#include <dirent.h>
#include <unistd.h>
#include <errno.h>
+#ifdef HAVE_LANGINFO_CODESET
+# include <langinfo.h>
+#endif
#include <lib-lib/mem.h>
#include <lib-lib/ascii.h>
#include "mutt.h"
#include "charset.h"
-
#ifndef EILSEQ
-# define EILSEQ EINVAL
+# define EILSEQ EINVAL
#endif
+char *Charset = NULL;
int Charset_is_utf8 = 0;
-/*
- * The following list has been created manually from the data under:
- * http://www.isi.edu/in-notes/iana/assignments/character-sets
- * Last update: 2000-09-07
- *
- * Note that it includes only the subset of character sets for which
- * a preferred MIME name is given.
- */
-
-static struct {
- const char *key;
- const char *pref;
-} PreferredMIMENames[] = {
- {
- "ansi_x3.4-1968", "us-ascii"}, {
- "iso-ir-6", "us-ascii"}, {
- "iso_646.irv:1991", "us-ascii"}, {
- "ascii", "us-ascii"}, {
- "iso646-us", "us-ascii"}, {
- "us", "us-ascii"}, {
- "ibm367", "us-ascii"}, {
- "cp367", "us-ascii"}, {
- "csASCII", "us-ascii"}, {
- "csISO2022KR", "iso-2022-kr"}, {
- "csEUCKR", "euc-kr"}, {
- "csISO2022JP", "iso-2022-jp"}, {
- "csISO2022JP2", "iso-2022-jp-2"}, {
- "ISO_8859-1:1987", "iso-8859-1"}, {
- "iso-ir-100", "iso-8859-1"}, {
- "iso_8859-1", "iso-8859-1"}, {
- "latin1", "iso-8859-1"}, {
- "l1", "iso-8859-1"}, {
- "IBM819", "iso-8859-1"}, {
- "CP819", "iso-8859-1"}, {
- "csISOLatin1", "iso-8859-1"}, {
- "ISO_8859-2:1987", "iso-8859-2"}, {
- "iso-ir-101", "iso-8859-2"}, {
- "iso_8859-2", "iso-8859-2"}, {
- "latin2", "iso-8859-2"}, {
- "l2", "iso-8859-2"}, {
- "csISOLatin2", "iso-8859-2"}, {
- "ISO_8859-3:1988", "iso-8859-3"}, {
- "iso-ir-109", "iso-8859-3"}, {
- "ISO_8859-3", "iso-8859-3"}, {
- "latin3", "iso-8859-3"}, {
- "l3", "iso-8859-3"}, {
- "csISOLatin3", "iso-8859-3"}, {
- "ISO_8859-4:1988", "iso-8859-4"}, {
- "iso-ir-110", "iso-8859-4"}, {
- "ISO_8859-4", "iso-8859-4"}, {
- "latin4", "iso-8859-4"}, {
- "l4", "iso-8859-4"}, {
- "csISOLatin4", "iso-8859-4"}, {
- "ISO_8859-6:1987", "iso-8859-6"}, {
- "iso-ir-127", "iso-8859-6"}, {
- "iso_8859-6", "iso-8859-6"}, {
- "ECMA-114", "iso-8859-6"}, {
- "ASMO-708", "iso-8859-6"}, {
- "arabic", "iso-8859-6"}, {
- "csISOLatinArabic", "iso-8859-6"}, {
- "ISO_8859-7:1987", "iso-8859-7"}, {
- "iso-ir-126", "iso-8859-7"}, {
- "ISO_8859-7", "iso-8859-7"}, {
- "ELOT_928", "iso-8859-7"}, {
- "ECMA-118", "iso-8859-7"}, {
- "greek", "iso-8859-7"}, {
- "greek8", "iso-8859-7"}, {
- "csISOLatinGreek", "iso-8859-7"}, {
- "ISO_8859-8:1988", "iso-8859-8"}, {
- "iso-ir-138", "iso-8859-8"}, {
- "ISO_8859-8", "iso-8859-8"}, {
- "hebrew", "iso-8859-8"}, {
- "csISOLatinHebrew", "iso-8859-8"}, {
- "ISO_8859-5:1988", "iso-8859-5"}, {
- "iso-ir-144", "iso-8859-5"}, {
- "ISO_8859-5", "iso-8859-5"}, {
- "cyrillic", "iso-8859-5"}, {
- "csISOLatinCyrillic", "iso8859-5"}, {
- "ISO_8859-9:1989", "iso-8859-9"}, {
- "iso-ir-148", "iso-8859-9"}, {
- "ISO_8859-9", "iso-8859-9"}, {
- "latin5", "iso-8859-9"}, /* this is not a bug */
- {
- "l5", "iso-8859-9"}, {
- "csISOLatin5", "iso-8859-9"}, {
- "ISO_8859-10:1992", "iso-8859-10"}, {
- "iso-ir-157", "iso-8859-10"}, {
- "latin6", "iso-8859-10"}, /* this is not a bug */
- {
- "l6", "iso-8859-10"}, {
- "csISOLatin6", "iso-8859-10"}, {
- "csKOI8r", "koi8-r"}, {
- "MS_Kanji", "Shift_JIS"}, /* Note the underscore! */
- {
- "csShiftJis", "Shift_JIS"}, {
- "Extended_UNIX_Code_Packed_Format_for_Japanese", "EUC-JP"}, {
- "csEUCPkdFmtJapanese", "EUC-JP"}, {
- "csGB2312", "gb2312"}, {
- "csbig5", "big5"},
- /*
- * End of official brain damage. What follows has been taken
- * from glibc's localedata files.
- */
- {
- "iso_8859-13", "iso-8859-13"}, {
- "iso-ir-179", "iso-8859-13"}, {
- "latin7", "iso-8859-13"}, /* this is not a bug */
- {
- "l7", "iso-8859-13"}, {
- "iso_8859-14", "iso-8859-14"}, {
- "latin8", "iso-8859-14"}, /* this is not a bug */
- {
- "l8", "iso-8859-14"}, {
- "iso_8859-15", "iso-8859-15"}, {
- "latin9", "iso-8859-15"}, /* this is not a bug */
- /* Suggested by Ionel Mugurel Ciobica <tgakic@sg10.chem.tue.nl> */
- {
- "latin0", "iso-8859-15"}, /* this is not a bug */
- {
- "iso_8859-16", "iso-8859-16"}, {
- "latin10", "iso-8859-16"}, /* this is not a bug */
- /*
- * David Champion <dgc@uchicago.edu> has observed this with
- * nl_langinfo under SunOS 5.8.
- */
- {
- "646", "us-ascii"},
- /*
- * http://www.sun.com/software/white-papers/wp-unicode/
- */
- {
- "eucJP", "euc-jp"}, {
- "PCK", "Shift_JIS"}, {
- "ko_KR-euc", "euc-kr"}, {
- "zh_TW-big5", "big5"},
- /* seems to be common on some systems */
- {
- "sjis", "Shift_JIS"}, {
- "euc-jp-ms", "eucJP-ms"},
- /*
- * If you happen to encounter system-specific brain-damage with
- * respect to character set naming, please add it above this
- * comment, and submit a patch to <mutt-dev@mutt.org>.
- */
- /* End of aliases. Please keep this line last. */
- {
- NULL, NULL}
-};
-
+void charset_initialize(void)
+{
#ifdef HAVE_LANGINFO_CODESET
-# include <langinfo.h>
+ char buff[LONG_STRING];
+ char buff2[LONG_STRING];
+ m_strcpy(buff, sizeof(buff), nl_langinfo(CODESET));
+ mutt_canonical_charset(buff2, sizeof(buff2), buff);
-void mutt_set_langinfo_charset (void)
-{
- char buff[LONG_STRING];
- char buff2[LONG_STRING];
-
- m_strcpy(buff, sizeof(buff), nl_langinfo(CODESET));
- mutt_canonical_charset (buff2, sizeof (buff2), buff);
+ /* finally, set $charset */
+ if (!m_strisempty(buff2)) {
+ m_strreplace(&Charset, buff2);
+ } else
+#endif
+ m_strreplace(&Charset, "iso-8859-1");
- /* finally, set $charset */
- if (!(Charset = m_strdup(buff2)))
- Charset = m_strdup("iso-8859-1");
+ Charset_is_utf8 = !strcmp(Charset, "utf-8");
+#ifdef HAVE_BIND_TEXTDOMAIN_CODESET
+ bind_textdomain_codeset(PACKAGE, Charset);
+#endif
}
-#else
-void mutt_set_langinfo_charset (void)
+#include "charset.gperf"
+
+void mutt_canonical_charset(char *dest, ssize_t dlen, const char *name)
{
- Charset = m_strdup("iso-8859-1");
+ const struct cset_pair *cp;
+ char scratch[LONG_STRING];
+ const char *p;
+ int i = 0;
+
+ // canonize name: only keep a-z0-9 and dots, put into lowercase
+ for (p = name; *p && *p != ':' && i < ssizeof(scratch) - 1; p++) {
+ if (isalnum(*p) || *p== '.') {
+ scratch[i++] = tolower((unsigned char)*p);
+ }
+ }
+ scratch[i] = '\0';
+
+ cp = mutt_canonical_charset_aux(scratch, strlen(scratch));
+ if (cp) {
+ m_strcpy(dest, dlen, cp->pref);
+ } else {
+ m_strcpy(dest, dlen, name);
+ m_strtolower(dest);
+ }
}
-#endif
-
-void mutt_canonical_charset (char *dest, ssize_t dlen, const char *name)
+static int mutt_chscmp(const char *s, const char *chs)
{
- ssize_t i;
- char *p;
- char scratch[LONG_STRING];
-
- /* catch some common iso-8859-something misspellings */
- if (!ascii_strncasecmp (name, "8859", 4) && name[4] != '-')
- snprintf (scratch, sizeof (scratch), "iso-8859-%s", name + 4);
- else if (!ascii_strncasecmp (name, "8859-", 5))
- snprintf (scratch, sizeof (scratch), "iso-8859-%s", name + 5);
- else if (!ascii_strncasecmp (name, "iso8859", 7) && name[7] != '-')
- snprintf (scratch, sizeof (scratch), "iso_8859-%s", name + 7);
- else if (!ascii_strncasecmp (name, "iso8859-", 8))
- snprintf (scratch, sizeof (scratch), "iso_8859-%s", name + 8);
- else
- m_strcpy(scratch, sizeof(scratch), NONULL(name));
-
- for (i = 0; PreferredMIMENames[i].key; i++)
- if (!ascii_strcasecmp (scratch, PreferredMIMENames[i].key) ||
- !m_strcasecmp(scratch, PreferredMIMENames[i].key)) {
- m_strcpy(dest, dlen, PreferredMIMENames[i].pref);
- return;
- }
+ char buffer[STRING];
- m_strcpy(dest, dlen, scratch);
+ if (!s)
+ return 0;
- /* for cosmetics' sake, transform to lowercase. */
- for (p = dest; *p; p++)
- *p = ascii_tolower (*p);
+ mutt_canonical_charset(buffer, sizeof(buffer), s);
+ return !strcmp(buffer, chs);
}
-int mutt_chscmp (const char *s, const char *chs)
+int charset_is_utf8(const char *s)
{
- char buffer[STRING];
-
- if (!s)
- return 0;
+ return mutt_chscmp(s, "utf-8");
+}
- mutt_canonical_charset (buffer, sizeof (buffer), s);
- return !ascii_strcasecmp (buffer, chs);
+int charset_is_us_ascii(const char *s)
+{
+ return mutt_chscmp(s, "us-ascii");
}
const char **inrepls = NULL;
const char *outrepl = NULL;
- if (mutt_is_utf8 (to))
+ if (charset_is_utf8 (to))
outrepl = "\357\277\275";
- else if (mutt_is_utf8 (from))
+ else if (charset_is_utf8 (from))
inrepls = repls;
else
outrepl = "?";
fc->p = fc->ob = fc->bufo;
fc->ib = fc->bufi;
fc->ibl = 0;
- fc->inrepls = mutt_is_utf8 (to) ? repls : repls + 1;
+ fc->inrepls = charset_is_utf8 (to) ? repls : repls + 1;
}
else
fc = p_new(struct fgetconv_s, 1);
return -1;
}
-void mutt_set_charset (char *charset)
-{
- char buffer[STRING];
-
- mutt_canonical_charset (buffer, sizeof (buffer), charset);
-
- Charset_is_utf8 = 0;
- if (!strcmp (buffer, "utf-8"))
- Charset_is_utf8 = 1;
-
-#ifdef HAVE_BIND_TEXTDOMAIN_CODESET
- bind_textdomain_codeset (PACKAGE, buffer);
-#endif
-}
-
-wchar_t replacement_char (void)
+wchar_t replacement_char(void)
{
return Charset_is_utf8 ? 0xfffd : '?';
}