git.distorted.org.uk Git - u/mdw/putty/blob - charset/charset.h

   1 /*
   2  * charset.h - header file for general character set conversion
   3  * routines.
   4  */
   5
   6 #ifndef charset_charset_h
   7 #define charset_charset_h
   8
   9 #include <stddef.h>
  10
  11 /*
  12  * Enumeration that lists all the multibyte or single-byte
  13  * character sets known to this library.
  14  */
  15 typedef enum {
  16     CS_NONE,                           /* used for reporting errors, etc */
  17     CS_ISO8859_1,
  18     CS_ISO8859_1_X11,                  /* X font encoding with VT100 glyphs */
  19     CS_ISO8859_2,
  20     CS_ISO8859_3,
  21     CS_ISO8859_4,
  22     CS_ISO8859_5,
  23     CS_ISO8859_6,
  24     CS_ISO8859_7,
  25     CS_ISO8859_8,
  26     CS_ISO8859_9,
  27     CS_ISO8859_10,
  28     CS_ISO8859_11,
  29     CS_ISO8859_13,
  30     CS_ISO8859_14,
  31     CS_ISO8859_15,
  32     CS_ISO8859_16,
  33     CS_CP437,
  34     CS_CP850,
  35     CS_CP1250,
  36     CS_CP1251,
  37     CS_CP1252,
  38     CS_CP1253,
  39     CS_CP1254,
  40     CS_CP1255,
  41     CS_CP1256,
  42     CS_CP1257,
  43     CS_CP1258,
  44     CS_KOI8_R,
  45     CS_KOI8_U,
  46     CS_MAC_ROMAN,
  47     CS_MAC_TURKISH,
  48     CS_MAC_CROATIAN,
  49     CS_MAC_ICELAND,
  50     CS_MAC_ROMANIAN,
  51     CS_MAC_GREEK,
  52     CS_MAC_CYRILLIC,
  53     CS_MAC_THAI,
  54     CS_MAC_CENTEURO,
  55     CS_MAC_SYMBOL,
  56     CS_MAC_DINGBATS,
  57     CS_MAC_ROMAN_OLD,
  58     CS_MAC_CROATIAN_OLD,
  59     CS_MAC_ICELAND_OLD,
  60     CS_MAC_ROMANIAN_OLD,
  61     CS_MAC_GREEK_OLD,
  62     CS_MAC_CYRILLIC_OLD,
  63     CS_MAC_UKRAINE,
  64     CS_MAC_VT100,
  65     CS_MAC_VT100_OLD,
  66     CS_VISCII,
  67     CS_HP_ROMAN8,
  68     CS_DEC_MCS,
  69     CS_UTF8
  70 } charset_t;
  71
  72 typedef struct {
  73     unsigned long s0;
  74 } charset_state;
  75
  76 /*
  77  * Routine to convert a MB/SB character set to Unicode.
  78  *
  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.
  87  *
  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.
  92  *
  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).
  98  */
  99
 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);
 103
 104 /*
 105  * Routine to convert Unicode to an MB/SB character set.
 106  *
 107  * This routine accepts some number of Unicode characters, updates
 108  * a state variable, and outputs some number of bytes.
 109  *
 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.
 115  *
 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
 120  * output charset).
 121  */
 122
 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);
 126
 127 /*
 128  * Convert X11 encoding names to and from our charset identifiers.
 129  */
 130 const char *charset_to_xenc(int charset);
 131 int charset_from_xenc(const char *name);
 132
 133 /*
 134  * Convert MIME encoding names to and from our charset identifiers.
 135  */
 136 const char *charset_to_mimeenc(int charset);
 137 int charset_from_mimeenc(const char *name);
 138
 139 /*
 140  * Convert our own encoding names to and from our charset
 141  * identifiers.
 142  */
 143 const char *charset_to_localenc(int charset);
 144 int charset_from_localenc(const char *name);
 145 int charset_localenc_nth(int n);
 146
 147 /*
 148  * Convert Mac OS script/region/font to our charset identifiers.
 149  */
 150 int charset_from_macenc(int script, int region, int sysvers,
 151                         const char *fontname);
 152
 153 #endif /* charset_charset_h */