2 * charset.h - header file for general character set conversion
6 #ifndef charset_charset_h
7 #define charset_charset_h
12 * Enumeration that lists all the multibyte or single-byte
13 * character sets known to this library.
16 CS_NONE, /* used for reporting errors, etc */
18 CS_ISO8859_1_X11, /* X font encoding with VT100 glyphs */
58 * Routine to convert a MB/SB character set to Unicode.
60 * This routine accepts some number of bytes, updates a state
61 * variable, and outputs some number of Unicode characters. There
62 * are no guarantees. You can't even guarantee that at most one
63 * Unicode character will be output per byte you feed in; for
64 * example, suppose you're reading UTF-8, you've seen E1 80, and
65 * then you suddenly see FE. Now you need to output _two_ error
66 * characters - one for the incomplete sequence E1 80, and one for
67 * the completely invalid UTF-8 byte FE.
69 * Returns the number of wide characters output; will never output
70 * more than the size of the buffer (as specified on input).
71 * Advances the `input' pointer and decrements `inlen', to indicate
72 * how far along the input string it got.
74 * The sequence of `errlen' wide characters pointed to by `errstr'
75 * will be used to indicate a conversion error. If `errstr' is
76 * NULL, `errlen' will be ignored, and the library will choose
77 * something sensible to do on its own. For Unicode, this will be
78 * U+FFFD (REPLACEMENT CHARACTER).
81 int charset_to_unicode(char **input, int *inlen, wchar_t *output, int outlen,
82 int charset, charset_state *state,
83 const wchar_t *errstr, int errlen);
86 * Routine to convert Unicode to an MB/SB character set.
88 * This routine accepts some number of Unicode characters, updates
89 * a state variable, and outputs some number of bytes.
91 * Returns the number of bytes characters output; will never output
92 * more than the size of the buffer (as specified on input), and
93 * will never output a partial MB character. Advances the `input'
94 * pointer and decrements `inlen', to indicate how far along the
95 * input string it got.
97 * The sequence of `errlen' characters pointed to by `errstr' will
98 * be used to indicate a conversion error. If `errstr' is NULL,
99 * `errlen' will be ignored, and the library will choose something
100 * sensible to do on its own (which will vary depending on the
104 int charset_from_unicode(wchar_t **input, int *inlen, char *output, int outlen,
105 int charset, charset_state *state,
106 const char *errstr, int errlen);
109 * Convert X11 encoding names to and from our charset identifiers.
111 const char *charset_to_xenc(int charset);
112 int charset_from_xenc(const char *name);
115 * Convert MIME encoding names to and from our charset identifiers.
117 const char *charset_to_mimeenc(int charset);
118 int charset_from_mimeenc(const char *name);
120 #endif /* charset_charset_h */