libguile/strings.c

   1 /* Copyright (C) 1995,1996,1998,2000,2001 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 /* Software engineering face-lift by Greg J. Badros, 11-Dec-1999,
  43    gjb@cs.washington.edu, http://www.cs.washington.edu/homes/gjb */
  44
  45 \f
  46
  47 #include <string.h>
  48
  49 #include "libguile/_scm.h"
  50 #include "libguile/chars.h"
  51 #include "libguile/root.h"
  52 #include "libguile/strings.h"
  53 #include "libguile/validate.h"
  54 \f
  55
  56 /* {Strings}
  57  */
  58
  59 SCM_DEFINE (scm_string_p, "string?", 1, 0, 0,
  60             (SCM obj),
  61             "Return @code{#t} iff @var{obj} is a string, else returns\n"
  62             "@code{#f}.")
  63 #define FUNC_NAME s_scm_string_p
  64 {
  65   return SCM_BOOL (SCM_STRINGP (obj));
  66 }
  67 #undef FUNC_NAME
  68
  69 #if SCM_DEBUG_DEPRECATED == 0
  70
  71 /* The concept of read-only strings will disappear in next release
  72  * of Guile.
  73  */
  74
  75 SCM_DEFINE (scm_read_only_string_p, "read-only-string?", 1, 0, 0,
  76             (SCM obj),
  77             "Return @code{#t} if @var{obj} is either a string or a symbol,\n"
  78             "otherwise return @code{#f}.")
  79 #define FUNC_NAME s_scm_read_only_string_p
  80 {
  81   return SCM_BOOL(SCM_ROSTRINGP (obj));
  82 }
  83 #undef FUNC_NAME
  84
  85 #endif /* DEPRECATED */
  86
  87 SCM_REGISTER_PROC (s_scm_list_to_string, "list->string", 1, 0, 0, scm_string);
  88
  89 SCM_DEFINE (scm_string, "string", 0, 0, 1,
  90             (SCM chrs),
  91             "@deffnx primitive list->string chrs\n"
  92             "Return a newly allocated string composed of the arguments,\n"
  93             "@var{chrs}.")
  94 #define FUNC_NAME s_scm_string
  95 {
  96   SCM result;
  97
  98   {
  99     long i = scm_ilength (chrs);
 100
 101     SCM_ASSERT (i >= 0, chrs, SCM_ARGn, FUNC_NAME);
 102     result = scm_allocate_string (i);
 103   }
 104
 105   {
 106     unsigned char *data = SCM_STRING_UCHARS (result);
 107
 108     while (!SCM_NULLP (chrs))
 109       {
 110         SCM elt = SCM_CAR (chrs);
 111
 112         SCM_VALIDATE_CHAR (SCM_ARGn, elt);
 113         *data++ = SCM_CHAR (elt);
 114         chrs = SCM_CDR (chrs);
 115       }
 116   }
 117   return result;
 118 }
 119 #undef FUNC_NAME
 120
 121 #if (SCM_DEBUG_DEPRECATED == 0)
 122
 123 SCM
 124 scm_makstr (size_t len, int dummy)
 125 #define FUNC_NAME "scm_makstr"
 126 {
 127   SCM s;
 128   char *mem;
 129
 130   SCM_ASSERT_RANGE (1, scm_long2num (len), len <= SCM_STRING_MAX_LENGTH);
 131
 132   mem = (char *) scm_must_malloc (len + 1, FUNC_NAME);
 133   mem[len] = 0;
 134
 135   SCM_NEWCELL (s);
 136   SCM_SET_STRING_CHARS (s, mem);
 137   SCM_SET_STRING_LENGTH (s, len);
 138
 139   return s;
 140 }
 141 #undef FUNC_NAME
 142
 143 #endif  /* SCM_DEBUG_DEPRECATED == 0 */
 144
 145 /* converts C scm_array of strings to SCM scm_list of strings. */
 146 /* If argc < 0, a null terminated scm_array is assumed. */
 147
 148 SCM
 149 scm_makfromstrs (int argc, char **argv)
 150 {
 151   int i = argc;
 152   SCM lst = SCM_EOL;
 153   if (0 > i)
 154     for (i = 0; argv[i]; i++);
 155   while (i--)
 156     lst = scm_cons (scm_mem2string (argv[i], strlen (argv[i])), lst);
 157   return lst;
 158 }
 159
 160
 161 /* This function must only be applied to memory obtained via malloc,
 162    since the GC is going to apply `free' to it when the string is
 163    dropped.
 164
 165    Also, s[len] must be `\0', since we promise that strings are
 166    null-terminated.  Perhaps we could handle non-null-terminated
 167    strings by claiming they're shared substrings of a string we just
 168    made up.  */
 169 SCM
 170 scm_take_str (char *s, size_t len)
 171 #define FUNC_NAME "scm_take_str"
 172 {
 173   SCM answer;
 174
 175   SCM_ASSERT_RANGE (2, scm_ulong2num (len), len <= SCM_STRING_MAX_LENGTH);
 176
 177   SCM_NEWCELL (answer);
 178   SCM_SET_STRING_CHARS (answer, s);
 179   SCM_SET_STRING_LENGTH (answer, len);
 180   scm_done_malloc (len + 1);
 181
 182   return answer;
 183 }
 184 #undef FUNC_NAME
 185
 186
 187 /* `s' must be a malloc'd string.  See scm_take_str.  */
 188 SCM
 189 scm_take0str (char *s)
 190 {
 191   return scm_take_str (s, strlen (s));
 192 }
 193
 194 #if (SCM_DEBUG_DEPRECATED == 0)
 195
 196 SCM
 197 scm_makfromstr (const char *src, size_t len, int dummy SCM_UNUSED)
 198 {
 199   scm_c_issue_deprecation_warning ("`scm_makfromstr' is deprecated. "
 200                                    "Use `scm_mem2string' instead.");
 201
 202   return scm_mem2string (src, len);
 203 }
 204
 205 #endif
 206
 207 SCM
 208 scm_mem2string (const char *src, size_t len)
 209 {
 210   SCM s = scm_allocate_string (len);
 211   char *dst = SCM_STRING_CHARS (s);
 212
 213   while (len--)
 214     *dst++ = *src++;
 215   return s;
 216 }
 217
 218 SCM
 219 scm_makfrom0str (const char *src)
 220 {
 221   if (!src) return SCM_BOOL_F;
 222   return scm_mem2string (src, strlen (src));
 223 }
 224
 225
 226 SCM
 227 scm_makfrom0str_opt (const char *src)
 228 {
 229   return scm_makfrom0str (src);
 230 }
 231
 232
 233 SCM
 234 scm_allocate_string (size_t len)
 235 #define FUNC_NAME "scm_allocate_string"
 236 {
 237   char *mem;
 238   SCM s;
 239
 240   SCM_ASSERT_RANGE (1, scm_long2num (len), len <= SCM_STRING_MAX_LENGTH);
 241
 242   mem = (char *) scm_must_malloc (len + 1, FUNC_NAME);
 243   mem[len] = 0;
 244
 245   SCM_NEWCELL (s);
 246   SCM_SET_STRING_CHARS (s, mem);
 247   SCM_SET_STRING_LENGTH (s, len);
 248
 249   return s;
 250 }
 251 #undef FUNC_NAME
 252
 253
 254 SCM_DEFINE (scm_make_string, "make-string", 1, 1, 0,
 255             (SCM k, SCM chr),
 256             "Return a newly allocated string of\n"
 257             "length @var{k}.  If @var{chr} is given, then all elements of\n"
 258             "the string are initialized to @var{chr}, otherwise the contents\n"
 259             "of the @var{string} are unspecified.\n")
 260 #define FUNC_NAME s_scm_make_string
 261 {
 262   if (SCM_INUMP (k))
 263     {
 264       long int i = SCM_INUM (k);
 265       SCM res;
 266
 267       SCM_ASSERT_RANGE (1, k, i >= 0);
 268
 269       res = scm_allocate_string (i);
 270       if (!SCM_UNBNDP (chr))
 271         {
 272           unsigned char *dst;
 273
 274           SCM_VALIDATE_CHAR (2, chr);
 275
 276           dst = SCM_STRING_UCHARS (res);
 277           memset (dst, SCM_CHAR (chr), i);
 278         }
 279
 280       return res;
 281     }
 282   else if (SCM_BIGP (k))
 283     SCM_OUT_OF_RANGE (1, k);
 284   else
 285     SCM_WRONG_TYPE_ARG (1, k);
 286 }
 287 #undef FUNC_NAME
 288
 289
 290 SCM_DEFINE (scm_string_length, "string-length", 1, 0, 0,
 291             (SCM string),
 292             "Return the number of characters in @var{string}.")
 293 #define FUNC_NAME s_scm_string_length
 294 {
 295   SCM_VALIDATE_STRING (1, string);
 296   return SCM_MAKINUM (SCM_STRING_LENGTH (string));
 297 }
 298 #undef FUNC_NAME
 299
 300 SCM_DEFINE (scm_string_ref, "string-ref", 2, 0, 0,
 301             (SCM str, SCM k),
 302             "Return character @var{k} of @var{str} using zero-origin\n"
 303             "indexing. @var{k} must be a valid index of @var{str}.")
 304 #define FUNC_NAME s_scm_string_ref
 305 {
 306   long idx;
 307
 308   SCM_VALIDATE_STRING (1, str);
 309   SCM_VALIDATE_INUM_COPY (2, k, idx);
 310   SCM_ASSERT_RANGE (2, k, idx >= 0 && idx < SCM_STRING_LENGTH (str));
 311   return SCM_MAKE_CHAR (SCM_STRING_UCHARS (str)[idx]);
 312 }
 313 #undef FUNC_NAME
 314
 315
 316 SCM_DEFINE (scm_string_set_x, "string-set!", 3, 0, 0,
 317             (SCM str, SCM k, SCM chr),
 318             "Store @var{chr} in element @var{k} of @var{str} and return\n"
 319             "an unspecified value. @var{k} must be a valid index of\n"
 320             "@var{str}.")
 321 #define FUNC_NAME s_scm_string_set_x
 322 {
 323 #if (SCM_DEBUG_DEPRECATED == 0)
 324   SCM_VALIDATE_RWSTRING (1, str);
 325 #else
 326   SCM_VALIDATE_STRING (1, str);
 327 #endif
 328   SCM_VALIDATE_INUM_RANGE (2,k,0,SCM_STRING_LENGTH(str));
 329   SCM_VALIDATE_CHAR (3,chr);
 330   SCM_STRING_UCHARS (str)[SCM_INUM (k)] = SCM_CHAR (chr);
 331   return SCM_UNSPECIFIED;
 332 }
 333 #undef FUNC_NAME
 334
 335
 336 SCM_DEFINE (scm_substring, "substring", 2, 1, 0,
 337             (SCM str, SCM start, SCM end),
 338             "Return a newly allocated string formed from the characters\n"
 339             "of @var{str} beginning with index @var{start} (inclusive) and\n"
 340             "ending with index @var{end} (exclusive).\n"
 341             "@var{str} must be a string, @var{start} and @var{end} must be\n"
 342             "exact integers satisfying:\n\n"
 343             "0 <= @var{start} <= @var{end} <= (string-length @var{str}).")
 344 #define FUNC_NAME s_scm_substring
 345 {
 346   long int from;
 347   long int to;
 348   SCM substr;
 349
 350   SCM_VALIDATE_STRING (1, str);
 351   SCM_VALIDATE_INUM (2, start);
 352   SCM_VALIDATE_INUM_DEF (3, end, SCM_STRING_LENGTH (str));
 353
 354   from = SCM_INUM (start);
 355   SCM_ASSERT_RANGE (2, start, 0 <= from && from <= SCM_STRING_LENGTH (str));
 356   to = SCM_INUM (end);
 357   SCM_ASSERT_RANGE (3, end, from <= to && to <= SCM_STRING_LENGTH (str));
 358
 359   substr = scm_mem2string (&SCM_STRING_CHARS (str)[from], to - from);
 360   scm_remember_upto_here_1 (str);
 361   return substr;
 362 }
 363 #undef FUNC_NAME
 364
 365
 366 SCM_DEFINE (scm_string_append, "string-append", 0, 0, 1,
 367             (SCM args),
 368             "Return a newly allocated string whose characters form the\n"
 369             "concatenation of the given strings, @var{args}.")
 370 #define FUNC_NAME s_scm_string_append
 371 {
 372   SCM res;
 373   size_t i = 0;
 374   register SCM l, s;
 375   register unsigned char *data;
 376
 377   SCM_VALIDATE_REST_ARGUMENT (args);
 378   for (l = args; !SCM_NULLP (l); l = SCM_CDR (l)) {
 379     s = SCM_CAR (l);
 380     SCM_VALIDATE_STRING (SCM_ARGn,s);
 381     i += SCM_STRING_LENGTH (s);
 382   }
 383   res = scm_allocate_string (i);
 384   data = SCM_STRING_UCHARS (res);
 385   for (l = args; !SCM_NULLP (l);l = SCM_CDR (l)) {
 386     s = SCM_CAR (l);
 387     for (i = 0;i<SCM_STRING_LENGTH (s);i++) *data++ = SCM_STRING_UCHARS (s)[i];
 388   }
 389   return res;
 390 }
 391 #undef FUNC_NAME
 392
 393 #if SCM_DEBUG_DEPRECATED == 0
 394
 395 /* Explicit shared substrings will disappear from Guile.
 396  *
 397  * Instead, "normal" strings will be implemented using sharing
 398  * internally, combined with a copy-on-write strategy.
 399  */
 400
 401 SCM_DEFINE (scm_make_shared_substring, "make-shared-substring", 1, 2, 0,
 402             (SCM str, SCM start, SCM end),
 403             "Return a shared substring of @var{str}.  The arguments are the\n"
 404             "same as for the @code{substring} function: the shared substring\n"
 405             "returned includes all of the text from @var{str} between\n"
 406             "indexes @var{start} (inclusive) and @var{end} (exclusive).  If\n"
 407             "@var{end} is omitted, it defaults to the end of @var{str}.  The\n"
 408             "shared substring returned by @code{make-shared-substring}\n"
 409             "occupies the same storage space as @var{str}.")
 410 #define FUNC_NAME s_scm_make_shared_substring
 411 {
 412   long f;
 413   long t;
 414   SCM answer;
 415   SCM len_str;
 416
 417   SCM_VALIDATE_ROSTRING (1,str);
 418   SCM_VALIDATE_INUM_DEF_COPY (2,start,0,f);
 419   SCM_VALIDATE_INUM_DEF_COPY (3,end,SCM_ROLENGTH(str),t);
 420
 421   SCM_ASSERT_RANGE (2,start,(f >= 0));
 422   SCM_ASSERT_RANGE (3,end, (f <= t) && (t <= SCM_ROLENGTH (str)));
 423
 424   SCM_NEWCELL (answer);
 425   SCM_NEWCELL (len_str);
 426
 427   SCM_DEFER_INTS;
 428   if (SCM_SUBSTRP (str))
 429     {
 430       long offset;
 431       offset = SCM_INUM (SCM_SUBSTR_OFFSET (str));
 432       f += offset;
 433       t += offset;
 434       SCM_SETCAR (len_str, SCM_MAKINUM (f));
 435       SCM_SETCDR (len_str, SCM_SUBSTR_STR (str));
 436       SCM_SETCDR (answer, len_str);
 437       SCM_SETLENGTH (answer, t - f, scm_tc7_substring);
 438     }
 439   else
 440     {
 441       SCM_SETCAR (len_str, SCM_MAKINUM (f));
 442       SCM_SETCDR (len_str, str);
 443       SCM_SETCDR (answer, len_str);
 444       SCM_SETLENGTH (answer, t - f, scm_tc7_substring);
 445     }
 446   SCM_ALLOW_INTS;
 447   return answer;
 448 }
 449 #undef FUNC_NAME
 450
 451 #endif /* DEPRECATED */
 452
 453 void
 454 scm_init_strings ()
 455 {
 456   scm_nullstr = scm_allocate_string (0);
 457
 458 #ifndef SCM_MAGIC_SNARFER
 459 #include "libguile/strings.x"
 460 #endif
 461 }
 462
 463
 464 /*
 465   Local Variables:
 466   c-file-style: "gnu"
 467   End:
 468 */