1 /* GLIB - Library of useful routines for C programming
2 * Copyright (C) 1995-1997 Peter Mattis, Spencer Kimball and Josh MacDonald
4 * This library is free software; you can redistribute it and/or
5 * modify it under the terms of the GNU Lesser General Public
6 * License as published by the Free Software Foundation; either
7 * version 2 of the License, or (at your option) any later version.
9 * This library is distributed in the hope that it will be useful,
10 * but WITHOUT ANY WARRANTY; without even the implied warranty of
11 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
12 * Lesser General Public License for more details.
14 * You should have received a copy of the GNU Lesser General Public
15 * License along with this library; if not, write to the
16 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
17 * Boston, MA 02111-1307, USA.
21 * Modified by the GLib Team and others 1997-2000. See the AUTHORS
22 * file for a list of people on the GLib Team. See the ChangeLog
23 * files for a list of changes. These files are distributed with
24 * GLib at ftp://ftp.gtk.org/pub/gtk/.
40 #include <ctype.h> /* For tolower() */
42 #ifdef HAVE_XLOCALE_H
\r
43 /* Needed on BSD/OS X for e.g. strtod_l */
\r
44 #include <xlocale.h>
\r
51 /* do not include <unistd.h> here, it may interfere with g_strsignal() */
53 #include "gstrfuncs.h"
56 #include "gprintfint.h"
61 * SECTION:string_utils
62 * @title: String Utility Functions
63 * @short_description: various string-related functions
65 * This section describes a number of utility functions for creating,
66 * duplicating, and manipulating strings.
68 * Note that the functions g_printf(), g_fprintf(), g_sprintf(),
69 * g_snprintf(), g_vprintf(), g_vfprintf(), g_vsprintf() and g_vsnprintf()
70 * are declared in the header <filename>gprintf.h</filename> which is
71 * <emphasis>not</emphasis> included in <filename>glib.h</filename>
72 * (otherwise using <filename>glib.h</filename> would drag in
73 * <filename>stdio.h</filename>), so you'll have to explicitly include
74 * <literal><glib/gprintf.h></literal> in order to use the GLib
77 * <para id="string-precision">While you may use the printf() functions
78 * to format UTF-8 strings, notice that the precision of a
79 * <literal>%Ns</literal> parameter is interpreted as the
80 * number of <emphasis>bytes</emphasis>, not <emphasis>characters</emphasis>
81 * to print. On top of that, the GNU libc implementation of the printf()
82 * functions has the "feature" that it checks that the string given for
83 * the <literal>%Ns</literal> parameter consists of a whole number
84 * of characters in the current encoding. So, unless you are sure you are
85 * always going to be in an UTF-8 locale or your know your text is restricted
86 * to ASCII, avoid using <literal>%Ns</literal>. If your intention is
87 * to format strings for a certain number of columns, then
88 * <literal>%Ns</literal> is not a correct solution anyway, since it
89 * fails to take wide characters (see g_unichar_iswide()) into account.
97 * Determines whether a character is alphanumeric.
99 * Unlike the standard C library isalnum() function, this only
100 * recognizes standard ASCII letters and ignores the locale,
101 * returning %FALSE for all non-ASCII characters. Also, unlike
102 * the standard library function, this takes a <type>char</type>,
103 * not an <type>int</type>, so don't call it on %EOF, but no need to
104 * cast to #guchar before passing a possibly non-ASCII character in.
106 * Returns: %TRUE if @c is an ASCII alphanumeric character
113 * Determines whether a character is alphabetic (i.e. a letter).
115 * Unlike the standard C library isalpha() function, this only
116 * recognizes standard ASCII letters and ignores the locale,
117 * returning %FALSE for all non-ASCII characters. Also, unlike
118 * the standard library function, this takes a <type>char</type>,
119 * not an <type>int</type>, so don't call it on %EOF, but no need to
120 * cast to #guchar before passing a possibly non-ASCII character in.
122 * Returns: %TRUE if @c is an ASCII alphabetic character
129 * Determines whether a character is a control character.
131 * Unlike the standard C library iscntrl() function, this only
132 * recognizes standard ASCII control characters and ignores the
133 * locale, returning %FALSE for all non-ASCII characters. Also,
134 * unlike the standard library function, this takes a <type>char</type>,
135 * not an <type>int</type>, so don't call it on %EOF, but no need to
136 * cast to #guchar before passing a possibly non-ASCII character in.
138 * Returns: %TRUE if @c is an ASCII control character.
145 * Determines whether a character is digit (0-9).
147 * Unlike the standard C library isdigit() function, this takes
148 * a <type>char</type>, not an <type>int</type>, so don't call it
149 * on %EOF, but no need to cast to #guchar before passing a possibly
150 * non-ASCII character in.
152 * Returns: %TRUE if @c is an ASCII digit.
159 * Determines whether a character is a printing character and not a space.
161 * Unlike the standard C library isgraph() function, this only
162 * recognizes standard ASCII characters and ignores the locale,
163 * returning %FALSE for all non-ASCII characters. Also, unlike
164 * the standard library function, this takes a <type>char</type>,
165 * not an <type>int</type>, so don't call it on %EOF, but no need
166 * to cast to #guchar before passing a possibly non-ASCII character in.
168 * Returns: %TRUE if @c is an ASCII printing character other than space.
175 * Determines whether a character is an ASCII lower case letter.
177 * Unlike the standard C library islower() function, this only
178 * recognizes standard ASCII letters and ignores the locale,
179 * returning %FALSE for all non-ASCII characters. Also, unlike
180 * the standard library function, this takes a <type>char</type>,
181 * not an <type>int</type>, so don't call it on %EOF, but no need
182 * to worry about casting to #guchar before passing a possibly
183 * non-ASCII character in.
185 * Returns: %TRUE if @c is an ASCII lower case letter
192 * Determines whether a character is a printing character.
194 * Unlike the standard C library isprint() function, this only
195 * recognizes standard ASCII characters and ignores the locale,
196 * returning %FALSE for all non-ASCII characters. Also, unlike
197 * the standard library function, this takes a <type>char</type>,
198 * not an <type>int</type>, so don't call it on %EOF, but no need
199 * to cast to #guchar before passing a possibly non-ASCII character in.
201 * Returns: %TRUE if @c is an ASCII printing character.
208 * Determines whether a character is a punctuation character.
210 * Unlike the standard C library ispunct() function, this only
211 * recognizes standard ASCII letters and ignores the locale,
212 * returning %FALSE for all non-ASCII characters. Also, unlike
213 * the standard library function, this takes a <type>char</type>,
214 * not an <type>int</type>, so don't call it on %EOF, but no need to
215 * cast to #guchar before passing a possibly non-ASCII character in.
217 * Returns: %TRUE if @c is an ASCII punctuation character.
224 * Determines whether a character is a white-space character.
226 * Unlike the standard C library isspace() function, this only
227 * recognizes standard ASCII white-space and ignores the locale,
228 * returning %FALSE for all non-ASCII characters. Also, unlike
229 * the standard library function, this takes a <type>char</type>,
230 * not an <type>int</type>, so don't call it on %EOF, but no need to
231 * cast to #guchar before passing a possibly non-ASCII character in.
233 * Returns: %TRUE if @c is an ASCII white-space character
240 * Determines whether a character is an ASCII upper case letter.
242 * Unlike the standard C library isupper() function, this only
243 * recognizes standard ASCII letters and ignores the locale,
244 * returning %FALSE for all non-ASCII characters. Also, unlike
245 * the standard library function, this takes a <type>char</type>,
246 * not an <type>int</type>, so don't call it on %EOF, but no need to
247 * worry about casting to #guchar before passing a possibly non-ASCII
250 * Returns: %TRUE if @c is an ASCII upper case letter
257 * Determines whether a character is a hexadecimal-digit character.
259 * Unlike the standard C library isxdigit() function, this takes
260 * a <type>char</type>, not an <type>int</type>, so don't call it
261 * on %EOF, but no need to cast to #guchar before passing a
262 * possibly non-ASCII character in.
264 * Returns: %TRUE if @c is an ASCII hexadecimal-digit character.
268 * G_ASCII_DTOSTR_BUF_SIZE:
270 * A good size for a buffer to be passed into g_ascii_dtostr().
271 * It is guaranteed to be enough for all output of that function
272 * on systems with 64bit IEEE-compatible doubles.
274 * The typical usage would be something like:
276 * char buf[G_ASCII_DTOSTR_BUF_SIZE];
278 * fprintf (out, "value=%s\n", g_ascii_dtostr (buf, sizeof (buf), value));
284 * @string: a string to remove the leading and trailing whitespace from
286 * Removes leading and trailing whitespace from a string.
287 * See g_strchomp() and g_strchug().
295 * The standard delimiters, used in g_strdelimit().
298 static const guint16 ascii_table_data[256] = {
299 0x004, 0x004, 0x004, 0x004, 0x004, 0x004, 0x004, 0x004,
300 0x004, 0x104, 0x104, 0x004, 0x104, 0x104, 0x004, 0x004,
301 0x004, 0x004, 0x004, 0x004, 0x004, 0x004, 0x004, 0x004,
302 0x004, 0x004, 0x004, 0x004, 0x004, 0x004, 0x004, 0x004,
303 0x140, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0,
304 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0,
305 0x459, 0x459, 0x459, 0x459, 0x459, 0x459, 0x459, 0x459,
306 0x459, 0x459, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0,
307 0x0d0, 0x653, 0x653, 0x653, 0x653, 0x653, 0x653, 0x253,
308 0x253, 0x253, 0x253, 0x253, 0x253, 0x253, 0x253, 0x253,
309 0x253, 0x253, 0x253, 0x253, 0x253, 0x253, 0x253, 0x253,
310 0x253, 0x253, 0x253, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x0d0,
311 0x0d0, 0x473, 0x473, 0x473, 0x473, 0x473, 0x473, 0x073,
312 0x073, 0x073, 0x073, 0x073, 0x073, 0x073, 0x073, 0x073,
313 0x073, 0x073, 0x073, 0x073, 0x073, 0x073, 0x073, 0x073,
314 0x073, 0x073, 0x073, 0x0d0, 0x0d0, 0x0d0, 0x0d0, 0x004
315 /* the upper 128 are all zeroes */
318 const guint16 * const g_ascii_table = ascii_table_data;
320 #ifdef HAVE_NEWLOCALE
324 static gsize initialized = FALSE;
325 static locale_t C_locale = NULL;
327 if (g_once_init_enter (&initialized))
329 C_locale = newlocale (LC_ALL_MASK, "C", NULL);
330 g_once_init_leave (&initialized, TRUE);
339 * @str: the string to duplicate
341 * Duplicates a string. If @str is %NULL it returns %NULL.
342 * The returned string should be freed with g_free()
343 * when no longer needed.
345 * Returns: a newly-allocated copy of @str
348 g_strdup (const gchar *str)
355 length = strlen (str) + 1;
356 new_str = g_new (char, length);
357 memcpy (new_str, str, length);
367 * @mem: the memory to copy.
368 * @byte_size: the number of bytes to copy.
370 * Allocates @byte_size bytes of memory, and copies @byte_size bytes into it
371 * from @mem. If @mem is %NULL it returns %NULL.
373 * Returns: a pointer to the newly-allocated copy of the memory, or %NULL if @mem
377 g_memdup (gconstpointer mem,
384 new_mem = g_malloc (byte_size);
385 memcpy (new_mem, mem, byte_size);
395 * @str: the string to duplicate
396 * @n: the maximum number of bytes to copy from @str
398 * Duplicates the first @n bytes of a string, returning a newly-allocated
399 * buffer @n + 1 bytes long which will always be nul-terminated.
400 * If @str is less than @n bytes long the buffer is padded with nuls.
401 * If @str is %NULL it returns %NULL.
402 * The returned value should be freed when no longer needed.
405 * To copy a number of characters from a UTF-8 encoded string, use
406 * g_utf8_strncpy() instead.
409 * Returns: a newly-allocated buffer containing the first @n bytes
410 * of @str, nul-terminated
413 g_strndup (const gchar *str,
420 new_str = g_new (gchar, n + 1);
421 strncpy (new_str, str, n);
432 * @length: the length of the new string
433 * @fill_char: the byte to fill the string with
435 * Creates a new string @length bytes long filled with @fill_char.
436 * The returned string should be freed when no longer needed.
438 * Returns: a newly-allocated string filled the @fill_char
441 g_strnfill (gsize length,
446 str = g_new (gchar, length + 1);
447 memset (str, (guchar)fill_char, length);
455 * @dest: destination buffer.
456 * @src: source string.
458 * Copies a nul-terminated string into the dest buffer, include the
459 * trailing nul, and return a pointer to the trailing nul byte.
460 * This is useful for concatenating multiple strings together
461 * without having to repeatedly scan for the end.
463 * Return value: a pointer to trailing nul byte.
466 g_stpcpy (gchar *dest,
470 g_return_val_if_fail (dest != NULL, NULL);
471 g_return_val_if_fail (src != NULL, NULL);
472 return stpcpy (dest, src);
474 register gchar *d = dest;
475 register const gchar *s = src;
477 g_return_val_if_fail (dest != NULL, NULL);
478 g_return_val_if_fail (src != NULL, NULL);
481 while (*s++ != '\0');
489 * @format: a standard printf() format string, but notice
490 * <link linkend="string-precision">string precision pitfalls</link>
491 * @args: the list of parameters to insert into the format string
493 * Similar to the standard C vsprintf() function but safer, since it
494 * calculates the maximum space required and allocates memory to hold
495 * the result. The returned string should be freed with g_free() when
498 * See also g_vasprintf(), which offers the same functionality, but
499 * additionally returns the length of the allocated string.
501 * Returns: a newly-allocated string holding the result
504 g_strdup_vprintf (const gchar *format,
507 gchar *string = NULL;
509 g_vasprintf (&string, format, args);
516 * @format: a standard printf() format string, but notice
517 * <link linkend="string-precision">string precision pitfalls</link>
518 * @...: the parameters to insert into the format string
520 * Similar to the standard C sprintf() function but safer, since it
521 * calculates the maximum space required and allocates memory to hold
522 * the result. The returned string should be freed with g_free() when no
525 * Returns: a newly-allocated string holding the result
528 g_strdup_printf (const gchar *format,
534 va_start (args, format);
535 buffer = g_strdup_vprintf (format, args);
543 * @string1: the first string to add, which must not be %NULL
544 * @...: a %NULL-terminated list of strings to append to the string
546 * Concatenates all of the given strings into one long string.
547 * The returned string should be freed with g_free() when no longer needed.
549 * Note that this function is usually not the right function to use to
550 * assemble a translated message from pieces, since proper translation
551 * often requires the pieces to be reordered.
553 * <warning><para>The variable argument list <emphasis>must</emphasis> end
554 * with %NULL. If you forget the %NULL, g_strconcat() will start appending
555 * random memory junk to your string.</para></warning>
557 * Returns: a newly-allocated string containing all the string arguments
560 g_strconcat (const gchar *string1, ...)
571 l = 1 + strlen (string1);
572 va_start (args, string1);
573 s = va_arg (args, gchar*);
577 s = va_arg (args, gchar*);
581 concat = g_new (gchar, l);
584 ptr = g_stpcpy (ptr, string1);
585 va_start (args, string1);
586 s = va_arg (args, gchar*);
589 ptr = g_stpcpy (ptr, s);
590 s = va_arg (args, gchar*);
599 * @nptr: the string to convert to a numeric value.
600 * @endptr: if non-%NULL, it returns the character after
601 * the last character used in the conversion.
603 * Converts a string to a #gdouble value.
604 * It calls the standard strtod() function to handle the conversion, but
605 * if the string is not completely converted it attempts the conversion
606 * again with g_ascii_strtod(), and returns the best match.
608 * This function should seldom be used. The normal situation when reading
609 * numbers not for human consumption is to use g_ascii_strtod(). Only when
610 * you know that you must expect both locale formatted and C formatted numbers
611 * should you use this. Make sure that you don't pass strings such as comma
612 * separated lists of values, since the commas may be interpreted as a decimal
613 * point in some locales, causing unexpected results.
615 * Return value: the #gdouble value.
618 g_strtod (const gchar *nptr,
626 g_return_val_if_fail (nptr != NULL, 0);
631 val_1 = strtod (nptr, &fail_pos_1);
633 if (fail_pos_1 && fail_pos_1[0] != 0)
634 val_2 = g_ascii_strtod (nptr, &fail_pos_2);
636 if (!fail_pos_1 || fail_pos_1[0] == 0 || fail_pos_1 >= fail_pos_2)
639 *endptr = fail_pos_1;
645 *endptr = fail_pos_2;
652 * @nptr: the string to convert to a numeric value.
653 * @endptr: if non-%NULL, it returns the character after
654 * the last character used in the conversion.
656 * Converts a string to a #gdouble value.
658 * This function behaves like the standard strtod() function
659 * does in the C locale. It does this without actually changing
660 * the current locale, since that would not be thread-safe.
661 * A limitation of the implementation is that this function
662 * will still accept localized versions of infinities and NANs.
664 * This function is typically used when reading configuration
665 * files or other non-user input that should be locale independent.
666 * To handle input from the user you should normally use the
667 * locale-sensitive system strtod() function.
669 * To convert from a #gdouble to a string in a locale-insensitive
670 * way, use g_ascii_dtostr().
672 * If the correct value would cause overflow, plus or minus %HUGE_VAL
673 * is returned (according to the sign of the value), and %ERANGE is
674 * stored in %errno. If the correct value would cause underflow,
675 * zero is returned and %ERANGE is stored in %errno.
677 * This function resets %errno before calling strtod() so that
678 * you can reliably detect overflow and underflow.
680 * Return value: the #gdouble value.
683 g_ascii_strtod (const gchar *nptr,
688 struct lconv *locale_data;
689 const char *decimal_point;
690 int decimal_point_len;
691 const char *p, *decimal_point_pos;
692 const char *end = NULL; /* Silence gcc */
695 g_return_val_if_fail (nptr != NULL, 0);
700 return strtod_l (nptr, endptr, get_C_locale ());
705 locale_data = localeconv ();
706 decimal_point = locale_data->decimal_point;
707 decimal_point_len = strlen (decimal_point);
709 g_assert (decimal_point_len != 0);
711 decimal_point_pos = NULL;
714 if (decimal_point[0] != '.' ||
715 decimal_point[1] != 0)
718 /* Skip leading space */
719 while (g_ascii_isspace (*p))
722 /* Skip leading optional sign */
723 if (*p == '+' || *p == '-')
727 (p[1] == 'x' || p[1] == 'X'))
730 /* HEX - find the (optional) decimal point */
732 while (g_ascii_isxdigit (*p))
736 decimal_point_pos = p++;
738 while (g_ascii_isxdigit (*p))
741 if (*p == 'p' || *p == 'P')
743 if (*p == '+' || *p == '-')
745 while (g_ascii_isdigit (*p))
750 else if (g_ascii_isdigit (*p) || *p == '.')
752 while (g_ascii_isdigit (*p))
756 decimal_point_pos = p++;
758 while (g_ascii_isdigit (*p))
761 if (*p == 'e' || *p == 'E')
763 if (*p == '+' || *p == '-')
765 while (g_ascii_isdigit (*p))
770 /* For the other cases, we need not convert the decimal point */
773 if (decimal_point_pos)
777 /* We need to convert the '.' to the locale specific decimal point */
778 copy = g_malloc (end - nptr + 1 + decimal_point_len);
781 memcpy (c, nptr, decimal_point_pos - nptr);
782 c += decimal_point_pos - nptr;
783 memcpy (c, decimal_point, decimal_point_len);
784 c += decimal_point_len;
785 memcpy (c, decimal_point_pos + 1, end - (decimal_point_pos + 1));
786 c += end - (decimal_point_pos + 1);
790 val = strtod (copy, &fail_pos);
791 strtod_errno = errno;
795 if (fail_pos - copy > decimal_point_pos - nptr)
796 fail_pos = (char *)nptr + (fail_pos - copy) - (decimal_point_len - 1);
798 fail_pos = (char *)nptr + (fail_pos - copy);
808 copy = g_malloc (end - (char *)nptr + 1);
809 memcpy (copy, nptr, end - nptr);
810 *(copy + (end - (char *)nptr)) = 0;
813 val = strtod (copy, &fail_pos);
814 strtod_errno = errno;
818 fail_pos = (char *)nptr + (fail_pos - copy);
826 val = strtod (nptr, &fail_pos);
827 strtod_errno = errno;
833 errno = strtod_errno;
842 * @buffer: A buffer to place the resulting string in
843 * @buf_len: The length of the buffer.
844 * @d: The #gdouble to convert
846 * Converts a #gdouble to a string, using the '.' as
849 * This functions generates enough precision that converting
850 * the string back using g_ascii_strtod() gives the same machine-number
851 * (on machines with IEEE compatible 64bit doubles). It is
852 * guaranteed that the size of the resulting string will never
853 * be larger than @G_ASCII_DTOSTR_BUF_SIZE bytes.
855 * Return value: The pointer to the buffer with the converted string.
858 g_ascii_dtostr (gchar *buffer,
862 return g_ascii_formatd (buffer, buf_len, "%.17g", d);
867 * @buffer: A buffer to place the resulting string in
868 * @buf_len: The length of the buffer.
869 * @format: The printf()-style format to use for the
870 * code to use for converting.
871 * @d: The #gdouble to convert
873 * Converts a #gdouble to a string, using the '.' as
874 * decimal point. To format the number you pass in
875 * a printf()-style format string. Allowed conversion
876 * specifiers are 'e', 'E', 'f', 'F', 'g' and 'G'.
878 * If you just want to want to serialize the value into a
879 * string, use g_ascii_dtostr().
881 * Return value: The pointer to the buffer with the converted string.
884 g_ascii_formatd (gchar *buffer,
889 #ifdef HAVE_USELOCALE
892 old_locale = uselocale (get_C_locale ());
893 _g_snprintf (buffer, buf_len, format, d);
894 uselocale (old_locale);
898 struct lconv *locale_data;
899 const char *decimal_point;
900 int decimal_point_len;
905 g_return_val_if_fail (buffer != NULL, NULL);
906 g_return_val_if_fail (format[0] == '%', NULL);
907 g_return_val_if_fail (strpbrk (format + 1, "'l%") == NULL, NULL);
909 format_char = format[strlen (format) - 1];
911 g_return_val_if_fail (format_char == 'e' || format_char == 'E' ||
912 format_char == 'f' || format_char == 'F' ||
913 format_char == 'g' || format_char == 'G',
916 if (format[0] != '%')
919 if (strpbrk (format + 1, "'l%"))
922 if (!(format_char == 'e' || format_char == 'E' ||
923 format_char == 'f' || format_char == 'F' ||
924 format_char == 'g' || format_char == 'G'))
927 _g_snprintf (buffer, buf_len, format, d);
929 locale_data = localeconv ();
930 decimal_point = locale_data->decimal_point;
931 decimal_point_len = strlen (decimal_point);
933 g_assert (decimal_point_len != 0);
935 if (decimal_point[0] != '.' ||
936 decimal_point[1] != 0)
940 while (g_ascii_isspace (*p))
943 if (*p == '+' || *p == '-')
946 while (isdigit ((guchar)*p))
949 if (strncmp (p, decimal_point, decimal_point_len) == 0)
953 if (decimal_point_len > 1)
955 rest_len = strlen (p + (decimal_point_len-1));
956 memmove (p, p + (decimal_point_len-1), rest_len);
966 #define ISSPACE(c) ((c) == ' ' || (c) == '\f' || (c) == '\n' || \
967 (c) == '\r' || (c) == '\t' || (c) == '\v')
968 #define ISUPPER(c) ((c) >= 'A' && (c) <= 'Z')
969 #define ISLOWER(c) ((c) >= 'a' && (c) <= 'z')
970 #define ISALPHA(c) (ISUPPER (c) || ISLOWER (c))
971 #define TOUPPER(c) (ISLOWER (c) ? (c) - 'a' + 'A' : (c))
972 #define TOLOWER(c) (ISUPPER (c) ? (c) - 'A' + 'a' : (c))
974 #if !defined(HAVE_STRTOLL_L) || !defined(HAVE_STRTOULL_L)
977 g_parse_long_long (const gchar *nptr,
978 const gchar **endptr,
982 /* this code is based on on the strtol(3) code from GNU libc released under
983 * the GNU Lesser General Public License.
985 * Copyright (C) 1991,92,94,95,96,97,98,99,2000,01,02
986 * Free Software Foundation, Inc.
992 const gchar *s, *save;
995 g_return_val_if_fail (nptr != NULL, 0);
998 if (base == 1 || base > 36)
1008 /* Skip white space. */
1009 while (ISSPACE (*s))
1012 if (G_UNLIKELY (!*s))
1015 /* Check for a sign. */
1024 /* Recognize number prefix and if BASE is zero, figure it out ourselves. */
1027 if ((base == 0 || base == 16) && TOUPPER (s[1]) == 'X')
1038 /* Save the pointer so we can check later if anything happened. */
1040 cutoff = G_MAXUINT64 / base;
1041 cutlim = G_MAXUINT64 % base;
1048 if (c >= '0' && c <= '9')
1050 else if (ISALPHA (c))
1051 c = TOUPPER (c) - 'A' + 10;
1056 /* Check for overflow. */
1057 if (ui64 > cutoff || (ui64 == cutoff && c > cutlim))
1066 /* Check if anything actually happened. */
1070 /* Store in ENDPTR the address of one character
1071 past the last character we converted. */
1075 if (G_UNLIKELY (overflow))
1084 /* We must handle a special case here: the base is 0 or 16 and the
1085 first two characters are '0' and 'x', but the rest are no
1086 hexadecimal digits. This is no error case. We return 0 and
1087 ENDPTR points to the `x`. */
1090 if (save - nptr >= 2 && TOUPPER (save[-1]) == 'X'
1092 *endptr = &save[-1];
1094 /* There was no number to convert. */
1103 * @nptr: the string to convert to a numeric value.
1104 * @endptr: if non-%NULL, it returns the character after
1105 * the last character used in the conversion.
1106 * @base: to be used for the conversion, 2..36 or 0
1108 * Converts a string to a #guint64 value.
1109 * This function behaves like the standard strtoull() function
1110 * does in the C locale. It does this without actually
1111 * changing the current locale, since that would not be
1114 * This function is typically used when reading configuration
1115 * files or other non-user input that should be locale independent.
1116 * To handle input from the user you should normally use the
1117 * locale-sensitive system strtoull() function.
1119 * If the correct value would cause overflow, %G_MAXUINT64
1120 * is returned, and %ERANGE is stored in %errno. If the base is
1121 * outside the valid range, zero is returned, and %EINVAL is stored
1122 * in %errno. If the string conversion fails, zero is returned, and
1123 * @endptr returns @nptr (if @endptr is non-%NULL).
1125 * Return value: the #guint64 value or zero on error.
1130 g_ascii_strtoull (const gchar *nptr,
1134 #ifdef HAVE_STRTOULL_L
1135 return strtoull_l (nptr, endptr, base, get_C_locale ());
1140 result = g_parse_long_long (nptr, (const gchar **) endptr, base, &negative);
1142 /* Return the result of the appropriate sign. */
1143 return negative ? -result : result;
1149 * @nptr: the string to convert to a numeric value.
1150 * @endptr: if non-%NULL, it returns the character after
1151 * the last character used in the conversion.
1152 * @base: to be used for the conversion, 2..36 or 0
1154 * Converts a string to a #gint64 value.
1155 * This function behaves like the standard strtoll() function
1156 * does in the C locale. It does this without actually
1157 * changing the current locale, since that would not be
1160 * This function is typically used when reading configuration
1161 * files or other non-user input that should be locale independent.
1162 * To handle input from the user you should normally use the
1163 * locale-sensitive system strtoll() function.
1165 * If the correct value would cause overflow, %G_MAXINT64 or %G_MININT64
1166 * is returned, and %ERANGE is stored in %errno. If the base is
1167 * outside the valid range, zero is returned, and %EINVAL is stored
1168 * in %errno. If the string conversion fails, zero is returned, and
1169 * @endptr returns @nptr (if @endptr is non-%NULL).
1171 * Return value: the #gint64 value or zero on error.
1176 g_ascii_strtoll (const gchar *nptr,
1180 #ifdef HAVE_STRTOLL_L
1181 return strtoll_l (nptr, endptr, base, get_C_locale ());
1186 result = g_parse_long_long (nptr, (const gchar **) endptr, base, &negative);
1188 if (negative && result > (guint64) G_MININT64)
1193 else if (!negative && result > (guint64) G_MAXINT64)
1199 return - (gint64) result;
1201 return (gint64) result;
1207 * @errnum: the system error number. See the standard C %errno
1210 * Returns a string corresponding to the given error code, e.g.
1211 * "no such process". You should use this function in preference to
1212 * strerror(), because it returns a string in UTF-8 encoding, and since
1213 * not all platforms support the strerror() function.
1215 * Returns: a UTF-8 string describing the error code. If the error code
1216 * is unknown, it returns "unknown error (<code>)".
1219 g_strerror (gint errnum)
1225 gint saved_errno = errno;
1227 msg = tofree = NULL;
1229 #ifdef HAVE_STRERROR
1230 msg = strerror (errnum);
1231 if (!g_get_charset (NULL))
1232 msg = tofree = g_locale_to_utf8 (msg, -1, NULL, NULL, NULL);
1238 _g_sprintf (msg, "unknown error (%d)", errnum);
1241 ret = g_intern_string (msg);
1243 errno = saved_errno;
1249 * @signum: the signal number. See the <literal>signal</literal>
1252 * Returns a string describing the given signal, e.g. "Segmentation fault".
1253 * You should use this function in preference to strsignal(), because it
1254 * returns a string in UTF-8 encoding, and since not all platforms support
1255 * the strsignal() function.
1257 * Returns: a UTF-8 string describing the signal. If the signal is unknown,
1258 * it returns "unknown signal (<signum>)".
1261 g_strsignal (gint signum)
1267 msg = tofree = NULL;
1269 #ifdef HAVE_STRSIGNAL
1270 msg = strsignal (signum);
1271 if (!g_get_charset (NULL))
1272 msg = tofree = g_locale_to_utf8 (msg, -1, NULL, NULL, NULL);
1276 msg = tofree = g_strdup_printf ("unknown signal (%d)", signum);
1277 ret = g_intern_string (msg);
1283 /* Functions g_strlcpy and g_strlcat were originally developed by
1284 * Todd C. Miller <Todd.Miller@courtesan.com> to simplify writing secure code.
1285 * See http://www.openbsd.org/cgi-bin/man.cgi?query=strlcpy
1286 * for more information.
1290 /* Use the native ones, if available; they might be implemented in assembly */
1292 g_strlcpy (gchar *dest,
1296 g_return_val_if_fail (dest != NULL, 0);
1297 g_return_val_if_fail (src != NULL, 0);
1299 return strlcpy (dest, src, dest_size);
1303 g_strlcat (gchar *dest,
1307 g_return_val_if_fail (dest != NULL, 0);
1308 g_return_val_if_fail (src != NULL, 0);
1310 return strlcat (dest, src, dest_size);
1313 #else /* ! HAVE_STRLCPY */
1316 * @dest: destination buffer
1317 * @src: source buffer
1318 * @dest_size: length of @dest in bytes
1320 * Portability wrapper that calls strlcpy() on systems which have it,
1321 * and emulates strlcpy() otherwise. Copies @src to @dest; @dest is
1322 * guaranteed to be nul-terminated; @src must be nul-terminated;
1323 * @dest_size is the buffer size, not the number of chars to copy.
1325 * At most dest_size - 1 characters will be copied. Always nul-terminates
1326 * (unless dest_size == 0). This function does <emphasis>not</emphasis>
1327 * allocate memory. Unlike strncpy(), this function doesn't pad dest (so
1328 * it's often faster). It returns the size of the attempted result,
1329 * strlen (src), so if @retval >= @dest_size, truncation occurred.
1331 * <note><para>Caveat: strlcpy() is supposedly more secure than
1332 * strcpy() or strncpy(), but if you really want to avoid screwups,
1333 * g_strdup() is an even better idea.</para></note>
1335 * Returns: length of @src
1338 g_strlcpy (gchar *dest,
1342 register gchar *d = dest;
1343 register const gchar *s = src;
1344 register gsize n = dest_size;
1346 g_return_val_if_fail (dest != NULL, 0);
1347 g_return_val_if_fail (src != NULL, 0);
1349 /* Copy as many bytes as will fit */
1350 if (n != 0 && --n != 0)
1353 register gchar c = *s++;
1361 /* If not enough room in dest, add NUL and traverse rest of src */
1370 return s - src - 1; /* count does not include NUL */
1375 * @dest: destination buffer, already containing one nul-terminated string
1376 * @src: source buffer
1377 * @dest_size: length of @dest buffer in bytes (not length of existing string
1380 * Portability wrapper that calls strlcat() on systems which have it,
1381 * and emulates it otherwise. Appends nul-terminated @src string to @dest,
1382 * guaranteeing nul-termination for @dest. The total size of @dest won't
1383 * exceed @dest_size.
1385 * At most dest_size - 1 characters will be copied.
1386 * Unlike strncat, dest_size is the full size of dest, not the space left over.
1387 * This function does NOT allocate memory.
1388 * This always NUL terminates (unless siz == 0 or there were no NUL characters
1389 * in the dest_size characters of dest to start with).
1391 * <note><para>Caveat: this is supposedly a more secure alternative to
1392 * strcat() or strncat(), but for real security g_strconcat() is harder
1393 * to mess up.</para></note>
1395 * Returns: size of attempted result, which is MIN (dest_size, strlen
1396 * (original dest)) + strlen (src), so if retval >= dest_size,
1397 * truncation occurred.
1400 g_strlcat (gchar *dest,
1404 register gchar *d = dest;
1405 register const gchar *s = src;
1406 register gsize bytes_left = dest_size;
1407 gsize dlength; /* Logically, MIN (strlen (d), dest_size) */
1409 g_return_val_if_fail (dest != NULL, 0);
1410 g_return_val_if_fail (src != NULL, 0);
1412 /* Find the end of dst and adjust bytes left but don't go past end */
1413 while (*d != 0 && bytes_left-- != 0)
1416 bytes_left = dest_size - dlength;
1418 if (bytes_left == 0)
1419 return dlength + strlen (s);
1423 if (bytes_left != 1)
1432 return dlength + (s - src); /* count does not include NUL */
1434 #endif /* ! HAVE_STRLCPY */
1439 * @len: length of @str in bytes, or -1 if @str is nul-terminated.
1441 * Converts all upper case ASCII letters to lower case ASCII letters.
1443 * Return value: a newly-allocated string, with all the upper case
1444 * characters in @str converted to lower case, with
1445 * semantics that exactly match g_ascii_tolower(). (Note
1446 * that this is unlike the old g_strdown(), which modified
1447 * the string in place.)
1450 g_ascii_strdown (const gchar *str,
1455 g_return_val_if_fail (str != NULL, NULL);
1460 result = g_strndup (str, len);
1461 for (s = result; *s; s++)
1462 *s = g_ascii_tolower (*s);
1470 * @len: length of @str in bytes, or -1 if @str is nul-terminated.
1472 * Converts all lower case ASCII letters to upper case ASCII letters.
1474 * Return value: a newly allocated string, with all the lower case
1475 * characters in @str converted to upper case, with
1476 * semantics that exactly match g_ascii_toupper(). (Note
1477 * that this is unlike the old g_strup(), which modified
1478 * the string in place.)
1481 g_ascii_strup (const gchar *str,
1486 g_return_val_if_fail (str != NULL, NULL);
1491 result = g_strndup (str, len);
1492 for (s = result; *s; s++)
1493 *s = g_ascii_toupper (*s);
1500 * @string: the string to convert.
1502 * Converts a string to lower case.
1504 * Return value: the string
1506 * Deprecated:2.2: This function is totally broken for the reasons discussed
1507 * in the g_strncasecmp() docs - use g_ascii_strdown() or g_utf8_strdown()
1511 g_strdown (gchar *string)
1515 g_return_val_if_fail (string != NULL, NULL);
1517 s = (guchar *) string;
1526 return (gchar *) string;
1531 * @string: the string to convert.
1533 * Converts a string to upper case.
1535 * Return value: the string
1537 * Deprecated:2.2: This function is totally broken for the reasons discussed
1538 * in the g_strncasecmp() docs - use g_ascii_strup() or g_utf8_strup() instead.
1541 g_strup (gchar *string)
1545 g_return_val_if_fail (string != NULL, NULL);
1547 s = (guchar *) string;
1556 return (gchar *) string;
1561 * @string: the string to reverse
1563 * Reverses all of the bytes in a string. For example,
1564 * <literal>g_strreverse ("abcdef")</literal> will result
1567 * Note that g_strreverse() doesn't work on UTF-8 strings
1568 * containing multibyte characters. For that purpose, use
1569 * g_utf8_strreverse().
1571 * Returns: the same pointer passed in as @string
1574 g_strreverse (gchar *string)
1576 g_return_val_if_fail (string != NULL, NULL);
1580 register gchar *h, *t;
1583 t = string + strlen (string) - 1;
1602 * @c: any character.
1604 * Convert a character to ASCII lower case.
1606 * Unlike the standard C library tolower() function, this only
1607 * recognizes standard ASCII letters and ignores the locale, returning
1608 * all non-ASCII characters unchanged, even if they are lower case
1609 * letters in a particular character set. Also unlike the standard
1610 * library function, this takes and returns a char, not an int, so
1611 * don't call it on %EOF but no need to worry about casting to #guchar
1612 * before passing a possibly non-ASCII character in.
1614 * Return value: the result of converting @c to lower case.
1615 * If @c is not an ASCII upper case letter,
1616 * @c is returned unchanged.
1619 g_ascii_tolower (gchar c)
1621 return g_ascii_isupper (c) ? c - 'A' + 'a' : c;
1626 * @c: any character.
1628 * Convert a character to ASCII upper case.
1630 * Unlike the standard C library toupper() function, this only
1631 * recognizes standard ASCII letters and ignores the locale, returning
1632 * all non-ASCII characters unchanged, even if they are upper case
1633 * letters in a particular character set. Also unlike the standard
1634 * library function, this takes and returns a char, not an int, so
1635 * don't call it on %EOF but no need to worry about casting to #guchar
1636 * before passing a possibly non-ASCII character in.
1638 * Return value: the result of converting @c to upper case.
1639 * If @c is not an ASCII lower case letter,
1640 * @c is returned unchanged.
1643 g_ascii_toupper (gchar c)
1645 return g_ascii_islower (c) ? c - 'a' + 'A' : c;
1649 * g_ascii_digit_value:
1650 * @c: an ASCII character.
1652 * Determines the numeric value of a character as a decimal
1653 * digit. Differs from g_unichar_digit_value() because it takes
1654 * a char, so there's no worry about sign extension if characters
1657 * Return value: If @c is a decimal digit (according to
1658 * g_ascii_isdigit()), its numeric value. Otherwise, -1.
1661 g_ascii_digit_value (gchar c)
1663 if (g_ascii_isdigit (c))
1669 * g_ascii_xdigit_value:
1670 * @c: an ASCII character.
1672 * Determines the numeric value of a character as a hexidecimal
1673 * digit. Differs from g_unichar_xdigit_value() because it takes
1674 * a char, so there's no worry about sign extension if characters
1677 * Return value: If @c is a hex digit (according to
1678 * g_ascii_isxdigit()), its numeric value. Otherwise, -1.
1681 g_ascii_xdigit_value (gchar c)
1683 if (c >= 'A' && c <= 'F')
1684 return c - 'A' + 10;
1685 if (c >= 'a' && c <= 'f')
1686 return c - 'a' + 10;
1687 return g_ascii_digit_value (c);
1691 * g_ascii_strcasecmp:
1692 * @s1: string to compare with @s2.
1693 * @s2: string to compare with @s1.
1695 * Compare two strings, ignoring the case of ASCII characters.
1697 * Unlike the BSD strcasecmp() function, this only recognizes standard
1698 * ASCII letters and ignores the locale, treating all non-ASCII
1699 * bytes as if they are not letters.
1701 * This function should be used only on strings that are known to be
1702 * in encodings where the bytes corresponding to ASCII letters always
1703 * represent themselves. This includes UTF-8 and the ISO-8859-*
1704 * charsets, but not for instance double-byte encodings like the
1705 * Windows Codepage 932, where the trailing bytes of double-byte
1706 * characters include all ASCII letters. If you compare two CP932
1707 * strings using this function, you will get false matches.
1709 * Return value: 0 if the strings match, a negative value if @s1 < @s2,
1710 * or a positive value if @s1 > @s2.
1713 g_ascii_strcasecmp (const gchar *s1,
1718 g_return_val_if_fail (s1 != NULL, 0);
1719 g_return_val_if_fail (s2 != NULL, 0);
1723 c1 = (gint)(guchar) TOLOWER (*s1);
1724 c2 = (gint)(guchar) TOLOWER (*s2);
1730 return (((gint)(guchar) *s1) - ((gint)(guchar) *s2));
1734 * g_ascii_strncasecmp:
1735 * @s1: string to compare with @s2.
1736 * @s2: string to compare with @s1.
1737 * @n: number of characters to compare.
1739 * Compare @s1 and @s2, ignoring the case of ASCII characters and any
1740 * characters after the first @n in each string.
1742 * Unlike the BSD strcasecmp() function, this only recognizes standard
1743 * ASCII letters and ignores the locale, treating all non-ASCII
1744 * characters as if they are not letters.
1746 * The same warning as in g_ascii_strcasecmp() applies: Use this
1747 * function only on strings known to be in encodings where bytes
1748 * corresponding to ASCII letters always represent themselves.
1750 * Return value: 0 if the strings match, a negative value if @s1 < @s2,
1751 * or a positive value if @s1 > @s2.
1754 g_ascii_strncasecmp (const gchar *s1,
1760 g_return_val_if_fail (s1 != NULL, 0);
1761 g_return_val_if_fail (s2 != NULL, 0);
1763 while (n && *s1 && *s2)
1766 c1 = (gint)(guchar) TOLOWER (*s1);
1767 c2 = (gint)(guchar) TOLOWER (*s2);
1774 return (((gint) (guchar) *s1) - ((gint) (guchar) *s2));
1782 * @s2: a string to compare with @s1.
1784 * A case-insensitive string comparison, corresponding to the standard
1785 * strcasecmp() function on platforms which support it.
1787 * Return value: 0 if the strings match, a negative value if @s1 < @s2,
1788 * or a positive value if @s1 > @s2.
1790 * Deprecated:2.2: See g_strncasecmp() for a discussion of why this function
1791 * is deprecated and how to replace it.
1794 g_strcasecmp (const gchar *s1,
1797 #ifdef HAVE_STRCASECMP
1798 g_return_val_if_fail (s1 != NULL, 0);
1799 g_return_val_if_fail (s2 != NULL, 0);
1801 return strcasecmp (s1, s2);
1805 g_return_val_if_fail (s1 != NULL, 0);
1806 g_return_val_if_fail (s2 != NULL, 0);
1810 /* According to A. Cox, some platforms have islower's that
1811 * don't work right on non-uppercase
1813 c1 = isupper ((guchar)*s1) ? tolower ((guchar)*s1) : *s1;
1814 c2 = isupper ((guchar)*s2) ? tolower ((guchar)*s2) : *s2;
1820 return (((gint)(guchar) *s1) - ((gint)(guchar) *s2));
1827 * @s2: a string to compare with @s1.
1828 * @n: the maximum number of characters to compare.
1830 * A case-insensitive string comparison, corresponding to the standard
1831 * strncasecmp() function on platforms which support it.
1832 * It is similar to g_strcasecmp() except it only compares the first @n
1833 * characters of the strings.
1835 * Return value: 0 if the strings match, a negative value if @s1 < @s2,
1836 * or a positive value if @s1 > @s2.
1838 * Deprecated:2.2: The problem with g_strncasecmp() is that it does the
1839 * comparison by calling toupper()/tolower(). These functions are
1840 * locale-specific and operate on single bytes. However, it is impossible
1841 * to handle things correctly from an I18N standpoint by operating on
1842 * bytes, since characters may be multibyte. Thus g_strncasecmp() is
1843 * broken if your string is guaranteed to be ASCII, since it's
1844 * locale-sensitive, and it's broken if your string is localized, since
1845 * it doesn't work on many encodings at all, including UTF-8, EUC-JP,
1848 * There are therefore two replacement functions: g_ascii_strncasecmp(),
1849 * which only works on ASCII and is not locale-sensitive, and
1850 * g_utf8_casefold(), which is good for case-insensitive sorting of UTF-8.
1853 g_strncasecmp (const gchar *s1,
1857 #ifdef HAVE_STRNCASECMP
1858 return strncasecmp (s1, s2, n);
1862 g_return_val_if_fail (s1 != NULL, 0);
1863 g_return_val_if_fail (s2 != NULL, 0);
1865 while (n && *s1 && *s2)
1868 /* According to A. Cox, some platforms have islower's that
1869 * don't work right on non-uppercase
1871 c1 = isupper ((guchar)*s1) ? tolower ((guchar)*s1) : *s1;
1872 c2 = isupper ((guchar)*s2) ? tolower ((guchar)*s2) : *s2;
1879 return (((gint) (guchar) *s1) - ((gint) (guchar) *s2));
1887 * @string: the string to convert
1888 * @delimiters: a string containing the current delimiters, or %NULL
1889 * to use the standard delimiters defined in #G_STR_DELIMITERS
1890 * @new_delimiter: the new delimiter character
1892 * Converts any delimiter characters in @string to @new_delimiter.
1893 * Any characters in @string which are found in @delimiters are
1894 * changed to the @new_delimiter character. Modifies @string in place,
1895 * and returns @string itself, not a copy. The return value is to
1896 * allow nesting such as
1898 * g_ascii_strup (g_strdelimit (str, "abc", '?'))
1904 g_strdelimit (gchar *string,
1905 const gchar *delimiters,
1910 g_return_val_if_fail (string != NULL, NULL);
1913 delimiters = G_STR_DELIMITERS;
1915 for (c = string; *c; c++)
1917 if (strchr (delimiters, *c))
1926 * @string: a nul-terminated array of bytes
1927 * @valid_chars: bytes permitted in @string
1928 * @substitutor: replacement character for disallowed bytes
1930 * For each character in @string, if the character is not in
1931 * @valid_chars, replaces the character with @substitutor.
1932 * Modifies @string in place, and return @string itself, not
1933 * a copy. The return value is to allow nesting such as
1935 * g_ascii_strup (g_strcanon (str, "abc", '?'))
1941 g_strcanon (gchar *string,
1942 const gchar *valid_chars,
1947 g_return_val_if_fail (string != NULL, NULL);
1948 g_return_val_if_fail (valid_chars != NULL, NULL);
1950 for (c = string; *c; c++)
1952 if (!strchr (valid_chars, *c))
1961 * @source: a string to compress
1963 * Replaces all escaped characters with their one byte equivalent.
1965 * This function does the reverse conversion of g_strescape().
1967 * Returns: a newly-allocated copy of @source with all escaped
1968 * character compressed
1971 g_strcompress (const gchar *source)
1973 const gchar *p = source, *octal;
1974 gchar *dest = g_malloc (strlen (source) + 1);
1985 g_warning ("g_strcompress: trailing \\");
1987 case '0': case '1': case '2': case '3': case '4':
1988 case '5': case '6': case '7':
1991 while ((p < octal + 3) && (*p >= '0') && (*p <= '7'))
1993 *q = (*q * 8) + (*p - '0');
2014 default: /* Also handles \" and \\ */
2031 * @source: a string to escape
2032 * @exceptions: a string of characters not to escape in @source
2034 * Escapes the special characters '\b', '\f', '\n', '\r', '\t', '\'
2035 * and '"' in the string @source by inserting a '\' before
2036 * them. Additionally all characters in the range 0x01-0x1F (everything
2037 * below SPACE) and in the range 0x7F-0xFF (all non-ASCII chars) are
2038 * replaced with a '\' followed by their octal representation.
2039 * Characters supplied in @exceptions are not escaped.
2041 * g_strcompress() does the reverse conversion.
2043 * Returns: a newly-allocated copy of @source with certain
2044 * characters escaped. See above.
2047 g_strescape (const gchar *source,
2048 const gchar *exceptions)
2055 g_return_val_if_fail (source != NULL, NULL);
2057 p = (guchar *) source;
2058 /* Each source byte needs maximally four destination chars (\777) */
2059 q = dest = g_malloc (strlen (source) * 4 + 1);
2061 memset (excmap, 0, 256);
2064 guchar *e = (guchar *) exceptions;
2110 if ((*p < ' ') || (*p >= 0177))
2113 *q++ = '0' + (((*p) >> 6) & 07);
2114 *q++ = '0' + (((*p) >> 3) & 07);
2115 *q++ = '0' + ((*p) & 07);
2130 * @string: a string to remove the leading whitespace from
2132 * Removes leading whitespace from a string, by moving the rest
2133 * of the characters forward.
2135 * This function doesn't allocate or reallocate any memory;
2136 * it modifies @string in place. The pointer to @string is
2137 * returned to allow the nesting of functions.
2139 * Also see g_strchomp() and g_strstrip().
2144 g_strchug (gchar *string)
2148 g_return_val_if_fail (string != NULL, NULL);
2150 for (start = (guchar*) string; *start && g_ascii_isspace (*start); start++)
2153 g_memmove (string, start, strlen ((gchar *) start) + 1);
2160 * @string: a string to remove the trailing whitespace from
2162 * Removes trailing whitespace from a string.
2164 * This function doesn't allocate or reallocate any memory;
2165 * it modifies @string in place. The pointer to @string is
2166 * returned to allow the nesting of functions.
2168 * Also see g_strchug() and g_strstrip().
2173 g_strchomp (gchar *string)
2177 g_return_val_if_fail (string != NULL, NULL);
2179 len = strlen (string);
2182 if (g_ascii_isspace ((guchar) string[len]))
2193 * @string: a string to split
2194 * @delimiter: a string which specifies the places at which to split
2195 * the string. The delimiter is not included in any of the resulting
2196 * strings, unless @max_tokens is reached.
2197 * @max_tokens: the maximum number of pieces to split @string into.
2198 * If this is less than 1, the string is split completely.
2200 * Splits a string into a maximum of @max_tokens pieces, using the given
2201 * @delimiter. If @max_tokens is reached, the remainder of @string is
2202 * appended to the last token.
2204 * As a special case, the result of splitting the empty string "" is an empty
2205 * vector, not a vector containing a single string. The reason for this
2206 * special case is that being able to represent a empty vector is typically
2207 * more useful than consistent handling of empty elements. If you do need
2208 * to represent empty elements, you'll need to check for the empty string
2209 * before calling g_strsplit().
2211 * Return value: a newly-allocated %NULL-terminated array of strings. Use
2212 * g_strfreev() to free it.
2215 g_strsplit (const gchar *string,
2216 const gchar *delimiter,
2219 GSList *string_list = NULL, *slist;
2220 gchar **str_array, *s;
2222 const gchar *remainder;
2224 g_return_val_if_fail (string != NULL, NULL);
2225 g_return_val_if_fail (delimiter != NULL, NULL);
2226 g_return_val_if_fail (delimiter[0] != '\0', NULL);
2229 max_tokens = G_MAXINT;
2232 s = strstr (remainder, delimiter);
2235 gsize delimiter_len = strlen (delimiter);
2237 while (--max_tokens && s)
2241 len = s - remainder;
2242 string_list = g_slist_prepend (string_list,
2243 g_strndup (remainder, len));
2245 remainder = s + delimiter_len;
2246 s = strstr (remainder, delimiter);
2252 string_list = g_slist_prepend (string_list, g_strdup (remainder));
2255 str_array = g_new (gchar*, n + 1);
2257 str_array[n--] = NULL;
2258 for (slist = string_list; slist; slist = slist->next)
2259 str_array[n--] = slist->data;
2261 g_slist_free (string_list);
2268 * @string: The string to be tokenized
2269 * @delimiters: A nul-terminated string containing bytes that are used
2270 * to split the string.
2271 * @max_tokens: The maximum number of tokens to split @string into.
2272 * If this is less than 1, the string is split completely
2274 * Splits @string into a number of tokens not containing any of the characters
2275 * in @delimiter. A token is the (possibly empty) longest string that does not
2276 * contain any of the characters in @delimiters. If @max_tokens is reached, the
2277 * remainder is appended to the last token.
2279 * For example the result of g_strsplit_set ("abc:def/ghi", ":/", -1) is a
2280 * %NULL-terminated vector containing the three strings "abc", "def",
2283 * The result if g_strsplit_set (":def/ghi:", ":/", -1) is a %NULL-terminated
2284 * vector containing the four strings "", "def", "ghi", and "".
2286 * As a special case, the result of splitting the empty string "" is an empty
2287 * vector, not a vector containing a single string. The reason for this
2288 * special case is that being able to represent a empty vector is typically
2289 * more useful than consistent handling of empty elements. If you do need
2290 * to represent empty elements, you'll need to check for the empty string
2291 * before calling g_strsplit_set().
2293 * Note that this function works on bytes not characters, so it can't be used
2294 * to delimit UTF-8 strings for anything but ASCII characters.
2296 * Return value: a newly-allocated %NULL-terminated array of strings. Use
2297 * g_strfreev() to free it.
2302 g_strsplit_set (const gchar *string,
2303 const gchar *delimiters,
2306 gboolean delim_table[256];
2307 GSList *tokens, *list;
2310 const gchar *current;
2314 g_return_val_if_fail (string != NULL, NULL);
2315 g_return_val_if_fail (delimiters != NULL, NULL);
2318 max_tokens = G_MAXINT;
2320 if (*string == '\0')
2322 result = g_new (char *, 1);
2327 memset (delim_table, FALSE, sizeof (delim_table));
2328 for (s = delimiters; *s != '\0'; ++s)
2329 delim_table[*(guchar *)s] = TRUE;
2334 s = current = string;
2337 if (delim_table[*(guchar *)s] && n_tokens + 1 < max_tokens)
2339 token = g_strndup (current, s - current);
2340 tokens = g_slist_prepend (tokens, token);
2349 token = g_strndup (current, s - current);
2350 tokens = g_slist_prepend (tokens, token);
2353 result = g_new (gchar *, n_tokens + 1);
2355 result[n_tokens] = NULL;
2356 for (list = tokens; list != NULL; list = list->next)
2357 result[--n_tokens] = list->data;
2359 g_slist_free (tokens);
2366 * @str_array: a %NULL-terminated array of strings to free
2368 * Frees a %NULL-terminated array of strings, and the array itself.
2369 * If called on a %NULL value, g_strfreev() simply returns.
2372 g_strfreev (gchar **str_array)
2378 for (i = 0; str_array[i] != NULL; i++)
2379 g_free (str_array[i]);
2387 * @str_array: a %NULL-terminated array of strings
2389 * Copies %NULL-terminated array of strings. The copy is a deep copy;
2390 * the new array should be freed by first freeing each string, then
2391 * the array itself. g_strfreev() does this for you. If called
2392 * on a %NULL value, g_strdupv() simply returns %NULL.
2394 * Return value: a new %NULL-terminated array of strings.
2397 g_strdupv (gchar **str_array)
2405 while (str_array[i])
2408 retval = g_new (gchar*, i + 1);
2411 while (str_array[i])
2413 retval[i] = g_strdup (str_array[i]);
2426 * @separator: a string to insert between each of the strings, or %NULL
2427 * @str_array: a %NULL-terminated array of strings to join
2429 * Joins a number of strings together to form one long string, with the
2430 * optional @separator inserted between each of them. The returned string
2431 * should be freed with g_free().
2433 * Returns: a newly-allocated string containing all of the strings joined
2434 * together, with @separator between them
2437 g_strjoinv (const gchar *separator,
2443 g_return_val_if_fail (str_array != NULL, NULL);
2445 if (separator == NULL)
2452 gsize separator_len;
2454 separator_len = strlen (separator);
2455 /* First part, getting length */
2456 len = 1 + strlen (str_array[0]);
2457 for (i = 1; str_array[i] != NULL; i++)
2458 len += strlen (str_array[i]);
2459 len += separator_len * (i - 1);
2461 /* Second part, building string */
2462 string = g_new (gchar, len);
2463 ptr = g_stpcpy (string, *str_array);
2464 for (i = 1; str_array[i] != NULL; i++)
2466 ptr = g_stpcpy (ptr, separator);
2467 ptr = g_stpcpy (ptr, str_array[i]);
2471 string = g_strdup ("");
2478 * @separator: a string to insert between each of the strings, or %NULL
2479 * @...: a %NULL-terminated list of strings to join
2481 * Joins a number of strings together to form one long string, with the
2482 * optional @separator inserted between each of them. The returned string
2483 * should be freed with g_free().
2485 * Returns: a newly-allocated string containing all of the strings joined
2486 * together, with @separator between them
2489 g_strjoin (const gchar *separator,
2495 gsize separator_len;
2498 if (separator == NULL)
2501 separator_len = strlen (separator);
2503 va_start (args, separator);
2505 s = va_arg (args, gchar*);
2509 /* First part, getting length */
2510 len = 1 + strlen (s);
2512 s = va_arg (args, gchar*);
2515 len += separator_len + strlen (s);
2516 s = va_arg (args, gchar*);
2520 /* Second part, building string */
2521 string = g_new (gchar, len);
2523 va_start (args, separator);
2525 s = va_arg (args, gchar*);
2526 ptr = g_stpcpy (string, s);
2528 s = va_arg (args, gchar*);
2531 ptr = g_stpcpy (ptr, separator);
2532 ptr = g_stpcpy (ptr, s);
2533 s = va_arg (args, gchar*);
2537 string = g_strdup ("");
2547 * @haystack: a string
2548 * @haystack_len: the maximum length of @haystack. Note that -1 is
2549 * a valid length, if @haystack is nul-terminated, meaning it will
2550 * search through the whole string.
2551 * @needle: the string to search for
2553 * Searches the string @haystack for the first occurrence
2554 * of the string @needle, limiting the length of the search
2557 * Return value: a pointer to the found occurrence, or
2558 * %NULL if not found.
2561 g_strstr_len (const gchar *haystack,
2562 gssize haystack_len,
2563 const gchar *needle)
2565 g_return_val_if_fail (haystack != NULL, NULL);
2566 g_return_val_if_fail (needle != NULL, NULL);
2568 if (haystack_len < 0)
2569 return strstr (haystack, needle);
2572 const gchar *p = haystack;
2573 gsize needle_len = strlen (needle);
2577 if (needle_len == 0)
2578 return (gchar *)haystack;
2580 if (haystack_len < needle_len)
2583 end = haystack + haystack_len - needle_len;
2585 while (p <= end && *p)
2587 for (i = 0; i < needle_len; i++)
2588 if (p[i] != needle[i])
2603 * @haystack: a nul-terminated string
2604 * @needle: the nul-terminated string to search for
2606 * Searches the string @haystack for the last occurrence
2607 * of the string @needle.
2609 * Return value: a pointer to the found occurrence, or
2610 * %NULL if not found.
2613 g_strrstr (const gchar *haystack,
2614 const gchar *needle)
2621 g_return_val_if_fail (haystack != NULL, NULL);
2622 g_return_val_if_fail (needle != NULL, NULL);
2624 needle_len = strlen (needle);
2625 haystack_len = strlen (haystack);
2627 if (needle_len == 0)
2628 return (gchar *)haystack;
2630 if (haystack_len < needle_len)
2633 p = haystack + haystack_len - needle_len;
2635 while (p >= haystack)
2637 for (i = 0; i < needle_len; i++)
2638 if (p[i] != needle[i])
2652 * @haystack: a nul-terminated string
2653 * @haystack_len: the maximum length of @haystack
2654 * @needle: the nul-terminated string to search for
2656 * Searches the string @haystack for the last occurrence
2657 * of the string @needle, limiting the length of the search
2660 * Return value: a pointer to the found occurrence, or
2661 * %NULL if not found.
2664 g_strrstr_len (const gchar *haystack,
2665 gssize haystack_len,
2666 const gchar *needle)
2668 g_return_val_if_fail (haystack != NULL, NULL);
2669 g_return_val_if_fail (needle != NULL, NULL);
2671 if (haystack_len < 0)
2672 return g_strrstr (haystack, needle);
2675 gsize needle_len = strlen (needle);
2676 const gchar *haystack_max = haystack + haystack_len;
2677 const gchar *p = haystack;
2680 while (p < haystack_max && *p)
2683 if (p < haystack + needle_len)
2688 while (p >= haystack)
2690 for (i = 0; i < needle_len; i++)
2691 if (p[i] != needle[i])
2707 * @str: a nul-terminated string
2708 * @suffix: the nul-terminated suffix to look for
2710 * Looks whether the string @str ends with @suffix.
2712 * Return value: %TRUE if @str end with @suffix, %FALSE otherwise.
2717 g_str_has_suffix (const gchar *str,
2718 const gchar *suffix)
2723 g_return_val_if_fail (str != NULL, FALSE);
2724 g_return_val_if_fail (suffix != NULL, FALSE);
2726 str_len = strlen (str);
2727 suffix_len = strlen (suffix);
2729 if (str_len < suffix_len)
2732 return strcmp (str + str_len - suffix_len, suffix) == 0;
2737 * @str: a nul-terminated string
2738 * @prefix: the nul-terminated prefix to look for
2740 * Looks whether the string @str begins with @prefix.
2742 * Return value: %TRUE if @str begins with @prefix, %FALSE otherwise.
2747 g_str_has_prefix (const gchar *str,
2748 const gchar *prefix)
2753 g_return_val_if_fail (str != NULL, FALSE);
2754 g_return_val_if_fail (prefix != NULL, FALSE);
2756 str_len = strlen (str);
2757 prefix_len = strlen (prefix);
2759 if (str_len < prefix_len)
2762 return strncmp (str, prefix, prefix_len) == 0;
2767 * @str_array: a %NULL-terminated array of strings
2769 * Returns the length of the given %NULL-terminated
2770 * string array @str_array.
2772 * Return value: length of @str_array.
2777 g_strv_length (gchar **str_array)
2781 g_return_val_if_fail (str_array != NULL, 0);
2783 while (str_array[i])