ctype/ctype.h

   1 /* Copyright (C) 1991,92,93,95,96,97,98,99,2001,2002,2004,2007-2009,2011,2012
   2         Free Software Foundation, Inc.
   3    This file is part of the GNU C Library.
   4
   5    The GNU C Library is free software; you can redistribute it and/or
   6    modify it under the terms of the GNU Lesser General Public
   7    License as published by the Free Software Foundation; either
   8    version 2.1 of the License, or (at your option) any later version.
   9
  10    The GNU C Library is distributed in the hope that it will be useful,
  11    but WITHOUT ANY WARRANTY; without even the implied warranty of
  12    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  13    Lesser General Public License for more details.
  14
  15    You should have received a copy of the GNU Lesser General Public
  16    License along with the GNU C Library; if not, see
  17    <http://www.gnu.org/licenses/>.  */
  18
  19 /*
  20  *      ISO C99 Standard 7.4: Character handling        <ctype.h>
  21  */
  22
  23 #ifndef _CTYPE_H
  24 #define _CTYPE_H        1
  25
  26 #include <features.h>
  27 #include <bits/types.h>
  28
  29 __BEGIN_DECLS
  30
  31 #ifndef _ISbit
  32 /* These are all the characteristics of characters.
  33    If there get to be more than 16 distinct characteristics,
  34    many things must be changed that use `unsigned short int's.
  35
  36    The characteristics are stored always in network byte order (big
  37    endian).  We define the bit value interpretations here dependent on the
  38    machine's byte order.  */
  39
  40 # include <endian.h>
  41 # if __BYTE_ORDER == __BIG_ENDIAN
  42 #  define _ISbit(bit)   (1 << (bit))
  43 # else /* __BYTE_ORDER == __LITTLE_ENDIAN */
  44 #  define _ISbit(bit)   ((bit) < 8 ? ((1 << (bit)) << 8) : ((1 << (bit)) >> 8))
  45 # endif
  46
  47 enum
  48 {
  49   _ISupper = _ISbit (0),        /* UPPERCASE.  */
  50   _ISlower = _ISbit (1),        /* lowercase.  */
  51   _ISalpha = _ISbit (2),        /* Alphabetic.  */
  52   _ISdigit = _ISbit (3),        /* Numeric.  */
  53   _ISxdigit = _ISbit (4),       /* Hexadecimal numeric.  */
  54   _ISspace = _ISbit (5),        /* Whitespace.  */
  55   _ISprint = _ISbit (6),        /* Printing.  */
  56   _ISgraph = _ISbit (7),        /* Graphical.  */
  57   _ISblank = _ISbit (8),        /* Blank (usually SPC and TAB).  */
  58   _IScntrl = _ISbit (9),        /* Control character.  */
  59   _ISpunct = _ISbit (10),       /* Punctuation.  */
  60   _ISalnum = _ISbit (11)        /* Alphanumeric.  */
  61 };
  62 #endif /* ! _ISbit  */
  63
  64 /* These are defined in ctype-info.c.
  65    The declarations here must match those in localeinfo.h.
  66
  67    In the thread-specific locale model (see `uselocale' in <locale.h>)
  68    we cannot use global variables for these as was done in the past.
  69    Instead, the following accessor functions return the address of
  70    each variable, which is local to the current thread if multithreaded.
  71
  72    These point into arrays of 384, so they can be indexed by any `unsigned
  73    char' value [0,255]; by EOF (-1); or by any `signed char' value
  74    [-128,-1).  ISO C requires that the ctype functions work for `unsigned
  75    char' values and for EOF; we also support negative `signed char' values
  76    for broken old programs.  The case conversion arrays are of `int's
  77    rather than `unsigned char's because tolower (EOF) must be EOF, which
  78    doesn't fit into an `unsigned char'.  But today more important is that
  79    the arrays are also used for multi-byte character sets.  */
  80 extern const unsigned short int **__ctype_b_loc (void)
  81      __THROW __attribute__ ((__const__));
  82 extern const __int32_t **__ctype_tolower_loc (void)
  83      __THROW __attribute__ ((__const__));
  84 extern const __int32_t **__ctype_toupper_loc (void)
  85      __THROW __attribute__ ((__const__));
  86
  87
  88 #ifndef __cplusplus
  89 # define __isctype(c, type) \
  90   ((*__ctype_b_loc ())[(int) (c)] & (unsigned short int) type)
  91 #elif defined __USE_EXTERN_INLINES
  92 # define __isctype_f(type) \
  93   __extern_inline int                                                         \
  94   is##type (int __c) __THROW                                                  \
  95   {                                                                           \
  96     return (*__ctype_b_loc ())[(int) (__c)] & (unsigned short int) _IS##type; \
  97   }
  98 #endif
  99
 100 #define __isascii(c)    (((c) & ~0x7f) == 0)    /* If C is a 7 bit value.  */
 101 #define __toascii(c)    ((c) & 0x7f)            /* Mask off high bits.  */
 102
 103 #define __exctype(name) extern int name (int) __THROW
 104
 105 __BEGIN_NAMESPACE_STD
 106
 107 /* The following names are all functions:
 108      int isCHARACTERISTIC(int c);
 109    which return nonzero iff C has CHARACTERISTIC.
 110    For the meaning of the characteristic names, see the `enum' above.  */
 111 __exctype (isalnum);
 112 __exctype (isalpha);
 113 __exctype (iscntrl);
 114 __exctype (isdigit);
 115 __exctype (islower);
 116 __exctype (isgraph);
 117 __exctype (isprint);
 118 __exctype (ispunct);
 119 __exctype (isspace);
 120 __exctype (isupper);
 121 __exctype (isxdigit);
 122
 123
 124 /* Return the lowercase version of C.  */
 125 extern int tolower (int __c) __THROW;
 126
 127 /* Return the uppercase version of C.  */
 128 extern int toupper (int __c) __THROW;
 129
 130 __END_NAMESPACE_STD
 131
 132
 133 /* ISO C99 introduced one new function.  */
 134 #ifdef  __USE_ISOC99
 135 __BEGIN_NAMESPACE_C99
 136
 137 __exctype (isblank);
 138
 139 __END_NAMESPACE_C99
 140 #endif
 141
 142 #ifdef __USE_GNU
 143 /* Test C for a set of character classes according to MASK.  */
 144 extern int isctype (int __c, int __mask) __THROW;
 145 #endif
 146
 147 #if defined __USE_SVID || defined __USE_MISC || defined __USE_XOPEN
 148
 149 /* Return nonzero iff C is in the ASCII set
 150    (i.e., is no more than 7 bits wide).  */
 151 extern int isascii (int __c) __THROW;
 152
 153 /* Return the part of C that is in the ASCII set
 154    (i.e., the low-order 7 bits of C).  */
 155 extern int toascii (int __c) __THROW;
 156
 157 /* These are the same as `toupper' and `tolower' except that they do not
 158    check the argument for being in the range of a `char'.  */
 159 __exctype (_toupper);
 160 __exctype (_tolower);
 161 #endif /* Use SVID or use misc.  */
 162
 163 /* This code is needed for the optimized mapping functions.  */
 164 #define __tobody(c, f, a, args) \
 165   (__extension__                                                              \
 166    ({ int __res;                                                              \
 167       if (sizeof (c) > 1)                                                     \
 168         {                                                                     \
 169           if (__builtin_constant_p (c))                                       \
 170             {                                                                 \
 171               int __c = (c);                                                  \
 172               __res = __c < -128 || __c > 255 ? __c : (a)[__c];               \
 173             }                                                                 \
 174           else                                                                \
 175             __res = f args;                                                   \
 176         }                                                                     \
 177       else                                                                    \
 178         __res = (a)[(int) (c)];                                               \
 179       __res; }))
 180
 181 #if !defined __NO_CTYPE
 182 # ifdef __isctype_f
 183 __isctype_f (alnum)
 184 __isctype_f (alpha)
 185 __isctype_f (cntrl)
 186 __isctype_f (digit)
 187 __isctype_f (lower)
 188 __isctype_f (graph)
 189 __isctype_f (print)
 190 __isctype_f (punct)
 191 __isctype_f (space)
 192 __isctype_f (upper)
 193 __isctype_f (xdigit)
 194 #  ifdef __USE_ISOC99
 195 __isctype_f (blank)
 196 #  endif
 197 # elif defined __isctype
 198 # define isalnum(c)     __isctype((c), _ISalnum)
 199 # define isalpha(c)     __isctype((c), _ISalpha)
 200 # define iscntrl(c)     __isctype((c), _IScntrl)
 201 # define isdigit(c)     __isctype((c), _ISdigit)
 202 # define islower(c)     __isctype((c), _ISlower)
 203 # define isgraph(c)     __isctype((c), _ISgraph)
 204 # define isprint(c)     __isctype((c), _ISprint)
 205 # define ispunct(c)     __isctype((c), _ISpunct)
 206 # define isspace(c)     __isctype((c), _ISspace)
 207 # define isupper(c)     __isctype((c), _ISupper)
 208 # define isxdigit(c)    __isctype((c), _ISxdigit)
 209 #  ifdef __USE_ISOC99
 210 #   define isblank(c)   __isctype((c), _ISblank)
 211 #  endif
 212 # endif
 213
 214 # ifdef __USE_EXTERN_INLINES
 215 __extern_inline int
 216 __NTH (tolower (int __c))
 217 {
 218   return __c >= -128 && __c < 256 ? (*__ctype_tolower_loc ())[__c] : __c;
 219 }
 220
 221 __extern_inline int
 222 __NTH (toupper (int __c))
 223 {
 224   return __c >= -128 && __c < 256 ? (*__ctype_toupper_loc ())[__c] : __c;
 225 }
 226 # endif
 227
 228 # if __GNUC__ >= 2 && defined __OPTIMIZE__ && !defined __cplusplus
 229 #  define tolower(c)    __tobody (c, tolower, *__ctype_tolower_loc (), (c))
 230 #  define toupper(c)    __tobody (c, toupper, *__ctype_toupper_loc (), (c))
 231 # endif /* Optimizing gcc */
 232
 233 # if defined __USE_SVID || defined __USE_MISC || defined __USE_XOPEN
 234 #  define isascii(c)    __isascii (c)
 235 #  define toascii(c)    __toascii (c)
 236
 237 #  define _tolower(c)   ((int) (*__ctype_tolower_loc ())[(int) (c)])
 238 #  define _toupper(c)   ((int) (*__ctype_toupper_loc ())[(int) (c)])
 239 # endif
 240
 241 #endif /* Not __NO_CTYPE.  */
 242
 243
 244 #ifdef __USE_XOPEN2K8
 245 /* The concept of one static locale per category is not very well
 246    thought out.  Many applications will need to process its data using
 247    information from several different locales.  Another application is
 248    the implementation of the internationalization handling in the
 249    upcoming ISO C++ standard library.  To support this another set of
 250    the functions using locale data exist which have an additional
 251    argument.
 252
 253    Attention: all these functions are *not* standardized in any form.
 254    This is a proof-of-concept implementation.  */
 255
 256 /* Structure for reentrant locale using functions.  This is an
 257    (almost) opaque type for the user level programs.  */
 258 # include <xlocale.h>
 259
 260 /* These definitions are similar to the ones above but all functions
 261    take as an argument a handle for the locale which shall be used.  */
 262 #  define __isctype_l(c, type, locale) \
 263   ((locale)->__ctype_b[(int) (c)] & (unsigned short int) type)
 264
 265 # define __exctype_l(name)                                                    \
 266   extern int name (int, __locale_t) __THROW
 267
 268 /* The following names are all functions:
 269      int isCHARACTERISTIC(int c, locale_t *locale);
 270    which return nonzero iff C has CHARACTERISTIC.
 271    For the meaning of the characteristic names, see the `enum' above.  */
 272 __exctype_l (isalnum_l);
 273 __exctype_l (isalpha_l);
 274 __exctype_l (iscntrl_l);
 275 __exctype_l (isdigit_l);
 276 __exctype_l (islower_l);
 277 __exctype_l (isgraph_l);
 278 __exctype_l (isprint_l);
 279 __exctype_l (ispunct_l);
 280 __exctype_l (isspace_l);
 281 __exctype_l (isupper_l);
 282 __exctype_l (isxdigit_l);
 283
 284 __exctype_l (isblank_l);
 285
 286
 287 /* Return the lowercase version of C in locale L.  */
 288 extern int __tolower_l (int __c, __locale_t __l) __THROW;
 289 extern int tolower_l (int __c, __locale_t __l) __THROW;
 290
 291 /* Return the uppercase version of C.  */
 292 extern int __toupper_l (int __c, __locale_t __l) __THROW;
 293 extern int toupper_l (int __c, __locale_t __l) __THROW;
 294
 295 # if __GNUC__ >= 2 && defined __OPTIMIZE__ && !defined __cplusplus
 296 #  define __tolower_l(c, locale) \
 297   __tobody (c, __tolower_l, (locale)->__ctype_tolower, (c, locale))
 298 #  define __toupper_l(c, locale) \
 299   __tobody (c, __toupper_l, (locale)->__ctype_toupper, (c, locale))
 300 #  define tolower_l(c, locale)  __tolower_l ((c), (locale))
 301 #  define toupper_l(c, locale)  __toupper_l ((c), (locale))
 302 # endif /* Optimizing gcc */
 303
 304
 305 # ifndef __NO_CTYPE
 306 #  define __isalnum_l(c,l)      __isctype_l((c), _ISalnum, (l))
 307 #  define __isalpha_l(c,l)      __isctype_l((c), _ISalpha, (l))
 308 #  define __iscntrl_l(c,l)      __isctype_l((c), _IScntrl, (l))
 309 #  define __isdigit_l(c,l)      __isctype_l((c), _ISdigit, (l))
 310 #  define __islower_l(c,l)      __isctype_l((c), _ISlower, (l))
 311 #  define __isgraph_l(c,l)      __isctype_l((c), _ISgraph, (l))
 312 #  define __isprint_l(c,l)      __isctype_l((c), _ISprint, (l))
 313 #  define __ispunct_l(c,l)      __isctype_l((c), _ISpunct, (l))
 314 #  define __isspace_l(c,l)      __isctype_l((c), _ISspace, (l))
 315 #  define __isupper_l(c,l)      __isctype_l((c), _ISupper, (l))
 316 #  define __isxdigit_l(c,l)     __isctype_l((c), _ISxdigit, (l))
 317
 318 #  define __isblank_l(c,l)      __isctype_l((c), _ISblank, (l))
 319
 320 #  if defined __USE_SVID || defined __USE_MISC
 321 #   define __isascii_l(c,l)     ((l), __isascii (c))
 322 #   define __toascii_l(c,l)     ((l), __toascii (c))
 323 #  endif
 324
 325 #  define isalnum_l(c,l)        __isalnum_l ((c), (l))
 326 #  define isalpha_l(c,l)        __isalpha_l ((c), (l))
 327 #  define iscntrl_l(c,l)        __iscntrl_l ((c), (l))
 328 #  define isdigit_l(c,l)        __isdigit_l ((c), (l))
 329 #  define islower_l(c,l)        __islower_l ((c), (l))
 330 #  define isgraph_l(c,l)        __isgraph_l ((c), (l))
 331 #  define isprint_l(c,l)        __isprint_l ((c), (l))
 332 #  define ispunct_l(c,l)        __ispunct_l ((c), (l))
 333 #  define isspace_l(c,l)        __isspace_l ((c), (l))
 334 #  define isupper_l(c,l)        __isupper_l ((c), (l))
 335 #  define isxdigit_l(c,l)       __isxdigit_l ((c), (l))
 336
 337 #  define isblank_l(c,l)        __isblank_l ((c), (l))
 338
 339 #  if defined __USE_SVID || defined __USE_MISC
 340 #   define isascii_l(c,l)       __isascii_l ((c), (l))
 341 #   define toascii_l(c,l)       __toascii_l ((c), (l))
 342 #  endif
 343
 344 # endif /* Not __NO_CTYPE.  */
 345
 346 #endif /* Use POSIX 2008.  */
 347
 348 __END_DECLS
 349
 350 #endif /* ctype.h  */