libguile/symbols.c

   1 /* Copyright (C) 1995,1996,1997,1998,2000,2001, 2003, 2004, 2006 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., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  16  */
  17
  18
  19 \f
  20 #ifdef 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 #include "libguile/private-options.h"
  44
  45
  46 #ifdef HAVE_STRING_H
  47 #include <string.h>
  48 #endif
  49
  50 \f
  51
  52 static SCM symbols;
  53
  54 #ifdef GUILE_DEBUG
  55 SCM_DEFINE (scm_sys_symbols, "%symbols", 0, 0, 0,
  56             (),
  57             "Return the system symbol obarray.")
  58 #define FUNC_NAME s_scm_sys_symbols
  59 {
  60   return symbols;
  61 }
  62 #undef FUNC_NAME
  63 #endif
  64
  65 \f
  66
  67 /* {Symbols}
  68  */
  69
  70 /* In order to optimize reading speed, this function breaks part of
  71  * the hashtable abstraction.  The optimizations are:
  72  *
  73  * 1. The argument string can be compared directly to symbol objects
  74  *    without first creating an SCM string object.  (This would have
  75  *    been necessary if we had used the hashtable API in hashtab.h.)
  76  *
  77  * 2. We can use the raw hash value stored in scm_i_symbol_hash (sym)
  78  *    to speed up lookup.
  79  *
  80  * Both optimizations might be possible without breaking the
  81  * abstraction if the API in hashtab.c is improved.
  82  */
  83
  84 unsigned long
  85 scm_i_hash_symbol (SCM obj, unsigned long n, void *closure)
  86 {
  87   return scm_i_symbol_hash (obj) % n;
  88 }
  89
  90 static SCM
  91 lookup_interned_symbol (const char *name, size_t len,
  92                         unsigned long raw_hash)
  93 {
  94   /* Try to find the symbol in the symbols table */
  95   SCM l;
  96   unsigned long hash = raw_hash % SCM_HASHTABLE_N_BUCKETS (symbols);
  97
  98   for (l = SCM_HASHTABLE_BUCKET (symbols, hash);
  99        !scm_is_null (l);
 100        l = SCM_CDR (l))
 101     {
 102       SCM sym = SCM_CAAR (l);
 103       if (scm_i_symbol_hash (sym) == raw_hash
 104           && scm_i_symbol_length (sym) == len)
 105         {
 106           const char *chrs = scm_i_symbol_chars (sym);
 107           size_t i = len;
 108
 109           while (i != 0)
 110             {
 111               --i;
 112               if (name[i] != chrs[i])
 113                 goto next_symbol;
 114             }
 115
 116           return sym;
 117         }
 118     next_symbol:
 119       ;
 120     }
 121
 122   return SCM_BOOL_F;
 123 }
 124
 125 static SCM
 126 scm_i_c_mem2symbol (const char *name, size_t len)
 127 {
 128   SCM symbol;
 129   size_t raw_hash = scm_string_hash ((const unsigned char *) name, len);
 130   size_t hash = raw_hash % SCM_HASHTABLE_N_BUCKETS (symbols);
 131
 132   symbol = lookup_interned_symbol (name, len, raw_hash);
 133   if (symbol != SCM_BOOL_F)
 134     return symbol;
 135
 136   {
 137     /* The symbol was not found - create it. */
 138     SCM symbol = scm_i_c_make_symbol (name, len, 0, raw_hash,
 139                                       scm_cons (SCM_BOOL_F, SCM_EOL));
 140
 141     SCM slot = SCM_HASHTABLE_BUCKET (symbols, hash);
 142     SCM cell = scm_cons (symbol, SCM_UNDEFINED);
 143     SCM_SET_HASHTABLE_BUCKET (symbols, hash, scm_cons (cell, slot));
 144     SCM_HASHTABLE_INCREMENT (symbols);
 145     if (SCM_HASHTABLE_N_ITEMS (symbols) > SCM_HASHTABLE_UPPER (symbols))
 146       scm_i_rehash (symbols, scm_i_hash_symbol, 0, "scm_mem2symbol");
 147
 148     return symbol;
 149   }
 150 }
 151
 152 static SCM
 153 scm_i_mem2symbol (SCM str)
 154 {
 155   SCM symbol;
 156   const char *name = scm_i_string_chars (str);
 157   size_t len = scm_i_string_length (str);
 158   size_t raw_hash = scm_string_hash ((const unsigned char *) name, len);
 159   size_t hash = raw_hash % SCM_HASHTABLE_N_BUCKETS (symbols);
 160
 161   symbol = lookup_interned_symbol (name, len, raw_hash);
 162   if (symbol != SCM_BOOL_F)
 163     return symbol;
 164
 165   {
 166     /* The symbol was not found - create it. */
 167     SCM symbol = scm_i_make_symbol (str, 0, raw_hash,
 168                                     scm_cons (SCM_BOOL_F, SCM_EOL));
 169
 170     SCM slot = SCM_HASHTABLE_BUCKET (symbols, hash);
 171     SCM cell = scm_cons (symbol, SCM_UNDEFINED);
 172     SCM_SET_HASHTABLE_BUCKET (symbols, hash, scm_cons (cell, slot));
 173     SCM_HASHTABLE_INCREMENT (symbols);
 174     if (SCM_HASHTABLE_N_ITEMS (symbols) > SCM_HASHTABLE_UPPER (symbols))
 175       scm_i_rehash (symbols, scm_i_hash_symbol, 0, "scm_mem2symbol");
 176
 177     return symbol;
 178   }
 179 }
 180
 181
 182 static SCM
 183 scm_i_mem2uninterned_symbol (SCM str)
 184 {
 185   const char *name = scm_i_string_chars (str);
 186   size_t len = scm_i_string_length (str);
 187   size_t raw_hash = scm_string_hash ((const unsigned char *) name, len);
 188
 189   return scm_i_make_symbol (str, SCM_I_F_SYMBOL_UNINTERNED,
 190                             raw_hash, scm_cons (SCM_BOOL_F, SCM_EOL));
 191 }
 192
 193 SCM_DEFINE (scm_symbol_p, "symbol?", 1, 0, 0,
 194             (SCM obj),
 195             "Return @code{#t} if @var{obj} is a symbol, otherwise return\n"
 196             "@code{#f}.")
 197 #define FUNC_NAME s_scm_symbol_p
 198 {
 199   return scm_from_bool (scm_is_symbol (obj));
 200 }
 201 #undef FUNC_NAME
 202
 203 SCM_DEFINE (scm_symbol_interned_p, "symbol-interned?", 1, 0, 0,
 204             (SCM symbol),
 205             "Return @code{#t} if @var{symbol} is interned, otherwise return\n"
 206             "@code{#f}.")
 207 #define FUNC_NAME s_scm_symbol_interned_p
 208 {
 209   SCM_VALIDATE_SYMBOL (1, symbol);
 210   return scm_from_bool (scm_i_symbol_is_interned (symbol));
 211 }
 212 #undef FUNC_NAME
 213
 214 SCM_DEFINE (scm_make_symbol, "make-symbol", 1, 0, 0,
 215             (SCM name),
 216             "Return a new uninterned symbol with the name @var{name}.  "
 217             "The returned symbol is guaranteed to be unique and future "
 218             "calls to @code{string->symbol} will not return it.")
 219 #define FUNC_NAME s_scm_make_symbol
 220 {
 221   SCM_VALIDATE_STRING (1, name);
 222   return scm_i_mem2uninterned_symbol (name);
 223 }
 224 #undef FUNC_NAME
 225
 226 SCM_DEFINE (scm_symbol_to_string, "symbol->string", 1, 0, 0,
 227            (SCM s),
 228             "Return the name of @var{symbol} as a string.  If the symbol was\n"
 229             "part of an object returned as the value of a literal expression\n"
 230             "(section @pxref{Literal expressions,,,r5rs, The Revised^5\n"
 231             "Report on Scheme}) or by a call to the @code{read} procedure,\n"
 232             "and its name contains alphabetic characters, then the string\n"
 233             "returned will contain characters in the implementation's\n"
 234             "preferred standard case---some implementations will prefer\n"
 235             "upper case, others lower case.  If the symbol was returned by\n"
 236             "@code{string->symbol}, the case of characters in the string\n"
 237             "returned will be the same as the case in the string that was\n"
 238             "passed to @code{string->symbol}.  It is an error to apply\n"
 239             "mutation procedures like @code{string-set!} to strings returned\n"
 240             "by this procedure.\n"
 241             "\n"
 242             "The following examples assume that the implementation's\n"
 243             "standard case is lower case:\n"
 244             "\n"
 245             "@lisp\n"
 246             "(symbol->string 'flying-fish)    @result{} \"flying-fish\"\n"
 247             "(symbol->string 'Martin)         @result{}  \"martin\"\n"
 248             "(symbol->string\n"
 249             "   (string->symbol \"Malvina\")) @result{} \"Malvina\"\n"
 250             "@end lisp")
 251 #define FUNC_NAME s_scm_symbol_to_string
 252 {
 253   SCM_VALIDATE_SYMBOL (1, s);
 254   return scm_i_symbol_substring (s, 0, scm_i_symbol_length (s));
 255 }
 256 #undef FUNC_NAME
 257
 258
 259 SCM_DEFINE (scm_string_to_symbol, "string->symbol", 1, 0, 0,
 260             (SCM string),
 261             "Return the symbol whose name is @var{string}. This procedure\n"
 262             "can create symbols with names containing special characters or\n"
 263             "letters in the non-standard case, but it is usually a bad idea\n"
 264             "to create such symbols because in some implementations of\n"
 265             "Scheme they cannot be read as themselves.  See\n"
 266             "@code{symbol->string}.\n"
 267             "\n"
 268             "The following examples assume that the implementation's\n"
 269             "standard case is lower case:\n"
 270             "\n"
 271             "@lisp\n"
 272             "(eq? 'mISSISSIppi 'mississippi) @result{} #t\n"
 273             "(string->symbol \"mISSISSIppi\") @result{} @r{the symbol with name \"mISSISSIppi\"}\n"
 274             "(eq? 'bitBlt (string->symbol \"bitBlt\")) @result{} #f\n"
 275             "(eq? 'JollyWog\n"
 276             "  (string->symbol (symbol->string 'JollyWog))) @result{} #t\n"
 277             "(string=? \"K. Harper, M.D.\"\n"
 278             "  (symbol->string\n"
 279             "    (string->symbol \"K. Harper, M.D.\"))) @result{}#t\n"
 280             "@end lisp")
 281 #define FUNC_NAME s_scm_string_to_symbol
 282 {
 283   SCM_VALIDATE_STRING (1, string);
 284   return scm_i_mem2symbol (string);
 285 }
 286 #undef FUNC_NAME
 287
 288 SCM_DEFINE (scm_string_ci_to_symbol, "string-ci->symbol", 1, 0, 0,
 289             (SCM str),
 290             "Return the symbol whose name is @var{str}.  @var{str} is\n"
 291             "converted to lowercase before the conversion is done, if Guile\n"
 292             "is currently reading symbols case-insensitively.")
 293 #define FUNC_NAME s_scm_string_ci_to_symbol
 294 {
 295   return scm_string_to_symbol (SCM_CASE_INSENSITIVE_P
 296                                ? scm_string_downcase(str)
 297                                : str);
 298 }
 299 #undef FUNC_NAME
 300
 301 #define MAX_PREFIX_LENGTH 30
 302
 303 SCM_DEFINE (scm_gensym, "gensym", 0, 1, 0,
 304             (SCM prefix),
 305             "Create a new symbol with a name constructed from a prefix and\n"
 306             "a counter value. The string @var{prefix} can be specified as\n"
 307             "an optional argument. Default prefix is @code{ g}.  The counter\n"
 308             "is increased by 1 at each call. There is no provision for\n"
 309             "resetting the counter.")
 310 #define FUNC_NAME s_scm_gensym
 311 {
 312   static int gensym_counter = 0;
 313
 314   SCM suffix, name;
 315   int n, n_digits;
 316   char buf[SCM_INTBUFLEN];
 317
 318   if (SCM_UNBNDP (prefix))
 319     prefix = scm_from_locale_string (" g");
 320
 321   /* mutex in case another thread looks and incs at the exact same moment */
 322   scm_i_scm_pthread_mutex_lock (&scm_i_misc_mutex);
 323   n = gensym_counter++;
 324   scm_i_pthread_mutex_unlock (&scm_i_misc_mutex);
 325
 326   n_digits = scm_iint2str (n, 10, buf);
 327   suffix = scm_from_locale_stringn (buf, n_digits);
 328   name = scm_string_append (scm_list_2 (prefix, suffix));
 329   return scm_string_to_symbol (name);
 330 }
 331 #undef FUNC_NAME
 332
 333 SCM_DEFINE (scm_symbol_hash, "symbol-hash", 1, 0, 0,
 334             (SCM symbol),
 335             "Return a hash value for @var{symbol}.")
 336 #define FUNC_NAME s_scm_symbol_hash
 337 {
 338   SCM_VALIDATE_SYMBOL (1, symbol);
 339   return scm_from_ulong (scm_i_symbol_hash (symbol));
 340 }
 341 #undef FUNC_NAME
 342
 343 SCM_DEFINE (scm_symbol_fref, "symbol-fref", 1, 0, 0,
 344            (SCM s),
 345             "Return the contents of @var{symbol}'s @dfn{function slot}.")
 346 #define FUNC_NAME s_scm_symbol_fref
 347 {
 348   SCM_VALIDATE_SYMBOL (1, s);
 349   return SCM_CAR (SCM_CELL_OBJECT_3 (s));
 350 }
 351 #undef FUNC_NAME
 352
 353
 354 SCM_DEFINE (scm_symbol_pref, "symbol-pref", 1, 0, 0,
 355            (SCM s),
 356             "Return the @dfn{property list} currently associated with @var{symbol}.")
 357 #define FUNC_NAME s_scm_symbol_pref
 358 {
 359   SCM_VALIDATE_SYMBOL (1, s);
 360   return SCM_CDR (SCM_CELL_OBJECT_3 (s));
 361 }
 362 #undef FUNC_NAME
 363
 364
 365 SCM_DEFINE (scm_symbol_fset_x, "symbol-fset!", 2, 0, 0,
 366            (SCM s, SCM val),
 367             "Change the binding of @var{symbol}'s function slot.")
 368 #define FUNC_NAME s_scm_symbol_fset_x
 369 {
 370   SCM_VALIDATE_SYMBOL (1, s);
 371   SCM_SETCAR (SCM_CELL_OBJECT_3 (s), val);
 372   return SCM_UNSPECIFIED;
 373 }
 374 #undef FUNC_NAME
 375
 376
 377 SCM_DEFINE (scm_symbol_pset_x, "symbol-pset!", 2, 0, 0,
 378            (SCM s, SCM val),
 379             "Change the binding of @var{symbol}'s property slot.")
 380 #define FUNC_NAME s_scm_symbol_pset_x
 381 {
 382   SCM_VALIDATE_SYMBOL (1, s);
 383   SCM_SETCDR (SCM_CELL_OBJECT_3 (s), val);
 384   return SCM_UNSPECIFIED;
 385 }
 386 #undef FUNC_NAME
 387
 388 SCM
 389 scm_from_locale_symbol (const char *sym)
 390 {
 391   return scm_i_c_mem2symbol (sym, strlen (sym));
 392 }
 393
 394 SCM
 395 scm_from_locale_symboln (const char *sym, size_t len)
 396 {
 397   return scm_i_c_mem2symbol (sym, len);
 398 }
 399
 400 SCM
 401 scm_take_locale_symboln (char *sym, size_t len)
 402 {
 403   SCM res;
 404   unsigned long raw_hash;
 405
 406   if (len == (size_t)-1)
 407     len = strlen (sym);
 408   else
 409     {
 410       /* Ensure STR is null terminated.  A realloc for 1 extra byte should
 411          often be satisfied from the alignment padding after the block, with
 412          no actual data movement.  */
 413       sym = scm_realloc (sym, len+1);
 414       sym[len] = '\0';
 415     }
 416
 417   raw_hash = scm_string_hash ((unsigned char *)sym, len);
 418   res = lookup_interned_symbol (sym, len, raw_hash);
 419   if (res != SCM_BOOL_F)
 420     {
 421       free (sym);
 422       return res;
 423     }
 424
 425   res = scm_i_c_take_symbol (sym, len, 0, raw_hash,
 426                              scm_cons (SCM_BOOL_F, SCM_EOL));
 427
 428   return res;
 429 }
 430
 431 SCM
 432 scm_take_locale_symbol (char *sym)
 433 {
 434   return scm_take_locale_symboln (sym, (size_t)-1);
 435 }
 436
 437 void
 438 scm_symbols_prehistory ()
 439 {
 440   symbols = scm_make_weak_key_hash_table (scm_from_int (2139));
 441   scm_permanent_object (symbols);
 442 }
 443
 444
 445 void
 446 scm_init_symbols ()
 447 {
 448 #include "libguile/symbols.x"
 449 }
 450
 451 /*
 452   Local Variables:
 453   c-file-style: "gnu"
 454   End:
 455 */