1 /* Copyright (C) 1991-2015 Free Software Foundation, Inc.
2 This file is part of the GNU C Library.
3
4 The GNU C 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 2.1 of the License, or (at your option) any later version.
8
9 The GNU C 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.
13
14 You should have received a copy of the GNU Lesser General Public
15 License along with the GNU C Library; if not, see
16 <http://www.gnu.org/licenses/>. */
17
18 /*
19 * ISO C99 Standard 7.4: Character handling <ctype.h>
20 */
21
22 #ifndef _CTYPE_H
23 #define _CTYPE_H 1
24
25 #include <features.h>
26 #include <bits/types.h>
27
28 __BEGIN_DECLS
29
30 #ifndef _ISbit
31 /* These are all the characteristics of characters.
32 If there get to be more than 16 distinct characteristics,
33 many things must be changed that use `unsigned short int's.
34
35 The characteristics are stored always in network byte order (big
36 endian). We define the bit value interpretations here dependent on the
37 machine's byte order. */
38
39 # include <endian.h>
40 # if __BYTE_ORDER == __BIG_ENDIAN
41 # define _ISbit(bit) (1 << (bit))
42 # else /* __BYTE_ORDER == __LITTLE_ENDIAN */
43 # define _ISbit(bit) ((bit) < 8 ? ((1 << (bit)) << 8) : ((1 << (bit)) >> 8))
44 # endif
45
46 enum
47 {
48 _ISupper = _ISbit (0), /* UPPERCASE. */
49 _ISlower = _ISbit (1), /* lowercase. */
50 _ISalpha = _ISbit (2), /* Alphabetic. */
51 _ISdigit = _ISbit (3), /* Numeric. */
52 _ISxdigit = _ISbit (4), /* Hexadecimal numeric. */
53 _ISspace = _ISbit (5), /* Whitespace. */
54 _ISprint = _ISbit (6), /* Printing. */
55 _ISgraph = _ISbit (7), /* Graphical. */
56 _ISblank = _ISbit (8), /* Blank (usually SPC and TAB). */
57 _IScntrl = _ISbit (9), /* Control character. */
58 _ISpunct = _ISbit (10), /* Punctuation. */
59 _ISalnum = _ISbit (11) /* Alphanumeric. */
60 };
61 #endif /* ! _ISbit */
62
63 /* These are defined in ctype-info.c.
64 The declarations here must match those in localeinfo.h.
65
66 In the thread-specific locale model (see `uselocale' in <locale.h>)
67 we cannot use global variables for these as was done in the past.
68 Instead, the following accessor functions return the address of
69 each variable, which is local to the current thread if multithreaded.
70
71 These point into arrays of 384, so they can be indexed by any `unsigned
72 char' value [0,255]; by EOF (-1); or by any `signed char' value
73 [-128,-1). ISO C requires that the ctype functions work for `unsigned
74 char' values and for EOF; we also support negative `signed char' values
75 for broken old programs. The case conversion arrays are of `int's
76 rather than `unsigned char's because tolower (EOF) must be EOF, which
77 doesn't fit into an `unsigned char'. But today more important is that
78 the arrays are also used for multi-byte character sets. */
79 extern const unsigned short int **__ctype_b_loc (void)
80 __THROW __attribute__ ((__const__));
81 extern const __int32_t **__ctype_tolower_loc (void)
82 __THROW __attribute__ ((__const__));
83 extern const __int32_t **__ctype_toupper_loc (void)
84 __THROW __attribute__ ((__const__));
85
86
87 #ifndef __cplusplus
88 # define __isctype(c, type) \
89 ((*__ctype_b_loc ())[(int) (c)] & (unsigned short int) type)
90 #elif defined __USE_EXTERN_INLINES
91 # define __isctype_f(type) \
92 __extern_inline int \
93 is##type (int __c) __THROW \
94 { \
95 return (*__ctype_b_loc ())[(int) (__c)] & (unsigned short int) _IS##type; \
96 }
97 #endif
98
99 #define __isascii(c) (((c) & ~0x7f) == 0) /* If C is a 7 bit value. */
100 #define __toascii(c) ((c) & 0x7f) /* Mask off high bits. */
101
102 #define __exctype(name) extern int name (int) __THROW
103
104 __BEGIN_NAMESPACE_STD
105
106 /* The following names are all functions:
107 int isCHARACTERISTIC(int c);
108 which return nonzero iff C has CHARACTERISTIC.
109 For the meaning of the characteristic names, see the `enum' above. */
110 __exctype (isalnum);
111 __exctype (isalpha);
112 __exctype (iscntrl);
113 __exctype (isdigit);
114 __exctype (islower);
115 __exctype (isgraph);
116 __exctype (isprint);
117 __exctype (ispunct);
118 __exctype (isspace);
119 __exctype (isupper);
120 __exctype (isxdigit);
121
122
123 /* Return the lowercase version of C. */
124 extern int tolower (int __c) __THROW;
125
126 /* Return the uppercase version of C. */
127 extern int toupper (int __c) __THROW;
128
129 __END_NAMESPACE_STD
130
131
132 /* ISO C99 introduced one new function. */
133 #ifdef __USE_ISOC99
134 __BEGIN_NAMESPACE_C99
135
136 __exctype (isblank);
137
138 __END_NAMESPACE_C99
139 #endif
140
141 #ifdef __USE_GNU
142 /* Test C for a set of character classes according to MASK. */
143 extern int isctype (int __c, int __mask) __THROW;
144 #endif
145
146 #if defined __USE_MISC || defined __USE_XOPEN
147
148 /* Return nonzero iff C is in the ASCII set
149 (i.e., is no more than 7 bits wide). */
150 extern int isascii (int __c) __THROW;
151
152 /* Return the part of C that is in the ASCII set
153 (i.e., the low-order 7 bits of C). */
154 extern int toascii (int __c) __THROW;
155
156 /* These are the same as `toupper' and `tolower' except that they do not
157 check the argument for being in the range of a `char'. */
158 __exctype (_toupper);
159 __exctype (_tolower);
160 #endif /* Use X/Open or use misc. */
161
162 /* This code is needed for the optimized mapping functions. */
163 #define __tobody(c, f, a, args) \
164 (__extension__ \
165 ({ int __res; \
166 if (sizeof (c) > 1) \
167 { \
168 if (__builtin_constant_p (c)) \
169 { \
170 int __c = (c); \
171 __res = __c < -128 || __c > 255 ? __c : (a)[__c]; \
172 } \
173 else \
174 __res = f args; \
175 } \
176 else \
177 __res = (a)[(int) (c)]; \
178 __res; }))
179
180 #if !defined __NO_CTYPE
181 # ifdef __isctype_f
182 __isctype_f (alnum)
183 __isctype_f (alpha)
184 __isctype_f (cntrl)
185 __isctype_f (digit)
186 __isctype_f (lower)
187 __isctype_f (graph)
188 __isctype_f (print)
189 __isctype_f (punct)
190 __isctype_f (space)
191 __isctype_f (upper)
192 __isctype_f (xdigit)
193 # ifdef __USE_ISOC99
194 __isctype_f (blank)
195 # endif
196 # elif defined __isctype
197 # define isalnum(c) __isctype((c), _ISalnum)
198 # define isalpha(c) __isctype((c), _ISalpha)
199 # define iscntrl(c) __isctype((c), _IScntrl)
200 # define isdigit(c) __isctype((c), _ISdigit)
201 # define islower(c) __isctype((c), _ISlower)
202 # define isgraph(c) __isctype((c), _ISgraph)
203 # define isprint(c) __isctype((c), _ISprint)
204 # define ispunct(c) __isctype((c), _ISpunct)
205 # define isspace(c) __isctype((c), _ISspace)
206 # define isupper(c) __isctype((c), _ISupper)
207 # define isxdigit(c) __isctype((c), _ISxdigit)
208 # ifdef __USE_ISOC99
209 # define isblank(c) __isctype((c), _ISblank)
210 # endif
211 # endif
212
213 # ifdef __USE_EXTERN_INLINES
214 __extern_inline int
215 __NTH (tolower (int __c))
216 {
217 return __c >= -128 && __c < 256 ? (*__ctype_tolower_loc ())[__c] : __c;
218 }
219
220 __extern_inline int
221 __NTH (toupper (int __c))
222 {
223 return __c >= -128 && __c < 256 ? (*__ctype_toupper_loc ())[__c] : __c;
224 }
225 # endif
226
227 # if __GNUC__ >= 2 && defined __OPTIMIZE__ && !defined __cplusplus
228 # define tolower(c) __tobody (c, tolower, *__ctype_tolower_loc (), (c))
229 # define toupper(c) __tobody (c, toupper, *__ctype_toupper_loc (), (c))
230 # endif /* Optimizing gcc */
231
232 # if defined __USE_MISC || defined __USE_XOPEN
233 # define isascii(c) __isascii (c)
234 # define toascii(c) __toascii (c)
235
236 # define _tolower(c) ((int) (*__ctype_tolower_loc ())[(int) (c)])
237 # define _toupper(c) ((int) (*__ctype_toupper_loc ())[(int) (c)])
238 # endif
239
240 #endif /* Not __NO_CTYPE. */
241
242
243 #ifdef __USE_XOPEN2K8
244 /* The concept of one static locale per category is not very well
245 thought out. Many applications will need to process its data using
246 information from several different locales. Another application is
247 the implementation of the internationalization handling in the
248 upcoming ISO C++ standard library. To support this another set of
249 the functions using locale data exist which have an additional
250 argument.
251
252 Attention: all these functions are *not* standardized in any form.
253 This is a proof-of-concept implementation. */
254
255 /* Structure for reentrant locale using functions. This is an
256 (almost) opaque type for the user level programs. */
257 # include <xlocale.h>
258
259 /* These definitions are similar to the ones above but all functions
260 take as an argument a handle for the locale which shall be used. */
261 # define __isctype_l(c, type, locale) \
262 ((locale)->__ctype_b[(int) (c)] & (unsigned short int) type)
263
264 # define __exctype_l(name) \
265 extern int name (int, __locale_t) __THROW
266
267 /* The following names are all functions:
268 int isCHARACTERISTIC(int c, locale_t *locale);
269 which return nonzero iff C has CHARACTERISTIC.
270 For the meaning of the characteristic names, see the `enum' above. */
271 __exctype_l (isalnum_l);
272 __exctype_l (isalpha_l);
273 __exctype_l (iscntrl_l);
274 __exctype_l (isdigit_l);
275 __exctype_l (islower_l);
276 __exctype_l (isgraph_l);
277 __exctype_l (isprint_l);
278 __exctype_l (ispunct_l);
279 __exctype_l (isspace_l);
280 __exctype_l (isupper_l);
281 __exctype_l (isxdigit_l);
282
283 __exctype_l (isblank_l);
284
285
286 /* Return the lowercase version of C in locale L. */
287 extern int __tolower_l (int __c, __locale_t __l) __THROW;
288 extern int tolower_l (int __c, __locale_t __l) __THROW;
289
290 /* Return the uppercase version of C. */
291 extern int __toupper_l (int __c, __locale_t __l) __THROW;
292 extern int toupper_l (int __c, __locale_t __l) __THROW;
293
294 # if __GNUC__ >= 2 && defined __OPTIMIZE__ && !defined __cplusplus
295 # define __tolower_l(c, locale) \
296 __tobody (c, __tolower_l, (locale)->__ctype_tolower, (c, locale))
297 # define __toupper_l(c, locale) \
298 __tobody (c, __toupper_l, (locale)->__ctype_toupper, (c, locale))
299 # define tolower_l(c, locale) __tolower_l ((c), (locale))
300 # define toupper_l(c, locale) __toupper_l ((c), (locale))
301 # endif /* Optimizing gcc */
302
303
304 # ifndef __NO_CTYPE
305 # define __isalnum_l(c,l) __isctype_l((c), _ISalnum, (l))
306 # define __isalpha_l(c,l) __isctype_l((c), _ISalpha, (l))
307 # define __iscntrl_l(c,l) __isctype_l((c), _IScntrl, (l))
308 # define __isdigit_l(c,l) __isctype_l((c), _ISdigit, (l))
309 # define __islower_l(c,l) __isctype_l((c), _ISlower, (l))
310 # define __isgraph_l(c,l) __isctype_l((c), _ISgraph, (l))
311 # define __isprint_l(c,l) __isctype_l((c), _ISprint, (l))
312 # define __ispunct_l(c,l) __isctype_l((c), _ISpunct, (l))
313 # define __isspace_l(c,l) __isctype_l((c), _ISspace, (l))
314 # define __isupper_l(c,l) __isctype_l((c), _ISupper, (l))
315 # define __isxdigit_l(c,l) __isctype_l((c), _ISxdigit, (l))
316
317 # define __isblank_l(c,l) __isctype_l((c), _ISblank, (l))
318
319 # ifdef __USE_MISC
320 # define __isascii_l(c,l) ((l), __isascii (c))
321 # define __toascii_l(c,l) ((l), __toascii (c))
322 # endif
323
324 # define isalnum_l(c,l) __isalnum_l ((c), (l))
325 # define isalpha_l(c,l) __isalpha_l ((c), (l))
326 # define iscntrl_l(c,l) __iscntrl_l ((c), (l))
327 # define isdigit_l(c,l) __isdigit_l ((c), (l))
328 # define islower_l(c,l) __islower_l ((c), (l))
329 # define isgraph_l(c,l) __isgraph_l ((c), (l))
330 # define isprint_l(c,l) __isprint_l ((c), (l))
331 # define ispunct_l(c,l) __ispunct_l ((c), (l))
332 # define isspace_l(c,l) __isspace_l ((c), (l))
333 # define isupper_l(c,l) __isupper_l ((c), (l))
334 # define isxdigit_l(c,l) __isxdigit_l ((c), (l))
335
336 # define isblank_l(c,l) __isblank_l ((c), (l))
337
338 # ifdef __USE_MISC
339 # define isascii_l(c,l) __isascii_l ((c), (l))
340 # define toascii_l(c,l) __toascii_l ((c), (l))
341 # endif
342
343 # endif /* Not __NO_CTYPE. */
344
345 #endif /* Use POSIX 2008. */
346
347 __END_DECLS
348
349 #endif /* ctype.h */
350