X-Git-Url: https://git.distorted.org.uk/~mdw/sgt/putty/blobdiff_plain/8772ac697da9af645ce4ce9d73b6406f7d126f49..93511b79b8f756347a81cf174ed99bf900d81042:/unix/uxucs.c diff --git a/unix/uxucs.c b/unix/uxucs.c index 5b211498..5d3d5af0 100644 --- a/unix/uxucs.c +++ b/unix/uxucs.c @@ -8,6 +8,7 @@ #include #include "putty.h" +#include "charset.h" #include "terminal.h" #include "misc.h" @@ -20,14 +21,14 @@ int is_dbcs_leadbyte(int codepage, char byte) return 0; /* we don't do DBCS */ } -int mb_to_wc(int codepage, int flags, char *mbstr, int mblen, +int mb_to_wc(int codepage, int flags, const char *mbstr, int mblen, wchar_t *wcstr, int wclen) { if (codepage == DEFAULT_CODEPAGE) { int n = 0; - mbstate_t state = { 0 }; + mbstate_t state; - setlocale(LC_CTYPE, ""); + memset(&state, 0, sizeof state); while (mblen > 0) { size_t i = mbrtowc(wcstr+n, mbstr, (size_t)mblen, &state); @@ -38,8 +39,6 @@ int mb_to_wc(int codepage, int flags, char *mbstr, int mblen, mblen -= i; } - setlocale(LC_CTYPE, "C"); - return n; } else if (codepage == CS_NONE) { int n = 0; @@ -57,8 +56,9 @@ int mb_to_wc(int codepage, int flags, char *mbstr, int mblen, NULL, NULL, 0); } -int wc_to_mb(int codepage, int flags, wchar_t *wcstr, int wclen, - char *mbstr, int mblen, char *defchr, int *defused) +int wc_to_mb(int codepage, int flags, const wchar_t *wcstr, int wclen, + char *mbstr, int mblen, char *defchr, int *defused, + struct unicode_data *ucsdata) { /* FIXME: we should remove the defused param completely... */ if (defused) @@ -66,10 +66,10 @@ int wc_to_mb(int codepage, int flags, wchar_t *wcstr, int wclen, if (codepage == DEFAULT_CODEPAGE) { char output[MB_LEN_MAX]; - mbstate_t state = { 0 }; + mbstate_t state; int n = 0; - setlocale(LC_CTYPE, ""); + memset(&state, 0, sizeof state); while (wclen > 0) { int i = wcrtomb(output, wcstr[0], &state); @@ -81,8 +81,6 @@ int wc_to_mb(int codepage, int flags, wchar_t *wcstr, int wclen, wclen--; } - setlocale(LC_CTYPE, "C"); - return n; } else if (codepage == CS_NONE) { int n = 0; @@ -104,7 +102,8 @@ int wc_to_mb(int codepage, int flags, wchar_t *wcstr, int wclen, /* * Return value is TRUE if pterm is to run in direct-to-font mode. */ -int init_ucs(char *linecharset, int font_charset) +int init_ucs(struct unicode_data *ucsdata, char *linecharset, + int utf8_override, int font_charset, int vtmode) { int i, ret = 0; @@ -114,15 +113,30 @@ int init_ucs(char *linecharset, int font_charset) * support at all. So we set this to something which will never * be used. */ - font_codepage = -1; + ucsdata->font_codepage = -1; /* - * line_codepage should be decoded from the specification in - * cfg. + * If utf8_override is set and the POSIX locale settings + * dictate a UTF-8 character set, then just go straight for + * UTF-8. */ - line_codepage = charset_from_mimeenc(linecharset); - if (line_codepage == CS_NONE) - line_codepage = charset_from_xenc(linecharset); + ucsdata->line_codepage = CS_NONE; + if (utf8_override) { + const char *s; + if (((s = getenv("LC_ALL")) && *s) || + ((s = getenv("LC_CTYPE")) && *s) || + ((s = getenv("LANG")) && *s)) { + if (strstr(s, "UTF-8")) + ucsdata->line_codepage = CS_UTF8; + } + } + + /* + * Failing that, line_codepage should be decoded from the + * specification in conf. + */ + if (ucsdata->line_codepage == CS_NONE) + ucsdata->line_codepage = decode_codepage(linecharset); /* * If line_codepage is _still_ CS_NONE, we assume we're using @@ -131,10 +145,10 @@ int init_ucs(char *linecharset, int font_charset) * font we were given had an incomprehensible charset - then we * fall back to using the D800 page. */ - if (line_codepage == CS_NONE) - line_codepage = font_charset; + if (ucsdata->line_codepage == CS_NONE) + ucsdata->line_codepage = font_charset; - if (line_codepage == CS_NONE) + if (ucsdata->line_codepage == CS_NONE) ret = 1; /* @@ -142,19 +156,21 @@ int init_ucs(char *linecharset, int font_charset) * in the line codepage into Unicode. */ for (i = 0; i < 256; i++) { - char c[1], *p; + char c[1]; + const char *p; wchar_t wc[1]; int len; c[0] = i; p = c; len = 1; - if (line_codepage == CS_NONE) - unitab_line[i] = 0xD800 | i; - else if (1 == charset_to_unicode(&p, &len, wc, 1, line_codepage, + if (ucsdata->line_codepage == CS_NONE) + ucsdata->unitab_line[i] = 0xD800 | i; + else if (1 == charset_to_unicode(&p, &len, wc, 1, + ucsdata->line_codepage, NULL, L"", 0)) - unitab_line[i] = wc[0]; + ucsdata->unitab_line[i] = wc[0]; else - unitab_line[i] = 0xFFFD; + ucsdata->unitab_line[i] = 0xFFFD; } /* @@ -174,10 +190,20 @@ int init_ucs(char *linecharset, int font_charset) 0x23bb, 0x2500, 0x23bc, 0x23bd, 0x251c, 0x2524, 0x2534, 0x252c, 0x2502, 0x2264, 0x2265, 0x03c0, 0x2260, 0x00a3, 0x00b7, 0x0020 }; + static const wchar_t unitab_xterm_poorman[32] = + L"*#****o~**+++++-----++++|****L. "; + + const wchar_t *ptr; + + if (vtmode == VT_POORMAN) + ptr = unitab_xterm_poorman; + else + ptr = unitab_xterm_std; + if (i >= 0x5F && i < 0x7F) - unitab_xterm[i] = unitab_xterm_std[i & 0x1F]; + ucsdata->unitab_xterm[i] = ptr[i & 0x1F]; else - unitab_xterm[i] = unitab_line[i]; + ucsdata->unitab_xterm[i] = ucsdata->unitab_line[i]; } /* @@ -185,16 +211,17 @@ int init_ucs(char *linecharset, int font_charset) * simply CP437. */ for (i = 0; i < 256; i++) { - char c[1], *p; + char c[1]; + const char *p; wchar_t wc[1]; int len; c[0] = i; p = c; len = 1; if (1 == charset_to_unicode(&p, &len, wc, 1, CS_CP437, NULL, L"", 0)) - unitab_scoacs[i] = wc[0]; + ucsdata->unitab_scoacs[i] = wc[0]; else - unitab_scoacs[i] = 0xFFFD; + ucsdata->unitab_scoacs[i] = 0xFFFD; } /* @@ -205,13 +232,40 @@ int init_ucs(char *linecharset, int font_charset) * used in this way will be IBM or MS code pages anyway.) */ for (i = 0; i < 256; i++) { - int lineval = unitab_line[i]; + int lineval = ucsdata->unitab_line[i]; if (lineval < ' ' || (lineval >= 0x7F && lineval < 0xA0) || (lineval >= 0xD800 && lineval < 0xD820) || (lineval == 0xD87F)) - unitab_ctrl[i] = i; + ucsdata->unitab_ctrl[i] = i; else - unitab_ctrl[i] = 0xFF; + ucsdata->unitab_ctrl[i] = 0xFF; } return ret; } + +const char *cp_name(int codepage) +{ + if (codepage == CS_NONE) + return "Use font encoding"; + return charset_to_localenc(codepage); +} + +const char *cp_enumerate(int index) +{ + int charset; + charset = charset_localenc_nth(index); + if (charset == CS_NONE) { + /* "Use font encoding" comes after all the named charsets */ + if (charset_localenc_nth(index-1) != CS_NONE) + return "Use font encoding"; + return NULL; + } + return charset_to_localenc(charset); +} + +int decode_codepage(char *cp_name) +{ + if (!cp_name || !*cp_name) + return CS_UTF8; + return charset_from_localenc(cp_name); +}