Tweak previous change
[emacs.git] / lib / c-ctype.h
blobad589b5c2099c956c449fe95cb3393127a8f7c68
1 /* Character handling in C locale.
3 These functions work like the corresponding functions in <ctype.h>,
4 except that they have the C (POSIX) locale hardwired, whereas the
5 <ctype.h> functions' behaviour depends on the current locale set via
6 setlocale.
8 Copyright (C) 2000-2003, 2006, 2008-2013 Free Software Foundation, Inc.
10 This program is free software; you can redistribute it and/or modify
11 it under the terms of the GNU General Public License as published by
12 the Free Software Foundation; either version 3 of the License, or
13 (at your option) any later version.
15 This program is distributed in the hope that it will be useful,
16 but WITHOUT ANY WARRANTY; without even the implied warranty of
17 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18 GNU General Public License for more details.
20 You should have received a copy of the GNU General Public License
21 along with this program; if not, see <http://www.gnu.org/licenses/>. */
23 #ifndef C_CTYPE_H
24 #define C_CTYPE_H
26 #include <stdbool.h>
29 #ifdef __cplusplus
30 extern "C" {
31 #endif
34 /* The functions defined in this file assume the "C" locale and a character
35 set without diacritics (ASCII-US or EBCDIC-US or something like that).
36 Even if the "C" locale on a particular system is an extension of the ASCII
37 character set (like on BeOS, where it is UTF-8, or on AmigaOS, where it
38 is ISO-8859-1), the functions in this file recognize only the ASCII
39 characters. */
42 /* Check whether the ASCII optimizations apply. */
44 /* ANSI C89 (and ISO C99 5.2.1.3 too) already guarantees that
45 '0', '1', ..., '9' have consecutive integer values. */
46 #define C_CTYPE_CONSECUTIVE_DIGITS 1
48 #if ('A' <= 'Z') \
49 && ('A' + 1 == 'B') && ('B' + 1 == 'C') && ('C' + 1 == 'D') \
50 && ('D' + 1 == 'E') && ('E' + 1 == 'F') && ('F' + 1 == 'G') \
51 && ('G' + 1 == 'H') && ('H' + 1 == 'I') && ('I' + 1 == 'J') \
52 && ('J' + 1 == 'K') && ('K' + 1 == 'L') && ('L' + 1 == 'M') \
53 && ('M' + 1 == 'N') && ('N' + 1 == 'O') && ('O' + 1 == 'P') \
54 && ('P' + 1 == 'Q') && ('Q' + 1 == 'R') && ('R' + 1 == 'S') \
55 && ('S' + 1 == 'T') && ('T' + 1 == 'U') && ('U' + 1 == 'V') \
56 && ('V' + 1 == 'W') && ('W' + 1 == 'X') && ('X' + 1 == 'Y') \
57 && ('Y' + 1 == 'Z')
58 #define C_CTYPE_CONSECUTIVE_UPPERCASE 1
59 #endif
61 #if ('a' <= 'z') \
62 && ('a' + 1 == 'b') && ('b' + 1 == 'c') && ('c' + 1 == 'd') \
63 && ('d' + 1 == 'e') && ('e' + 1 == 'f') && ('f' + 1 == 'g') \
64 && ('g' + 1 == 'h') && ('h' + 1 == 'i') && ('i' + 1 == 'j') \
65 && ('j' + 1 == 'k') && ('k' + 1 == 'l') && ('l' + 1 == 'm') \
66 && ('m' + 1 == 'n') && ('n' + 1 == 'o') && ('o' + 1 == 'p') \
67 && ('p' + 1 == 'q') && ('q' + 1 == 'r') && ('r' + 1 == 's') \
68 && ('s' + 1 == 't') && ('t' + 1 == 'u') && ('u' + 1 == 'v') \
69 && ('v' + 1 == 'w') && ('w' + 1 == 'x') && ('x' + 1 == 'y') \
70 && ('y' + 1 == 'z')
71 #define C_CTYPE_CONSECUTIVE_LOWERCASE 1
72 #endif
74 #if (' ' == 32) && ('!' == 33) && ('"' == 34) && ('#' == 35) \
75 && ('%' == 37) && ('&' == 38) && ('\'' == 39) && ('(' == 40) \
76 && (')' == 41) && ('*' == 42) && ('+' == 43) && (',' == 44) \
77 && ('-' == 45) && ('.' == 46) && ('/' == 47) && ('0' == 48) \
78 && ('1' == 49) && ('2' == 50) && ('3' == 51) && ('4' == 52) \
79 && ('5' == 53) && ('6' == 54) && ('7' == 55) && ('8' == 56) \
80 && ('9' == 57) && (':' == 58) && (';' == 59) && ('<' == 60) \
81 && ('=' == 61) && ('>' == 62) && ('?' == 63) && ('A' == 65) \
82 && ('B' == 66) && ('C' == 67) && ('D' == 68) && ('E' == 69) \
83 && ('F' == 70) && ('G' == 71) && ('H' == 72) && ('I' == 73) \
84 && ('J' == 74) && ('K' == 75) && ('L' == 76) && ('M' == 77) \
85 && ('N' == 78) && ('O' == 79) && ('P' == 80) && ('Q' == 81) \
86 && ('R' == 82) && ('S' == 83) && ('T' == 84) && ('U' == 85) \
87 && ('V' == 86) && ('W' == 87) && ('X' == 88) && ('Y' == 89) \
88 && ('Z' == 90) && ('[' == 91) && ('\\' == 92) && (']' == 93) \
89 && ('^' == 94) && ('_' == 95) && ('a' == 97) && ('b' == 98) \
90 && ('c' == 99) && ('d' == 100) && ('e' == 101) && ('f' == 102) \
91 && ('g' == 103) && ('h' == 104) && ('i' == 105) && ('j' == 106) \
92 && ('k' == 107) && ('l' == 108) && ('m' == 109) && ('n' == 110) \
93 && ('o' == 111) && ('p' == 112) && ('q' == 113) && ('r' == 114) \
94 && ('s' == 115) && ('t' == 116) && ('u' == 117) && ('v' == 118) \
95 && ('w' == 119) && ('x' == 120) && ('y' == 121) && ('z' == 122) \
96 && ('{' == 123) && ('|' == 124) && ('}' == 125) && ('~' == 126)
97 /* The character set is ASCII or one of its variants or extensions, not EBCDIC.
98 Testing the value of '\n' and '\r' is not relevant. */
99 #define C_CTYPE_ASCII 1
100 #endif
103 /* Function declarations. */
105 /* Unlike the functions in <ctype.h>, which require an argument in the range
106 of the 'unsigned char' type, the functions here operate on values that are
107 in the 'unsigned char' range or in the 'char' range. In other words,
108 when you have a 'char' value, you need to cast it before using it as
109 argument to a <ctype.h> function:
111 const char *s = ...;
112 if (isalpha ((unsigned char) *s)) ...
114 but you don't need to cast it for the functions defined in this file:
116 const char *s = ...;
117 if (c_isalpha (*s)) ...
120 extern bool c_isascii (int c) _GL_ATTRIBUTE_CONST; /* not locale dependent */
122 extern bool c_isalnum (int c) _GL_ATTRIBUTE_CONST;
123 extern bool c_isalpha (int c) _GL_ATTRIBUTE_CONST;
124 extern bool c_isblank (int c) _GL_ATTRIBUTE_CONST;
125 extern bool c_iscntrl (int c) _GL_ATTRIBUTE_CONST;
126 extern bool c_isdigit (int c) _GL_ATTRIBUTE_CONST;
127 extern bool c_islower (int c) _GL_ATTRIBUTE_CONST;
128 extern bool c_isgraph (int c) _GL_ATTRIBUTE_CONST;
129 extern bool c_isprint (int c) _GL_ATTRIBUTE_CONST;
130 extern bool c_ispunct (int c) _GL_ATTRIBUTE_CONST;
131 extern bool c_isspace (int c) _GL_ATTRIBUTE_CONST;
132 extern bool c_isupper (int c) _GL_ATTRIBUTE_CONST;
133 extern bool c_isxdigit (int c) _GL_ATTRIBUTE_CONST;
135 extern int c_tolower (int c) _GL_ATTRIBUTE_CONST;
136 extern int c_toupper (int c) _GL_ATTRIBUTE_CONST;
139 #if (defined __GNUC__ && !defined __STRICT_ANSI__ && defined __OPTIMIZE__ \
140 && !defined __OPTIMIZE_SIZE__ && !defined NO_C_CTYPE_MACROS)
142 /* ASCII optimizations. */
144 #undef c_isascii
145 #define c_isascii(c) \
146 ({ int __c = (c); \
147 (__c >= 0x00 && __c <= 0x7f); \
150 #if C_CTYPE_CONSECUTIVE_DIGITS \
151 && C_CTYPE_CONSECUTIVE_UPPERCASE && C_CTYPE_CONSECUTIVE_LOWERCASE
152 #if C_CTYPE_ASCII
153 #undef c_isalnum
154 #define c_isalnum(c) \
155 ({ int __c = (c); \
156 ((__c >= '0' && __c <= '9') \
157 || ((__c & ~0x20) >= 'A' && (__c & ~0x20) <= 'Z')); \
159 #else
160 #undef c_isalnum
161 #define c_isalnum(c) \
162 ({ int __c = (c); \
163 ((__c >= '0' && __c <= '9') \
164 || (__c >= 'A' && __c <= 'Z') \
165 || (__c >= 'a' && __c <= 'z')); \
167 #endif
168 #endif
170 #if C_CTYPE_CONSECUTIVE_UPPERCASE && C_CTYPE_CONSECUTIVE_LOWERCASE
171 #if C_CTYPE_ASCII
172 #undef c_isalpha
173 #define c_isalpha(c) \
174 ({ int __c = (c); \
175 ((__c & ~0x20) >= 'A' && (__c & ~0x20) <= 'Z'); \
177 #else
178 #undef c_isalpha
179 #define c_isalpha(c) \
180 ({ int __c = (c); \
181 ((__c >= 'A' && __c <= 'Z') || (__c >= 'a' && __c <= 'z')); \
183 #endif
184 #endif
186 #undef c_isblank
187 #define c_isblank(c) \
188 ({ int __c = (c); \
189 (__c == ' ' || __c == '\t'); \
192 #if C_CTYPE_ASCII
193 #undef c_iscntrl
194 #define c_iscntrl(c) \
195 ({ int __c = (c); \
196 ((__c & ~0x1f) == 0 || __c == 0x7f); \
198 #endif
200 #if C_CTYPE_CONSECUTIVE_DIGITS
201 #undef c_isdigit
202 #define c_isdigit(c) \
203 ({ int __c = (c); \
204 (__c >= '0' && __c <= '9'); \
206 #endif
208 #if C_CTYPE_CONSECUTIVE_LOWERCASE
209 #undef c_islower
210 #define c_islower(c) \
211 ({ int __c = (c); \
212 (__c >= 'a' && __c <= 'z'); \
214 #endif
216 #if C_CTYPE_ASCII
217 #undef c_isgraph
218 #define c_isgraph(c) \
219 ({ int __c = (c); \
220 (__c >= '!' && __c <= '~'); \
222 #endif
224 #if C_CTYPE_ASCII
225 #undef c_isprint
226 #define c_isprint(c) \
227 ({ int __c = (c); \
228 (__c >= ' ' && __c <= '~'); \
230 #endif
232 #if C_CTYPE_ASCII
233 #undef c_ispunct
234 #define c_ispunct(c) \
235 ({ int _c = (c); \
236 (c_isgraph (_c) && ! c_isalnum (_c)); \
238 #endif
240 #undef c_isspace
241 #define c_isspace(c) \
242 ({ int __c = (c); \
243 (__c == ' ' || __c == '\t' \
244 || __c == '\n' || __c == '\v' || __c == '\f' || __c == '\r'); \
247 #if C_CTYPE_CONSECUTIVE_UPPERCASE
248 #undef c_isupper
249 #define c_isupper(c) \
250 ({ int __c = (c); \
251 (__c >= 'A' && __c <= 'Z'); \
253 #endif
255 #if C_CTYPE_CONSECUTIVE_DIGITS \
256 && C_CTYPE_CONSECUTIVE_UPPERCASE && C_CTYPE_CONSECUTIVE_LOWERCASE
257 #if C_CTYPE_ASCII
258 #undef c_isxdigit
259 #define c_isxdigit(c) \
260 ({ int __c = (c); \
261 ((__c >= '0' && __c <= '9') \
262 || ((__c & ~0x20) >= 'A' && (__c & ~0x20) <= 'F')); \
264 #else
265 #undef c_isxdigit
266 #define c_isxdigit(c) \
267 ({ int __c = (c); \
268 ((__c >= '0' && __c <= '9') \
269 || (__c >= 'A' && __c <= 'F') \
270 || (__c >= 'a' && __c <= 'f')); \
272 #endif
273 #endif
275 #if C_CTYPE_CONSECUTIVE_UPPERCASE && C_CTYPE_CONSECUTIVE_LOWERCASE
276 #undef c_tolower
277 #define c_tolower(c) \
278 ({ int __c = (c); \
279 (__c >= 'A' && __c <= 'Z' ? __c - 'A' + 'a' : __c); \
281 #undef c_toupper
282 #define c_toupper(c) \
283 ({ int __c = (c); \
284 (__c >= 'a' && __c <= 'z' ? __c - 'a' + 'A' : __c); \
286 #endif
288 #endif /* optimizing for speed */
291 #ifdef __cplusplus
293 #endif
295 #endif /* C_CTYPE_H */