X-Git-Url: https://git.distorted.org.uk/~mdw/sgt/charset/blobdiff_plain/b063a840d1dded0455a70fc3e71ef8f92e8644ab..49152469924ab8bc1bb6cad6152297fc85e95be1:/charset.h diff --git a/charset.h b/charset.h index 5d64877..144187f 100644 --- a/charset.h +++ b/charset.h @@ -33,6 +33,7 @@ typedef enum { CS_ISO8859_16, CS_CP437, CS_CP850, + CS_CP866, CS_CP1250, CS_CP1251, CS_CP1252, @@ -85,6 +86,7 @@ typedef enum { CS_HZ, CS_CP949, CS_PDF, + CS_PSSTD, CS_CTEXT, CS_ISO2022, CS_BS4730, @@ -133,6 +135,13 @@ extern const charset_state charset_init_state; * NULL, `errlen' will be ignored, and the library will choose * something sensible to do on its own. For Unicode, this will be * U+FFFD (REPLACEMENT CHARACTER). + * + * `output' may be NULL, in which case the entire translation will + * be performed in theory (e.g. a dry run to work out how much + * space needs to be allocated for the real thing). `outlen' may + * also be negative, indicating an unlimited buffer length + * (although this is almost certainly unwise if `output' is _not_ + * NULL). */ int charset_to_unicode(const char **input, int *inlen, @@ -163,6 +172,13 @@ int charset_to_unicode(const char **input, int *inlen, * If `input' is NULL, this routine will output the necessary bytes * to reset the encoding state in any way which might be required * at the end of an output piece of text. + * + * `output' may be NULL, in which case the entire translation will + * be performed in theory (e.g. a dry run to work out how much + * space needs to be allocated for the real thing). `outlen' may + * also be negative, indicating an unlimited buffer length + * (although this is almost certainly unwise if `output' is _not_ + * NULL). */ int charset_from_unicode(const wchar_t **input, int *inlen, @@ -196,6 +212,13 @@ int charset_from_macenc(int script, int region, int sysvers, const char *fontname); /* + * Convert GNU Emacs coding system symbol to and from our charset + * identifiers. + */ +const char *charset_to_emacsenc(int charset); +int charset_from_emacsenc(const char *name); + +/* * Upgrade a charset identifier to a superset charset which is * often confused with it. For example, people whose MUAs report * their mail as ASCII or ISO8859-1 often in practice turn out to