-/* Copyright (C) 1995,1996,1998 Free Software Foundation, Inc.
+/* Copyright (C) 1995,1996,1998,2000,2001 Free Software Foundation, Inc.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* whether to permit this exception to apply to your modifications.
* If you do not wish that, delete this exception notice. */
-/* Software engineering face-lift by Greg J. Badros, 11-Dec-1999,
- gjb@cs.washington.edu, http://www.cs.washington.edu/homes/gjb */
\f
-#include <stdio.h>
-#include "_scm.h"
-#include "chars.h"
+#include <string.h>
+
+#include "libguile/_scm.h"
+#include "libguile/chars.h"
+#include "libguile/root.h"
+#include "libguile/strings.h"
+#include "libguile/deprecation.h"
+#include "libguile/validate.h"
-#include "strings.h"
-#include "scm_validate.h"
\f
/* {Strings}
*/
SCM_DEFINE (scm_string_p, "string?", 1, 0, 0,
- (SCM x),
-"")
+ (SCM obj),
+ "Return @code{#t} if @var{obj} is a string, else @code{#f}.")
#define FUNC_NAME s_scm_string_p
{
- if (SCM_IMP (x))
- return SCM_BOOL_F;
- return SCM_BOOL(SCM_STRINGP (x));
+ return SCM_BOOL (SCM_STRINGP (obj));
}
#undef FUNC_NAME
-SCM_DEFINE (scm_read_only_string_p, "read-only-string?", 1, 0, 0,
- (SCM x),
-"Return true of OBJ can be read as a string,
-
-This illustrates the difference between @code{string?} and
-@code{read-only-string?}:
-@example
-(string? \"a string\") @result{} #t
-(string? 'a-symbol) @result{} #f
+SCM_REGISTER_PROC (s_scm_list_to_string, "list->string", 1, 0, 0, scm_string);
-(read-only-string? \"a string\") @result{} #t
-(read-only-string? 'a-symbol) @result{} #t
-@end example")
-#define FUNC_NAME s_scm_read_only_string_p
+SCM_DEFINE (scm_string, "string", 0, 0, 1,
+ (SCM chrs),
+ "@deffnx {Scheme Procedure} list->string chrs\n"
+ "Return a newly allocated string composed of the arguments,\n"
+ "@var{chrs}.")
+#define FUNC_NAME s_scm_string
{
- if (SCM_IMP (x))
- return SCM_BOOL_F;
- return SCM_BOOL(SCM_ROSTRINGP (x));
-}
-#undef FUNC_NAME
+ SCM result;
-SCM_REGISTER_PROC(s_list_to_string, "list->string", 1, 0, 0, scm_string);
+ {
+ long i = scm_ilength (chrs);
+ SCM_ASSERT (i >= 0, chrs, SCM_ARG1, FUNC_NAME);
+ result = scm_allocate_string (i);
+ }
-SCM_DEFINE (scm_string, "string", 0, 0, 1,
- (SCM chrs),
-"")
-#define FUNC_NAME s_scm_string
-{
- SCM res;
- register unsigned char *data;
- long i;
- long len;
- SCM_DEFER_INTS;
- i = scm_ilength (chrs);
- if (i < 0)
- {
- SCM_ALLOW_INTS;
- SCM_ASSERT (0, chrs, SCM_ARG1, FUNC_NAME);
- }
- len = 0;
{
- SCM s;
-
- for (len = 0, s = chrs; s != SCM_EOL; s = SCM_CDR (s))
- if (SCM_ICHRP (SCM_CAR (s)))
- len += 1;
- else if (SCM_ROSTRINGP (SCM_CAR (s)))
- len += SCM_ROLENGTH (SCM_CAR (s));
- else
- {
- SCM_ALLOW_INTS;
- SCM_ASSERT (0, s, SCM_ARG1, FUNC_NAME);
- }
+ unsigned char *data = SCM_STRING_UCHARS (result);
+
+ while (!SCM_NULLP (chrs))
+ {
+ SCM elt = SCM_CAR (chrs);
+
+ SCM_VALIDATE_CHAR (SCM_ARGn, elt);
+ *data++ = SCM_CHAR (elt);
+ chrs = SCM_CDR (chrs);
+ }
}
- res = scm_makstr (len, 0);
- data = SCM_UCHARS (res);
- for (;SCM_NNULLP (chrs);chrs = SCM_CDR (chrs))
- {
- if (SCM_ICHRP (SCM_CAR (chrs)))
- *data++ = SCM_ICHR (SCM_CAR (chrs));
- else
- {
- int l;
- char * c;
- l = SCM_ROLENGTH (SCM_CAR (chrs));
- c = SCM_ROCHARS (SCM_CAR (chrs));
- while (l)
- {
- --l;
- *data++ = *c++;
- }
- }
- }
- SCM_ALLOW_INTS;
- return res;
+ return result;
}
#undef FUNC_NAME
-SCM
-scm_makstr (long len, int slots)
-{
- SCM s;
- SCM * mem;
- SCM_NEWCELL (s);
- --slots;
- SCM_REDEFER_INTS;
- mem = (SCM *)scm_must_malloc (sizeof (SCM) * (slots + 1) + len + 1,
- "scm_makstr");
- if (slots >= 0)
- {
- int x;
- mem[slots] = (SCM)mem;
- for (x = 0; x < slots; ++x)
- mem[x] = SCM_BOOL_F;
- }
- SCM_SETCHARS (s, (char *) (mem + slots + 1));
- SCM_SETLENGTH (s, len, scm_tc7_string);
- SCM_REALLOW_INTS;
- SCM_CHARS (s)[len] = 0;
- return s;
-}
-
/* converts C scm_array of strings to SCM scm_list of strings. */
/* If argc < 0, a null terminated scm_array is assumed. */
-
SCM
scm_makfromstrs (int argc, char **argv)
{
if (0 > i)
for (i = 0; argv[i]; i++);
while (i--)
- lst = scm_cons (scm_makfromstr (argv[i], (scm_sizet) strlen (argv[i]), 0), lst);
+ lst = scm_cons (scm_mem2string (argv[i], strlen (argv[i])), lst);
return lst;
}
strings by claiming they're shared substrings of a string we just
made up. */
SCM
-scm_take_str (char *s, int len)
+scm_take_str (char *s, size_t len)
+#define FUNC_NAME "scm_take_str"
{
SCM answer;
- SCM_NEWCELL (answer);
- SCM_DEFER_INTS;
- SCM_SETLENGTH (answer, len, scm_tc7_string);
- scm_done_malloc (len + 1);
- SCM_SETCHARS (answer, s);
- SCM_ALLOW_INTS;
+
+ SCM_ASSERT_RANGE (2, scm_ulong2num (len), len <= SCM_STRING_MAX_LENGTH);
+
+ answer = scm_cell (SCM_MAKE_STRING_TAG (len), (scm_t_bits) s);
+ scm_gc_register_collectable_memory (s, len+1, "string");
+
return answer;
}
+#undef FUNC_NAME
+
/* `s' must be a malloc'd string. See scm_take_str. */
SCM
SCM
-scm_makfromstr (const char *src, scm_sizet len, int slots)
+scm_mem2string (const char *src, size_t len)
{
- SCM s;
- register char *dst;
- s = scm_makstr ((long) len, slots);
- dst = SCM_CHARS (s);
+ SCM s = scm_allocate_string (len);
+ char *dst = SCM_STRING_CHARS (s);
+
while (len--)
*dst++ = *src++;
return s;
}
+SCM
+scm_str2string (const char *src)
+{
+ return scm_mem2string (src, strlen (src));
+}
+
SCM
scm_makfrom0str (const char *src)
{
if (!src) return SCM_BOOL_F;
- return scm_makfromstr (src, (scm_sizet) strlen (src), 0);
+ return scm_mem2string (src, strlen (src));
}
}
+SCM
+scm_allocate_string (size_t len)
+#define FUNC_NAME "scm_allocate_string"
+{
+ char *mem;
+ SCM s;
+
+ SCM_ASSERT_RANGE (1, scm_long2num (len), len <= SCM_STRING_MAX_LENGTH);
+
+ mem = (char *) scm_gc_malloc (len + 1, "string");
+ mem[len] = 0;
+
+ s = scm_cell (SCM_MAKE_STRING_TAG (len), (scm_t_bits) mem);
+
+ return s;
+}
+#undef FUNC_NAME
SCM_DEFINE (scm_make_string, "make-string", 1, 1, 0,
- (SCM k, SCM chr),
-"")
+ (SCM k, SCM chr),
+ "Return a newly allocated string of\n"
+ "length @var{k}. If @var{chr} is given, then all elements of\n"
+ "the string are initialized to @var{chr}, otherwise the contents\n"
+ "of the @var{string} are unspecified.")
#define FUNC_NAME s_scm_make_string
{
- SCM res;
- register long i;
- SCM_VALIDATE_INUM_MIN_COPY (1,k,0,i);
- res = scm_makstr (i, 0);
- if (!SCM_UNBNDP (chr))
+ if (SCM_INUMP (k))
{
- SCM_VALIDATE_CHAR (2,chr);
- {
- unsigned char *dst = SCM_UCHARS (res);
- char c = SCM_ICHR (chr);
-
- memset (dst, c, i);
- }
+ long int i = SCM_INUM (k);
+ SCM res;
+
+ SCM_ASSERT_RANGE (1, k, i >= 0);
+
+ res = scm_allocate_string (i);
+ if (!SCM_UNBNDP (chr))
+ {
+ unsigned char *dst;
+
+ SCM_VALIDATE_CHAR (2, chr);
+
+ dst = SCM_STRING_UCHARS (res);
+ memset (dst, SCM_CHAR (chr), i);
+ }
+
+ return res;
}
- return res;
+ else if (SCM_BIGP (k))
+ SCM_OUT_OF_RANGE (1, k);
+ else
+ SCM_WRONG_TYPE_ARG (1, k);
}
#undef FUNC_NAME
+
SCM_DEFINE (scm_string_length, "string-length", 1, 0, 0,
- (SCM str),
-"")
+ (SCM string),
+ "Return the number of characters in @var{string}.")
#define FUNC_NAME s_scm_string_length
{
- SCM_VALIDATE_ROSTRING (1,str);
- return SCM_MAKINUM (SCM_ROLENGTH (str));
+ SCM_VALIDATE_STRING (1, string);
+ return SCM_MAKINUM (SCM_STRING_LENGTH (string));
}
#undef FUNC_NAME
-SCM_DEFINE (scm_string_ref, "string-ref", 1, 1, 0,
- (SCM str, SCM k),
-"")
+SCM_DEFINE (scm_string_ref, "string-ref", 2, 0, 0,
+ (SCM str, SCM k),
+ "Return character @var{k} of @var{str} using zero-origin\n"
+ "indexing. @var{k} must be a valid index of @var{str}.")
#define FUNC_NAME s_scm_string_ref
{
- SCM_VALIDATE_ROSTRING (1,str);
- SCM_VALIDATE_INUM_DEF (2,k,0);
- SCM_ASSERT_RANGE (2,k,SCM_INUM (k) < SCM_ROLENGTH (str) && SCM_INUM (k) >= 0);
- return SCM_MAKICHR (SCM_ROUCHARS (str)[SCM_INUM (k)]);
+ long idx;
+
+ SCM_VALIDATE_STRING (1, str);
+ SCM_VALIDATE_INUM_COPY (2, k, idx);
+ SCM_ASSERT_RANGE (2, k, idx >= 0 && idx < SCM_STRING_LENGTH (str));
+ return SCM_MAKE_CHAR (SCM_STRING_UCHARS (str)[idx]);
}
#undef FUNC_NAME
+
SCM_DEFINE (scm_string_set_x, "string-set!", 3, 0, 0,
- (SCM str, SCM k, SCM chr),
-"")
+ (SCM str, SCM k, SCM chr),
+ "Store @var{chr} in element @var{k} of @var{str} and return\n"
+ "an unspecified value. @var{k} must be a valid index of\n"
+ "@var{str}.")
#define FUNC_NAME s_scm_string_set_x
{
- SCM_VALIDATE_RWSTRING (1,str);
- SCM_VALIDATE_INUM_RANGE (2,k,0,SCM_LENGTH(str));
- SCM_VALIDATE_CHAR (3,chr);
- SCM_UCHARS (str)[SCM_INUM (k)] = SCM_ICHR (chr);
+ SCM_VALIDATE_STRING (1, str);
+ SCM_VALIDATE_INUM_RANGE (2, k,0, SCM_STRING_LENGTH(str));
+ SCM_VALIDATE_CHAR (3, chr);
+ SCM_STRING_UCHARS (str)[SCM_INUM (k)] = SCM_CHAR (chr);
return SCM_UNSPECIFIED;
}
#undef FUNC_NAME
-
SCM_DEFINE (scm_substring, "substring", 2, 1, 0,
- (SCM str, SCM start, SCM end),
-"")
+ (SCM str, SCM start, SCM end),
+ "Return a newly allocated string formed from the characters\n"
+ "of @var{str} beginning with index @var{start} (inclusive) and\n"
+ "ending with index @var{end} (exclusive).\n"
+ "@var{str} must be a string, @var{start} and @var{end} must be\n"
+ "exact integers satisfying:\n\n"
+ "0 <= @var{start} <= @var{end} <= (string-length @var{str}).")
#define FUNC_NAME s_scm_substring
{
- long l;
- SCM_VALIDATE_ROSTRING (1,str);
- SCM_VALIDATE_INUM (2,start);
- SCM_VALIDATE_INUM_DEF (3,end,SCM_ROLENGTH(str));
- SCM_ASSERT_RANGE (2,start,SCM_INUM (start) <= SCM_ROLENGTH (str));
- SCM_ASSERT_RANGE (2,end,SCM_INUM (end) <= SCM_ROLENGTH (str));
- l = SCM_INUM (end)-SCM_INUM (start);
- SCM_ASSERT (l >= 0, SCM_MAKINUM (l), SCM_OUTOFRANGE, FUNC_NAME);
- return scm_makfromstr (&SCM_ROCHARS (str)[SCM_INUM (start)], (scm_sizet)l, 0);
+ long int from;
+ long int to;
+ SCM substr;
+
+ SCM_VALIDATE_STRING (1, str);
+ SCM_VALIDATE_INUM (2, start);
+ SCM_VALIDATE_INUM_DEF (3, end, SCM_STRING_LENGTH (str));
+
+ from = SCM_INUM (start);
+ SCM_ASSERT_RANGE (2, start, 0 <= from && from <= SCM_STRING_LENGTH (str));
+ to = SCM_INUM (end);
+ SCM_ASSERT_RANGE (3, end, from <= to && to <= SCM_STRING_LENGTH (str));
+
+ substr = scm_mem2string (&SCM_STRING_CHARS (str)[from], to - from);
+ scm_remember_upto_here_1 (str);
+ return substr;
}
#undef FUNC_NAME
+
SCM_DEFINE (scm_string_append, "string-append", 0, 0, 1,
- (SCM args),
-"")
+ (SCM args),
+ "Return a newly allocated string whose characters form the\n"
+ "concatenation of the given strings, @var{args}.")
#define FUNC_NAME s_scm_string_append
{
SCM res;
- register long i = 0;
+ size_t i = 0;
register SCM l, s;
register unsigned char *data;
- for (l = args;SCM_CONSP (l);) {
+
+ SCM_VALIDATE_REST_ARGUMENT (args);
+ for (l = args; !SCM_NULLP (l); l = SCM_CDR (l)) {
s = SCM_CAR (l);
- SCM_VALIDATE_ROSTRING (SCM_ARGn,s);
- i += SCM_ROLENGTH (s);
- l = SCM_CDR (l);
+ SCM_VALIDATE_STRING (SCM_ARGn, s);
+ i += SCM_STRING_LENGTH (s);
}
- SCM_ASSERT (SCM_NULLP (l), args, SCM_ARGn, FUNC_NAME);
- res = scm_makstr (i, 0);
- data = SCM_UCHARS (res);
- for (l = args;SCM_NIMP (l);l = SCM_CDR (l)) {
+ res = scm_allocate_string (i);
+ data = SCM_STRING_UCHARS (res);
+ for (l = args; !SCM_NULLP (l);l = SCM_CDR (l)) {
s = SCM_CAR (l);
- for (i = 0;i<SCM_ROLENGTH (s);i++) *data++ = SCM_ROUCHARS (s)[i];
+ for (i = 0;i<SCM_STRING_LENGTH (s);i++) *data++ = SCM_STRING_UCHARS (s)[i];
}
return res;
}
#undef FUNC_NAME
-SCM_DEFINE (scm_make_shared_substring, "make-shared-substring", 1, 2, 0,
- (SCM str, SCM frm, SCM to),
-"Return a shared substring of @var{str}. The semantics are the same as
-for the @code{substring} function: the shared substring returned
-includes all of the text from @var{str} between indexes @var{start}
-(inclusive) and @var{end} (exclusive). If @var{end} is omitted, it
-defaults to the end of @var{str}. The shared substring returned by
-@code{make-shared-substring} occupies the same storage space as
-@var{str}.")
-#define FUNC_NAME s_scm_make_shared_substring
-{
- long f;
- long t;
- SCM answer;
- SCM len_str;
- SCM_VALIDATE_ROSTRING (1,str);
- SCM_VALIDATE_INUM_DEF_COPY (2,frm,0,f);
- SCM_VALIDATE_INUM_DEF_COPY (3,to,SCM_ROLENGTH(str),t);
+/* Converts the given Scheme string OBJ into a C string, containing a copy
+ of OBJ's content with a trailing null byte. If LENP is non-NULL, set
+ *LENP to the string's length.
+
+ When STR is non-NULL it receives the copy and is returned by the function,
+ otherwise new memory is allocated and the caller is responsible for
+ freeing it via free(). If out of memory, NULL is returned.
- SCM_ASSERT_RANGE (2,frm,(f >= 0));
- SCM_ASSERT_RANGE (3,to, (f <= t) && (t <= SCM_ROLENGTH (str)));
+ Note that Scheme strings may contain arbitrary data, including null
+ characters. This means that null termination is not a reliable way to
+ determine the length of the returned value. However, the function always
+ copies the complete contents of OBJ, and sets *LENP to the length of the
+ scheme string (if LENP is non-null). */
+#define FUNC_NAME "scm_c_string2str"
+char *
+scm_c_string2str (SCM obj, char *str, size_t *lenp)
+{
+ size_t len;
- SCM_NEWCELL (answer);
- SCM_NEWCELL (len_str);
+ SCM_ASSERT (SCM_STRINGP (obj), obj, SCM_ARG1, FUNC_NAME);
+ len = SCM_STRING_LENGTH (obj);
- SCM_DEFER_INTS;
- if (SCM_SUBSTRP (str))
- {
- long offset;
- offset = SCM_INUM (SCM_SUBSTR_OFFSET (str));
- f += offset;
- t += offset;
- SCM_SETCAR (len_str, SCM_MAKINUM (f));
- SCM_SETCDR (len_str, SCM_SUBSTR_STR (str));
- SCM_SETCDR (answer, len_str);
- SCM_SETLENGTH (answer, t - f, scm_tc7_substring);
- }
- else
+ if (str == NULL)
{
- SCM_SETCAR (len_str, SCM_MAKINUM (f));
- SCM_SETCDR (len_str, str);
- SCM_SETCDR (answer, len_str);
- SCM_SETLENGTH (answer, t - f, scm_tc7_substring);
+ /* FIXME: Should we use exported wrappers for malloc (and free), which
+ * allow windows DLLs to call the correct freeing function? */
+ str = (char *) malloc ((len + 1) * sizeof (char));
+ if (str == NULL)
+ return NULL;
}
- SCM_ALLOW_INTS;
- return answer;
+
+ memcpy (str, SCM_STRING_CHARS (obj), len);
+ scm_remember_upto_here_1 (obj);
+ str[len] = '\0';
+
+ if (lenp != NULL)
+ *lenp = len;
+
+ return str;
}
#undef FUNC_NAME
+
+/* Copy LEN characters at START from the Scheme string OBJ to memory
+ at STR. START is an index into OBJ; zero means the beginning of
+ the string. STR has already been allocated by the caller.
+
+ If START + LEN is off the end of OBJ, silently truncate the source
+ region to fit the string. If truncation occurs, the corresponding
+ area of STR is left unchanged. */
+#define FUNC_NAME "scm_c_substring2str"
+char *
+scm_c_substring2str (SCM obj, char *str, size_t start, size_t len)
+{
+ size_t src_length, effective_length;
+
+ SCM_ASSERT (SCM_STRINGP (obj), obj, SCM_ARG2, FUNC_NAME);
+ src_length = SCM_STRING_LENGTH (obj);
+ effective_length = (len + start <= src_length) ? len : src_length - start;
+ memcpy (str, SCM_STRING_CHARS (obj) + start, effective_length);
+ scm_remember_upto_here_1 (obj);
+ return str;
+}
+#undef FUNC_NAME
+
+
void
scm_init_strings ()
{
-#include "strings.x"
+ scm_nullstr = scm_allocate_string (0);
+
+#include "libguile/strings.x"
}
+
+/*
+ Local Variables:
+ c-file-style: "gnu"
+ End:
+*/