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);
  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_is_null (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, 0, 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   size_t raw_hash = scm_string_hash ((const unsigned char *) name, len);
 148
 149   return scm_i_make_symbol (str, SCM_I_F_SYMBOL_UNINTERNED,
 150                             raw_hash, scm_cons (SCM_BOOL_F, SCM_EOL));
 151 }
 152
 153 SCM_DEFINE (scm_symbol_p, "symbol?", 1, 0, 0,
 154             (SCM obj),
 155             "Return @code{#t} if @var{obj} is a symbol, otherwise return\n"
 156             "@code{#f}.")
 157 #define FUNC_NAME s_scm_symbol_p
 158 {
 159   return scm_from_bool (scm_is_symbol (obj));
 160 }
 161 #undef FUNC_NAME
 162
 163 SCM_DEFINE (scm_symbol_interned_p, "symbol-interned?", 1, 0, 0,
 164             (SCM symbol),
 165             "Return @code{#t} if @var{symbol} is interned, otherwise return\n"
 166             "@code{#f}.")
 167 #define FUNC_NAME s_scm_symbol_interned_p
 168 {
 169   SCM_VALIDATE_SYMBOL (1, symbol);
 170   return scm_from_bool (scm_i_symbol_is_interned (symbol));
 171 }
 172 #undef FUNC_NAME
 173
 174 SCM_DEFINE (scm_make_symbol, "make-symbol", 1, 0, 0,
 175             (SCM name),
 176             "Return a new uninterned symbol with the name @var{name}.  "
 177             "The returned symbol is guaranteed to be unique and future "
 178             "calls to @code{string->symbol} will not return it.")
 179 #define FUNC_NAME s_scm_make_symbol
 180 {
 181   SCM_VALIDATE_STRING (1, name);
 182   return scm_i_mem2uninterned_symbol (name);
 183 }
 184 #undef FUNC_NAME
 185
 186 SCM_DEFINE (scm_symbol_to_string, "symbol->string", 1, 0, 0,
 187            (SCM s),
 188             "Return the name of @var{symbol} as a string.  If the symbol was\n"
 189             "part of an object returned as the value of a literal expression\n"
 190             "(section @pxref{Literal expressions,,,r5rs, The Revised^5\n"
 191             "Report on Scheme}) or by a call to the @code{read} procedure,\n"
 192             "and its name contains alphabetic characters, then the string\n"
 193             "returned will contain characters in the implementation's\n"
 194             "preferred standard case---some implementations will prefer\n"
 195             "upper case, others lower case.  If the symbol was returned by\n"
 196             "@code{string->symbol}, the case of characters in the string\n"
 197             "returned will be the same as the case in the string that was\n"
 198             "passed to @code{string->symbol}.  It is an error to apply\n"
 199             "mutation procedures like @code{string-set!} to strings returned\n"
 200             "by this procedure.\n"
 201             "\n"
 202             "The following examples assume that the implementation's\n"
 203             "standard case is lower case:\n"
 204             "\n"
 205             "@lisp\n"
 206             "(symbol->string 'flying-fish)    @result{} \"flying-fish\"\n"
 207             "(symbol->string 'Martin)         @result{}  \"martin\"\n"
 208             "(symbol->string\n"
 209             "   (string->symbol \"Malvina\")) @result{} \"Malvina\"\n"
 210             "@end lisp")
 211 #define FUNC_NAME s_scm_symbol_to_string
 212 {
 213   SCM_VALIDATE_SYMBOL (1, s);
 214   return scm_i_symbol_substring (s, 0, scm_i_symbol_length (s));
 215 }
 216 #undef FUNC_NAME
 217
 218
 219 SCM_DEFINE (scm_string_to_symbol, "string->symbol", 1, 0, 0,
 220             (SCM string),
 221             "Return the symbol whose name is @var{string}. This procedure\n"
 222             "can create symbols with names containing special characters or\n"
 223             "letters in the non-standard case, but it is usually a bad idea\n"
 224             "to create such symbols because in some implementations of\n"
 225             "Scheme they cannot be read as themselves.  See\n"
 226             "@code{symbol->string}.\n"
 227             "\n"
 228             "The following examples assume that the implementation's\n"
 229             "standard case is lower case:\n"
 230             "\n"
 231             "@lisp\n"
 232             "(eq? 'mISSISSIppi 'mississippi) @result{} #t\n"
 233             "(string->symbol \"mISSISSIppi\") @result{} @r{the symbol with name \"mISSISSIppi\"}\n"
 234             "(eq? 'bitBlt (string->symbol \"bitBlt\")) @result{} #f\n"
 235             "(eq? 'JollyWog\n"
 236             "  (string->symbol (symbol->string 'JollyWog))) @result{} #t\n"
 237             "(string=? \"K. Harper, M.D.\"\n"
 238             "  (symbol->string\n"
 239             "    (string->symbol \"K. Harper, M.D.\"))) @result{}#t\n"
 240             "@end lisp")
 241 #define FUNC_NAME s_scm_string_to_symbol
 242 {
 243   SCM_VALIDATE_STRING (1, string);
 244   return scm_i_mem2symbol (string);
 245 }
 246 #undef FUNC_NAME
 247
 248 SCM_DEFINE (scm_string_ci_to_symbol, "string-ci->symbol", 1, 0, 0,
 249             (SCM str),
 250             "Return the symbol whose name is @var{str}.  @var{str} is\n"
 251             "converted to lowercase before the conversion is done, if Guile\n"
 252             "is currently reading symbols case-insensitively.")
 253 #define FUNC_NAME s_scm_string_ci_to_symbol
 254 {
 255   return scm_string_to_symbol (SCM_CASE_INSENSITIVE_P
 256                                ? scm_string_downcase(str)
 257                                : str);
 258 }
 259 #undef FUNC_NAME
 260
 261 #define MAX_PREFIX_LENGTH 30
 262
 263 SCM_DEFINE (scm_gensym, "gensym", 0, 1, 0,
 264             (SCM prefix),
 265             "Create a new symbol with a name constructed from a prefix and\n"
 266             "a counter value. The string @var{prefix} can be specified as\n"
 267             "an optional argument. Default prefix is @code{ g}.  The counter\n"
 268             "is increased by 1 at each call. There is no provision for\n"
 269             "resetting the counter.")
 270 #define FUNC_NAME s_scm_gensym
 271 {
 272   static int gensym_counter = 0;
 273
 274   SCM suffix, name;
 275   int n, n_digits;
 276   char buf[SCM_INTBUFLEN];
 277
 278   if (SCM_UNBNDP (prefix))
 279     prefix = scm_from_locale_string (" g");
 280
 281   /* mutex in case another thread looks and incs at the exact same moment */
 282   scm_mutex_lock (&scm_i_misc_mutex);
 283   n = gensym_counter++;
 284   scm_mutex_unlock (&scm_i_misc_mutex);
 285
 286   n_digits = scm_iint2str (n, 10, buf);
 287   suffix = scm_from_locale_stringn (buf, n_digits);
 288   name = scm_string_append (scm_list_2 (prefix, suffix));
 289   return scm_string_to_symbol (name);
 290 }
 291 #undef FUNC_NAME
 292
 293 SCM_DEFINE (scm_symbol_hash, "symbol-hash", 1, 0, 0,
 294             (SCM symbol),
 295             "Return a hash value for @var{symbol}.")
 296 #define FUNC_NAME s_scm_symbol_hash
 297 {
 298   SCM_VALIDATE_SYMBOL (1, symbol);
 299   return scm_from_ulong (scm_i_symbol_hash (symbol));
 300 }
 301 #undef FUNC_NAME
 302
 303 SCM_DEFINE (scm_symbol_fref, "symbol-fref", 1, 0, 0,
 304            (SCM s),
 305             "Return the contents of @var{symbol}'s @dfn{function slot}.")
 306 #define FUNC_NAME s_scm_symbol_fref
 307 {
 308   SCM_VALIDATE_SYMBOL (1, s);
 309   return SCM_CAR (SCM_CELL_OBJECT_3 (s));
 310 }
 311 #undef FUNC_NAME
 312
 313
 314 SCM_DEFINE (scm_symbol_pref, "symbol-pref", 1, 0, 0,
 315            (SCM s),
 316             "Return the @dfn{property list} currently associated with @var{symbol}.")
 317 #define FUNC_NAME s_scm_symbol_pref
 318 {
 319   SCM_VALIDATE_SYMBOL (1, s);
 320   return SCM_CDR (SCM_CELL_OBJECT_3 (s));
 321 }
 322 #undef FUNC_NAME
 323
 324
 325 SCM_DEFINE (scm_symbol_fset_x, "symbol-fset!", 2, 0, 0,
 326            (SCM s, SCM val),
 327             "Change the binding of @var{symbol}'s function slot.")
 328 #define FUNC_NAME s_scm_symbol_fset_x
 329 {
 330   SCM_VALIDATE_SYMBOL (1, s);
 331   SCM_SETCAR (SCM_CELL_OBJECT_3 (s), val);
 332   return SCM_UNSPECIFIED;
 333 }
 334 #undef FUNC_NAME
 335
 336
 337 SCM_DEFINE (scm_symbol_pset_x, "symbol-pset!", 2, 0, 0,
 338            (SCM s, SCM val),
 339             "Change the binding of @var{symbol}'s property slot.")
 340 #define FUNC_NAME s_scm_symbol_pset_x
 341 {
 342   SCM_VALIDATE_SYMBOL (1, s);
 343   SCM_SETCDR (SCM_CELL_OBJECT_3 (s), val);
 344   return SCM_UNSPECIFIED;
 345 }
 346 #undef FUNC_NAME
 347
 348 SCM
 349 scm_from_locale_symbol (const char *sym)
 350 {
 351   return scm_string_to_symbol (scm_from_locale_string (sym));
 352 }
 353
 354 SCM
 355 scm_from_locale_symboln (const char *sym, size_t len)
 356 {
 357   return scm_string_to_symbol (scm_from_locale_stringn (sym, len));
 358 }
 359
 360 void
 361 scm_symbols_prehistory ()
 362 {
 363   symbols = scm_make_weak_key_hash_table (scm_from_int (2139));
 364   scm_permanent_object (symbols);
 365 }
 366
 367
 368 void
 369 scm_init_symbols ()
 370 {
 371 #include "libguile/symbols.x"
 372 }
 373
 374 /*
 375   Local Variables:
 376   c-file-style: "gnu"
 377   End:
 378 */