SCM_HASHTABLE_INCREMENT (symbols);
 135     if (SCM_HASHTABLE_N_ITEMS (symbols) > SCM_HASHTABLE_UPPER (symbols))
 136       scm_i_rehash (symbols, scm_i_hash_symbol, 0, "scm_mem2symbol");
 137
 138     return symbol;
 139   }
 140 }
 141
 142 static SCM
 143 scm_i_mem2uninterned_symbol (SCM str)
 144 {
 145   const char *name = scm_i_string_chars (str);
 146   size_t len = scm_i_string_length (str);
 147
 148   size_t raw_hash = (scm_string_hash ((const unsigned char *) name, len)/2
 149                      + SCM_T_BITS_MAX/2 + 1);
 150
 151   return scm_i_make_symbol (str, raw_hash,
 152                             scm_cons (SCM_BOOL_F, SCM_EOL));
 153 }
 154
 155 SCM_DEFINE (scm_symbol_p, "symbol?", 1, 0, 0,
 156             (SCM obj),
 157             "Return @code{#t} if @var{obj} is a symbol, otherwise return\n"
 158             "@code{#f}.")
 159 #define FUNC_NAME s_scm_symbol_p
 160 {
 161   return scm_from_bool (scm_is_symbol (obj));
 162 }
 163 #undef FUNC_NAME
 164
 165 SCM_DEFINE (scm_symbol_interned_p, "symbol-interned?", 1, 0, 0,
 166             (SCM symbol),
 167             "Return @code{#t} if @var{symbol} is interned, otherwise return\n"
 168             "@code{#f}.")
 169 #define FUNC_NAME s_scm_symbol_interned_p
 170 {
 171   SCM_VALIDATE_SYMBOL (1, symbol);
 172   return scm_from_bool (scm_i_symbol_is_interned (symbol));
 173 }
 174 #undef FUNC_NAME
 175
 176 SCM_DEFINE (scm_make_symbol, "make-symbol", 1, 0, 0,
 177             (SCM name),
 178             "Return a new uninterned symbol with the name @var{name}.  "
 179             "The returned symbol is guaranteed to be unique and future "
 180             "calls to @code{string->symbol} will not return it.")
 181 #define FUNC_NAME s_scm_make_symbol
 182 {
 183   SCM_VALIDATE_STRING (1, name);
 184   return scm_i_mem2uninterned_symbol (name);
 185 }
 186 #undef FUNC_NAME
 187
 188 SCM_DEFINE (scm_symbol_to_string, "symbol->string", 1, 0, 0,
 189            (SCM s),
 190             "Return the name of @var{symbol} as a string.  If the symbol was\n"
 191             "part of an object returned as the value of a literal expression\n"
 192             "(section @pxref{Literal expressions,,,r5rs, The Revised^5\n"
 193             "Report on Scheme}) or by a call to the @code{read} procedure,\n"
 194             "and its name contains alphabetic characters, then the string\n"
 195             "returned will contain characters in the implementation's\n"
 196             "preferred standard case---some implementations will prefer\n"
 197             "upper case, others lower case.  If the symbol was returned by\n"
 198             "@code{string->symbol}, the case of characters in the string\n"
 199             "returned will be the same as the case in the string that was\n"
 200             "passed to @code{string->symbol}.  It is an error to apply\n"
 201             "mutation procedures like @code{string-set!} to strings returned\n"
 202             "by this procedure.\n"
 203             "\n"
 204             "The following examples assume that the implementation's\n"
 205             "standard case is lower case:\n"
 206             "\n"
 207             "@lisp\n"
 208             "(symbol->string 'flying-fish)    @result{} \"flying-fish\"\n"
 209             "(symbol->string 'Martin)         @result{}  \"martin\"\n"
 210             "(symbol->string\n"
 211             "   (string->symbol \"Malvina\")) @result{} \"Malvina\"\n"
 212             "@end lisp")
 213 #define FUNC_NAME s_scm_symbol_to_string
 214 {
 215   SCM_VALIDATE_SYMBOL (1, s);
 216   return scm_i_symbol_substring (s, 0, scm_i_symbol_length (s));
 217 }
 218 #undef FUNC_NAME
 219
 220
 221 SCM_DEFINE (scm_string_to_symbol, "string->symbol", 1, 0, 0,
 222             (SCM string),
 223             "Return the symbol whose name is @var{string}. This procedure\n"
 224             "can create symbols with names containing special characters or\n"
 225             "letters in the non-standard case, but it is usually a bad idea\n"
 226             "to create such symbols because in some implementations of\n"
 227             "Scheme they cannot be read as themselves.  See\n"
 228             "@code{symbol->string}.\n"
 229             "\n"
 230             "The following examples assume that the implementation's\n"
 231             "standard case is lower case:\n"
 232             "\n"
 233             "@lisp\n"
 234             "(eq? 'mISSISSIppi 'mississippi) @result{} #t\n"
 235             "(string->symbol \"mISSISSIppi\") @result{} @r{the symbol with name \"mISSISSIppi\"}\n"
 236             "(eq? 'bitBlt (string->symbol \"bitBlt\")) @result{} #f\n"
 237             "(eq? 'JollyWog\n"
 238             "  (string->symbol (symbol->string 'JollyWog))) @result{} #t\n"
 239             "(string=? \"K. Harper, M.D.\"\n"
 240             "  (symbol->string\n"
 241             "    (string->symbol \"K. Harper, M.D.\"))) @result{}#t\n"
 242             "@end lisp")
 243 #define FUNC_NAME s_scm_string_to_symbol
 244 {
 245   SCM_VALIDATE_STRING (1, string);
 246   return scm_i_mem2symbol (string);
 247 }
 248 #undef FUNC_NAME
 249
 250 SCM_DEFINE (scm_string_ci_to_symbol, "string-ci->symbol", 1, 0, 0,
 251             (SCM str),
 252             "Return the symbol whose name is @var{str}.  @var{str} is\n"
 253             "converted to lowercase before the conversion is done, if Guile\n"
 254             "is currently reading symbols case-insensitively.")
 255 #define FUNC_NAME s_scm_string_ci_to_symbol
 256 {
 257   return scm_string_to_symbol (SCM_CASE_INSENSITIVE_P
 258                                ? scm_string_downcase(str)
 259                                : str);
 260 }
 261 #undef FUNC_NAME
 262
 263 #define MAX_PREFIX_LENGTH 30
 264
 265 SCM_DEFINE (scm_gensym, "gensym", 0, 1, 0,
 266             (SCM prefix),
 267             "Create a new symbol with a name constructed from a prefix and\n"
 268             "a counter value. The string @var{prefix} can be specified as\n"
 269             "an optional argument. Default prefix is @code{ g}.  The counter\n"
 270             "is increased by 1 at each call. There is no provision for\n"
 271             "resetting the counter.")
 272 #define FUNC_NAME s_scm_gensym
 273 {
 274   static int gensym_counter = 0;
 275
 276   SCM suffix, name;
 277   int n, n_digits;
 278   char buf[SCM_INTBUFLEN];
 279
 280   if (SCM_UNBNDP (prefix))
 281     prefix = scm_from_locale_string (" g");
 282
 283   /* mutex in case another thread looks and incs at the exact same moment */
 284   scm_mutex_lock (&scm_i_misc_mutex);
 285   n = gensym_counter++;
 286   scm_mutex_unlock (&scm_i_misc_mutex);
 287
 288   n_digits = scm_iint2str (n, 10, buf);
 289   suffix = scm_from_locale_stringn (buf, n_digits);
 290   name = scm_string_append (scm_list_2 (prefix, suffix));
 291   return scm_string_to_symbol (name);
 292 }
 293 #undef FUNC_NAME
 294
 295 SCM_DEFINE (scm_symbol_hash, "symbol-hash", 1, 0, 0,
 296             (SCM symbol),
 297             "Return a hash value for @var{symbol}.")
 298 #define FUNC_NAME s_scm_symbol_hash
 299 {
 300   SCM_VALIDATE_SYMBOL (1, symbol);
 301   return scm_from_ulong (scm_i_symbol_hash (symbol));
 302 }
 303 #undef FUNC_NAME
 304
 305 SCM_DEFINE (scm_symbol_fref, "symbol-fref", 1, 0, 0,
 306            (SCM s),
 307             "Return the contents of @var{symbol}'s @dfn{function slot}.")
 308 #define FUNC_NAME s_scm_symbol_fref
 309 {
 310   SCM_VALIDATE_SYMBOL (1, s);
 311   return SCM_CAR (SCM_CELL_OBJECT_3 (s));
 312 }
 313 #undef FUNC_NAME
 314
 315
 316 SCM_DEFINE (scm_symbol_pref, "symbol-pref", 1, 0, 0,
 317            (SCM s),
 318             "Return the @dfn{property list} currently associated with @var{symbol}.")
 319 #define FUNC_NAME s_scm_symbol_pref
 320 {
 321   SCM_VALIDATE_SYMBOL (1, s);
 322   return SCM_CDR (SCM_CELL_OBJECT_3 (s));
 323 }
 324 #undef FUNC_NAME
 325
 326
 327 SCM_DEFINE (scm_symbol_fset_x, "symbol-fset!", 2, 0, 0,
 328            (SCM s, SCM val),
 329             "Change the binding of @var{symbol}'s function slot.")
 330 #define FUNC_NAME s_scm_symbol_fset_x
 331 {
 332   SCM_VALIDATE_SYMBOL (1, s);
 333   SCM_SETCAR (SCM_CELL_OBJECT_3 (s), val);
 334   return SCM_UNSPECIFIED;
 335 }
 336 #undef FUNC_NAME
 337
 338
 339 SCM_DEFINE (scm_symbol_pset_x, "symbol-pset!", 2, 0, 0,
 340            (SCM s, SCM val),
 341             "Change the binding of @var{symbol}'s property slot.")
 342 #define FUNC_NAME s_scm_symbol_pset_x
 343 {
 344   SCM_VALIDATE_SYMBOL (1, s);
 345   SCM_SETCDR (SCM_CELL_OBJECT_3 (s), val);
 346   return SCM_UNSPECIFIED;
 347 }
 348 #undef FUNC_NAME
 349
 350 SCM
 351 scm_from_locale_symbol (const char *sym)
 352 {
 353   return scm_string_to_symbol (scm_from_locale_string (sym));
 354 }
 355
 356 SCM
 357 scm_from_locale_symboln (const char *sym, size_t len)
 358 {
 359   return scm_string_to_symbol (scm_from_locale_stringn (sym, len));
 360 }
 361
 362 void
 363 scm_symbols_prehistory ()
 364 {
 365   symbols = scm_make_weak_key_hash_table (scm_from_int (2139));
 366   scm_permanent_object (symbols);
 367 }
 368
 369
 370 void
 371 scm_init_symbols ()
 372 {
 373 #include "libguile/symbols.x"
 374 }
 375
 376 /*
 377   Local Variables:
 378   c-file-style: "gnu"
 379   End:
 380 */