2 * UCW Library -- Character Types
4 * (c) 1997--2004 Martin Mares <mj@ucw.cz>
6 * This software may be freely distributed and used according to the terms
7 * of the GNU Lesser General Public License.
10 #ifndef _UCW_CHARTYPE_H
11 #define _UCW_CHARTYPE_H
14 * We define our own routines to classify 8-bit characters (based on US-ASCII charset).
15 * This way we bypass most possible problems with different compilation environments.
17 * All functions and macros accept any numbers and if it is necessary, they simply ignore higher bits.
18 * It does not matter whether a parameter is signed or unsigned. Parameters are evaluated exactly once,
19 * so they can have side-effects.
22 #define _C_UPPER 1 /* Upper-case letters */
23 #define _C_LOWER 2 /* Lower-case letters */
24 #define _C_PRINT 4 /* Printable */
25 #define _C_DIGIT 8 /* Digits */
26 #define _C_CTRL 16 /* Control characters */
27 #define _C_XDIGIT 32 /* Hexadecimal digits */
28 #define _C_BLANK 64 /* White spaces (spaces, tabs, newlines) */
29 #define _C_INNER 128 /* `inner punctuation' -- underscore etc. */
31 #define _C_ALPHA (_C_UPPER | _C_LOWER)
32 #define _C_ALNUM (_C_ALPHA | _C_DIGIT)
33 #define _C_WORD (_C_ALNUM | _C_INNER)
34 #define _C_WSTART (_C_ALPHA | _C_INNER)
36 extern const byte ucw_c_cat[256], ucw_c_upper[256], ucw_c_lower[256];
38 #define Category(x) (ucw_c_cat[(byte)(x)])
39 #define Ccat(x,y) (Category(x) & y)
41 #define Cupper(x) Ccat(x, _C_UPPER) /** Checks for an upper-case character (`A-Z`). **/
42 #define Clower(x) Ccat(x, _C_LOWER) /** Checks for a lower-case character (`a-z`). **/
43 #define Calpha(x) Ccat(x, _C_ALPHA) /** Checks for an alphabetic character (`a-z`, `A-Z`). **/
44 #define Calnum(x) Ccat(x, _C_ALNUM) /** Checks for an alpha-numeric character (`a-z`, `A-Z`, `0-9`). */
45 #define Cprint(x) Ccat(x, _C_PRINT) /** Checks for printable characters, including 8-bit values (`\t`, `0x20-0x7E`, `0x80-0xFF`). **/
46 #define Cdigit(x) Ccat(x, _C_DIGIT) /** Checks for a digit (`0-9`). **/
47 #define Cxdigit(x) Ccat(x, _C_XDIGIT) /** Checks for a hexadecimal digit (`0-9`, `a-f`, `A-F`). **/
48 #define Cword(x) Ccat(x, _C_WORD) /** Checks for an alpha-numeric character or an inner punctation (`a-z`, `A-Z`, `0-9`, `_`). **/
49 #define Cblank(x) Ccat(x, _C_BLANK) /** Checks for a white space (`0x20`, `\t`, `\n`, `\r`, `0x8`, `0xC`). **/
50 #define Cctrl(x) Ccat(x, _C_CTRL) /** Checks for control characters (`0x0-0x1F`, `0x7F`). **/
51 #define Cspace(x) Cblank(x)
53 #define Cupcase(x) (ucw_c_upper[(byte)(x)]) /** Convert a letter to upper case, leave non-letter characters unchanged. **/
54 #define Clocase(x) (ucw_c_lower[(byte)(x)]) /** Convert a letter to lower case, leave non-letter characters unchanged. **/
57 * Compute the value of a valid hexadecimal character (ie. passed the @Cxdigit() check).
59 static inline uns Cxvalue(byte x)
61 return (x < (uns)'A') ? x - '0' : (x & 0xdf) - 'A' + 10;