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 */
77 * Routine to convert a MB/SB character set to Unicode.
79 * This routine accepts some number of bytes, updates a state
80 * variable, and outputs some number of Unicode characters. There
81 * are no guarantees. You can't even guarantee that at most one
82 * Unicode character will be output per byte you feed in; for
83 * example, suppose you're reading UTF-8, you've seen E1 80, and
84 * then you suddenly see FE. Now you need to output _two_ error
85 * characters - one for the incomplete sequence E1 80, and one for
86 * the completely invalid UTF-8 byte FE.
88 * Returns the number of wide characters output; will never output
89 * more than the size of the buffer (as specified on input).
90 * Advances the `input' pointer and decrements `inlen', to indicate
91 * how far along the input string it got.
93 * The sequence of `errlen' wide characters pointed to by `errstr'
94 * will be used to indicate a conversion error. If `errstr' is
95 * NULL, `errlen' will be ignored, and the library will choose
96 * something sensible to do on its own. For Unicode, this will be
97 * U+FFFD (REPLACEMENT CHARACTER).
100 int charset_to_unicode(char **input, int *inlen, wchar_t *output, int outlen,
101 int charset, charset_state *state,
102 const wchar_t *errstr, int errlen);
105 * Routine to convert Unicode to an MB/SB character set.
107 * This routine accepts some number of Unicode characters, updates
108 * a state variable, and outputs some number of bytes.
110 * Returns the number of bytes characters output; will never output
111 * more than the size of the buffer (as specified on input), and
112 * will never output a partial MB character. Advances the `input'
113 * pointer and decrements `inlen', to indicate how far along the
114 * input string it got.
116 * The sequence of `errlen' characters pointed to by `errstr' will
117 * be used to indicate a conversion error. If `errstr' is NULL,
118 * `errlen' will be ignored, and the library will choose something
119 * sensible to do on its own (which will vary depending on the
123 int charset_from_unicode(wchar_t **input, int *inlen, char *output, int outlen,
124 int charset, charset_state *state,
125 const char *errstr, int errlen);
128 * Convert X11 encoding names to and from our charset identifiers.
130 const char *charset_to_xenc(int charset);
131 int charset_from_xenc(const char *name);
134 * Convert MIME encoding names to and from our charset identifiers.
136 const char *charset_to_mimeenc(int charset);
137 int charset_from_mimeenc(const char *name);
140 * Convert our own encoding names to and from our charset
143 const char *charset_to_localenc(int charset);
144 int charset_from_localenc(const char *name);
145 int charset_localenc_nth(int n);
148 * Convert Mac OS script/region/font to our charset identifiers.
150 int charset_from_macenc(int script, int region, int sysvers,
151 const char *fontname);
153 #endif /* charset_charset_h */