| 8a731dfa |
1 | /* |
| 2 | * locale.c: try very hard to figure out the libcharset charset |
| 3 | * identifier corresponding to the current C library locale. |
| 4 | * |
| 5 | * This function works by calling nl_langinfo(CODESET) if |
| 6 | * available; failing that, it will try to figure it out manually |
| 7 | * by examining the locale environment variables. Code for the |
| 8 | * latter is adapted from Markus Kuhn's public-domain |
| 9 | * implementation of nl_langinfo(CODESET), available at |
| 10 | * |
| 11 | * http://www.cl.cam.ac.uk/~mgk25/ucs/langinfo.c |
| 12 | */ |
| 13 | |
| 14 | #include <stdlib.h> |
| 15 | #include <string.h> |
| 16 | #include "charset.h" |
| 17 | |
| 18 | #ifdef HAS_LANGINFO_CODESET |
| 19 | #include "langinfo.h" |
| 20 | #endif |
| 21 | |
| 22 | int charset_from_locale(void) |
| 23 | { |
| 24 | char *l, *p; |
| 25 | |
| 26 | #ifdef HAS_LANGINFO_CODESET |
| 27 | int charset; |
| 28 | char const *csname; |
| 29 | |
| 30 | csname = nl_langinfo(CODESET); |
| 31 | if (csname && |
| 32 | (charset = charset_from_localenc(csname)) != CS_NONE) |
| 33 | return charset; |
| 34 | #endif |
| 35 | |
| 36 | if (((l = getenv("LC_ALL")) && *l) || |
| 37 | ((l = getenv("LC_CTYPE")) && *l) || |
| 38 | ((l = getenv("LANG")) && *l)) { |
| 39 | /* check standardized locales */ |
| 40 | if (!strcmp(l, "C") || !strcmp(l, "POSIX")) |
| 41 | return CS_ASCII; |
| 42 | /* check for encoding name fragment */ |
| 43 | if (strstr(l, "UTF") || strstr(l, "utf")) |
| 44 | return CS_UTF8; |
| 45 | if ((p = strstr(l, "8859-"))) { |
| 46 | char buf[16]; |
| 47 | int charset; |
| 48 | |
| 49 | memcpy(buf, "ISO-8859-\0\0", 12); |
| 50 | p += 5; |
| 51 | if ((*p) >= '0' && (*p) <= '9') { |
| 52 | buf[9] = *p++; |
| 53 | if ((*p) >= '0' && (*p) <= '9') buf[10] = *p++; |
| 54 | if ((charset = charset_from_localenc(buf)) != CS_NONE) |
| 55 | return charset; |
| 56 | } |
| 57 | } |
| 58 | if (strstr(l, "KOI8-RU")) return CS_KOI8_RU; |
| 59 | if (strstr(l, "KOI8-R")) return CS_KOI8_R; |
| 60 | if (strstr(l, "KOI8-U")) return CS_KOI8_U; |
| 61 | /* if (strstr(l, "620")) return "TIS-620"; */ |
| 62 | if (strstr(l, "2312")) return CS_EUC_CN; |
| 63 | /* if (strstr(l, "HKSCS")) return "Big5HKSCS"; */ |
| 64 | if (strstr(l, "Big5") || strstr(l, "BIG5")) return CS_BIG5; |
| 65 | /* if (strstr(l, "GBK")) return "GBK"; */ |
| 66 | /* if (strstr(l, "18030")) return "GB18030"; */ |
| 67 | if (strstr(l, "Shift_JIS") || strstr(l, "SJIS")) return CS_SHIFT_JIS; |
| 68 | /* check for conclusive modifier */ |
| 69 | if (strstr(l, "euro")) return CS_ISO8859_15; |
| 70 | /* check for language (and perhaps country) codes */ |
| 71 | if (strstr(l, "zh_TW")) return CS_BIG5; |
| 72 | /* if (strstr(l, "zh_HK")) return "Big5HKSCS"; */ |
| 73 | if (strstr(l, "zh")) return CS_EUC_CN; |
| 74 | if (strstr(l, "ja")) return CS_EUC_JP; |
| 75 | if (strstr(l, "ko")) return CS_EUC_KR; |
| 76 | if (strstr(l, "ru")) return CS_KOI8_R; |
| 77 | if (strstr(l, "uk")) return CS_KOI8_U; |
| 78 | if (strstr(l, "pl") || strstr(l, "hr") || |
| 79 | strstr(l, "hu") || strstr(l, "cs") || |
| 80 | strstr(l, "sk") || strstr(l, "sl")) return CS_ISO8859_2; |
| 81 | if (strstr(l, "eo") || strstr(l, "mt")) return CS_ISO8859_3; |
| 82 | if (strstr(l, "el")) return CS_ISO8859_7; |
| 83 | if (strstr(l, "he")) return CS_ISO8859_8; |
| 84 | if (strstr(l, "tr")) return CS_ISO8859_9; |
| 85 | /* if (strstr(l, "th")) return "TIS-620"; */ |
| 86 | if (strstr(l, "lt")) return CS_ISO8859_13; |
| 87 | if (strstr(l, "cy")) return CS_ISO8859_14; |
| 88 | if (strstr(l, "ro")) return CS_ISO8859_2; /* or ISO-8859-16 */ |
| 89 | if (strstr(l, "am") || strstr(l, "vi")) return CS_UTF8; |
| 90 | return CS_ISO8859_1; |
| 91 | } |
| 92 | return CS_ASCII; |
| 93 | } |
~mdw / sgt / charset / blame