libguile/strings.c

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