• Main Page
  • Modules
  • Data Structures
  • Files
  • File List
  • Globals

missing/langinfo.c

Go to the documentation of this file.
00001 /* -*- c-file-style: "gnu" -*- */
00002 /*
00003  * This is a quick-and-dirty emulator of the nl_langinfo(CODESET)
00004  * function defined in the Single Unix Specification for those systems
00005  * (FreeBSD, etc.) that don't have one yet. It behaves as if it had
00006  * been called after setlocale(LC_CTYPE, ""), that is it looks at
00007  * the locale environment variables.
00008  *
00009  * http://www.opengroup.org/onlinepubs/7908799/xsh/langinfo.h.html
00010  *
00011  * Please extend it as needed and suggest improvements to the author.
00012  * This emulator will hopefully become redundant soon as
00013  * nl_langinfo(CODESET) becomes more widely implemented.
00014  *
00015  * Since the proposed Li18nux encoding name registry is still not mature,
00016  * the output follows the MIME registry where possible:
00017  *
00018  *   http://www.iana.org/assignments/character-sets
00019  *
00020  * A possible autoconf test for the availability of nl_langinfo(CODESET)
00021  * can be found in
00022  *
00023  *   http://www.cl.cam.ac.uk/~mgk25/unicode.html#activate
00024  *
00025  * Markus.Kuhn@cl.cam.ac.uk -- 2002-03-11
00026  * Permission to use, copy, modify, and distribute this software
00027  * for any purpose and without fee is hereby granted. The author
00028  * disclaims all warranties with regard to this software.
00029  *
00030  * Latest version:
00031  *
00032  *   http://www.cl.cam.ac.uk/~mgk25/ucs/langinfo.c
00033  */
00034 
00035 #include <stdlib.h>
00036 #include <string.h>
00037 #if defined _WIN32 || defined __CYGWIN__
00038 #include <windows.h>
00039 #if defined _WIN32
00040 #define strncasecmp strnicmp
00041 #endif
00042 #endif
00043 #ifdef HAVE_LANGINFO_H
00044 #include "langinfo.h"
00045 #endif
00046 
00047 #define C_CODESET "US-ASCII"     /* Return this as the encoding of the
00048                                   * C/POSIX locale. Could as well one day
00049                                   * become "UTF-8". */
00050 
00051 #if defined _WIN32 || defined __CYGWIN__
00052 #define JA_CODESET "Windows-31J"
00053 #else
00054 #define JA_CODESET "EUC-JP"
00055 #endif
00056 
00057 #define digit(x) ((x) >= '0' && (x) <= '9')
00058 #define strstart(s, n) (strncasecmp(s, n, strlen(n)) == 0)
00059 
00060 static char buf[16];
00061 
00062 const char *
00063 nl_langinfo_codeset(void)
00064 {
00065   const char *l, *p;
00066   int n;
00067   
00068   if (((l = getenv("LC_ALL"))   && *l) ||
00069       ((l = getenv("LC_CTYPE")) && *l) ||
00070       ((l = getenv("LANG"))     && *l)) {
00071     /* check standardized locales */
00072     if (!strcmp(l, "C") || !strcmp(l, "POSIX"))
00073       return C_CODESET;
00074     /* check for encoding name fragment */
00075     p = strchr(l, '.');
00076     if (!p++) p = l;
00077     if (strstart(p, "UTF"))
00078         return "UTF-8";
00079     if ((n = 5, strstart(p, "8859-")) || (n = 9, strstart(p, "ISO-8859-"))) {
00080       if (digit(p[n])) {
00081         p += n;
00082         memcpy(buf, "ISO-8859-\0\0", 12);
00083         buf[9] = *p++;
00084         if (digit(*p)) buf[10] = *p++;
00085         return buf;
00086       }
00087     }
00088     if (strstart(p, "KOI8-R")) return "KOI8-R";
00089     if (strstart(p, "KOI8-U")) return "KOI8-U";
00090     if (strstart(p, "620")) return "TIS-620";
00091     if (strstart(p, "2312")) return "GB2312";
00092     if (strstart(p, "HKSCS")) return "Big5HKSCS";   /* no MIME charset */
00093     if (strstart(p, "BIG5")) return "Big5";
00094     if (strstart(p, "GBK")) return "GBK";           /* no MIME charset */
00095     if (strstart(p, "18030")) return "GB18030";     /* no MIME charset */
00096     if (strstart(p, "Shift_JIS") || strstart(p, "SJIS")) return "Windows-31J";
00097     /* check for conclusive modifier */
00098     if (strstart(p, "euro")) return "ISO-8859-15";
00099     /* check for language (and perhaps country) codes */
00100     if (strstart(l, "zh_TW")) return "Big5";
00101     if (strstart(l, "zh_HK")) return "Big5HKSCS";   /* no MIME charset */
00102     if (strstart(l, "zh")) return "GB2312";
00103     if (strstart(l, "ja")) return JA_CODESET;
00104     if (strstart(l, "ko")) return "EUC-KR";
00105     if (strstart(l, "ru")) return "KOI8-R";
00106     if (strstart(l, "uk")) return "KOI8-U";
00107     if (strstart(l, "pl") || strstart(l, "hr") ||
00108         strstart(l, "hu") || strstart(l, "cs") ||
00109         strstart(l, "sk") || strstart(l, "sl")) return "ISO-8859-2";
00110     if (strstart(l, "eo") || strstart(l, "mt")) return "ISO-8859-3";
00111     if (strstart(l, "el")) return "ISO-8859-7";
00112     if (strstart(l, "he")) return "ISO-8859-8";
00113     if (strstart(l, "tr")) return "ISO-8859-9";
00114     if (strstart(l, "th")) return "TIS-620";      /* or ISO-8859-11 */
00115     if (strstart(l, "lt")) return "ISO-8859-13";
00116     if (strstart(l, "cy")) return "ISO-8859-14";
00117     if (strstart(l, "ro")) return "ISO-8859-2";   /* or ISO-8859-16 */
00118     if (strstart(l, "am") || strstart(l, "vi")) return "UTF-8";
00119     /* Send me further rules if you like, but don't forget that we are
00120      * *only* interested in locale naming conventions on platforms
00121      * that do not already provide an nl_langinfo(CODESET) implementation. */
00122   }
00123   return NULL;
00124 }
00125 
00126 #ifdef HAVE_LANGINFO_H
00127 char *nl_langinfo(nl_item item)
00128 {
00129   const char *codeset;
00130   if (item != CODESET)
00131     return NULL;
00132   codeset = nl_langinfo_codeset();
00133   if (!codeset) codeset = C_CODESET;
00134   return (char *)codeset;
00135 }
00136 #endif
00137 
00138 /* For a demo, compile with "gcc -W -Wall -o langinfo -D TEST langinfo.c" */
00139 
00140 #ifdef TEST
00141 #include <stdio.h>
00142 int main()
00143 {
00144   printf("%s\n", nl_langinfo(CODESET));
00145   return 0;
00146 }
00147 #endif
00148 

Generated on Wed Sep 8 2010 09:55:59 for Ruby by  doxygen 1.7.1