#include <errno.h>
#include <lib-lib/mem.h>
+#include <lib-lib/ascii.h>
+#include <lib-lib/str.h>
+#include <lib-lib/macros.h>
#include "mutt.h"
#include "charset.h"
-#include "ascii.h"
-
-#include "lib/mem.h"
-#include "lib/intl.h"
-#include "lib/str.h"
#ifndef EILSEQ
# define EILSEQ EINVAL
#endif
+char *Charset;
+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
*/
static struct {
- const char *key;
- const char *pref;
+ 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}
+ {"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"},
+ {"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"},
+ {"l6", "iso-8859-10"},
+ {"csISOLatin6", "iso-8859-10"},
+ {"csKOI8r", "koi8-r"},
+ {"MS_Kanji", "Shift_JIS"},
+ {"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"},
+ {"l7", "iso-8859-13"},
+ {"iso_8859-14", "iso-8859-14"},
+ {"latin8", "iso-8859-14"},
+ {"l8", "iso-8859-14"},
+ {"iso_8859-15", "iso-8859-15"},
+ {"latin9", "iso-8859-15"},
+ {"latin0", "iso-8859-15"},
+ {"iso_8859-16", "iso-8859-16"},
+ {"latin10", "iso-8859-16"},
+ {"646", "us-ascii"},
+ {"eucJP", "euc-jp"},
+ {"PCK", "Shift_JIS"},
+ {"ko_KR-euc", "euc-kr"},
+ {"zh_TW-big5", "big5"},
+ {"sjis", "Shift_JIS"},
+ {"euc-jp-ms", "eucJP-ms"},
+ {NULL, NULL}
};
#ifdef HAVE_LANGINFO_CODESET
-# include <langinfo.h>
-
+# include <langinfo.h>
+#endif
void mutt_set_langinfo_charset (void)
{
- char buff[LONG_STRING];
- char buff2[LONG_STRING];
-
- strfcpy (buff, nl_langinfo (CODESET), sizeof (buff));
- mutt_canonical_charset (buff2, sizeof (buff2), buff);
-
- /* finally, set $charset */
- if (!(Charset = str_dup (buff2)))
- Charset = str_dup ("iso-8859-1");
-}
+#ifdef HAVE_LANGINFO_CODESET
+ char buff[LONG_STRING];
+ char buff2[LONG_STRING];
-#else
+ m_strcpy(buff, sizeof(buff), nl_langinfo(CODESET));
+ mutt_canonical_charset (buff2, sizeof (buff2), buff);
-void mutt_set_langinfo_charset (void)
-{
- Charset = str_dup ("iso-8859-1");
+ /* finally, set $charset */
+ if (!(Charset = m_strdup(buff2)))
+#endif
+ Charset = m_strdup("iso-8859-1");
}
-#endif
-void mutt_canonical_charset (char *dest, size_t dlen, const char *name)
+void mutt_canonical_charset(char *dest, ssize_t dlen, const char *name)
{
- size_t i;
+ ssize_t i;
char *p;
char scratch[LONG_STRING];
else if (!ascii_strncasecmp (name, "iso8859-", 8))
snprintf (scratch, sizeof (scratch), "iso_8859-%s", name + 8);
else
- strfcpy (scratch, NONULL (name), sizeof (scratch));
+ m_strcpy(scratch, sizeof(scratch), NONULL(name));
for (i = 0; PreferredMIMENames[i].key; i++)
if (!ascii_strcasecmp (scratch, PreferredMIMENames[i].key) ||
- !str_casecmp (scratch, PreferredMIMENames[i].key)) {
- strfcpy (dest, PreferredMIMENames[i].pref, dlen);
+ !m_strcasecmp(scratch, PreferredMIMENames[i].key)) {
+ m_strcpy(dest, dlen, PreferredMIMENames[i].pref);
return;
}
- strfcpy (dest, scratch, dlen);
+ m_strcpy(dest, dlen, scratch);
/* for cosmetics' sake, transform to lowercase. */
for (p = dest; *p; p++)
* if you're supplying an outrepl, the target charset should be.
*/
-size_t mutt_iconv (iconv_t cd, const char **inbuf, size_t * inbytesleft,
- char **outbuf, size_t * outbytesleft,
+ssize_t mutt_iconv(iconv_t cd, const char **inbuf, ssize_t *inbytesleft,
+ char **outbuf, ssize_t *outbytesleft,
const char **inrepls, const char *outrepl)
{
- size_t ret = 0, ret1;
+ ssize_t ret = 0, ret1;
const char *ib = *inbuf;
- size_t ibl = *inbytesleft;
+ ssize_t ibl = *inbytesleft;
char *ob = *outbuf;
- size_t obl = *outbytesleft;
+ ssize_t obl = *outbytesleft;
for (;;) {
ret1 = my_iconv(cd, &ib, &ibl, &ob, &obl);
- if (ret1 != (size_t) - 1)
+ if (ret1 != -1)
ret += ret1;
if (ibl && obl && errno == EILSEQ) {
if (inrepls) {
for (t = inrepls; *t; t++) {
const char *ib1 = *t;
- size_t ibl1 = str_len (*t);
+ ssize_t ibl1 = m_strlen(*t);
char *ob1 = ob;
- size_t obl1 = obl;
+ ssize_t obl1 = obl;
my_iconv(cd, &ib1, &ibl1, &ob1, &obl1);
if (!ibl1) {
outrepl = "?";
my_iconv(cd, 0, 0, &ob, &obl);
if (obl) {
- int n = str_len (outrepl);
+ ssize_t n = m_strlen(outrepl);
if (n > obl) {
outrepl = "?";
int len;
const char *ib;
char *buf, *ob;
- size_t ibl, obl;
+ ssize_t ibl, obl;
const char **inrepls = NULL;
const char *outrepl = NULL;
else
outrepl = "?";
- len = str_len (s);
+ len = m_strlen(s);
ib = s, ibl = len + 1;
obl = MB_LEN_MAX * ibl;
ob = buf = xmalloc(obl + 1);
p_delete(ps);
*ps = buf;
-
- str_adjust (ps);
return 0;
}
else
char *p;
char *ob;
char *ib;
- size_t ibl;
+ ssize_t ibl;
const char **inrepls;
};
fc->inrepls = mutt_is_utf8 (to) ? repls : repls + 1;
}
else
- fc = p_new(struct fgetconv_not, 1);
+ fc = p_new(struct fgetconv_s, 1);
fc->file = file;
fc->cd = cd;
return (FGETCONV *) fc;
}
-char *fgetconvs (char *buf, size_t l, FGETCONV * _fc)
+char *fgetconvs (char *buf, ssize_t l, FGETCONV * _fc)
{
int c;
- size_t r;
+ ssize_t r;
for (r = 0; r + 1 < l;) {
if ((c = fgetconv (_fc)) == EOF)
/* Try to convert some more */
fc->p = fc->ob = fc->bufo;
if (fc->ibl) {
- size_t obl = sizeof (fc->bufo);
+ ssize_t obl = ssizeof(fc->bufo);
my_iconv(fc->cd, (const char **) &fc->ib, &fc->ibl, &fc->ob, &obl);
if (fc->p < fc->ob)
/* Try harder this time to convert some */
if (fc->ibl) {
- size_t obl = sizeof (fc->bufo);
+ ssize_t obl = ssizeof(fc->bufo);
mutt_iconv (fc->cd, (const char **) &fc->ib, &fc->ibl, &fc->ob,
&obl, fc->inrepls, 0);
const char *c, *c1;
c = charset;
- if (!str_len (c))
+ if (!m_strlen(c))
return "us-ascii";
if (!(c1 = strchr (c, ':')))
return ((char*) charset);
- strfcpy (fcharset, c, c1 - c + 1);
+ m_strcpy(fcharset, c1 - c + 1, c);
return fcharset;
}
-static size_t convert_string (const char *f, size_t flen,
+static ssize_t convert_string (const char *f, ssize_t flen,
const char *from, const char *to,
- char **t, size_t * tlen)
+ char **t, ssize_t * tlen)
{
iconv_t cd;
char *buf, *ob;
- size_t obl, n;
+ ssize_t obl;
+ ssize_t n;
int e;
cd = mutt_iconv_open (to, from, 0);
if (cd == (iconv_t) (-1))
- return (size_t) (-1);
+ return -1;
obl = 4 * flen + 1;
ob = buf = xmalloc(obl);
n = my_iconv(cd, &f, &flen, &ob, &obl);
- if (n == (size_t) (-1) || my_iconv(cd, 0, 0, &ob, &obl) == (size_t) (-1)) {
+ if (n < 0 || my_iconv(cd, 0, 0, &ob, &obl) < 0) {
e = errno;
p_delete(&buf);
iconv_close (cd);
errno = e;
- return (size_t) (-1);
+ return -1;
}
*ob = '\0';
*tlen = ob - buf;
- mem_realloc (&buf, ob - buf + 1);
+ p_realloc(&buf, ob - buf + 1);
*t = buf;
iconv_close (cd);
char *u = *ps;
char *s = NULL;
char *fromcode;
- size_t m, n;
- size_t ulen = str_len (*ps);
- size_t slen;
+ ssize_t m, n;
+ ssize_t ulen = m_strlen(*ps);
+ ssize_t slen;
if (!u || !*u)
return 0;
c1 = strchr (c, ':');
- n = c1 ? c1 - c : str_len (c);
+ n = c1 ? c1 - c : m_strlen(c);
if (!n)
continue;
fromcode = p_dupstr(c, n);
m = convert_string (u, ulen, fromcode, Charset, &s, &slen);
p_delete(&fromcode);
- if (m != (size_t) (-1)) {
+ if (m != -1) {
p_delete(ps);
*ps = s;
return 0;
}
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)
+{
+ return Charset_is_utf8 ? 0xfffd : '?';
+}