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 "_scm.h"
  49 #include "chars.h"
  50
  51 #include "strings.h"
  52 #include "validate.h"
  53 \f
  54
  55 /* {Strings}
  56  */
  57
  58 SCM_DEFINE (scm_string_p, "string?", 1, 0, 0,
  59            (SCM obj),
  60             "Returns #t iff OBJ is a string, else returns #f.")
  61 #define FUNC_NAME s_scm_string_p
  62 {
  63   return SCM_BOOL(SCM_STRINGP (obj));
  64 }
  65 #undef FUNC_NAME
  66
  67 SCM_DEFINE (scm_read_only_string_p, "read-only-string?", 1, 0, 0,
  68            (SCM x),
  69             "Return true if OBJ can be read as a string,\n\n"
  70             "This illustrates the difference between @code{string?} and\n"
  71             "@code{read-only-string?}:\n\n"
  72             "@example\n"
  73             "(string? \"a string\") @result{} #t\n"
  74             "(string? 'a-symbol) @result{} #f\n\n"
  75             "(read-only-string? \"a string\") @result{} #t\n"
  76             "(read-only-string? 'a-symbol) @result{} #t\n"
  77             "@end example")
  78 #define FUNC_NAME s_scm_read_only_string_p
  79 {
  80   return SCM_BOOL(SCM_ROSTRINGP (x));
  81 }
  82 #undef FUNC_NAME
  83
  84 SCM_REGISTER_PROC (s_scm_list_to_string, "list->string", 1, 0, 0, scm_string);
  85
  86 SCM_DEFINE (scm_string, "string", 0, 0, 1,
  87             (SCM chrs),
  88             "Returns a newly allocated string composed of the arguments, CHRS.")
  89 #define FUNC_NAME s_scm_string
  90 {
  91   SCM result;
  92
  93   {
  94     long i = scm_ilength (chrs);
  95
  96     SCM_ASSERT (i >= 0, chrs, SCM_ARGn, FUNC_NAME);
  97     result = scm_makstr (i, 0);
  98   }
  99
 100   {
 101     unsigned char *data = SCM_UCHARS (result);
 102
 103     while (SCM_NNULLP (chrs))
 104       {
 105         SCM elt = SCM_CAR (chrs);
 106
 107         SCM_VALIDATE_CHAR (SCM_ARGn, elt);
 108         *data++ = SCM_CHAR (elt);
 109         chrs = SCM_CDR (chrs);
 110       }
 111   }
 112   return result;
 113 }
 114 #undef FUNC_NAME
 115
 116 SCM
 117 scm_makstr (long len, int slots)
 118 {
 119   SCM s;
 120   SCM * mem;
 121   SCM_NEWCELL (s);
 122   --slots;
 123   SCM_REDEFER_INTS;
 124   mem = (SCM *)scm_must_malloc (sizeof (SCM) * (slots + 1) + len + 1,
 125                                 "scm_makstr");
 126   if (slots >= 0)
 127     {
 128       int x;
 129       mem[slots] = (SCM)mem;
 130       for (x = 0; x < slots; ++x)
 131         mem[x] = SCM_BOOL_F;
 132     }
 133   SCM_SETCHARS (s, (char *) (mem + slots + 1));
 134   SCM_SETLENGTH (s, len, scm_tc7_string);
 135   SCM_REALLOW_INTS;
 136   SCM_CHARS (s)[len] = 0;
 137   return s;
 138 }
 139
 140 /* converts C scm_array of strings to SCM scm_list of strings. */
 141 /* If argc < 0, a null terminated scm_array is assumed. */
 142
 143 SCM
 144 scm_makfromstrs (int argc, char **argv)
 145 {
 146   int i = argc;
 147   SCM lst = SCM_EOL;
 148   if (0 > i)
 149     for (i = 0; argv[i]; i++);
 150   while (i--)
 151     lst = scm_cons (scm_makfromstr (argv[i], (scm_sizet) strlen (argv[i]), 0), lst);
 152   return lst;
 153 }
 154
 155
 156 /* This function must only be applied to memory obtained via malloc,
 157    since the GC is going to apply `free' to it when the string is
 158    dropped.
 159
 160    Also, s[len] must be `\0', since we promise that strings are
 161    null-terminated.  Perhaps we could handle non-null-terminated
 162    strings by claiming they're shared substrings of a string we just
 163    made up.  */
 164 SCM
 165 scm_take_str (char *s, int len)
 166 {
 167   SCM answer;
 168   SCM_NEWCELL (answer);
 169   SCM_DEFER_INTS;
 170   SCM_SETLENGTH (answer, len, scm_tc7_string);
 171   scm_done_malloc (len + 1);
 172   SCM_SETCHARS (answer, s);
 173   SCM_ALLOW_INTS;
 174   return answer;
 175 }
 176
 177 /* `s' must be a malloc'd string.  See scm_take_str.  */
 178 SCM
 179 scm_take0str (char *s)
 180 {
 181   return scm_take_str (s, strlen (s));
 182 }
 183
 184 SCM
 185 scm_makfromstr (const char *src, scm_sizet len, int slots)
 186 {
 187   SCM s = scm_makstr (len, slots);
 188   char *dst = SCM_CHARS (s);
 189
 190   while (len--)
 191     *dst++ = *src++;
 192   return s;
 193 }
 194
 195 SCM
 196 scm_makfrom0str (const char *src)
 197 {
 198   if (!src) return SCM_BOOL_F;
 199   return scm_makfromstr (src, (scm_sizet) strlen (src), 0);
 200 }
 201
 202
 203 SCM
 204 scm_makfrom0str_opt (const char *src)
 205 {
 206   return scm_makfrom0str (src);
 207 }
 208
 209
 210
 211
 212 SCM_DEFINE (scm_make_string, "make-string", 1, 1, 0,
 213             (SCM k, SCM chr),
 214             "Returns a newly allocated string of\n"
 215             "length K.  If CHR is given, then all elements of the string\n"
 216             "are initialized to CHR, otherwise the contents of the\n"
 217             "STRING are unspecified.\n")
 218 #define FUNC_NAME s_scm_make_string
 219 {
 220   SCM res;
 221   register long i;
 222   SCM_VALIDATE_INUM_MIN_COPY (1,k,0,i);
 223   res = scm_makstr (i, 0);
 224   if (!SCM_UNBNDP (chr))
 225     {
 226       SCM_VALIDATE_CHAR (2,chr);
 227       {
 228         unsigned char *dst = SCM_UCHARS (res);
 229         char c = SCM_CHAR (chr);
 230
 231         memset (dst, c, i);
 232       }
 233     }
 234   return res;
 235 }
 236 #undef FUNC_NAME
 237
 238 SCM_DEFINE (scm_string_length, "string-length", 1, 0, 0,
 239            (SCM string),
 240             "Returns the number of characters in STRING")
 241 #define FUNC_NAME s_scm_string_length
 242 {
 243   SCM_VALIDATE_ROSTRING (1,string);
 244   return SCM_MAKINUM (SCM_ROLENGTH (string));
 245 }
 246 #undef FUNC_NAME
 247
 248 SCM_DEFINE (scm_string_ref, "string-ref", 2, 0, 0,
 249             (SCM str, SCM k),
 250             "Returns character K of STR using zero-origin indexing.\n"
 251             "K must be a valid index of STR.")
 252 #define FUNC_NAME s_scm_string_ref
 253 {
 254   int idx;
 255
 256   SCM_VALIDATE_ROSTRING (1, str);
 257   SCM_VALIDATE_INUM_COPY (2, k, idx);
 258   SCM_ASSERT_RANGE (2, k, idx >= 0 && idx < SCM_ROLENGTH (str));
 259   return SCM_MAKE_CHAR (SCM_ROUCHARS (str)[idx]);
 260 }
 261 #undef FUNC_NAME
 262
 263 SCM_DEFINE (scm_string_set_x, "string-set!", 3, 0, 0,
 264             (SCM str, SCM k, SCM chr),
 265             "Stores CHR in element K of STRING and returns an unspecified value.\n"
 266             "K must be a valid index of STR.")
 267 #define FUNC_NAME s_scm_string_set_x
 268 {
 269   SCM_VALIDATE_RWSTRING (1,str);
 270   SCM_VALIDATE_INUM_RANGE (2,k,0,SCM_LENGTH(str));
 271   SCM_VALIDATE_CHAR (3,chr);
 272   SCM_UCHARS (str)[SCM_INUM (k)] = SCM_CHAR (chr);
 273   return SCM_UNSPECIFIED;
 274 }
 275 #undef FUNC_NAME
 276
 277
 278
 279 SCM_DEFINE (scm_substring, "substring", 2, 1, 0,
 280            (SCM str, SCM start, SCM end),
 281             "Returns a newly allocated string formed from the characters\n"
 282             "of STR beginning with index START (inclusive) and ending with\n"
 283             "index END (exclusive).\n"
 284             "STR must be a string, START and END must be exact integers satisfying:\n\n"
 285             "0 <= START <= END <= (string-length STR).")
 286 #define FUNC_NAME s_scm_substring
 287 {
 288   long l;
 289   SCM_VALIDATE_ROSTRING (1,str);
 290   SCM_VALIDATE_INUM (2,start);
 291   SCM_VALIDATE_INUM_DEF (3,end,SCM_ROLENGTH(str));
 292   SCM_ASSERT_RANGE (2,start,SCM_INUM (start) <= SCM_ROLENGTH (str));
 293   SCM_ASSERT_RANGE (2,end,SCM_INUM (end) <= SCM_ROLENGTH (str));
 294   l = SCM_INUM (end)-SCM_INUM (start);
 295   SCM_ASSERT (l >= 0, SCM_MAKINUM (l), SCM_OUTOFRANGE, FUNC_NAME);
 296   return scm_makfromstr (&SCM_ROCHARS (str)[SCM_INUM (start)], (scm_sizet)l, 0);
 297 }
 298 #undef FUNC_NAME
 299
 300 SCM_DEFINE (scm_string_append, "string-append", 0, 0, 1,
 301             (SCM args),
 302             "Returns a newly allocated string whose characters form the\n"
 303             "concatenation of the given strings, ARGS.")
 304 #define FUNC_NAME s_scm_string_append
 305 {
 306   SCM res;
 307   register long i = 0;
 308   register SCM l, s;
 309   register unsigned char *data;
 310   for (l = args;SCM_CONSP (l);) {
 311     s = SCM_CAR (l);
 312     SCM_VALIDATE_ROSTRING (SCM_ARGn,s);
 313     i += SCM_ROLENGTH (s);
 314     l = SCM_CDR (l);
 315   }
 316   SCM_ASSERT (SCM_NULLP (l), args, SCM_ARGn, FUNC_NAME);
 317   res = scm_makstr (i, 0);
 318   data = SCM_UCHARS (res);
 319   for (l = args;SCM_NIMP (l);l = SCM_CDR (l)) {
 320     s = SCM_CAR (l);
 321     for (i = 0;i<SCM_ROLENGTH (s);i++) *data++ = SCM_ROUCHARS (s)[i];
 322   }
 323   return res;
 324 }
 325 #undef FUNC_NAME
 326
 327 SCM_DEFINE (scm_make_shared_substring, "make-shared-substring", 1, 2, 0,
 328            (SCM str, SCM frm, SCM to),
 329             "Return a shared substring of @var{str}.  The semantics are the same as\n"
 330             "for the @code{substring} function: the shared substring returned\n"
 331             "includes all of the text from @var{str} between indexes @var{start}\n"
 332             "(inclusive) and @var{end} (exclusive).  If @var{end} is omitted, it\n"
 333             "defaults to the end of @var{str}.  The shared substring returned by\n"
 334             "@code{make-shared-substring} occupies the same storage space as\n"
 335             "@var{str}.")
 336 #define FUNC_NAME s_scm_make_shared_substring
 337 {
 338   long f;
 339   long t;
 340   SCM answer;
 341   SCM len_str;
 342
 343   SCM_VALIDATE_ROSTRING (1,str);
 344   SCM_VALIDATE_INUM_DEF_COPY (2,frm,0,f);
 345   SCM_VALIDATE_INUM_DEF_COPY (3,to,SCM_ROLENGTH(str),t);
 346
 347   SCM_ASSERT_RANGE (2,frm,(f >= 0));
 348   SCM_ASSERT_RANGE (3,to, (f <= t) && (t <= SCM_ROLENGTH (str)));
 349
 350   SCM_NEWCELL (answer);
 351   SCM_NEWCELL (len_str);
 352
 353   SCM_DEFER_INTS;
 354   if (SCM_SUBSTRP (str))
 355     {
 356       long offset;
 357       offset = SCM_INUM (SCM_SUBSTR_OFFSET (str));
 358       f += offset;
 359       t += offset;
 360       SCM_SETCAR (len_str, SCM_MAKINUM (f));
 361       SCM_SETCDR (len_str, SCM_SUBSTR_STR (str));
 362       SCM_SETCDR (answer, len_str);
 363       SCM_SETLENGTH (answer, t - f, scm_tc7_substring);
 364     }
 365   else
 366     {
 367       SCM_SETCAR (len_str, SCM_MAKINUM (f));
 368       SCM_SETCDR (len_str, str);
 369       SCM_SETCDR (answer, len_str);
 370       SCM_SETLENGTH (answer, t - f, scm_tc7_substring);
 371     }
 372   SCM_ALLOW_INTS;
 373   return answer;
 374 }
 375 #undef FUNC_NAME
 376
 377 void
 378 scm_init_strings ()
 379 {
 380 #include "strings.x"
 381 }
 382