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_VISCII,
  48     CS_HP_ROMAN8,
  49     CS_DEC_MCS,
  50     CS_UTF8
  51 } charset_t;
  52
  53 typedef struct {
  54     unsigned long s0;
  55 } charset_state;
  56
  57 /*
  58  * Routine to convert a MB/SB character set to Unicode.
  59  *
  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.
  68  *
  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.
  73  *
  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).
  79  */
  80
  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);
  84
  85 /*
  86  * Routine to convert Unicode to an MB/SB character set.
  87  *
  88  * This routine accepts some number of Unicode characters, updates
  89  * a state variable, and outputs some number of bytes.
  90  *
  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.
  96  *
  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
 101  * output charset).
 102  */
 103
 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);
 107
 108 /*
 109  * Convert X11 encoding names to and from our charset identifiers.
 110  */
 111 const char *charset_to_xenc(int charset);
 112 int charset_from_xenc(const char *name);
 113
 114 /*
 115  * Convert MIME encoding names to and from our charset identifiers.
 116  */
 117 const char *charset_to_mimeenc(int charset);
 118 int charset_from_mimeenc(const char *name);
 119
 120 #endif /* charset_charset_h */