1 /* guniprop.c - Unicode character properties.
3 * Copyright (C) 1999 Tom Tromey
4 * Copyright (C) 2000 Red Hat, Inc.
6 * This library is free software; you can redistribute it and/or
7 * modify it under the terms of the GNU Lesser General Public
8 * License as published by the Free Software Foundation; either
9 * version 2 of the License, or (at your option) any later version.
11 * This library is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14 * Lesser General Public License for more details.
16 * You should have received a copy of the GNU Lesser General Public
17 * License along with this library; if not, see <http://www.gnu.org/licenses/>.
29 #include "gtestutils.h"
32 #include "gunichartables.h"
33 #include "gmirroringtable.h"
34 #include "gscripttable.h"
35 #include "gunicodeprivate.h"
40 #define ATTR_TABLE(Page) (((Page) <= G_UNICODE_LAST_PAGE_PART1) \
41 ? attr_table_part1[Page] \
42 : attr_table_part2[(Page) - 0xe00])
44 #define ATTTABLE(Page, Char) \
45 ((ATTR_TABLE(Page) == G_UNICODE_MAX_TABLE_INDEX) ? 0 : (attr_data[ATTR_TABLE(Page)][Char]))
47 #define TTYPE_PART1(Page, Char) \
48 ((type_table_part1[Page] >= G_UNICODE_MAX_TABLE_INDEX) \
49 ? (type_table_part1[Page] - G_UNICODE_MAX_TABLE_INDEX) \
50 : (type_data[type_table_part1[Page]][Char]))
52 #define TTYPE_PART2(Page, Char) \
53 ((type_table_part2[Page] >= G_UNICODE_MAX_TABLE_INDEX) \
54 ? (type_table_part2[Page] - G_UNICODE_MAX_TABLE_INDEX) \
55 : (type_data[type_table_part2[Page]][Char]))
58 (((Char) <= G_UNICODE_LAST_CHAR_PART1) \
59 ? TTYPE_PART1 ((Char) >> 8, (Char) & 0xff) \
60 : (((Char) >= 0xe0000 && (Char) <= G_UNICODE_LAST_CHAR) \
61 ? TTYPE_PART2 (((Char) - 0xe0000) >> 8, (Char) & 0xff) \
62 : G_UNICODE_UNASSIGNED))
65 #define IS(Type, Class) (((guint)1 << (Type)) & (Class))
66 #define OR(Type, Rest) (((guint)1 << (Type)) | (Rest))
70 #define ISALPHA(Type) IS ((Type), \
71 OR (G_UNICODE_LOWERCASE_LETTER, \
72 OR (G_UNICODE_UPPERCASE_LETTER, \
73 OR (G_UNICODE_TITLECASE_LETTER, \
74 OR (G_UNICODE_MODIFIER_LETTER, \
75 OR (G_UNICODE_OTHER_LETTER, 0))))))
77 #define ISALDIGIT(Type) IS ((Type), \
78 OR (G_UNICODE_DECIMAL_NUMBER, \
79 OR (G_UNICODE_LETTER_NUMBER, \
80 OR (G_UNICODE_OTHER_NUMBER, \
81 OR (G_UNICODE_LOWERCASE_LETTER, \
82 OR (G_UNICODE_UPPERCASE_LETTER, \
83 OR (G_UNICODE_TITLECASE_LETTER, \
84 OR (G_UNICODE_MODIFIER_LETTER, \
85 OR (G_UNICODE_OTHER_LETTER, 0)))))))))
87 #define ISMARK(Type) IS ((Type), \
88 OR (G_UNICODE_NON_SPACING_MARK, \
89 OR (G_UNICODE_SPACING_MARK, \
90 OR (G_UNICODE_ENCLOSING_MARK, 0))))
92 #define ISZEROWIDTHTYPE(Type) IS ((Type), \
93 OR (G_UNICODE_NON_SPACING_MARK, \
94 OR (G_UNICODE_ENCLOSING_MARK, \
95 OR (G_UNICODE_FORMAT, 0))))
99 * @c: a Unicode character
101 * Determines whether a character is alphanumeric.
102 * Given some UTF-8 text, obtain a character value
103 * with g_utf8_get_char().
105 * Returns: %TRUE if @c is an alphanumeric character
108 g_unichar_isalnum (gunichar c
)
110 return ISALDIGIT (TYPE (c
)) ? TRUE
: FALSE
;
115 * @c: a Unicode character
117 * Determines whether a character is alphabetic (i.e. a letter).
118 * Given some UTF-8 text, obtain a character value with
121 * Returns: %TRUE if @c is an alphabetic character
124 g_unichar_isalpha (gunichar c
)
126 return ISALPHA (TYPE (c
)) ? TRUE
: FALSE
;
132 * @c: a Unicode character
134 * Determines whether a character is a control character.
135 * Given some UTF-8 text, obtain a character value with
138 * Returns: %TRUE if @c is a control character
141 g_unichar_iscntrl (gunichar c
)
143 return TYPE (c
) == G_UNICODE_CONTROL
;
148 * @c: a Unicode character
150 * Determines whether a character is numeric (i.e. a digit). This
151 * covers ASCII 0-9 and also digits in other languages/scripts. Given
152 * some UTF-8 text, obtain a character value with g_utf8_get_char().
154 * Returns: %TRUE if @c is a digit
157 g_unichar_isdigit (gunichar c
)
159 return TYPE (c
) == G_UNICODE_DECIMAL_NUMBER
;
165 * @c: a Unicode character
167 * Determines whether a character is printable and not a space
168 * (returns %FALSE for control characters, format characters, and
169 * spaces). g_unichar_isprint() is similar, but returns %TRUE for
170 * spaces. Given some UTF-8 text, obtain a character value with
173 * Returns: %TRUE if @c is printable unless it's a space
176 g_unichar_isgraph (gunichar c
)
179 OR (G_UNICODE_CONTROL
,
180 OR (G_UNICODE_FORMAT
,
181 OR (G_UNICODE_UNASSIGNED
,
182 OR (G_UNICODE_SURROGATE
,
183 OR (G_UNICODE_SPACE_SEPARATOR
,
189 * @c: a Unicode character
191 * Determines whether a character is a lowercase letter.
192 * Given some UTF-8 text, obtain a character value with
195 * Returns: %TRUE if @c is a lowercase letter
198 g_unichar_islower (gunichar c
)
200 return TYPE (c
) == G_UNICODE_LOWERCASE_LETTER
;
206 * @c: a Unicode character
208 * Determines whether a character is printable.
209 * Unlike g_unichar_isgraph(), returns %TRUE for spaces.
210 * Given some UTF-8 text, obtain a character value with
213 * Returns: %TRUE if @c is printable
216 g_unichar_isprint (gunichar c
)
219 OR (G_UNICODE_CONTROL
,
220 OR (G_UNICODE_FORMAT
,
221 OR (G_UNICODE_UNASSIGNED
,
222 OR (G_UNICODE_SURROGATE
,
228 * @c: a Unicode character
230 * Determines whether a character is punctuation or a symbol.
231 * Given some UTF-8 text, obtain a character value with
234 * Returns: %TRUE if @c is a punctuation or symbol character
237 g_unichar_ispunct (gunichar c
)
240 OR (G_UNICODE_CONNECT_PUNCTUATION
,
241 OR (G_UNICODE_DASH_PUNCTUATION
,
242 OR (G_UNICODE_CLOSE_PUNCTUATION
,
243 OR (G_UNICODE_FINAL_PUNCTUATION
,
244 OR (G_UNICODE_INITIAL_PUNCTUATION
,
245 OR (G_UNICODE_OTHER_PUNCTUATION
,
246 OR (G_UNICODE_OPEN_PUNCTUATION
,
247 OR (G_UNICODE_CURRENCY_SYMBOL
,
248 OR (G_UNICODE_MODIFIER_SYMBOL
,
249 OR (G_UNICODE_MATH_SYMBOL
,
250 OR (G_UNICODE_OTHER_SYMBOL
,
251 0)))))))))))) ? TRUE
: FALSE
;
256 * @c: a Unicode character
258 * Determines whether a character is a space, tab, or line separator
259 * (newline, carriage return, etc.). Given some UTF-8 text, obtain a
260 * character value with g_utf8_get_char().
262 * (Note: don't use this to do word breaking; you have to use
263 * Pango or equivalent to get word breaking right, the algorithm
264 * is fairly complex.)
266 * Returns: %TRUE if @c is a space character
269 g_unichar_isspace (gunichar c
)
273 /* special-case these since Unicode thinks they are not spaces */
284 OR (G_UNICODE_SPACE_SEPARATOR
,
285 OR (G_UNICODE_LINE_SEPARATOR
,
286 OR (G_UNICODE_PARAGRAPH_SEPARATOR
,
287 0)))) ? TRUE
: FALSE
;
295 * @c: a Unicode character
297 * Determines whether a character is a mark (non-spacing mark,
298 * combining mark, or enclosing mark in Unicode speak).
299 * Given some UTF-8 text, obtain a character value
300 * with g_utf8_get_char().
302 * Note: in most cases where isalpha characters are allowed,
303 * ismark characters should be allowed to as they are essential
304 * for writing most European languages as well as many non-Latin
307 * Returns: %TRUE if @c is a mark character
312 g_unichar_ismark (gunichar c
)
314 return ISMARK (TYPE (c
));
319 * @c: a Unicode character
321 * Determines if a character is uppercase.
323 * Returns: %TRUE if @c is an uppercase character
326 g_unichar_isupper (gunichar c
)
328 return TYPE (c
) == G_UNICODE_UPPERCASE_LETTER
;
333 * @c: a Unicode character
335 * Determines if a character is titlecase. Some characters in
336 * Unicode which are composites, such as the DZ digraph
337 * have three case variants instead of just two. The titlecase
338 * form is used at the beginning of a word where only the
339 * first letter is capitalized. The titlecase form of the DZ
340 * digraph is U+01F2 LATIN CAPITAL LETTTER D WITH SMALL LETTER Z.
342 * Returns: %TRUE if the character is titlecase
345 g_unichar_istitle (gunichar c
)
348 for (i
= 0; i
< G_N_ELEMENTS (title_table
); ++i
)
349 if (title_table
[i
][0] == c
)
355 * g_unichar_isxdigit:
356 * @c: a Unicode character.
358 * Determines if a character is a hexidecimal digit.
360 * Returns: %TRUE if the character is a hexadecimal digit
363 g_unichar_isxdigit (gunichar c
)
365 return ((c
>= 'a' && c
<= 'f')
366 || (c
>= 'A' && c
<= 'F')
367 || (TYPE (c
) == G_UNICODE_DECIMAL_NUMBER
));
371 * g_unichar_isdefined:
372 * @c: a Unicode character
374 * Determines if a given character is assigned in the Unicode
377 * Returns: %TRUE if the character has an assigned value
380 g_unichar_isdefined (gunichar c
)
383 OR (G_UNICODE_UNASSIGNED
,
384 OR (G_UNICODE_SURROGATE
,
389 * g_unichar_iszerowidth:
390 * @c: a Unicode character
392 * Determines if a given character typically takes zero width when rendered.
393 * The return value is %TRUE for all non-spacing and enclosing marks
394 * (e.g., combining accents), format characters, zero-width
395 * space, but not U+00AD SOFT HYPHEN.
397 * A typical use of this function is with one of g_unichar_iswide() or
398 * g_unichar_iswide_cjk() to determine the number of cells a string occupies
399 * when displayed on a grid display (terminals). However, note that not all
400 * terminals support zero-width rendering of zero-width marks.
402 * Returns: %TRUE if the character has zero width
407 g_unichar_iszerowidth (gunichar c
)
409 if (G_UNLIKELY (c
== 0x00AD))
412 if (G_UNLIKELY (ISZEROWIDTHTYPE (TYPE (c
))))
415 if (G_UNLIKELY ((c
>= 0x1160 && c
< 0x1200) ||
423 interval_compare (const void *key
, const void *elt
)
425 gunichar c
= GPOINTER_TO_UINT (key
);
426 struct Interval
*interval
= (struct Interval
*)elt
;
428 if (c
< interval
->start
)
430 if (c
> interval
->end
)
436 #define G_WIDTH_TABLE_MIDPOINT (G_N_ELEMENTS (g_unicode_width_table_wide) / 2)
438 static inline gboolean
439 g_unichar_iswide_bsearch (gunichar ch
)
442 int upper
= G_N_ELEMENTS (g_unicode_width_table_wide
) + 1;
443 static int saved_mid
= G_WIDTH_TABLE_MIDPOINT
;
448 if (ch
< g_unicode_width_table_wide
[mid
].start
)
450 else if (ch
> g_unicode_width_table_wide
[mid
].end
)
455 mid
= (lower
+ upper
) / 2;
457 while (lower
<= upper
);
464 * @c: a Unicode character
466 * Determines if a character is typically rendered in a double-width
469 * Returns: %TRUE if the character is wide
472 g_unichar_iswide (gunichar c
)
474 if (c
< g_unicode_width_table_wide
[0].start
)
477 return g_unichar_iswide_bsearch (c
);
482 * g_unichar_iswide_cjk:
483 * @c: a Unicode character
485 * Determines if a character is typically rendered in a double-width
486 * cell under legacy East Asian locales. If a character is wide according to
487 * g_unichar_iswide(), then it is also reported wide with this function, but
488 * the converse is not necessarily true. See the
489 * [Unicode Standard Annex #11](http://www.unicode.org/reports/tr11/)
492 * If a character passes the g_unichar_iswide() test then it will also pass
493 * this test, but not the other way around. Note that some characters may
494 * pass both this test and g_unichar_iszerowidth().
496 * Returns: %TRUE if the character is wide in legacy East Asian locales
501 g_unichar_iswide_cjk (gunichar c
)
503 if (g_unichar_iswide (c
))
506 if (bsearch (GUINT_TO_POINTER (c
),
507 g_unicode_width_table_ambiguous
,
508 G_N_ELEMENTS (g_unicode_width_table_ambiguous
),
509 sizeof g_unicode_width_table_ambiguous
[0],
519 * @c: a Unicode character
521 * Converts a character to uppercase.
523 * Returns: the result of converting @c to uppercase.
524 * If @c is not an lowercase or titlecase character,
525 * or has no upper case equivalent @c is returned unchanged.
528 g_unichar_toupper (gunichar c
)
531 if (t
== G_UNICODE_LOWERCASE_LETTER
)
533 gunichar val
= ATTTABLE (c
>> 8, c
& 0xff);
534 if (val
>= 0x1000000)
536 const gchar
*p
= special_case_table
+ val
- 0x1000000;
537 val
= g_utf8_get_char (p
);
539 /* Some lowercase letters, e.g., U+000AA, FEMININE ORDINAL INDICATOR,
540 * do not have an uppercase equivalent, in which case val will be
543 return val
? val
: c
;
545 else if (t
== G_UNICODE_TITLECASE_LETTER
)
548 for (i
= 0; i
< G_N_ELEMENTS (title_table
); ++i
)
550 if (title_table
[i
][0] == c
)
551 return title_table
[i
][1] ? title_table
[i
][1] : c
;
559 * @c: a Unicode character.
561 * Converts a character to lower case.
563 * Returns: the result of converting @c to lower case.
564 * If @c is not an upperlower or titlecase character,
565 * or has no lowercase equivalent @c is returned unchanged.
568 g_unichar_tolower (gunichar c
)
571 if (t
== G_UNICODE_UPPERCASE_LETTER
)
573 gunichar val
= ATTTABLE (c
>> 8, c
& 0xff);
574 if (val
>= 0x1000000)
576 const gchar
*p
= special_case_table
+ val
- 0x1000000;
577 return g_utf8_get_char (p
);
581 /* Not all uppercase letters are guaranteed to have a lowercase
582 * equivalent. If this is the case, val will be zero. */
583 return val
? val
: c
;
586 else if (t
== G_UNICODE_TITLECASE_LETTER
)
589 for (i
= 0; i
< G_N_ELEMENTS (title_table
); ++i
)
591 if (title_table
[i
][0] == c
)
592 return title_table
[i
][2];
600 * @c: a Unicode character
602 * Converts a character to the titlecase.
604 * Returns: the result of converting @c to titlecase.
605 * If @c is not an uppercase or lowercase character,
606 * @c is returned unchanged.
609 g_unichar_totitle (gunichar c
)
612 for (i
= 0; i
< G_N_ELEMENTS (title_table
); ++i
)
614 if (title_table
[i
][0] == c
|| title_table
[i
][1] == c
615 || title_table
[i
][2] == c
)
616 return title_table
[i
][0];
619 if (TYPE (c
) == G_UNICODE_LOWERCASE_LETTER
)
620 return g_unichar_toupper (c
);
626 * g_unichar_digit_value:
627 * @c: a Unicode character
629 * Determines the numeric value of a character as a decimal
632 * Returns: If @c is a decimal digit (according to
633 * g_unichar_isdigit()), its numeric value. Otherwise, -1.
636 g_unichar_digit_value (gunichar c
)
638 if (TYPE (c
) == G_UNICODE_DECIMAL_NUMBER
)
639 return ATTTABLE (c
>> 8, c
& 0xff);
644 * g_unichar_xdigit_value:
645 * @c: a Unicode character
647 * Determines the numeric value of a character as a hexidecimal
650 * Returns: If @c is a hex digit (according to
651 * g_unichar_isxdigit()), its numeric value. Otherwise, -1.
654 g_unichar_xdigit_value (gunichar c
)
656 if (c
>= 'A' && c
<= 'F')
658 if (c
>= 'a' && c
<= 'f')
660 if (TYPE (c
) == G_UNICODE_DECIMAL_NUMBER
)
661 return ATTTABLE (c
>> 8, c
& 0xff);
667 * @c: a Unicode character
669 * Classifies a Unicode character by type.
671 * Returns: the type of the character.
674 g_unichar_type (gunichar c
)
680 * Case mapping functions
690 get_locale_type (void)
693 char *tem
= g_win32_getlocale ();
700 const char *locale
= setlocale (LC_CTYPE
, NULL
);
703 return LOCALE_NORMAL
;
709 if (locale
[1] == 'z')
710 return LOCALE_TURKIC
;
713 if (locale
[1] == 't')
714 return LOCALE_LITHUANIAN
;
717 if (locale
[1] == 'r')
718 return LOCALE_TURKIC
;
722 return LOCALE_NORMAL
;
726 output_marks (const char **p_inout
,
730 const char *p
= *p_inout
;
735 gunichar c
= g_utf8_get_char (p
);
737 if (ISMARK (TYPE (c
)))
739 if (!remove_dot
|| c
!= 0x307 /* COMBINING DOT ABOVE */)
740 len
+= g_unichar_to_utf8 (c
, out_buffer
? out_buffer
+ len
: NULL
);
741 p
= g_utf8_next_char (p
);
752 output_special_case (gchar
*out_buffer
,
757 const gchar
*p
= special_case_table
+ offset
;
760 if (type
!= G_UNICODE_TITLECASE_LETTER
)
761 p
= g_utf8_next_char (p
);
768 memcpy (out_buffer
, p
, len
);
774 real_toupper (const gchar
*str
,
777 LocaleType locale_type
)
779 const gchar
*p
= str
;
780 const char *last
= NULL
;
782 gboolean last_was_i
= FALSE
;
784 while ((max_len
< 0 || p
< str
+ max_len
) && *p
)
786 gunichar c
= g_utf8_get_char (p
);
791 p
= g_utf8_next_char (p
);
793 if (locale_type
== LOCALE_LITHUANIAN
)
801 /* Nasty, need to remove any dot above. Though
802 * I think only E WITH DOT ABOVE occurs in practice
803 * which could simplify this considerably.
806 gunichar decomp
[G_UNICHAR_MAX_DECOMPOSITION_LENGTH
];
808 decomp_len
= g_unichar_fully_decompose (c
, FALSE
, decomp
, G_N_ELEMENTS (decomp
));
809 for (i
=0; i
< decomp_len
; i
++)
811 if (decomp
[i
] != 0x307 /* COMBINING DOT ABOVE */)
812 len
+= g_unichar_to_utf8 (g_unichar_toupper (decomp
[i
]), out_buffer
? out_buffer
+ len
: NULL
);
815 len
+= output_marks (&p
, out_buffer
? out_buffer
+ len
: NULL
, TRUE
);
825 if (locale_type
== LOCALE_TURKIC
&& c
== 'i')
827 /* i => LATIN CAPITAL LETTER I WITH DOT ABOVE */
828 len
+= g_unichar_to_utf8 (0x130, out_buffer
? out_buffer
+ len
: NULL
);
830 else if (c
== 0x0345) /* COMBINING GREEK YPOGEGRAMMENI */
832 /* Nasty, need to move it after other combining marks .. this would go away if
833 * we normalized first.
835 len
+= output_marks (&p
, out_buffer
? out_buffer
+ len
: NULL
, FALSE
);
837 /* And output as GREEK CAPITAL LETTER IOTA */
838 len
+= g_unichar_to_utf8 (0x399, out_buffer
? out_buffer
+ len
: NULL
);
841 OR (G_UNICODE_LOWERCASE_LETTER
,
842 OR (G_UNICODE_TITLECASE_LETTER
,
845 val
= ATTTABLE (c
>> 8, c
& 0xff);
847 if (val
>= 0x1000000)
849 len
+= output_special_case (out_buffer
? out_buffer
+ len
: NULL
, val
- 0x1000000, t
,
850 t
== G_UNICODE_LOWERCASE_LETTER
? 0 : 1);
854 if (t
== G_UNICODE_TITLECASE_LETTER
)
857 for (i
= 0; i
< G_N_ELEMENTS (title_table
); ++i
)
859 if (title_table
[i
][0] == c
)
861 val
= title_table
[i
][1];
867 /* Some lowercase letters, e.g., U+000AA, FEMININE ORDINAL INDICATOR,
868 * do not have an uppercase equivalent, in which case val will be
870 len
+= g_unichar_to_utf8 (val
? val
: c
, out_buffer
? out_buffer
+ len
: NULL
);
875 gsize char_len
= g_utf8_skip
[*(guchar
*)last
];
878 memcpy (out_buffer
+ len
, last
, char_len
);
890 * @str: a UTF-8 encoded string
891 * @len: length of @str, in bytes, or -1 if @str is nul-terminated.
893 * Converts all Unicode characters in the string that have a case
894 * to uppercase. The exact manner that this is done depends
895 * on the current locale, and may result in the number of
896 * characters in the string increasing. (For instance, the
897 * German ess-zet will be changed to SS.)
899 * Returns: a newly allocated string, with all characters
900 * converted to uppercase.
903 g_utf8_strup (const gchar
*str
,
907 LocaleType locale_type
;
910 g_return_val_if_fail (str
!= NULL
, NULL
);
912 locale_type
= get_locale_type ();
915 * We use a two pass approach to keep memory management simple
917 result_len
= real_toupper (str
, len
, NULL
, locale_type
);
918 result
= g_malloc (result_len
+ 1);
919 real_toupper (str
, len
, result
, locale_type
);
920 result
[result_len
] = '\0';
925 /* traverses the string checking for characters with combining class == 230
926 * until a base character is found */
928 has_more_above (const gchar
*str
)
930 const gchar
*p
= str
;
931 gint combining_class
;
935 combining_class
= g_unichar_combining_class (g_utf8_get_char (p
));
936 if (combining_class
== 230)
938 else if (combining_class
== 0)
941 p
= g_utf8_next_char (p
);
948 real_tolower (const gchar
*str
,
951 LocaleType locale_type
)
953 const gchar
*p
= str
;
954 const char *last
= NULL
;
957 while ((max_len
< 0 || p
< str
+ max_len
) && *p
)
959 gunichar c
= g_utf8_get_char (p
);
964 p
= g_utf8_next_char (p
);
966 if (locale_type
== LOCALE_TURKIC
&& c
== 'I')
968 if (g_utf8_get_char (p
) == 0x0307)
970 /* I + COMBINING DOT ABOVE => i (U+0069) */
971 len
+= g_unichar_to_utf8 (0x0069, out_buffer
? out_buffer
+ len
: NULL
);
972 p
= g_utf8_next_char (p
);
976 /* I => LATIN SMALL LETTER DOTLESS I */
977 len
+= g_unichar_to_utf8 (0x131, out_buffer
? out_buffer
+ len
: NULL
);
980 /* Introduce an explicit dot above when lowercasing capital I's and J's
981 * whenever there are more accents above. [SpecialCasing.txt] */
982 else if (locale_type
== LOCALE_LITHUANIAN
&&
983 (c
== 0x00cc || c
== 0x00cd || c
== 0x0128))
985 len
+= g_unichar_to_utf8 (0x0069, out_buffer
? out_buffer
+ len
: NULL
);
986 len
+= g_unichar_to_utf8 (0x0307, out_buffer
? out_buffer
+ len
: NULL
);
991 len
+= g_unichar_to_utf8 (0x0300, out_buffer
? out_buffer
+ len
: NULL
);
994 len
+= g_unichar_to_utf8 (0x0301, out_buffer
? out_buffer
+ len
: NULL
);
997 len
+= g_unichar_to_utf8 (0x0303, out_buffer
? out_buffer
+ len
: NULL
);
1001 else if (locale_type
== LOCALE_LITHUANIAN
&&
1002 (c
== 'I' || c
== 'J' || c
== 0x012e) &&
1005 len
+= g_unichar_to_utf8 (g_unichar_tolower (c
), out_buffer
? out_buffer
+ len
: NULL
);
1006 len
+= g_unichar_to_utf8 (0x0307, out_buffer
? out_buffer
+ len
: NULL
);
1008 else if (c
== 0x03A3) /* GREEK CAPITAL LETTER SIGMA */
1010 if ((max_len
< 0 || p
< str
+ max_len
) && *p
)
1012 gunichar next_c
= g_utf8_get_char (p
);
1013 int next_type
= TYPE(next_c
);
1015 /* SIGMA mapps differently depending on whether it is
1016 * final or not. The following simplified test would
1017 * fail in the case of combining marks following the
1018 * sigma, but I don't think that occurs in real text.
1019 * The test here matches that in ICU.
1021 if (ISALPHA (next_type
)) /* Lu,Ll,Lt,Lm,Lo */
1022 val
= 0x3c3; /* GREEK SMALL SIGMA */
1024 val
= 0x3c2; /* GREEK SMALL FINAL SIGMA */
1027 val
= 0x3c2; /* GREEK SMALL FINAL SIGMA */
1029 len
+= g_unichar_to_utf8 (val
, out_buffer
? out_buffer
+ len
: NULL
);
1032 OR (G_UNICODE_UPPERCASE_LETTER
,
1033 OR (G_UNICODE_TITLECASE_LETTER
,
1036 val
= ATTTABLE (c
>> 8, c
& 0xff);
1038 if (val
>= 0x1000000)
1040 len
+= output_special_case (out_buffer
? out_buffer
+ len
: NULL
, val
- 0x1000000, t
, 0);
1044 if (t
== G_UNICODE_TITLECASE_LETTER
)
1047 for (i
= 0; i
< G_N_ELEMENTS (title_table
); ++i
)
1049 if (title_table
[i
][0] == c
)
1051 val
= title_table
[i
][2];
1057 /* Not all uppercase letters are guaranteed to have a lowercase
1058 * equivalent. If this is the case, val will be zero. */
1059 len
+= g_unichar_to_utf8 (val
? val
: c
, out_buffer
? out_buffer
+ len
: NULL
);
1064 gsize char_len
= g_utf8_skip
[*(guchar
*)last
];
1067 memcpy (out_buffer
+ len
, last
, char_len
);
1079 * @str: a UTF-8 encoded string
1080 * @len: length of @str, in bytes, or -1 if @str is nul-terminated.
1082 * Converts all Unicode characters in the string that have a case
1083 * to lowercase. The exact manner that this is done depends
1084 * on the current locale, and may result in the number of
1085 * characters in the string changing.
1087 * Returns: a newly allocated string, with all characters
1088 * converted to lowercase.
1091 g_utf8_strdown (const gchar
*str
,
1095 LocaleType locale_type
;
1098 g_return_val_if_fail (str
!= NULL
, NULL
);
1100 locale_type
= get_locale_type ();
1103 * We use a two pass approach to keep memory management simple
1105 result_len
= real_tolower (str
, len
, NULL
, locale_type
);
1106 result
= g_malloc (result_len
+ 1);
1107 real_tolower (str
, len
, result
, locale_type
);
1108 result
[result_len
] = '\0';
1115 * @str: a UTF-8 encoded string
1116 * @len: length of @str, in bytes, or -1 if @str is nul-terminated.
1118 * Converts a string into a form that is independent of case. The
1119 * result will not correspond to any particular case, but can be
1120 * compared for equality or ordered with the results of calling
1121 * g_utf8_casefold() on other strings.
1123 * Note that calling g_utf8_casefold() followed by g_utf8_collate() is
1124 * only an approximation to the correct linguistic case insensitive
1125 * ordering, though it is a fairly good one. Getting this exactly
1126 * right would require a more sophisticated collation function that
1127 * takes case sensitivity into account. GLib does not currently
1128 * provide such a function.
1130 * Returns: a newly allocated string, that is a
1131 * case independent form of @str.
1134 g_utf8_casefold (const gchar
*str
,
1140 g_return_val_if_fail (str
!= NULL
, NULL
);
1142 result
= g_string_new (NULL
);
1144 while ((len
< 0 || p
< str
+ len
) && *p
)
1146 gunichar ch
= g_utf8_get_char (p
);
1149 int end
= G_N_ELEMENTS (casefold_table
);
1151 if (ch
>= casefold_table
[start
].ch
&&
1152 ch
<= casefold_table
[end
- 1].ch
)
1156 int half
= (start
+ end
) / 2;
1157 if (ch
== casefold_table
[half
].ch
)
1159 g_string_append (result
, casefold_table
[half
].data
);
1162 else if (half
== start
)
1164 else if (ch
> casefold_table
[half
].ch
)
1171 g_string_append_unichar (result
, g_unichar_tolower (ch
));
1174 p
= g_utf8_next_char (p
);
1177 return g_string_free (result
, FALSE
);
1181 * g_unichar_get_mirror_char:
1182 * @ch: a Unicode character
1183 * @mirrored_ch: location to store the mirrored character
1185 * In Unicode, some characters are "mirrored". This means that their
1186 * images are mirrored horizontally in text that is laid out from right
1187 * to left. For instance, "(" would become its mirror image, ")", in
1188 * right-to-left text.
1190 * If @ch has the Unicode mirrored property and there is another unicode
1191 * character that typically has a glyph that is the mirror image of @ch's
1192 * glyph and @mirrored_ch is set, it puts that character in the address
1193 * pointed to by @mirrored_ch. Otherwise the original character is put.
1195 * Returns: %TRUE if @ch has a mirrored character, %FALSE otherwise
1200 g_unichar_get_mirror_char (gunichar ch
,
1201 gunichar
*mirrored_ch
)
1206 mirrored
= GLIB_GET_MIRRORING(ch
);
1208 found
= ch
!= mirrored
;
1210 *mirrored_ch
= mirrored
;
1216 #define G_SCRIPT_TABLE_MIDPOINT (G_N_ELEMENTS (g_script_table) / 2)
1218 static inline GUnicodeScript
1219 g_unichar_get_script_bsearch (gunichar ch
)
1222 int upper
= G_N_ELEMENTS (g_script_table
) - 1;
1223 static int saved_mid
= G_SCRIPT_TABLE_MIDPOINT
;
1224 int mid
= saved_mid
;
1229 if (ch
< g_script_table
[mid
].start
)
1231 else if (ch
>= g_script_table
[mid
].start
+ g_script_table
[mid
].chars
)
1234 return g_script_table
[saved_mid
= mid
].script
;
1236 mid
= (lower
+ upper
) / 2;
1238 while (lower
<= upper
);
1240 return G_UNICODE_SCRIPT_UNKNOWN
;
1244 * g_unichar_get_script:
1245 * @ch: a Unicode character
1247 * Looks up the #GUnicodeScript for a particular character (as defined
1248 * by Unicode Standard Annex \#24). No check is made for @ch being a
1249 * valid Unicode character; if you pass in invalid character, the
1250 * result is undefined.
1252 * This function is equivalent to pango_script_for_unichar() and the
1253 * two are interchangeable.
1255 * Returns: the #GUnicodeScript for the character.
1260 g_unichar_get_script (gunichar ch
)
1262 if (ch
< G_EASY_SCRIPTS_RANGE
)
1263 return g_script_easy_table
[ch
];
1265 return g_unichar_get_script_bsearch (ch
);
1269 /* http://unicode.org/iso15924/ */
1270 static const guint32 iso15924_tags
[] =
1272 #define PACK(a,b,c,d) ((guint32)((((guint8)(a))<<24)|(((guint8)(b))<<16)|(((guint8)(c))<<8)|((guint8)(d))))
1274 PACK ('Z','y','y','y'), /* G_UNICODE_SCRIPT_COMMON */
1275 PACK ('Z','i','n','h'), /* G_UNICODE_SCRIPT_INHERITED */
1276 PACK ('A','r','a','b'), /* G_UNICODE_SCRIPT_ARABIC */
1277 PACK ('A','r','m','n'), /* G_UNICODE_SCRIPT_ARMENIAN */
1278 PACK ('B','e','n','g'), /* G_UNICODE_SCRIPT_BENGALI */
1279 PACK ('B','o','p','o'), /* G_UNICODE_SCRIPT_BOPOMOFO */
1280 PACK ('C','h','e','r'), /* G_UNICODE_SCRIPT_CHEROKEE */
1281 PACK ('C','o','p','t'), /* G_UNICODE_SCRIPT_COPTIC */
1282 PACK ('C','y','r','l'), /* G_UNICODE_SCRIPT_CYRILLIC */
1283 PACK ('D','s','r','t'), /* G_UNICODE_SCRIPT_DESERET */
1284 PACK ('D','e','v','a'), /* G_UNICODE_SCRIPT_DEVANAGARI */
1285 PACK ('E','t','h','i'), /* G_UNICODE_SCRIPT_ETHIOPIC */
1286 PACK ('G','e','o','r'), /* G_UNICODE_SCRIPT_GEORGIAN */
1287 PACK ('G','o','t','h'), /* G_UNICODE_SCRIPT_GOTHIC */
1288 PACK ('G','r','e','k'), /* G_UNICODE_SCRIPT_GREEK */
1289 PACK ('G','u','j','r'), /* G_UNICODE_SCRIPT_GUJARATI */
1290 PACK ('G','u','r','u'), /* G_UNICODE_SCRIPT_GURMUKHI */
1291 PACK ('H','a','n','i'), /* G_UNICODE_SCRIPT_HAN */
1292 PACK ('H','a','n','g'), /* G_UNICODE_SCRIPT_HANGUL */
1293 PACK ('H','e','b','r'), /* G_UNICODE_SCRIPT_HEBREW */
1294 PACK ('H','i','r','a'), /* G_UNICODE_SCRIPT_HIRAGANA */
1295 PACK ('K','n','d','a'), /* G_UNICODE_SCRIPT_KANNADA */
1296 PACK ('K','a','n','a'), /* G_UNICODE_SCRIPT_KATAKANA */
1297 PACK ('K','h','m','r'), /* G_UNICODE_SCRIPT_KHMER */
1298 PACK ('L','a','o','o'), /* G_UNICODE_SCRIPT_LAO */
1299 PACK ('L','a','t','n'), /* G_UNICODE_SCRIPT_LATIN */
1300 PACK ('M','l','y','m'), /* G_UNICODE_SCRIPT_MALAYALAM */
1301 PACK ('M','o','n','g'), /* G_UNICODE_SCRIPT_MONGOLIAN */
1302 PACK ('M','y','m','r'), /* G_UNICODE_SCRIPT_MYANMAR */
1303 PACK ('O','g','a','m'), /* G_UNICODE_SCRIPT_OGHAM */
1304 PACK ('I','t','a','l'), /* G_UNICODE_SCRIPT_OLD_ITALIC */
1305 PACK ('O','r','y','a'), /* G_UNICODE_SCRIPT_ORIYA */
1306 PACK ('R','u','n','r'), /* G_UNICODE_SCRIPT_RUNIC */
1307 PACK ('S','i','n','h'), /* G_UNICODE_SCRIPT_SINHALA */
1308 PACK ('S','y','r','c'), /* G_UNICODE_SCRIPT_SYRIAC */
1309 PACK ('T','a','m','l'), /* G_UNICODE_SCRIPT_TAMIL */
1310 PACK ('T','e','l','u'), /* G_UNICODE_SCRIPT_TELUGU */
1311 PACK ('T','h','a','a'), /* G_UNICODE_SCRIPT_THAANA */
1312 PACK ('T','h','a','i'), /* G_UNICODE_SCRIPT_THAI */
1313 PACK ('T','i','b','t'), /* G_UNICODE_SCRIPT_TIBETAN */
1314 PACK ('C','a','n','s'), /* G_UNICODE_SCRIPT_CANADIAN_ABORIGINAL */
1315 PACK ('Y','i','i','i'), /* G_UNICODE_SCRIPT_YI */
1316 PACK ('T','g','l','g'), /* G_UNICODE_SCRIPT_TAGALOG */
1317 PACK ('H','a','n','o'), /* G_UNICODE_SCRIPT_HANUNOO */
1318 PACK ('B','u','h','d'), /* G_UNICODE_SCRIPT_BUHID */
1319 PACK ('T','a','g','b'), /* G_UNICODE_SCRIPT_TAGBANWA */
1321 /* Unicode-4.0 additions */
1322 PACK ('B','r','a','i'), /* G_UNICODE_SCRIPT_BRAILLE */
1323 PACK ('C','p','r','t'), /* G_UNICODE_SCRIPT_CYPRIOT */
1324 PACK ('L','i','m','b'), /* G_UNICODE_SCRIPT_LIMBU */
1325 PACK ('O','s','m','a'), /* G_UNICODE_SCRIPT_OSMANYA */
1326 PACK ('S','h','a','w'), /* G_UNICODE_SCRIPT_SHAVIAN */
1327 PACK ('L','i','n','b'), /* G_UNICODE_SCRIPT_LINEAR_B */
1328 PACK ('T','a','l','e'), /* G_UNICODE_SCRIPT_TAI_LE */
1329 PACK ('U','g','a','r'), /* G_UNICODE_SCRIPT_UGARITIC */
1331 /* Unicode-4.1 additions */
1332 PACK ('T','a','l','u'), /* G_UNICODE_SCRIPT_NEW_TAI_LUE */
1333 PACK ('B','u','g','i'), /* G_UNICODE_SCRIPT_BUGINESE */
1334 PACK ('G','l','a','g'), /* G_UNICODE_SCRIPT_GLAGOLITIC */
1335 PACK ('T','f','n','g'), /* G_UNICODE_SCRIPT_TIFINAGH */
1336 PACK ('S','y','l','o'), /* G_UNICODE_SCRIPT_SYLOTI_NAGRI */
1337 PACK ('X','p','e','o'), /* G_UNICODE_SCRIPT_OLD_PERSIAN */
1338 PACK ('K','h','a','r'), /* G_UNICODE_SCRIPT_KHAROSHTHI */
1340 /* Unicode-5.0 additions */
1341 PACK ('Z','z','z','z'), /* G_UNICODE_SCRIPT_UNKNOWN */
1342 PACK ('B','a','l','i'), /* G_UNICODE_SCRIPT_BALINESE */
1343 PACK ('X','s','u','x'), /* G_UNICODE_SCRIPT_CUNEIFORM */
1344 PACK ('P','h','n','x'), /* G_UNICODE_SCRIPT_PHOENICIAN */
1345 PACK ('P','h','a','g'), /* G_UNICODE_SCRIPT_PHAGS_PA */
1346 PACK ('N','k','o','o'), /* G_UNICODE_SCRIPT_NKO */
1348 /* Unicode-5.1 additions */
1349 PACK ('K','a','l','i'), /* G_UNICODE_SCRIPT_KAYAH_LI */
1350 PACK ('L','e','p','c'), /* G_UNICODE_SCRIPT_LEPCHA */
1351 PACK ('R','j','n','g'), /* G_UNICODE_SCRIPT_REJANG */
1352 PACK ('S','u','n','d'), /* G_UNICODE_SCRIPT_SUNDANESE */
1353 PACK ('S','a','u','r'), /* G_UNICODE_SCRIPT_SAURASHTRA */
1354 PACK ('C','h','a','m'), /* G_UNICODE_SCRIPT_CHAM */
1355 PACK ('O','l','c','k'), /* G_UNICODE_SCRIPT_OL_CHIKI */
1356 PACK ('V','a','i','i'), /* G_UNICODE_SCRIPT_VAI */
1357 PACK ('C','a','r','i'), /* G_UNICODE_SCRIPT_CARIAN */
1358 PACK ('L','y','c','i'), /* G_UNICODE_SCRIPT_LYCIAN */
1359 PACK ('L','y','d','i'), /* G_UNICODE_SCRIPT_LYDIAN */
1361 /* Unicode-5.2 additions */
1362 PACK ('A','v','s','t'), /* G_UNICODE_SCRIPT_AVESTAN */
1363 PACK ('B','a','m','u'), /* G_UNICODE_SCRIPT_BAMUM */
1364 PACK ('E','g','y','p'), /* G_UNICODE_SCRIPT_EGYPTIAN_HIEROGLYPHS */
1365 PACK ('A','r','m','i'), /* G_UNICODE_SCRIPT_IMPERIAL_ARAMAIC */
1366 PACK ('P','h','l','i'), /* G_UNICODE_SCRIPT_INSCRIPTIONAL_PAHLAVI */
1367 PACK ('P','r','t','i'), /* G_UNICODE_SCRIPT_INSCRIPTIONAL_PARTHIAN */
1368 PACK ('J','a','v','a'), /* G_UNICODE_SCRIPT_JAVANESE */
1369 PACK ('K','t','h','i'), /* G_UNICODE_SCRIPT_KAITHI */
1370 PACK ('L','i','s','u'), /* G_UNICODE_SCRIPT_LISU */
1371 PACK ('M','t','e','i'), /* G_UNICODE_SCRIPT_MEETEI_MAYEK */
1372 PACK ('S','a','r','b'), /* G_UNICODE_SCRIPT_OLD_SOUTH_ARABIAN */
1373 PACK ('O','r','k','h'), /* G_UNICODE_SCRIPT_OLD_TURKIC */
1374 PACK ('S','a','m','r'), /* G_UNICODE_SCRIPT_SAMARITAN */
1375 PACK ('L','a','n','a'), /* G_UNICODE_SCRIPT_TAI_THAM */
1376 PACK ('T','a','v','t'), /* G_UNICODE_SCRIPT_TAI_VIET */
1378 /* Unicode-6.0 additions */
1379 PACK ('B','a','t','k'), /* G_UNICODE_SCRIPT_BATAK */
1380 PACK ('B','r','a','h'), /* G_UNICODE_SCRIPT_BRAHMI */
1381 PACK ('M','a','n','d'), /* G_UNICODE_SCRIPT_MANDAIC */
1383 /* Unicode-6.1 additions */
1384 PACK ('C','a','k','m'), /* G_UNICODE_SCRIPT_CHAKMA */
1385 PACK ('M','e','r','c'), /* G_UNICODE_SCRIPT_MEROITIC_CURSIVE */
1386 PACK ('M','e','r','o'), /* G_UNICODE_SCRIPT_MEROITIC_HIEROGLYPHS */
1387 PACK ('P','l','r','d'), /* G_UNICODE_SCRIPT_MIAO */
1388 PACK ('S','h','r','d'), /* G_UNICODE_SCRIPT_SHARADA */
1389 PACK ('S','o','r','a'), /* G_UNICODE_SCRIPT_SORA_SOMPENG */
1390 PACK ('T','a','k','r'), /* G_UNICODE_SCRIPT_TAKRI */
1392 /* Unicode 7.0 additions */
1393 PACK ('B','a','s','s'), /* G_UNICODE_SCRIPT_BASSA_VAH */
1394 PACK ('A','g','h','b'), /* G_UNICODE_SCRIPT_CAUCASIAN_ALBANIAN */
1395 PACK ('D','u','p','l'), /* G_UNICODE_SCRIPT_DUPLOYAN */
1396 PACK ('E','l','b','a'), /* G_UNICODE_SCRIPT_ELBASAN */
1397 PACK ('G','r','a','n'), /* G_UNICODE_SCRIPT_GRANTHA */
1398 PACK ('K','h','o','j'), /* G_UNICODE_SCRIPT_KHOJKI*/
1399 PACK ('S','i','n','d'), /* G_UNICODE_SCRIPT_KHUDAWADI */
1400 PACK ('L','i','n','a'), /* G_UNICODE_SCRIPT_LINEAR_A */
1401 PACK ('M','a','h','j'), /* G_UNICODE_SCRIPT_MAHAJANI */
1402 PACK ('M','a','n','u'), /* G_UNICODE_SCRIPT_MANICHAEAN */
1403 PACK ('M','e','n','d'), /* G_UNICODE_SCRIPT_MENDE_KIKAKUI */
1404 PACK ('M','o','d','i'), /* G_UNICODE_SCRIPT_MODI */
1405 PACK ('M','r','o','o'), /* G_UNICODE_SCRIPT_MRO */
1406 PACK ('N','b','a','t'), /* G_UNICODE_SCRIPT_NABATAEAN */
1407 PACK ('N','a','r','b'), /* G_UNICODE_SCRIPT_OLD_NORTH_ARABIAN */
1408 PACK ('P','e','r','m'), /* G_UNICODE_SCRIPT_OLD_PERMIC */
1409 PACK ('H','m','n','g'), /* G_UNICODE_SCRIPT_PAHAWH_HMONG */
1410 PACK ('P','a','l','m'), /* G_UNICODE_SCRIPT_PALMYRENE */
1411 PACK ('P','a','u','c'), /* G_UNICODE_SCRIPT_PAU_CIN_HAU */
1412 PACK ('P','h','l','p'), /* G_UNICODE_SCRIPT_PSALTER_PAHLAVI */
1413 PACK ('S','i','d','d'), /* G_UNICODE_SCRIPT_SIDDHAM */
1414 PACK ('T','i','r','h'), /* G_UNICODE_SCRIPT_TIRHUTA */
1415 PACK ('W','a','r','a'), /* G_UNICODE_SCRIPT_WARANG_CITI */
1417 /* Unicode 8.0 additions */
1418 PACK ('A','h','o','m'), /* G_UNICODE_SCRIPT_AHOM */
1419 PACK ('H','l','u','w'), /* G_UNICODE_SCRIPT_ANATOLIAN_HIEROGLYPHS */
1420 PACK ('H','a','t','r'), /* G_UNICODE_SCRIPT_HATRAN */
1421 PACK ('M','u','l','t'), /* G_UNICODE_SCRIPT_MULTANI */
1422 PACK ('H','u','n','g'), /* G_UNICODE_SCRIPT_OLD_HUNGARIAN */
1423 PACK ('S','g','n','w'), /* G_UNICODE_SCRIPT_SIGNWRITING */
1428 * g_unicode_script_to_iso15924:
1429 * @script: a Unicode script
1431 * Looks up the ISO 15924 code for @script. ISO 15924 assigns four-letter
1432 * codes to scripts. For example, the code for Arabic is 'Arab'. The
1433 * four letter codes are encoded as a @guint32 by this function in a
1434 * big-endian fashion. That is, the code returned for Arabic is
1435 * 0x41726162 (0x41 is ASCII code for 'A', 0x72 is ASCII code for 'r', etc).
1438 * [Codes for the representation of names of scripts](http://unicode.org/iso15924/codelists.html)
1441 * Returns: the ISO 15924 code for @script, encoded as an integer,
1442 * of zero if @script is %G_UNICODE_SCRIPT_INVALID_CODE or
1443 * ISO 15924 code 'Zzzz' (script code for UNKNOWN) if @script is not understood.
1448 g_unicode_script_to_iso15924 (GUnicodeScript script
)
1450 if (G_UNLIKELY (script
== G_UNICODE_SCRIPT_INVALID_CODE
))
1453 if (G_UNLIKELY (script
< 0 || script
>= (int) G_N_ELEMENTS (iso15924_tags
)))
1456 return iso15924_tags
[script
];
1460 * g_unicode_script_from_iso15924:
1461 * @iso15924: a Unicode script
1463 * Looks up the Unicode script for @iso15924. ISO 15924 assigns four-letter
1464 * codes to scripts. For example, the code for Arabic is 'Arab'.
1465 * This function accepts four letter codes encoded as a @guint32 in a
1466 * big-endian fashion. That is, the code expected for Arabic is
1467 * 0x41726162 (0x41 is ASCII code for 'A', 0x72 is ASCII code for 'r', etc).
1470 * [Codes for the representation of names of scripts](http://unicode.org/iso15924/codelists.html)
1473 * Returns: the Unicode script for @iso15924, or
1474 * of %G_UNICODE_SCRIPT_INVALID_CODE if @iso15924 is zero and
1475 * %G_UNICODE_SCRIPT_UNKNOWN if @iso15924 is unknown.
1480 g_unicode_script_from_iso15924 (guint32 iso15924
)
1485 return G_UNICODE_SCRIPT_INVALID_CODE
;
1487 for (i
= 0; i
< G_N_ELEMENTS (iso15924_tags
); i
++)
1488 if (iso15924_tags
[i
] == iso15924
)
1489 return (GUnicodeScript
) i
;
1491 return G_UNICODE_SCRIPT_UNKNOWN
;