static void
locale_parse_lang (const char *locale)
{
char *codeset, *lang;
if ((codeset = strchr (locale, '.')))
lang = g_strndup (locale, codeset - locale);
else
lang = g_strdup (locale);
/* validate the language */
if (strlen (lang) >= 2) {
if (lang[2] == '-' || lang[2] == '_') {
/* canonicalise the lang */
e_strdown (lang);
/* validate the country code */
if (strlen (lang + 3) > 2) {
/* invalid country code */
lang[2] = '\0';
} else {
lang[2] = '-';
e_strup (lang + 3);
}
} else if (lang[2] != '\0') {
/* invalid language */
g_free (lang);
lang = NULL;
}
locale_lang = lang;
} else {
/* invalid language */
locale_lang = NULL;
g_free (lang);
}
}
const char *e_iconv_charset_name(const char *charset)
{
char *name, *ret, *tmp;
if (charset == NULL)
return NULL;
/* The current longest charsetname in
http://www.iana.org/assignments/character-sets is 45, so 80
should be more than enough to avoid SIGSEVS caused by
specially huge malformed charset names */
if (strlen (charset) > 80)
return NULL;
name = g_alloca (strlen (charset) + 1);
strcpy (name, charset);
e_strdown (name);
e_iconv_init(TRUE);
ret = g_hash_table_lookup(iconv_charsets, name);
if (ret != NULL) {
UNLOCK();
return ret;
}
/* Unknown, try canonicalise some basic charset types to something that should work */
if (strncmp(name, "iso", 3) == 0) {
/* Convert iso-nnnn-n or isonnnn-n or iso_nnnn-n to iso-nnnn-n or isonnnn-n */
int iso, codepage;
char *p;
tmp = name + 3;
if (*tmp == '-' || *tmp == '_')
tmp++;
iso = strtoul (tmp, &p, 10);
if (iso == 10646) {
/* they all become ICONV_10646 */
ret = g_strdup (ICONV_10646);
} else {
tmp = p;
if (*tmp == '-' || *tmp == '_')
tmp++;
codepage = strtoul (tmp, &p, 10);
if (p > tmp) {
/* codepage is numeric */
#ifdef __aix__
if (codepage == 13)
ret = g_strdup ("IBM-921");
else
#endif /* __aix__ */
ret = g_strdup_printf (ICONV_ISO_D_FORMAT, iso, codepage);
} else {
/* codepage is a string - probably iso-2022-jp or something */
ret = g_strdup_printf (ICONV_ISO_S_FORMAT, iso, p);
}
}
} else if (strncmp(name, "windows-", 8) == 0) {
/* Convert windows-nnnnn or windows-cpnnnnn to cpnnnn */
tmp = name+8;
if (!strncmp(tmp, "cp", 2))
tmp+=2;
ret = g_strdup_printf("CP%s", tmp);
} else if (strncmp(name, "microsoft-", 10) == 0) {
/* Convert microsoft-nnnnn or microsoft-cpnnnnn to cpnnnn */
tmp = name+10;
if (!strncmp(tmp, "cp", 2))
tmp+=2;
ret = g_strdup_printf("CP%s", tmp);
} else {
/* Just assume its ok enough as is, case and all */
ret = g_strdup(charset);
}
g_hash_table_insert(iconv_charsets, g_strdup(name), ret);
UNLOCK();
return ret;
}
/* NOTE: Owns the lock on return if keep is TRUE ! */
static void
e_iconv_init(int keep)
{
char *from, *to, *locale;
int i;
LOCK();
if (iconv_charsets != NULL) {
if (!keep)
UNLOCK();
return;
}
iconv_charsets = g_hash_table_new(g_str_hash, g_str_equal);
for (i = 0; known_iconv_charsets[i].charset != NULL; i++) {
from = g_strdup(known_iconv_charsets[i].charset);
to = g_strdup(known_iconv_charsets[i].iconv_name);
e_strdown (from);
g_hash_table_insert(iconv_charsets, from, to);
}
e_dlist_init(&iconv_cache_list);
iconv_cache = g_hash_table_new(g_str_hash, g_str_equal);
iconv_cache_open = g_hash_table_new(NULL, NULL);
#ifndef G_OS_WIN32
locale = setlocale (LC_ALL, NULL);
#else
locale = g_win32_getlocale ();
#endif
if (!locale || !strcmp (locale, "C") || !strcmp (locale, "POSIX")) {
/* The locale "C" or "POSIX" is a portable locale; its
* LC_CTYPE part corresponds to the 7-bit ASCII character
* set.
*/
locale_charset = NULL;
locale_lang = NULL;
} else {
#ifdef G_OS_WIN32
g_get_charset (&locale_charset);
locale_charset = g_strdup (locale_charset);
e_strdown (locale_charset);
#else
#ifdef HAVE_CODESET
locale_charset = g_strdup (nl_langinfo (CODESET));
e_strdown (locale_charset);
#else
/* A locale name is typically of the form language[_terri-
* tory][.codeset][@modifier], where language is an ISO 639
* language code, territory is an ISO 3166 country code, and
* codeset is a character set or encoding identifier like
* ISO-8859-1 or UTF-8.
*/
char *codeset, *p;
codeset = strchr (locale, '.');
if (codeset) {
codeset++;
/* ; is a hack for debian systems and / is a hack for Solaris systems */
for (p = codeset; *p && !strchr ("@;/", *p); p++);
locale_charset = g_strndup (codeset, p - codeset);
e_strdown (locale_charset);
} else {
/* charset unknown */
locale_charset = NULL;
}
#endif
#endif /* !G_OS_WIN32 */
/* parse the locale lang */
locale_parse_lang (locale);
}
#ifdef G_OS_WIN32
g_free (locale);
#endif
if (!keep)
UNLOCK();
}
const gchar *
camel_iconv_charset_name (const gchar *charset)
{
gchar *name, *ret, *tmp;
if (charset == NULL)
return NULL;
name = g_alloca (strlen (charset) + 1);
strcpy (name, charset);
e_strdown (name);
iconv_init (TRUE);
ret = g_hash_table_lookup (iconv_charsets, name);
if (ret != NULL) {
G_UNLOCK (iconv);
return ret;
}
/* Unknown, try canonicalise some basic charset types to something that should work */
if (strncmp (name, "iso", 3) == 0) {
/* Convert iso-nnnn-n or isonnnn-n or iso_nnnn-n to iso-nnnn-n or isonnnn-n */
gint iso, codepage;
gchar *p;
tmp = name + 3;
if (*tmp == '-' || *tmp == '_')
tmp++;
iso = strtoul (tmp, &p, 10);
if (iso == 10646) {
/* they all become ICONV_10646 */
ret = g_strdup (ICONV_10646);
} else {
tmp = p;
if (*tmp == '-' || *tmp == '_')
tmp++;
codepage = strtoul (tmp, &p, 10);
if (p > tmp) {
/* codepage is numeric */
#ifdef __aix__
if (codepage == 13)
ret = g_strdup ("IBM-921");
else
#endif /* __aix__ */
ret = g_strdup_printf (ICONV_ISO_D_FORMAT, iso, codepage);
} else {
/* codepage is a string - probably iso-2022-jp or something */
ret = g_strdup_printf (ICONV_ISO_S_FORMAT, iso, p);
}
}
} else if (strncmp (name, "windows-", 8) == 0) {
/* Convert windows-nnnnn or windows-cpnnnnn to cpnnnn */
tmp = name + 8;
if (!strncmp (tmp, "cp", 2))
tmp+=2;
ret = g_strdup_printf ("CP%s", tmp);
} else if (strncmp (name, "microsoft-", 10) == 0) {
/* Convert microsoft-nnnnn or microsoft-cpnnnnn to cpnnnn */
tmp = name + 10;
if (!strncmp (tmp, "cp", 2))
tmp+=2;
ret = g_strdup_printf ("CP%s", tmp);
} else {
/* Just assume its ok enough as is, case and all */
ret = g_strdup (charset);
}
g_hash_table_insert (iconv_charsets, g_strdup (name), ret);
G_UNLOCK (iconv);
return ret;
}
