2 * The Single Unix Specification function nl_langinfo(CODESET)
3 * returns the name of the encoding used by the currently selected
6 * http://www.opengroup.org/onlinepubs/7908799/xsh/langinfo.h.html
8 * Unfortunately the encoding names are not yet standardized.
9 * This function knows about the encoding names used on many
10 * different systems and converts them where possible into
11 * the corresponding MIME charset name registered in
13 * http://www.iana.org/assignments/character-sets
15 * Please extend it as needed and suggest improvements to the author.
17 * Markus.Kuhn@cl.cam.ac.uk -- 2002-03-11
18 * Permission to use, copy, modify, and distribute this software
19 * for any purpose and without fee is hereby granted. The author
20 * disclaims all warranties with regard to this software.
24 * http://www.cl.cam.ac.uk/~mgk25/ucs/norm_charmap.c
27 #include "config.h" /* import AC_C_CONST effects */
28 #include "norm_charmap.h"
29 #include "fetchmail.h"
39 #define digit(x) ((x) >= '0' && (x) <= '9')
43 const char *norm_charmap(const char *name)
50 /* Many need no remapping, but they are listed here so you
51 * can see what output to expect, and modify for your needs
53 if (!strcmp(name, "UTF-8"))
55 if (!strcmp(name, "EUC-JP"))
57 if (!strcmp(name, "EUC-KR"))
59 if (!strcmp(name, "EUC-TW"))
61 if (!strcmp(name, "KOI8-R"))
63 if (!strcmp(name, "KOI8-U"))
65 if (!strcmp(name, "GBK"))
67 if (!strcmp(name, "GB2312"))
69 if (!strcmp(name, "GB18030"))
71 if (!strcmp(name, "VSCII"))
74 /* ASCII comes in many names */
75 if (!strcmp(name, "ASCII") ||
76 !strcmp(name, "US-ASCII") ||
77 !strcmp(name, "ANSI_X3.4-1968") ||
78 !strcmp(name, "646") ||
79 !strcmp(name, "ISO646") ||
80 !strcmp(name, "ISO_646.IRV"))
83 /* ISO 8859 will be converted to "ISO-8859-x" */
84 if ((p = strstr(name, "8859-"))) {
85 memcpy(buf, "ISO-8859-\0\0", 12);
89 if (digit(*p)) buf[10] = *p;
94 /* Windows code pages will be converted to "WINDOWS-12xx" */
95 if ((p = strstr(name, "CP12"))) {
96 memcpy(buf, "WINDOWS-12\0\0", 13);
100 if (digit(*p)) buf[11] = *p;
105 /* TIS-620 comes in at least the following two forms */
106 if (!strcmp(name, "TIS-620") ||
107 !strcmp(name, "TIS620.2533"))
108 return "ISO-8859-11";
110 /* For some, uppercase/lowercase might differ */
111 if (!strcmp(name, "Big5") || !strcmp(name, "BIG5"))
113 if (!strcmp(name, "Big5HKSCS") || !strcmp(name, "BIG5HKSCS"))
116 /* I don't know of any implementation of nl_langinfo(CODESET) out
117 * there that returns anything else (and I'm not even certain all of
118 * the above occur in the wild), but just in case, as a fallback,
119 * return the unmodified name. */
121 printf("**** Unknown encoding name '%s'!\n", name);
126 /* For a demo, compile with
128 * gcc -W -Wall -o norm_charmap -D TEST norm_charmap.c
130 * and then test it on all available locales with
132 * for i in `locale -a` ; do printf "$i: " ; LC_ALL=$i ./norm_charmap ; done
136 int main(int argc, char **argv)
142 setlocale(LC_CTYPE, "");
143 s = nl_langinfo(CODESET);
145 printf("%s -> %s\n", s, norm_charmap(s));