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() */
43 /* Needed on BSD/OS X for e.g. strtod_l */
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 <literal>EOF</literal>, 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 <literal>EOF</literal>, 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 <literal>EOF</literal>, 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 <literal>EOF</literal>, 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 <literal>EOF</literal>, 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 <literal>EOF</literal>, 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 <literal>EOF</literal>, 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 <literal>EOF</literal>, 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 <literal>EOF</literal>, 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 <literal>EOF</literal>, 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 <literal>EOF</literal>, 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 #if defined (HAVE_NEWLOCALE) && \
321 defined (HAVE_USELOCALE) && \
322 defined (HAVE_STRTOD_L) && \
323 defined (HAVE_STRTOULL_L) && \
324 defined (HAVE_STRTOLL_L)
325 #define USE_XLOCALE 1
332 static gsize initialized = FALSE;
333 static locale_t C_locale = NULL;
335 if (g_once_init_enter (&initialized))
337 C_locale = newlocale (LC_ALL_MASK, "C", NULL);
338 g_once_init_leave (&initialized, TRUE);
347 * @str: the string to duplicate
349 * Duplicates a string. If @str is %NULL it returns %NULL.
350 * The returned string should be freed with g_free()
351 * when no longer needed.
353 * Returns: a newly-allocated copy of @str
356 g_strdup (const gchar *str)
363 length = strlen (str) + 1;
364 new_str = g_new (char, length);
365 memcpy (new_str, str, length);
375 * @mem: the memory to copy.
376 * @byte_size: the number of bytes to copy.
378 * Allocates @byte_size bytes of memory, and copies @byte_size bytes into it
379 * from @mem. If @mem is %NULL it returns %NULL.
381 * Returns: a pointer to the newly-allocated copy of the memory, or %NULL if @mem
385 g_memdup (gconstpointer mem,
392 new_mem = g_malloc (byte_size);
393 memcpy (new_mem, mem, byte_size);
403 * @str: the string to duplicate
404 * @n: the maximum number of bytes to copy from @str
406 * Duplicates the first @n bytes of a string, returning a newly-allocated
407 * buffer @n + 1 bytes long which will always be nul-terminated.
408 * If @str is less than @n bytes long the buffer is padded with nuls.
409 * If @str is %NULL it returns %NULL.
410 * The returned value should be freed when no longer needed.
413 * To copy a number of characters from a UTF-8 encoded string, use
414 * g_utf8_strncpy() instead.
417 * Returns: a newly-allocated buffer containing the first @n bytes
418 * of @str, nul-terminated
421 g_strndup (const gchar *str,
428 new_str = g_new (gchar, n + 1);
429 strncpy (new_str, str, n);
440 * @length: the length of the new string
441 * @fill_char: the byte to fill the string with
443 * Creates a new string @length bytes long filled with @fill_char.
444 * The returned string should be freed when no longer needed.
446 * Returns: a newly-allocated string filled the @fill_char
449 g_strnfill (gsize length,
454 str = g_new (gchar, length + 1);
455 memset (str, (guchar)fill_char, length);
463 * @dest: destination buffer.
464 * @src: source string.
466 * Copies a nul-terminated string into the dest buffer, include the
467 * trailing nul, and return a pointer to the trailing nul byte.
468 * This is useful for concatenating multiple strings together
469 * without having to repeatedly scan for the end.
471 * Return value: a pointer to trailing nul byte.
474 g_stpcpy (gchar *dest,
478 g_return_val_if_fail (dest != NULL, NULL);
479 g_return_val_if_fail (src != NULL, NULL);
480 return stpcpy (dest, src);
482 register gchar *d = dest;
483 register const gchar *s = src;
485 g_return_val_if_fail (dest != NULL, NULL);
486 g_return_val_if_fail (src != NULL, NULL);
489 while (*s++ != '\0');
497 * @format: a standard printf() format string, but notice
498 * <link linkend="string-precision">string precision pitfalls</link>
499 * @args: the list of parameters to insert into the format string
501 * Similar to the standard C vsprintf() function but safer, since it
502 * calculates the maximum space required and allocates memory to hold
503 * the result. The returned string should be freed with g_free() when
506 * See also g_vasprintf(), which offers the same functionality, but
507 * additionally returns the length of the allocated string.
509 * Returns: a newly-allocated string holding the result
512 g_strdup_vprintf (const gchar *format,
515 gchar *string = NULL;
517 g_vasprintf (&string, format, args);
524 * @format: a standard printf() format string, but notice
525 * <link linkend="string-precision">string precision pitfalls</link>
526 * @...: the parameters to insert into the format string
528 * Similar to the standard C sprintf() function but safer, since it
529 * calculates the maximum space required and allocates memory to hold
530 * the result. The returned string should be freed with g_free() when no
533 * Returns: a newly-allocated string holding the result
536 g_strdup_printf (const gchar *format,
542 va_start (args, format);
543 buffer = g_strdup_vprintf (format, args);
551 * @string1: the first string to add, which must not be %NULL
552 * @...: a %NULL-terminated list of strings to append to the string
554 * Concatenates all of the given strings into one long string.
555 * The returned string should be freed with g_free() when no longer needed.
557 * Note that this function is usually not the right function to use to
558 * assemble a translated message from pieces, since proper translation
559 * often requires the pieces to be reordered.
561 * <warning><para>The variable argument list <emphasis>must</emphasis> end
562 * with %NULL. If you forget the %NULL, g_strconcat() will start appending
563 * random memory junk to your string.</para></warning>
565 * Returns: a newly-allocated string containing all the string arguments
568 g_strconcat (const gchar *string1, ...)
579 l = 1 + strlen (string1);
580 va_start (args, string1);
581 s = va_arg (args, gchar*);
585 s = va_arg (args, gchar*);
589 concat = g_new (gchar, l);
592 ptr = g_stpcpy (ptr, string1);
593 va_start (args, string1);
594 s = va_arg (args, gchar*);
597 ptr = g_stpcpy (ptr, s);
598 s = va_arg (args, gchar*);
607 * @nptr: the string to convert to a numeric value.
608 * @endptr: if non-%NULL, it returns the character after
609 * the last character used in the conversion.
611 * Converts a string to a #gdouble value.
612 * It calls the standard strtod() function to handle the conversion, but
613 * if the string is not completely converted it attempts the conversion
614 * again with g_ascii_strtod(), and returns the best match.
616 * This function should seldom be used. The normal situation when reading
617 * numbers not for human consumption is to use g_ascii_strtod(). Only when
618 * you know that you must expect both locale formatted and C formatted numbers
619 * should you use this. Make sure that you don't pass strings such as comma
620 * separated lists of values, since the commas may be interpreted as a decimal
621 * point in some locales, causing unexpected results.
623 * Return value: the #gdouble value.
626 g_strtod (const gchar *nptr,
634 g_return_val_if_fail (nptr != NULL, 0);
639 val_1 = strtod (nptr, &fail_pos_1);
641 if (fail_pos_1 && fail_pos_1[0] != 0)
642 val_2 = g_ascii_strtod (nptr, &fail_pos_2);
644 if (!fail_pos_1 || fail_pos_1[0] == 0 || fail_pos_1 >= fail_pos_2)
647 *endptr = fail_pos_1;
653 *endptr = fail_pos_2;
660 * @nptr: the string to convert to a numeric value.
661 * @endptr: if non-%NULL, it returns the character after
662 * the last character used in the conversion.
664 * Converts a string to a #gdouble value.
666 * This function behaves like the standard strtod() function
667 * does in the C locale. It does this without actually changing
668 * the current locale, since that would not be thread-safe.
669 * A limitation of the implementation is that this function
670 * will still accept localized versions of infinities and NANs.
672 * This function is typically used when reading configuration
673 * files or other non-user input that should be locale independent.
674 * To handle input from the user you should normally use the
675 * locale-sensitive system strtod() function.
677 * To convert from a #gdouble to a string in a locale-insensitive
678 * way, use g_ascii_dtostr().
680 * If the correct value would cause overflow, plus or minus <literal>HUGE_VAL</literal>
681 * is returned (according to the sign of the value), and <literal>ERANGE</literal> is
682 * stored in <literal>errno</literal>. If the correct value would cause underflow,
683 * zero is returned and <literal>ERANGE</literal> is stored in <literal>errno</literal>.
685 * This function resets <literal>errno</literal> before calling strtod() so that
686 * you can reliably detect overflow and underflow.
688 * Return value: the #gdouble value.
691 g_ascii_strtod (const gchar *nptr,
696 g_return_val_if_fail (nptr != NULL, 0);
700 return strtod_l (nptr, endptr, get_C_locale ());
707 struct lconv *locale_data;
709 const char *decimal_point;
710 int decimal_point_len;
711 const char *p, *decimal_point_pos;
712 const char *end = NULL; /* Silence gcc */
715 g_return_val_if_fail (nptr != NULL, 0);
720 locale_data = localeconv ();
721 decimal_point = locale_data->decimal_point;
722 decimal_point_len = strlen (decimal_point);
725 decimal_point_len = 1;
728 g_assert (decimal_point_len != 0);
730 decimal_point_pos = NULL;
733 if (decimal_point[0] != '.' ||
734 decimal_point[1] != 0)
737 /* Skip leading space */
738 while (g_ascii_isspace (*p))
741 /* Skip leading optional sign */
742 if (*p == '+' || *p == '-')
746 (p[1] == 'x' || p[1] == 'X'))
749 /* HEX - find the (optional) decimal point */
751 while (g_ascii_isxdigit (*p))
755 decimal_point_pos = p++;
757 while (g_ascii_isxdigit (*p))
760 if (*p == 'p' || *p == 'P')
762 if (*p == '+' || *p == '-')
764 while (g_ascii_isdigit (*p))
769 else if (g_ascii_isdigit (*p) || *p == '.')
771 while (g_ascii_isdigit (*p))
775 decimal_point_pos = p++;
777 while (g_ascii_isdigit (*p))
780 if (*p == 'e' || *p == 'E')
782 if (*p == '+' || *p == '-')
784 while (g_ascii_isdigit (*p))
789 /* For the other cases, we need not convert the decimal point */
792 if (decimal_point_pos)
796 /* We need to convert the '.' to the locale specific decimal point */
797 copy = g_malloc (end - nptr + 1 + decimal_point_len);
800 memcpy (c, nptr, decimal_point_pos - nptr);
801 c += decimal_point_pos - nptr;
802 memcpy (c, decimal_point, decimal_point_len);
803 c += decimal_point_len;
804 memcpy (c, decimal_point_pos + 1, end - (decimal_point_pos + 1));
805 c += end - (decimal_point_pos + 1);
809 val = strtod (copy, &fail_pos);
810 strtod_errno = errno;
814 if (fail_pos - copy > decimal_point_pos - nptr)
815 fail_pos = (char *)nptr + (fail_pos - copy) - (decimal_point_len - 1);
817 fail_pos = (char *)nptr + (fail_pos - copy);
827 copy = g_malloc (end - (char *)nptr + 1);
828 memcpy (copy, nptr, end - nptr);
829 *(copy + (end - (char *)nptr)) = 0;
832 val = strtod (copy, &fail_pos);
833 strtod_errno = errno;
837 fail_pos = (char *)nptr + (fail_pos - copy);
845 val = strtod (nptr, &fail_pos);
846 strtod_errno = errno;
852 errno = strtod_errno;
861 * @buffer: A buffer to place the resulting string in
862 * @buf_len: The length of the buffer.
863 * @d: The #gdouble to convert
865 * Converts a #gdouble to a string, using the '.' as
868 * This functions generates enough precision that converting
869 * the string back using g_ascii_strtod() gives the same machine-number
870 * (on machines with IEEE compatible 64bit doubles). It is
871 * guaranteed that the size of the resulting string will never
872 * be larger than @G_ASCII_DTOSTR_BUF_SIZE bytes.
874 * Return value: The pointer to the buffer with the converted string.
877 g_ascii_dtostr (gchar *buffer,
881 return g_ascii_formatd (buffer, buf_len, "%.17g", d);
886 * @buffer: A buffer to place the resulting string in
887 * @buf_len: The length of the buffer.
888 * @format: The printf()-style format to use for the
889 * code to use for converting.
890 * @d: The #gdouble to convert
892 * Converts a #gdouble to a string, using the '.' as
893 * decimal point. To format the number you pass in
894 * a printf()-style format string. Allowed conversion
895 * specifiers are 'e', 'E', 'f', 'F', 'g' and 'G'.
897 * If you just want to want to serialize the value into a
898 * string, use g_ascii_dtostr().
900 * Return value: The pointer to the buffer with the converted string.
903 g_ascii_formatd (gchar *buffer,
911 old_locale = uselocale (get_C_locale ());
912 #pragma GCC diagnostic push
913 #pragma GCC diagnostic ignored "-Wformat-nonliteral"
914 _g_snprintf (buffer, buf_len, format, d);
915 #pragma GCC diagnostic pop
916 uselocale (old_locale);
921 struct lconv *locale_data;
923 const char *decimal_point;
924 int decimal_point_len;
929 g_return_val_if_fail (buffer != NULL, NULL);
930 g_return_val_if_fail (format[0] == '%', NULL);
931 g_return_val_if_fail (strpbrk (format + 1, "'l%") == NULL, NULL);
933 format_char = format[strlen (format) - 1];
935 g_return_val_if_fail (format_char == 'e' || format_char == 'E' ||
936 format_char == 'f' || format_char == 'F' ||
937 format_char == 'g' || format_char == 'G',
940 if (format[0] != '%')
943 if (strpbrk (format + 1, "'l%"))
946 if (!(format_char == 'e' || format_char == 'E' ||
947 format_char == 'f' || format_char == 'F' ||
948 format_char == 'g' || format_char == 'G'))
951 _g_snprintf (buffer, buf_len, format, d);
954 locale_data = localeconv ();
955 decimal_point = locale_data->decimal_point;
956 decimal_point_len = strlen (decimal_point);
959 decimal_point_len = 1;
962 g_assert (decimal_point_len != 0);
964 if (decimal_point[0] != '.' ||
965 decimal_point[1] != 0)
969 while (g_ascii_isspace (*p))
972 if (*p == '+' || *p == '-')
975 while (isdigit ((guchar)*p))
978 if (strncmp (p, decimal_point, decimal_point_len) == 0)
982 if (decimal_point_len > 1)
984 rest_len = strlen (p + (decimal_point_len-1));
985 memmove (p, p + (decimal_point_len-1), rest_len);
995 #define ISSPACE(c) ((c) == ' ' || (c) == '\f' || (c) == '\n' || \
996 (c) == '\r' || (c) == '\t' || (c) == '\v')
997 #define ISUPPER(c) ((c) >= 'A' && (c) <= 'Z')
998 #define ISLOWER(c) ((c) >= 'a' && (c) <= 'z')
999 #define ISALPHA(c) (ISUPPER (c) || ISLOWER (c))
1000 #define TOUPPER(c) (ISLOWER (c) ? (c) - 'a' + 'A' : (c))
1001 #define TOLOWER(c) (ISUPPER (c) ? (c) - 'A' + 'a' : (c))
1006 g_parse_long_long (const gchar *nptr,
1007 const gchar **endptr,
1011 /* this code is based on on the strtol(3) code from GNU libc released under
1012 * the GNU Lesser General Public License.
1014 * Copyright (C) 1991,92,94,95,96,97,98,99,2000,01,02
1015 * Free Software Foundation, Inc.
1021 const gchar *s, *save;
1024 g_return_val_if_fail (nptr != NULL, 0);
1027 if (base == 1 || base > 36)
1037 /* Skip white space. */
1038 while (ISSPACE (*s))
1041 if (G_UNLIKELY (!*s))
1044 /* Check for a sign. */
1053 /* Recognize number prefix and if BASE is zero, figure it out ourselves. */
1056 if ((base == 0 || base == 16) && TOUPPER (s[1]) == 'X')
1067 /* Save the pointer so we can check later if anything happened. */
1069 cutoff = G_MAXUINT64 / base;
1070 cutlim = G_MAXUINT64 % base;
1077 if (c >= '0' && c <= '9')
1079 else if (ISALPHA (c))
1080 c = TOUPPER (c) - 'A' + 10;
1085 /* Check for overflow. */
1086 if (ui64 > cutoff || (ui64 == cutoff && c > cutlim))
1095 /* Check if anything actually happened. */
1099 /* Store in ENDPTR the address of one character
1100 past the last character we converted. */
1104 if (G_UNLIKELY (overflow))
1113 /* We must handle a special case here: the base is 0 or 16 and the
1114 first two characters are '0' and 'x', but the rest are no
1115 hexadecimal digits. This is no error case. We return 0 and
1116 ENDPTR points to the `x`. */
1119 if (save - nptr >= 2 && TOUPPER (save[-1]) == 'X'
1121 *endptr = &save[-1];
1123 /* There was no number to convert. */
1128 #endif /* !USE_XLOCALE */
1132 * @nptr: the string to convert to a numeric value.
1133 * @endptr: if non-%NULL, it returns the character after
1134 * the last character used in the conversion.
1135 * @base: to be used for the conversion, 2..36 or 0
1137 * Converts a string to a #guint64 value.
1138 * This function behaves like the standard strtoull() function
1139 * does in the C locale. It does this without actually
1140 * changing the current locale, since that would not be
1143 * This function is typically used when reading configuration
1144 * files or other non-user input that should be locale independent.
1145 * To handle input from the user you should normally use the
1146 * locale-sensitive system strtoull() function.
1148 * If the correct value would cause overflow, %G_MAXUINT64
1149 * is returned, and <literal>ERANGE</literal> is stored in <literal>errno</literal>.
1150 * If the base is outside the valid range, zero is returned, and
1151 * <literal>EINVAL</literal> is stored in <literal>errno</literal>.
1152 * If the string conversion fails, zero is returned, and @endptr returns
1153 * @nptr (if @endptr is non-%NULL).
1155 * Return value: the #guint64 value or zero on error.
1160 g_ascii_strtoull (const gchar *nptr,
1165 return strtoull_l (nptr, endptr, base, get_C_locale ());
1170 result = g_parse_long_long (nptr, (const gchar **) endptr, base, &negative);
1172 /* Return the result of the appropriate sign. */
1173 return negative ? -result : result;
1179 * @nptr: the string to convert to a numeric value.
1180 * @endptr: if non-%NULL, it returns the character after
1181 * the last character used in the conversion.
1182 * @base: to be used for the conversion, 2..36 or 0
1184 * Converts a string to a #gint64 value.
1185 * This function behaves like the standard strtoll() function
1186 * does in the C locale. It does this without actually
1187 * changing the current locale, since that would not be
1190 * This function is typically used when reading configuration
1191 * files or other non-user input that should be locale independent.
1192 * To handle input from the user you should normally use the
1193 * locale-sensitive system strtoll() function.
1195 * If the correct value would cause overflow, %G_MAXINT64 or %G_MININT64
1196 * is returned, and <literal>ERANGE</literal> is stored in <literal>errno</literal>.
1197 * If the base is outside the valid range, zero is returned, and
1198 * <literal>EINVAL</literal> is stored in <literal>errno</literal>. If the
1199 * string conversion fails, zero is returned, and @endptr returns @nptr
1200 * (if @endptr is non-%NULL).
1202 * Return value: the #gint64 value or zero on error.
1207 g_ascii_strtoll (const gchar *nptr,
1212 return strtoll_l (nptr, endptr, base, get_C_locale ());
1217 result = g_parse_long_long (nptr, (const gchar **) endptr, base, &negative);
1219 if (negative && result > (guint64) G_MININT64)
1224 else if (!negative && result > (guint64) G_MAXINT64)
1230 return - (gint64) result;
1232 return (gint64) result;
1238 * @errnum: the system error number. See the standard C %errno
1241 * Returns a string corresponding to the given error code, e.g.
1242 * "no such process". You should use this function in preference to
1243 * strerror(), because it returns a string in UTF-8 encoding, and since
1244 * not all platforms support the strerror() function.
1246 * Returns: a UTF-8 string describing the error code. If the error code
1247 * is unknown, it returns "unknown error (<code>)".
1250 g_strerror (gint errnum)
1256 gint saved_errno = errno;
1258 msg = tofree = NULL;
1260 #ifdef HAVE_STRERROR
1261 msg = strerror (errnum);
1262 if (!g_get_charset (NULL))
1263 msg = tofree = g_locale_to_utf8 (msg, -1, NULL, NULL, NULL);
1269 _g_sprintf (msg, "unknown error (%d)", errnum);
1272 ret = g_intern_string (msg);
1274 errno = saved_errno;
1280 * @signum: the signal number. See the <literal>signal</literal>
1283 * Returns a string describing the given signal, e.g. "Segmentation fault".
1284 * You should use this function in preference to strsignal(), because it
1285 * returns a string in UTF-8 encoding, and since not all platforms support
1286 * the strsignal() function.
1288 * Returns: a UTF-8 string describing the signal. If the signal is unknown,
1289 * it returns "unknown signal (<signum>)".
1292 g_strsignal (gint signum)
1298 msg = tofree = NULL;
1300 #ifdef HAVE_STRSIGNAL
1301 msg = strsignal (signum);
1302 if (!g_get_charset (NULL))
1303 msg = tofree = g_locale_to_utf8 (msg, -1, NULL, NULL, NULL);
1307 msg = tofree = g_strdup_printf ("unknown signal (%d)", signum);
1308 ret = g_intern_string (msg);
1314 /* Functions g_strlcpy and g_strlcat were originally developed by
1315 * Todd C. Miller <Todd.Miller@courtesan.com> to simplify writing secure code.
1316 * See http://www.openbsd.org/cgi-bin/man.cgi?query=strlcpy
1317 * for more information.
1321 /* Use the native ones, if available; they might be implemented in assembly */
1323 g_strlcpy (gchar *dest,
1327 g_return_val_if_fail (dest != NULL, 0);
1328 g_return_val_if_fail (src != NULL, 0);
1330 return strlcpy (dest, src, dest_size);
1334 g_strlcat (gchar *dest,
1338 g_return_val_if_fail (dest != NULL, 0);
1339 g_return_val_if_fail (src != NULL, 0);
1341 return strlcat (dest, src, dest_size);
1344 #else /* ! HAVE_STRLCPY */
1347 * @dest: destination buffer
1348 * @src: source buffer
1349 * @dest_size: length of @dest in bytes
1351 * Portability wrapper that calls strlcpy() on systems which have it,
1352 * and emulates strlcpy() otherwise. Copies @src to @dest; @dest is
1353 * guaranteed to be nul-terminated; @src must be nul-terminated;
1354 * @dest_size is the buffer size, not the number of chars to copy.
1356 * At most dest_size - 1 characters will be copied. Always nul-terminates
1357 * (unless dest_size == 0). This function does <emphasis>not</emphasis>
1358 * allocate memory. Unlike strncpy(), this function doesn't pad dest (so
1359 * it's often faster). It returns the size of the attempted result,
1360 * strlen (src), so if @retval >= @dest_size, truncation occurred.
1362 * <note><para>Caveat: strlcpy() is supposedly more secure than
1363 * strcpy() or strncpy(), but if you really want to avoid screwups,
1364 * g_strdup() is an even better idea.</para></note>
1366 * Returns: length of @src
1369 g_strlcpy (gchar *dest,
1373 register gchar *d = dest;
1374 register const gchar *s = src;
1375 register gsize n = dest_size;
1377 g_return_val_if_fail (dest != NULL, 0);
1378 g_return_val_if_fail (src != NULL, 0);
1380 /* Copy as many bytes as will fit */
1381 if (n != 0 && --n != 0)
1384 register gchar c = *s++;
1392 /* If not enough room in dest, add NUL and traverse rest of src */
1401 return s - src - 1; /* count does not include NUL */
1406 * @dest: destination buffer, already containing one nul-terminated string
1407 * @src: source buffer
1408 * @dest_size: length of @dest buffer in bytes (not length of existing string
1411 * Portability wrapper that calls strlcat() on systems which have it,
1412 * and emulates it otherwise. Appends nul-terminated @src string to @dest,
1413 * guaranteeing nul-termination for @dest. The total size of @dest won't
1414 * exceed @dest_size.
1416 * At most dest_size - 1 characters will be copied.
1417 * Unlike strncat, dest_size is the full size of dest, not the space left over.
1418 * This function does NOT allocate memory.
1419 * This always NUL terminates (unless siz == 0 or there were no NUL characters
1420 * in the dest_size characters of dest to start with).
1422 * <note><para>Caveat: this is supposedly a more secure alternative to
1423 * strcat() or strncat(), but for real security g_strconcat() is harder
1424 * to mess up.</para></note>
1426 * Returns: size of attempted result, which is MIN (dest_size, strlen
1427 * (original dest)) + strlen (src), so if retval >= dest_size,
1428 * truncation occurred.
1431 g_strlcat (gchar *dest,
1435 register gchar *d = dest;
1436 register const gchar *s = src;
1437 register gsize bytes_left = dest_size;
1438 gsize dlength; /* Logically, MIN (strlen (d), dest_size) */
1440 g_return_val_if_fail (dest != NULL, 0);
1441 g_return_val_if_fail (src != NULL, 0);
1443 /* Find the end of dst and adjust bytes left but don't go past end */
1444 while (*d != 0 && bytes_left-- != 0)
1447 bytes_left = dest_size - dlength;
1449 if (bytes_left == 0)
1450 return dlength + strlen (s);
1454 if (bytes_left != 1)
1463 return dlength + (s - src); /* count does not include NUL */
1465 #endif /* ! HAVE_STRLCPY */
1470 * @len: length of @str in bytes, or -1 if @str is nul-terminated.
1472 * Converts all upper case ASCII letters to lower case ASCII letters.
1474 * Return value: a newly-allocated string, with all the upper case
1475 * characters in @str converted to lower case, with
1476 * semantics that exactly match g_ascii_tolower(). (Note
1477 * that this is unlike the old g_strdown(), which modified
1478 * the string in place.)
1481 g_ascii_strdown (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_tolower (*s);
1501 * @len: length of @str in bytes, or -1 if @str is nul-terminated.
1503 * Converts all lower case ASCII letters to upper case ASCII letters.
1505 * Return value: a newly allocated string, with all the lower case
1506 * characters in @str converted to upper case, with
1507 * semantics that exactly match g_ascii_toupper(). (Note
1508 * that this is unlike the old g_strup(), which modified
1509 * the string in place.)
1512 g_ascii_strup (const gchar *str,
1517 g_return_val_if_fail (str != NULL, NULL);
1522 result = g_strndup (str, len);
1523 for (s = result; *s; s++)
1524 *s = g_ascii_toupper (*s);
1531 * @string: the string to convert.
1533 * Converts a string to lower 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_strdown() or g_utf8_strdown()
1542 g_strdown (gchar *string)
1546 g_return_val_if_fail (string != NULL, NULL);
1548 s = (guchar *) string;
1557 return (gchar *) string;
1562 * @string: the string to convert.
1564 * Converts a string to upper case.
1566 * Return value: the string
1568 * Deprecated:2.2: This function is totally broken for the reasons discussed
1569 * in the g_strncasecmp() docs - use g_ascii_strup() or g_utf8_strup() instead.
1572 g_strup (gchar *string)
1576 g_return_val_if_fail (string != NULL, NULL);
1578 s = (guchar *) string;
1587 return (gchar *) string;
1592 * @string: the string to reverse
1594 * Reverses all of the bytes in a string. For example,
1595 * <literal>g_strreverse ("abcdef")</literal> will result
1598 * Note that g_strreverse() doesn't work on UTF-8 strings
1599 * containing multibyte characters. For that purpose, use
1600 * g_utf8_strreverse().
1602 * Returns: the same pointer passed in as @string
1605 g_strreverse (gchar *string)
1607 g_return_val_if_fail (string != NULL, NULL);
1611 register gchar *h, *t;
1614 t = string + strlen (string) - 1;
1633 * @c: any character.
1635 * Convert a character to ASCII lower case.
1637 * Unlike the standard C library tolower() function, this only
1638 * recognizes standard ASCII letters and ignores the locale, returning
1639 * all non-ASCII characters unchanged, even if they are lower case
1640 * letters in a particular character set. Also unlike the standard
1641 * library function, this takes and returns a char, not an int, so
1642 * don't call it on <literal>EOF</literal> but no need to worry about casting to #guchar
1643 * before passing a possibly non-ASCII character in.
1645 * Return value: the result of converting @c to lower case.
1646 * If @c is not an ASCII upper case letter,
1647 * @c is returned unchanged.
1650 g_ascii_tolower (gchar c)
1652 return g_ascii_isupper (c) ? c - 'A' + 'a' : c;
1657 * @c: any character.
1659 * Convert a character to ASCII upper case.
1661 * Unlike the standard C library toupper() function, this only
1662 * recognizes standard ASCII letters and ignores the locale, returning
1663 * all non-ASCII characters unchanged, even if they are upper case
1664 * letters in a particular character set. Also unlike the standard
1665 * library function, this takes and returns a char, not an int, so
1666 * don't call it on <literal>EOF</literal> but no need to worry about casting to #guchar
1667 * before passing a possibly non-ASCII character in.
1669 * Return value: the result of converting @c to upper case.
1670 * If @c is not an ASCII lower case letter,
1671 * @c is returned unchanged.
1674 g_ascii_toupper (gchar c)
1676 return g_ascii_islower (c) ? c - 'a' + 'A' : c;
1680 * g_ascii_digit_value:
1681 * @c: an ASCII character.
1683 * Determines the numeric value of a character as a decimal
1684 * digit. Differs from g_unichar_digit_value() because it takes
1685 * a char, so there's no worry about sign extension if characters
1688 * Return value: If @c is a decimal digit (according to
1689 * g_ascii_isdigit()), its numeric value. Otherwise, -1.
1692 g_ascii_digit_value (gchar c)
1694 if (g_ascii_isdigit (c))
1700 * g_ascii_xdigit_value:
1701 * @c: an ASCII character.
1703 * Determines the numeric value of a character as a hexidecimal
1704 * digit. Differs from g_unichar_xdigit_value() because it takes
1705 * a char, so there's no worry about sign extension if characters
1708 * Return value: If @c is a hex digit (according to
1709 * g_ascii_isxdigit()), its numeric value. Otherwise, -1.
1712 g_ascii_xdigit_value (gchar c)
1714 if (c >= 'A' && c <= 'F')
1715 return c - 'A' + 10;
1716 if (c >= 'a' && c <= 'f')
1717 return c - 'a' + 10;
1718 return g_ascii_digit_value (c);
1722 * g_ascii_strcasecmp:
1723 * @s1: string to compare with @s2.
1724 * @s2: string to compare with @s1.
1726 * Compare two strings, ignoring the case of ASCII characters.
1728 * Unlike the BSD strcasecmp() function, this only recognizes standard
1729 * ASCII letters and ignores the locale, treating all non-ASCII
1730 * bytes as if they are not letters.
1732 * This function should be used only on strings that are known to be
1733 * in encodings where the bytes corresponding to ASCII letters always
1734 * represent themselves. This includes UTF-8 and the ISO-8859-*
1735 * charsets, but not for instance double-byte encodings like the
1736 * Windows Codepage 932, where the trailing bytes of double-byte
1737 * characters include all ASCII letters. If you compare two CP932
1738 * strings using this function, you will get false matches.
1740 * Return value: 0 if the strings match, a negative value if @s1 < @s2,
1741 * or a positive value if @s1 > @s2.
1744 g_ascii_strcasecmp (const gchar *s1,
1749 g_return_val_if_fail (s1 != NULL, 0);
1750 g_return_val_if_fail (s2 != NULL, 0);
1754 c1 = (gint)(guchar) TOLOWER (*s1);
1755 c2 = (gint)(guchar) TOLOWER (*s2);
1761 return (((gint)(guchar) *s1) - ((gint)(guchar) *s2));
1765 * g_ascii_strncasecmp:
1766 * @s1: string to compare with @s2.
1767 * @s2: string to compare with @s1.
1768 * @n: number of characters to compare.
1770 * Compare @s1 and @s2, ignoring the case of ASCII characters and any
1771 * characters after the first @n in each string.
1773 * Unlike the BSD strcasecmp() function, this only recognizes standard
1774 * ASCII letters and ignores the locale, treating all non-ASCII
1775 * characters as if they are not letters.
1777 * The same warning as in g_ascii_strcasecmp() applies: Use this
1778 * function only on strings known to be in encodings where bytes
1779 * corresponding to ASCII letters always represent themselves.
1781 * Return value: 0 if the strings match, a negative value if @s1 < @s2,
1782 * or a positive value if @s1 > @s2.
1785 g_ascii_strncasecmp (const gchar *s1,
1791 g_return_val_if_fail (s1 != NULL, 0);
1792 g_return_val_if_fail (s2 != NULL, 0);
1794 while (n && *s1 && *s2)
1797 c1 = (gint)(guchar) TOLOWER (*s1);
1798 c2 = (gint)(guchar) TOLOWER (*s2);
1805 return (((gint) (guchar) *s1) - ((gint) (guchar) *s2));
1813 * @s2: a string to compare with @s1.
1815 * A case-insensitive string comparison, corresponding to the standard
1816 * strcasecmp() function on platforms which support it.
1818 * Return value: 0 if the strings match, a negative value if @s1 < @s2,
1819 * or a positive value if @s1 > @s2.
1821 * Deprecated:2.2: See g_strncasecmp() for a discussion of why this function
1822 * is deprecated and how to replace it.
1825 g_strcasecmp (const gchar *s1,
1828 #ifdef HAVE_STRCASECMP
1829 g_return_val_if_fail (s1 != NULL, 0);
1830 g_return_val_if_fail (s2 != NULL, 0);
1832 return strcasecmp (s1, s2);
1836 g_return_val_if_fail (s1 != NULL, 0);
1837 g_return_val_if_fail (s2 != NULL, 0);
1841 /* According to A. Cox, some platforms have islower's that
1842 * don't work right on non-uppercase
1844 c1 = isupper ((guchar)*s1) ? tolower ((guchar)*s1) : *s1;
1845 c2 = isupper ((guchar)*s2) ? tolower ((guchar)*s2) : *s2;
1851 return (((gint)(guchar) *s1) - ((gint)(guchar) *s2));
1858 * @s2: a string to compare with @s1.
1859 * @n: the maximum number of characters to compare.
1861 * A case-insensitive string comparison, corresponding to the standard
1862 * strncasecmp() function on platforms which support it.
1863 * It is similar to g_strcasecmp() except it only compares the first @n
1864 * characters of the strings.
1866 * Return value: 0 if the strings match, a negative value if @s1 < @s2,
1867 * or a positive value if @s1 > @s2.
1869 * Deprecated:2.2: The problem with g_strncasecmp() is that it does the
1870 * comparison by calling toupper()/tolower(). These functions are
1871 * locale-specific and operate on single bytes. However, it is impossible
1872 * to handle things correctly from an I18N standpoint by operating on
1873 * bytes, since characters may be multibyte. Thus g_strncasecmp() is
1874 * broken if your string is guaranteed to be ASCII, since it's
1875 * locale-sensitive, and it's broken if your string is localized, since
1876 * it doesn't work on many encodings at all, including UTF-8, EUC-JP,
1879 * There are therefore two replacement techniques: g_ascii_strncasecmp(),
1880 * which only works on ASCII and is not locale-sensitive, and
1881 * g_utf8_casefold() followed by strcmp() on the resulting strings, which is
1882 * good for case-insensitive sorting of UTF-8.
1885 g_strncasecmp (const gchar *s1,
1889 #ifdef HAVE_STRNCASECMP
1890 return strncasecmp (s1, s2, n);
1894 g_return_val_if_fail (s1 != NULL, 0);
1895 g_return_val_if_fail (s2 != NULL, 0);
1897 while (n && *s1 && *s2)
1900 /* According to A. Cox, some platforms have islower's that
1901 * don't work right on non-uppercase
1903 c1 = isupper ((guchar)*s1) ? tolower ((guchar)*s1) : *s1;
1904 c2 = isupper ((guchar)*s2) ? tolower ((guchar)*s2) : *s2;
1911 return (((gint) (guchar) *s1) - ((gint) (guchar) *s2));
1919 * @string: the string to convert
1920 * @delimiters: (allow-none): a string containing the current delimiters, or %NULL
1921 * to use the standard delimiters defined in #G_STR_DELIMITERS
1922 * @new_delimiter: the new delimiter character
1924 * Converts any delimiter characters in @string to @new_delimiter.
1925 * Any characters in @string which are found in @delimiters are
1926 * changed to the @new_delimiter character. Modifies @string in place,
1927 * and returns @string itself, not a copy. The return value is to
1928 * allow nesting such as
1930 * g_ascii_strup (g_strdelimit (str, "abc", '?'))
1936 g_strdelimit (gchar *string,
1937 const gchar *delimiters,
1942 g_return_val_if_fail (string != NULL, NULL);
1945 delimiters = G_STR_DELIMITERS;
1947 for (c = string; *c; c++)
1949 if (strchr (delimiters, *c))
1958 * @string: a nul-terminated array of bytes
1959 * @valid_chars: bytes permitted in @string
1960 * @substitutor: replacement character for disallowed bytes
1962 * For each character in @string, if the character is not in
1963 * @valid_chars, replaces the character with @substitutor.
1964 * Modifies @string in place, and return @string itself, not
1965 * a copy. The return value is to allow nesting such as
1967 * g_ascii_strup (g_strcanon (str, "abc", '?'))
1973 g_strcanon (gchar *string,
1974 const gchar *valid_chars,
1979 g_return_val_if_fail (string != NULL, NULL);
1980 g_return_val_if_fail (valid_chars != NULL, NULL);
1982 for (c = string; *c; c++)
1984 if (!strchr (valid_chars, *c))
1993 * @source: a string to compress
1995 * Replaces all escaped characters with their one byte equivalent.
1997 * This function does the reverse conversion of g_strescape().
1999 * Returns: a newly-allocated copy of @source with all escaped
2000 * character compressed
2003 g_strcompress (const gchar *source)
2005 const gchar *p = source, *octal;
2009 g_return_val_if_fail (source != NULL, NULL);
2011 dest = g_malloc (strlen (source) + 1);
2022 g_warning ("g_strcompress: trailing \\");
2024 case '0': case '1': case '2': case '3': case '4':
2025 case '5': case '6': case '7':
2028 while ((p < octal + 3) && (*p >= '0') && (*p <= '7'))
2030 *q = (*q * 8) + (*p - '0');
2054 default: /* Also handles \" and \\ */
2071 * @source: a string to escape
2072 * @exceptions: a string of characters not to escape in @source
2074 * Escapes the special characters '\b', '\f', '\n', '\r', '\t', '\v', '\'
2075 * and '"' in the string @source by inserting a '\' before
2076 * them. Additionally all characters in the range 0x01-0x1F (everything
2077 * below SPACE) and in the range 0x7F-0xFF (all non-ASCII chars) are
2078 * replaced with a '\' followed by their octal representation.
2079 * Characters supplied in @exceptions are not escaped.
2081 * g_strcompress() does the reverse conversion.
2083 * Returns: a newly-allocated copy of @source with certain
2084 * characters escaped. See above.
2087 g_strescape (const gchar *source,
2088 const gchar *exceptions)
2095 g_return_val_if_fail (source != NULL, NULL);
2097 p = (guchar *) source;
2098 /* Each source byte needs maximally four destination chars (\777) */
2099 q = dest = g_malloc (strlen (source) * 4 + 1);
2101 memset (excmap, 0, 256);
2104 guchar *e = (guchar *) exceptions;
2154 if ((*p < ' ') || (*p >= 0177))
2157 *q++ = '0' + (((*p) >> 6) & 07);
2158 *q++ = '0' + (((*p) >> 3) & 07);
2159 *q++ = '0' + ((*p) & 07);
2174 * @string: a string to remove the leading whitespace from
2176 * Removes leading whitespace from a string, by moving the rest
2177 * of the characters forward.
2179 * This function doesn't allocate or reallocate any memory;
2180 * it modifies @string in place. The pointer to @string is
2181 * returned to allow the nesting of functions.
2183 * Also see g_strchomp() and g_strstrip().
2188 g_strchug (gchar *string)
2192 g_return_val_if_fail (string != NULL, NULL);
2194 for (start = (guchar*) string; *start && g_ascii_isspace (*start); start++)
2197 g_memmove (string, start, strlen ((gchar *) start) + 1);
2204 * @string: a string to remove the trailing whitespace from
2206 * Removes trailing whitespace from a string.
2208 * This function doesn't allocate or reallocate any memory;
2209 * it modifies @string in place. The pointer to @string is
2210 * returned to allow the nesting of functions.
2212 * Also see g_strchug() and g_strstrip().
2217 g_strchomp (gchar *string)
2221 g_return_val_if_fail (string != NULL, NULL);
2223 len = strlen (string);
2226 if (g_ascii_isspace ((guchar) string[len]))
2237 * @string: a string to split
2238 * @delimiter: a string which specifies the places at which to split
2239 * the string. The delimiter is not included in any of the resulting
2240 * strings, unless @max_tokens is reached.
2241 * @max_tokens: the maximum number of pieces to split @string into.
2242 * If this is less than 1, the string is split completely.
2244 * Splits a string into a maximum of @max_tokens pieces, using the given
2245 * @delimiter. If @max_tokens is reached, the remainder of @string is
2246 * appended to the last token.
2248 * As a special case, the result of splitting the empty string "" is an empty
2249 * vector, not a vector containing a single string. The reason for this
2250 * special case is that being able to represent a empty vector is typically
2251 * more useful than consistent handling of empty elements. If you do need
2252 * to represent empty elements, you'll need to check for the empty string
2253 * before calling g_strsplit().
2255 * Return value: a newly-allocated %NULL-terminated array of strings. Use
2256 * g_strfreev() to free it.
2259 g_strsplit (const gchar *string,
2260 const gchar *delimiter,
2263 GSList *string_list = NULL, *slist;
2264 gchar **str_array, *s;
2266 const gchar *remainder;
2268 g_return_val_if_fail (string != NULL, NULL);
2269 g_return_val_if_fail (delimiter != NULL, NULL);
2270 g_return_val_if_fail (delimiter[0] != '\0', NULL);
2273 max_tokens = G_MAXINT;
2276 s = strstr (remainder, delimiter);
2279 gsize delimiter_len = strlen (delimiter);
2281 while (--max_tokens && s)
2285 len = s - remainder;
2286 string_list = g_slist_prepend (string_list,
2287 g_strndup (remainder, len));
2289 remainder = s + delimiter_len;
2290 s = strstr (remainder, delimiter);
2296 string_list = g_slist_prepend (string_list, g_strdup (remainder));
2299 str_array = g_new (gchar*, n + 1);
2301 str_array[n--] = NULL;
2302 for (slist = string_list; slist; slist = slist->next)
2303 str_array[n--] = slist->data;
2305 g_slist_free (string_list);
2312 * @string: The string to be tokenized
2313 * @delimiters: A nul-terminated string containing bytes that are used
2314 * to split the string.
2315 * @max_tokens: The maximum number of tokens to split @string into.
2316 * If this is less than 1, the string is split completely
2318 * Splits @string into a number of tokens not containing any of the characters
2319 * in @delimiter. A token is the (possibly empty) longest string that does not
2320 * contain any of the characters in @delimiters. If @max_tokens is reached, the
2321 * remainder is appended to the last token.
2323 * For example the result of g_strsplit_set ("abc:def/ghi", ":/", -1) is a
2324 * %NULL-terminated vector containing the three strings "abc", "def",
2327 * The result if g_strsplit_set (":def/ghi:", ":/", -1) is a %NULL-terminated
2328 * vector containing the four strings "", "def", "ghi", and "".
2330 * As a special case, the result of splitting the empty string "" is an empty
2331 * vector, not a vector containing a single string. The reason for this
2332 * special case is that being able to represent a empty vector is typically
2333 * more useful than consistent handling of empty elements. If you do need
2334 * to represent empty elements, you'll need to check for the empty string
2335 * before calling g_strsplit_set().
2337 * Note that this function works on bytes not characters, so it can't be used
2338 * to delimit UTF-8 strings for anything but ASCII characters.
2340 * Return value: a newly-allocated %NULL-terminated array of strings. Use
2341 * g_strfreev() to free it.
2346 g_strsplit_set (const gchar *string,
2347 const gchar *delimiters,
2350 gboolean delim_table[256];
2351 GSList *tokens, *list;
2354 const gchar *current;
2358 g_return_val_if_fail (string != NULL, NULL);
2359 g_return_val_if_fail (delimiters != NULL, NULL);
2362 max_tokens = G_MAXINT;
2364 if (*string == '\0')
2366 result = g_new (char *, 1);
2371 memset (delim_table, FALSE, sizeof (delim_table));
2372 for (s = delimiters; *s != '\0'; ++s)
2373 delim_table[*(guchar *)s] = TRUE;
2378 s = current = string;
2381 if (delim_table[*(guchar *)s] && n_tokens + 1 < max_tokens)
2383 token = g_strndup (current, s - current);
2384 tokens = g_slist_prepend (tokens, token);
2393 token = g_strndup (current, s - current);
2394 tokens = g_slist_prepend (tokens, token);
2397 result = g_new (gchar *, n_tokens + 1);
2399 result[n_tokens] = NULL;
2400 for (list = tokens; list != NULL; list = list->next)
2401 result[--n_tokens] = list->data;
2403 g_slist_free (tokens);
2410 * @str_array: a %NULL-terminated array of strings to free
2412 * Frees a %NULL-terminated array of strings, and the array itself.
2413 * If called on a %NULL value, g_strfreev() simply returns.
2416 g_strfreev (gchar **str_array)
2422 for (i = 0; str_array[i] != NULL; i++)
2423 g_free (str_array[i]);
2431 * @str_array: a %NULL-terminated array of strings
2433 * Copies %NULL-terminated array of strings. The copy is a deep copy;
2434 * the new array should be freed by first freeing each string, then
2435 * the array itself. g_strfreev() does this for you. If called
2436 * on a %NULL value, g_strdupv() simply returns %NULL.
2438 * Return value: a new %NULL-terminated array of strings.
2441 g_strdupv (gchar **str_array)
2449 while (str_array[i])
2452 retval = g_new (gchar*, i + 1);
2455 while (str_array[i])
2457 retval[i] = g_strdup (str_array[i]);
2470 * @separator: (allow-none): a string to insert between each of the strings, or %NULL
2471 * @str_array: a %NULL-terminated array of strings to join
2473 * Joins a number of strings together to form one long string, with the
2474 * optional @separator inserted between each of them. The returned string
2475 * should be freed with g_free().
2477 * Returns: a newly-allocated string containing all of the strings joined
2478 * together, with @separator between them
2481 g_strjoinv (const gchar *separator,
2487 g_return_val_if_fail (str_array != NULL, NULL);
2489 if (separator == NULL)
2496 gsize separator_len;
2498 separator_len = strlen (separator);
2499 /* First part, getting length */
2500 len = 1 + strlen (str_array[0]);
2501 for (i = 1; str_array[i] != NULL; i++)
2502 len += strlen (str_array[i]);
2503 len += separator_len * (i - 1);
2505 /* Second part, building string */
2506 string = g_new (gchar, len);
2507 ptr = g_stpcpy (string, *str_array);
2508 for (i = 1; str_array[i] != NULL; i++)
2510 ptr = g_stpcpy (ptr, separator);
2511 ptr = g_stpcpy (ptr, str_array[i]);
2515 string = g_strdup ("");
2522 * @separator: (allow-none): a string to insert between each of the strings, or %NULL
2523 * @...: a %NULL-terminated list of strings to join
2525 * Joins a number of strings together to form one long string, with the
2526 * optional @separator inserted between each of them. The returned string
2527 * should be freed with g_free().
2529 * Returns: a newly-allocated string containing all of the strings joined
2530 * together, with @separator between them
2533 g_strjoin (const gchar *separator,
2539 gsize separator_len;
2542 if (separator == NULL)
2545 separator_len = strlen (separator);
2547 va_start (args, separator);
2549 s = va_arg (args, gchar*);
2553 /* First part, getting length */
2554 len = 1 + strlen (s);
2556 s = va_arg (args, gchar*);
2559 len += separator_len + strlen (s);
2560 s = va_arg (args, gchar*);
2564 /* Second part, building string */
2565 string = g_new (gchar, len);
2567 va_start (args, separator);
2569 s = va_arg (args, gchar*);
2570 ptr = g_stpcpy (string, s);
2572 s = va_arg (args, gchar*);
2575 ptr = g_stpcpy (ptr, separator);
2576 ptr = g_stpcpy (ptr, s);
2577 s = va_arg (args, gchar*);
2581 string = g_strdup ("");
2591 * @haystack: a string
2592 * @haystack_len: the maximum length of @haystack. Note that -1 is
2593 * a valid length, if @haystack is nul-terminated, meaning it will
2594 * search through the whole string.
2595 * @needle: the string to search for
2597 * Searches the string @haystack for the first occurrence
2598 * of the string @needle, limiting the length of the search
2601 * Return value: a pointer to the found occurrence, or
2602 * %NULL if not found.
2605 g_strstr_len (const gchar *haystack,
2606 gssize haystack_len,
2607 const gchar *needle)
2609 g_return_val_if_fail (haystack != NULL, NULL);
2610 g_return_val_if_fail (needle != NULL, NULL);
2612 if (haystack_len < 0)
2613 return strstr (haystack, needle);
2616 const gchar *p = haystack;
2617 gsize needle_len = strlen (needle);
2621 if (needle_len == 0)
2622 return (gchar *)haystack;
2624 if (haystack_len < needle_len)
2627 end = haystack + haystack_len - needle_len;
2629 while (p <= end && *p)
2631 for (i = 0; i < needle_len; i++)
2632 if (p[i] != needle[i])
2647 * @haystack: a nul-terminated string
2648 * @needle: the nul-terminated string to search for
2650 * Searches the string @haystack for the last occurrence
2651 * of the string @needle.
2653 * Return value: a pointer to the found occurrence, or
2654 * %NULL if not found.
2657 g_strrstr (const gchar *haystack,
2658 const gchar *needle)
2665 g_return_val_if_fail (haystack != NULL, NULL);
2666 g_return_val_if_fail (needle != NULL, NULL);
2668 needle_len = strlen (needle);
2669 haystack_len = strlen (haystack);
2671 if (needle_len == 0)
2672 return (gchar *)haystack;
2674 if (haystack_len < needle_len)
2677 p = haystack + haystack_len - needle_len;
2679 while (p >= haystack)
2681 for (i = 0; i < needle_len; i++)
2682 if (p[i] != needle[i])
2696 * @haystack: a nul-terminated string
2697 * @haystack_len: the maximum length of @haystack
2698 * @needle: the nul-terminated string to search for
2700 * Searches the string @haystack for the last occurrence
2701 * of the string @needle, limiting the length of the search
2704 * Return value: a pointer to the found occurrence, or
2705 * %NULL if not found.
2708 g_strrstr_len (const gchar *haystack,
2709 gssize haystack_len,
2710 const gchar *needle)
2712 g_return_val_if_fail (haystack != NULL, NULL);
2713 g_return_val_if_fail (needle != NULL, NULL);
2715 if (haystack_len < 0)
2716 return g_strrstr (haystack, needle);
2719 gsize needle_len = strlen (needle);
2720 const gchar *haystack_max = haystack + haystack_len;
2721 const gchar *p = haystack;
2724 while (p < haystack_max && *p)
2727 if (p < haystack + needle_len)
2732 while (p >= haystack)
2734 for (i = 0; i < needle_len; i++)
2735 if (p[i] != needle[i])
2751 * @str: a nul-terminated string
2752 * @suffix: the nul-terminated suffix to look for
2754 * Looks whether the string @str ends with @suffix.
2756 * Return value: %TRUE if @str end with @suffix, %FALSE otherwise.
2761 g_str_has_suffix (const gchar *str,
2762 const gchar *suffix)
2767 g_return_val_if_fail (str != NULL, FALSE);
2768 g_return_val_if_fail (suffix != NULL, FALSE);
2770 str_len = strlen (str);
2771 suffix_len = strlen (suffix);
2773 if (str_len < suffix_len)
2776 return strcmp (str + str_len - suffix_len, suffix) == 0;
2781 * @str: a nul-terminated string
2782 * @prefix: the nul-terminated prefix to look for
2784 * Looks whether the string @str begins with @prefix.
2786 * Return value: %TRUE if @str begins with @prefix, %FALSE otherwise.
2791 g_str_has_prefix (const gchar *str,
2792 const gchar *prefix)
2797 g_return_val_if_fail (str != NULL, FALSE);
2798 g_return_val_if_fail (prefix != NULL, FALSE);
2800 str_len = strlen (str);
2801 prefix_len = strlen (prefix);
2803 if (str_len < prefix_len)
2806 return strncmp (str, prefix, prefix_len) == 0;
2811 * @str_array: a %NULL-terminated array of strings
2813 * Returns the length of the given %NULL-terminated
2814 * string array @str_array.
2816 * Return value: length of @str_array.
2821 g_strv_length (gchar **str_array)
2825 g_return_val_if_fail (str_array != NULL, 0);
2827 while (str_array[i])