LXR linux/include/linux/nls.h

   1#ifndef _LINUX_NLS_H
   2#define _LINUX_NLS_H
   3
   4#include <linux/init.h>
   5
   6/* Unicode has changed over the years.  Unicode code points no longer
   7 * fit into 16 bits; as of Unicode 5 valid code points range from 0
   8 * to 0x10ffff (17 planes, where each plane holds 65536 code points).
   9 *
  10 * The original decision to represent Unicode characters as 16-bit
  11 * wchar_t values is now outdated.  But plane 0 still includes the
  12 * most commonly used characters, so we will retain it.  The newer
  13 * 32-bit unicode_t type can be used when it is necessary to
  14 * represent the full Unicode character set.
  15 */
  16
  17/* Plane-0 Unicode character */
  18typedef u16 wchar_t;
  19#define MAX_WCHAR_T     0xffff
  20
  21/* Arbitrary Unicode character */
  22typedef u32 unicode_t;
  23
  24struct nls_table {
  25        const char *charset;
  26        const char *alias;
  27        int (*uni2char) (wchar_t uni, unsigned char *out, int boundlen);
  28        int (*char2uni) (const unsigned char *rawstring, int boundlen,
  29                         wchar_t *uni);
  30        const unsigned char *charset2lower;
  31        const unsigned char *charset2upper;
  32        struct module *owner;
  33        struct nls_table *next;
  34};
  35
  36/* this value hold the maximum octet of charset */
  37#define NLS_MAX_CHARSET_SIZE 6 /* for UTF-8 */
  38
  39/* Byte order for UTF-16 strings */
  40enum utf16_endian {
  41        UTF16_HOST_ENDIAN,
  42        UTF16_LITTLE_ENDIAN,
  43        UTF16_BIG_ENDIAN
  44};
  45
  46/* nls_base.c */
  47extern int __register_nls(struct nls_table *, struct module *);
  48extern int unregister_nls(struct nls_table *);
  49extern struct nls_table *load_nls(char *);
  50extern void unload_nls(struct nls_table *);
  51extern struct nls_table *load_nls_default(void);
  52#define register_nls(nls) __register_nls((nls), THIS_MODULE)
  53
  54extern int utf8_to_utf32(const u8 *s, int len, unicode_t *pu);
  55extern int utf32_to_utf8(unicode_t u, u8 *s, int maxlen);
  56extern int utf8s_to_utf16s(const u8 *s, int len,
  57                enum utf16_endian endian, wchar_t *pwcs, int maxlen);
  58extern int utf16s_to_utf8s(const wchar_t *pwcs, int len,
  59                enum utf16_endian endian, u8 *s, int maxlen);
  60
  61static inline unsigned char nls_tolower(struct nls_table *t, unsigned char c)
  62{
  63        unsigned char nc = t->charset2lower[c];
  64
  65        return nc ? nc : c;
  66}
  67
  68static inline unsigned char nls_toupper(struct nls_table *t, unsigned char c)
  69{
  70        unsigned char nc = t->charset2upper[c];
  71
  72        return nc ? nc : c;
  73}
  74
  75static inline int nls_strnicmp(struct nls_table *t, const unsigned char *s1,
  76                const unsigned char *s2, int len)
  77{
  78        while (len--) {
  79                if (nls_tolower(t, *s1++) != nls_tolower(t, *s2++))
  80                        return 1;
  81        }
  82
  83        return 0;
  84}
  85
  86/*
  87 * nls_nullsize - return length of null character for codepage
  88 * @codepage - codepage for which to return length of NULL terminator
  89 *
  90 * Since we can't guarantee that the null terminator will be a particular
  91 * length, we have to check against the codepage. If there's a problem
  92 * determining it, assume a single-byte NULL terminator.
  93 */
  94static inline int
  95nls_nullsize(const struct nls_table *codepage)
  96{
  97        int charlen;
  98        char tmp[NLS_MAX_CHARSET_SIZE];
  99
 100        charlen = codepage->uni2char(0, tmp, NLS_MAX_CHARSET_SIZE);
 101
 102        return charlen > 0 ? charlen : 1;
 103}
 104
 105#define MODULE_ALIAS_NLS(name)  MODULE_ALIAS("nls_" __stringify(name))
 106
 107#endif /* _LINUX_NLS_H */
 108
 109