/*
* Sylpheed -- a GTK+ based, lightweight, and fast e-mail client
- * Copyright (C) 1999-2004 Hiroyuki Yamamoto
+ * Copyright (C) 1999-2005 Hiroyuki Yamamoto
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
*/
#ifdef HAVE_CONFIG_H
# include "config.h"
#endif
+#include "defs.h"
+
#include <glib.h>
+#include <glib/gi18n.h>
#include <string.h>
#include <ctype.h>
#include <stdlib.h>
# include <locale.h>
#endif
-#if HAVE_ICONV
-# include <iconv.h>
-#endif
+#include <iconv.h>
-#include "intl.h"
#include "codeconv.h"
#include "unmime.h"
#include "base64.h"
JIS_AUXKANJI
} JISState;
-#define SUBST_CHAR '_'
+#define SUBST_CHAR 0x5f;
#define ESC '\033'
#define iseuckanji(c) \
state = JIS_AUXKANJI; \
}
-void conv_jistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_jistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf);
+static void conv_euctojis(gchar *outbuf, gint outlen, const gchar *inbuf);
+static void conv_sjistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf);
+
+static void conv_jistoutf8(gchar *outbuf, gint outlen, const gchar *inbuf);
+static void conv_sjistoutf8(gchar *outbuf, gint outlen, const gchar *inbuf);
+static void conv_euctoutf8(gchar *outbuf, gint outlen, const gchar *inbuf);
+static void conv_anytoutf8(gchar *outbuf, gint outlen, const gchar *inbuf);
+
+static void conv_utf8toeuc(gchar *outbuf, gint outlen, const gchar *inbuf);
+static void conv_utf8tojis(gchar *outbuf, gint outlen, const gchar *inbuf);
+
+static void conv_unreadable_8bit(gchar *str);
+
+static void conv_jistodisp(gchar *outbuf, gint outlen, const gchar *inbuf);
+static void conv_sjistodisp(gchar *outbuf, gint outlen, const gchar *inbuf);
+static void conv_euctodisp(gchar *outbuf, gint outlen, const gchar *inbuf);
+
+static void conv_anytodisp(gchar *outbuf, gint outlen, const gchar *inbuf);
+static void conv_ustodisp(gchar *outbuf, gint outlen, const gchar *inbuf);
+static void conv_noconv(gchar *outbuf, gint outlen, const gchar *inbuf);
+
+static gboolean strict_mode = FALSE;
+
+void codeconv_set_strict(gboolean mode)
+{
+ strict_mode = mode;
+}
+
+static void conv_jistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
{
const guchar *in = inbuf;
guchar *out = outbuf;
return 1;
}
-void conv_euctojis(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_euctojis(gchar *outbuf, gint outlen, const gchar *inbuf)
{
const guchar *in = inbuf;
guchar *out = outbuf;
JISState state = JIS_ASCII;
while (*in != '\0') {
- if (isascii(*in)) {
+ if (IS_ASCII(*in)) {
K_OUT();
*out++ = *in++;
} else if (iseuckanji(*in)) {
K_OUT();
*out++ = SUBST_CHAR;
in++;
- if (*in != '\0' && !isascii(*in)) {
+ if (*in != '\0' && !IS_ASCII(*in)) {
*out++ = SUBST_CHAR;
in++;
}
} else {
K_OUT();
in++;
- if (*in != '\0' && !isascii(*in)) {
+ if (*in != '\0' && !IS_ASCII(*in)) {
*out++ = SUBST_CHAR;
in++;
}
*out++ = *in++ & 0x7f;
} else {
K_OUT();
- if (*in != '\0' && !isascii(*in)) {
+ if (*in != '\0' && !IS_ASCII(*in)) {
*out++ = SUBST_CHAR;
in++;
- if (*in != '\0' && !isascii(*in)) {
+ if (*in != '\0' && !IS_ASCII(*in)) {
*out++ = SUBST_CHAR;
in++;
}
*out = '\0';
}
-void conv_sjistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_sjistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
{
const guchar *in = inbuf;
guchar *out = outbuf;
while (*in != '\0') {
- if (isascii(*in)) {
+ if (IS_ASCII(*in)) {
*out++ = *in++;
} else if (issjiskanji1(*in)) {
if (issjiskanji2(*(in + 1))) {
} else {
*out++ = SUBST_CHAR;
in++;
- if (*in != '\0' && !isascii(*in)) {
+ if (*in != '\0' && !IS_ASCII(*in)) {
*out++ = SUBST_CHAR;
in++;
}
*out = '\0';
}
-void conv_anytoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_jistoutf8(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- switch (conv_guess_ja_encoding(inbuf)) {
- case C_ISO_2022_JP:
- conv_jistoeuc(outbuf, outlen, inbuf);
- break;
- case C_SHIFT_JIS:
- conv_sjistoeuc(outbuf, outlen, inbuf);
- break;
- default:
- strncpy2(outbuf, inbuf, outlen);
- break;
- }
+ gchar *eucstr;
+
+ Xalloca(eucstr, outlen, return);
+
+ conv_jistoeuc(eucstr, outlen, inbuf);
+ conv_euctoutf8(outbuf, outlen, eucstr);
}
-void conv_anytoutf8(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_sjistoutf8(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- gchar *tmpstr = NULL;
+ gchar *tmpstr;
- switch (conv_guess_ja_encoding(inbuf)) {
- case C_ISO_2022_JP:
- tmpstr = conv_codeset_strdup(inbuf, CS_ISO_2022_JP, CS_UTF_8);
- strncpy2(outbuf, tmpstr, outlen);
- g_free(tmpstr);
- break;
- case C_SHIFT_JIS:
- tmpstr = conv_codeset_strdup(inbuf, CS_SHIFT_JIS, CS_UTF_8);
+ tmpstr = conv_iconv_strdup(inbuf, CS_SHIFT_JIS, CS_UTF_8);
+ if (tmpstr) {
strncpy2(outbuf, tmpstr, outlen);
g_free(tmpstr);
- break;
- case C_EUC_JP:
- tmpstr = conv_codeset_strdup(inbuf, CS_EUC_JP, CS_UTF_8);
+ } else
+ strncpy2(outbuf, inbuf, outlen);
+}
+
+static void conv_euctoutf8(gchar *outbuf, gint outlen, const gchar *inbuf)
+{
+ static iconv_t cd = (iconv_t)-1;
+ static gboolean iconv_ok = TRUE;
+ gchar *tmpstr;
+
+ if (cd == (iconv_t)-1) {
+ if (!iconv_ok) {
+ strncpy2(outbuf, inbuf, outlen);
+ return;
+ }
+ cd = iconv_open(CS_UTF_8, CS_EUC_JP_MS);
+ if (cd == (iconv_t)-1) {
+ cd = iconv_open(CS_UTF_8, CS_EUC_JP);
+ if (cd == (iconv_t)-1) {
+ g_warning("conv_euctoutf8(): %s\n",
+ g_strerror(errno));
+ iconv_ok = FALSE;
+ strncpy2(outbuf, inbuf, outlen);
+ return;
+ }
+ }
+ }
+
+ tmpstr = conv_iconv_strdup_with_cd(inbuf, cd);
+ if (tmpstr) {
strncpy2(outbuf, tmpstr, outlen);
g_free(tmpstr);
- break;
- default:
+ } else
strncpy2(outbuf, inbuf, outlen);
- break;
- }
}
-void conv_anytojis(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_anytoutf8(gchar *outbuf, gint outlen, const gchar *inbuf)
{
switch (conv_guess_ja_encoding(inbuf)) {
+ case C_ISO_2022_JP:
+ conv_jistoutf8(outbuf, outlen, inbuf);
+ break;
+ case C_SHIFT_JIS:
+ conv_sjistoutf8(outbuf, outlen, inbuf);
+ break;
case C_EUC_JP:
- conv_euctojis(outbuf, outlen, inbuf);
+ conv_euctoutf8(outbuf, outlen, inbuf);
break;
default:
strncpy2(outbuf, inbuf, outlen);
}
}
-static gchar valid_eucjp_tbl[][96] = {
- /* 0xa2a0 - 0xa2ff */
- { 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0,
- 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1,
- 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1,
- 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0,
- 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, 0 },
-
- /* 0xa3a0 - 0xa3ff */
- { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0,
- 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0,
- 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0 },
-
- /* 0xa4a0 - 0xa4ff */
- { 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 },
-
- /* 0xa5a0 - 0xa5ff */
- { 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0 },
-
- /* 0xa6a0 - 0xa6ff */
- { 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0,
- 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 },
-
- /* 0xa7a0 - 0xa7ff */
- { 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 },
-
- /* 0xa8a0 - 0xa8ff */
- { 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
- 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
- 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 }
-};
-
-static gboolean isprintableeuckanji(guchar c1, guchar c2)
+static void conv_utf8toeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- if (c1 <= 0xa0 || c1 >= 0xf5)
- return FALSE;
- if (c2 <= 0xa0 || c2 == 0xff)
- return FALSE;
-
- if (c1 >= 0xa9 && c1 <= 0xaf)
- return FALSE;
-
- if (c1 >= 0xa2 && c1 <= 0xa8)
- return (gboolean)valid_eucjp_tbl[c1 - 0xa2][c2 - 0xa0];
-
- if (c1 == 0xcf) {
- if (c2 >= 0xd4 && c2 <= 0xfe)
- return FALSE;
- } else if (c1 == 0xf4) {
- if (c2 >= 0xa7 && c2 <= 0xfe)
- return FALSE;
+ static iconv_t cd = (iconv_t)-1;
+ static gboolean iconv_ok = TRUE;
+ gchar *tmpstr;
+
+ if (cd == (iconv_t)-1) {
+ if (!iconv_ok) {
+ strncpy2(outbuf, inbuf, outlen);
+ return;
+ }
+ cd = iconv_open(CS_EUC_JP_MS, CS_UTF_8);
+ if (cd == (iconv_t)-1) {
+ cd = iconv_open(CS_EUC_JP, CS_UTF_8);
+ if (cd == (iconv_t)-1) {
+ g_warning("conv_utf8toeuc(): %s\n",
+ g_strerror(errno));
+ iconv_ok = FALSE;
+ strncpy2(outbuf, inbuf, outlen);
+ return;
+ }
+ }
}
- return TRUE;
+ tmpstr = conv_iconv_strdup_with_cd(inbuf, cd);
+ if (tmpstr) {
+ strncpy2(outbuf, tmpstr, outlen);
+ g_free(tmpstr);
+ } else
+ strncpy2(outbuf, inbuf, outlen);
}
-void conv_unreadable_eucjp(gchar *str)
+static void conv_utf8tojis(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- register guchar *p = str;
+ gchar *eucstr;
- while (*p != '\0') {
- if (isascii(*p)) {
- /* convert CR+LF -> LF */
- if (*p == '\r' && *(p + 1) == '\n')
- memmove(p, p + 1, strlen(p));
- /* printable 7 bit code */
- p++;
- } else if (iseuckanji(*p)) {
- if (isprintableeuckanji(*p, *(p + 1))) {
- /* printable euc-jp code */
- p += 2;
- } else {
- /* substitute unprintable code */
- *p++ = SUBST_CHAR;
- if (*p != '\0') {
- if (isascii(*p))
- p++;
- else
- *p++ = SUBST_CHAR;
- }
- }
- } else if (iseuchwkana1(*p)) {
- if (iseuchwkana2(*(p + 1)))
- /* euc-jp hankaku kana */
- p += 2;
- else
- *p++ = SUBST_CHAR;
- } else if (iseucaux(*p)) {
- if (iseuckanji(*(p + 1)) && iseuckanji(*(p + 2))) {
- /* auxiliary kanji */
- p += 3;
- } else
- *p++ = SUBST_CHAR;
- } else
- /* substitute unprintable 1 byte code */
- *p++ = SUBST_CHAR;
- }
-}
-
-void conv_unreadable_8bit(gchar *str)
-{
- register guchar *p = str;
+ Xalloca(eucstr, outlen, return);
- while (*p != '\0') {
- /* convert CR+LF -> LF */
- if (*p == '\r' && *(p + 1) == '\n')
- memmove(p, p + 1, strlen(p));
- else if (!isascii(*p)) *p = SUBST_CHAR;
- p++;
- }
+ conv_utf8toeuc(eucstr, outlen, inbuf);
+ conv_euctojis(outbuf, outlen, eucstr);
}
-void conv_unreadable_latin(gchar *str)
+static void conv_unreadable_8bit(gchar *str)
{
register guchar *p = str;
/* convert CR+LF -> LF */
if (*p == '\r' && *(p + 1) == '\n')
memmove(p, p + 1, strlen(p));
- else if ((*p & 0xff) >= 0x7f && (*p & 0xff) <= 0x9f)
- *p = SUBST_CHAR;
+ else if (!IS_ASCII(*p)) *p = SUBST_CHAR;
p++;
}
}
-void conv_unreadable_locale(gchar *str)
-{
- switch (conv_get_current_charset()) {
- case C_US_ASCII:
- case C_ISO_8859_1:
- case C_ISO_8859_2:
- case C_ISO_8859_3:
- case C_ISO_8859_4:
- case C_ISO_8859_5:
- case C_ISO_8859_6:
- case C_ISO_8859_7:
- case C_ISO_8859_8:
- case C_ISO_8859_9:
- case C_ISO_8859_10:
- case C_ISO_8859_11:
- case C_ISO_8859_13:
- case C_ISO_8859_14:
- case C_ISO_8859_15:
- conv_unreadable_latin(str);
- break;
- case C_EUC_JP:
- conv_unreadable_eucjp(str);
- break;
- default:
- break;
- }
-}
-
#define NCV '\0'
void conv_mb_alnum(gchar *str)
if (guessed == C_US_ASCII)
return C_ISO_2022_JP;
p += 2;
- } else if (isascii(*p)) {
+ } else if (IS_ASCII(*p)) {
p++;
} else if (iseuckanji(*p) && iseuckanji(*(p + 1))) {
if (*p >= 0xfd && *p <= 0xfe)
return guessed;
}
-void conv_jistodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_jistodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- conv_jistoeuc(outbuf, outlen, inbuf);
- conv_unreadable_eucjp(outbuf);
+ conv_jistoutf8(outbuf, outlen, inbuf);
}
-void conv_sjistodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_sjistodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- conv_sjistoeuc(outbuf, outlen, inbuf);
- conv_unreadable_eucjp(outbuf);
+ conv_sjistoutf8(outbuf, outlen, inbuf);
}
-void conv_euctodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_euctodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- strncpy2(outbuf, inbuf, outlen);
- conv_unreadable_eucjp(outbuf);
+ conv_euctoutf8(outbuf, outlen, inbuf);
}
-#warning FIXME_GTK2
-#if 0
-void conv_anytodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
+void conv_utf8todisp(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- conv_anytoeuc(outbuf, outlen, inbuf);
- conv_unreadable_eucjp(outbuf);
+ if (g_utf8_validate(inbuf, -1, NULL) == TRUE)
+ strncpy2(outbuf, inbuf, outlen);
+ else
+ conv_ustodisp(outbuf, outlen, inbuf);
}
-#else
-void conv_anytodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
+
+static void conv_anytodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
{
conv_anytoutf8(outbuf, outlen, inbuf);
+ if (g_utf8_validate(outbuf, -1, NULL) != TRUE)
+ conv_unreadable_8bit(outbuf);
}
-#endif
-void conv_ustodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_ustodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
{
strncpy2(outbuf, inbuf, outlen);
conv_unreadable_8bit(outbuf);
}
-void conv_latintodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
+void conv_localetodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- strncpy2(outbuf, inbuf, outlen);
- conv_unreadable_latin(outbuf);
+ gchar *tmpstr;
+
+ codeconv_set_strict(TRUE);
+ tmpstr = conv_iconv_strdup(inbuf, conv_get_locale_charset_str(),
+ CS_INTERNAL);
+ codeconv_set_strict(FALSE);
+ if (tmpstr && g_utf8_validate(tmpstr, -1, NULL)) {
+ strncpy2(outbuf, tmpstr, outlen);
+ g_free(tmpstr);
+ return;
+ } else if (tmpstr && !g_utf8_validate(tmpstr, -1, NULL)) {
+ g_free(tmpstr);
+ codeconv_set_strict(TRUE);
+ tmpstr = conv_iconv_strdup(inbuf,
+ conv_get_locale_charset_str_no_utf8(),
+ CS_INTERNAL);
+ codeconv_set_strict(FALSE);
+ }
+ if (tmpstr && g_utf8_validate(tmpstr, -1, NULL)) {
+ strncpy2(outbuf, tmpstr, outlen);
+ g_free(tmpstr);
+ } else
+ conv_utf8todisp(outbuf, outlen, inbuf);
}
-void conv_localetodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_noconv(gchar *outbuf, gint outlen, const gchar *inbuf)
{
strncpy2(outbuf, inbuf, outlen);
- conv_unreadable_locale(outbuf);
}
-void conv_noconv(gchar *outbuf, gint outlen, const gchar *inbuf)
+static const gchar *
+conv_get_fallback_for_private_encoding(const gchar *encoding)
{
- strncpy2(outbuf, inbuf, outlen);
+ if (encoding && (encoding[0] == 'X' || encoding[0] == 'x') &&
+ encoding[1] == '-') {
+ if (!g_ascii_strcasecmp(encoding, CS_X_GBK))
+ return CS_GBK;
+ }
+
+ return encoding;
}
-CodeConverter *conv_code_converter_new(const gchar *charset)
+CodeConverter *conv_code_converter_new(const gchar *src_charset)
{
CodeConverter *conv;
+ src_charset = conv_get_fallback_for_private_encoding(src_charset);
+
conv = g_new0(CodeConverter, 1);
-#warning FIXME_GTK2
- conv->code_conv_func = conv_get_code_conv_func(charset, CS_UTF_8);
- conv->charset_str = g_strdup(charset);
- conv->charset = conv_get_charset_from_str(charset);
+ conv->code_conv_func = conv_get_code_conv_func(src_charset, NULL);
+ conv->charset_str = g_strdup(src_charset);
+ conv->charset = conv_get_charset_from_str(src_charset);
return conv;
}
gint conv_convert(CodeConverter *conv, gchar *outbuf, gint outlen,
const gchar *inbuf)
{
-#if HAVE_ICONV
if (conv->code_conv_func != conv_noconv)
conv->code_conv_func(outbuf, outlen, inbuf);
else {
gchar *str;
-#warning FIXME_GTK2
- str = conv_iconv_strdup(inbuf, conv->charset_str, CS_UTF_8);
+ str = conv_iconv_strdup(inbuf, conv->charset_str, NULL);
if (!str)
return -1;
else {
g_free(str);
}
}
-#else /* !HAVE_ICONV */
- conv->code_conv_func(outbuf, outlen, inbuf);
-#endif
return 0;
}
size_t len;
CodeConvFunc conv_func;
+ src_code = conv_get_fallback_for_private_encoding(src_code);
conv_func = conv_get_code_conv_func(src_code, dest_code);
if (conv_func != conv_noconv) {
len = (strlen(inbuf) + 1) * 3;
return g_realloc(buf, strlen(buf) + 1);
}
-#if HAVE_ICONV
return conv_iconv_strdup(inbuf, src_code, dest_code);
-#else
- return g_strdup(inbuf);
-#endif /* HAVE_ICONV */
}
CodeConvFunc conv_get_code_conv_func(const gchar *src_charset_str,
CharSet dest_charset;
if (!src_charset_str)
- src_charset = conv_get_current_charset();
+ src_charset = conv_get_locale_charset();
else
src_charset = conv_get_charset_from_str(src_charset_str);
/* auto detection mode */
if (!src_charset_str && !dest_charset_str) {
- if (src_charset == C_EUC_JP || src_charset == C_SHIFT_JIS)
+ if (conv_is_ja_locale())
return conv_anytodisp;
else
return conv_noconv;
if (dest_charset == C_US_ASCII)
return conv_ustodisp;
- else if (dest_charset == C_UTF_8 ||
- (dest_charset == C_AUTO &&
- conv_get_current_charset() == C_UTF_8))
- return conv_noconv;
switch (src_charset) {
- case C_ISO_2022_JP:
- case C_ISO_2022_JP_2:
- if (dest_charset == C_AUTO &&
- conv_get_current_charset() == C_EUC_JP)
- code_conv = conv_jistodisp;
- else if (dest_charset == C_EUC_JP)
- code_conv = conv_jistoeuc;
- break;
case C_US_ASCII:
- if (dest_charset == C_AUTO)
- code_conv = conv_ustodisp;
- break;
case C_ISO_8859_1:
case C_ISO_8859_2:
case C_ISO_8859_3:
case C_ISO_8859_13:
case C_ISO_8859_14:
case C_ISO_8859_15:
- if (dest_charset == C_AUTO &&
- (conv_get_current_charset() == src_charset ||
- MB_CUR_MAX > 1))
- code_conv = conv_latintodisp;
+ break;
+ case C_ISO_2022_JP:
+ case C_ISO_2022_JP_2:
+ case C_ISO_2022_JP_3:
+ if (dest_charset == C_AUTO)
+ code_conv = conv_jistodisp;
+ else if (dest_charset == C_EUC_JP)
+ code_conv = conv_jistoeuc;
+ else if (dest_charset == C_UTF_8)
+ code_conv = conv_jistoutf8;
break;
case C_SHIFT_JIS:
- if (dest_charset == C_AUTO &&
- conv_get_current_charset() == C_EUC_JP)
+ if (dest_charset == C_AUTO)
code_conv = conv_sjistodisp;
else if (dest_charset == C_EUC_JP)
code_conv = conv_sjistoeuc;
+ else if (dest_charset == C_UTF_8)
+ code_conv = conv_sjistoutf8;
break;
case C_EUC_JP:
- if (dest_charset == C_AUTO &&
- conv_get_current_charset() == C_EUC_JP)
+ if (dest_charset == C_AUTO)
code_conv = conv_euctodisp;
- else if (dest_charset == C_ISO_2022_JP ||
- dest_charset == C_ISO_2022_JP_2)
+ else if (dest_charset == C_ISO_2022_JP ||
+ dest_charset == C_ISO_2022_JP_2 ||
+ dest_charset == C_ISO_2022_JP_3)
code_conv = conv_euctojis;
+ else if (dest_charset == C_UTF_8)
+ code_conv = conv_euctoutf8;
+ break;
+ case C_UTF_8:
+ if (dest_charset == C_EUC_JP)
+ code_conv = conv_utf8toeuc;
+ else if (dest_charset == C_ISO_2022_JP ||
+ dest_charset == C_ISO_2022_JP_2 ||
+ dest_charset == C_ISO_2022_JP_3)
+ code_conv = conv_utf8tojis;
break;
default:
break;
return code_conv;
}
-#if HAVE_ICONV
gchar *conv_iconv_strdup(const gchar *inbuf,
const gchar *src_code, const gchar *dest_code)
{
iconv_t cd;
- const gchar *inbuf_p;
gchar *outbuf;
- gchar *outbuf_p;
- size_t in_size;
- size_t in_left;
- size_t out_size;
- size_t out_left;
- size_t n_conv;
- size_t len;
if (!src_code)
src_code = conv_get_outgoing_charset_str();
if (!dest_code)
- dest_code = conv_get_current_charset_str();
-
- /* don't convert if current codeset is US-ASCII */
- if (!strcasecmp(dest_code, CS_US_ASCII))
- return g_strdup(inbuf);
+ dest_code = CS_INTERNAL;
/* don't convert if src and dest codeset are identical */
if (!strcasecmp(src_code, dest_code))
return g_strdup(inbuf);
+ /* don't convert if current codeset is US-ASCII */
+ if (!strcasecmp(dest_code, CS_US_ASCII))
+ return g_strdup(inbuf);
+
cd = iconv_open(dest_code, src_code);
if (cd == (iconv_t)-1)
return NULL;
+ outbuf = conv_iconv_strdup_with_cd(inbuf, cd);
+
+ iconv_close(cd);
+
+ return outbuf;
+}
+
+gchar *conv_iconv_strdup_with_cd(const gchar *inbuf, iconv_t cd)
+{
+ const gchar *inbuf_p;
+ gchar *outbuf;
+ gchar *outbuf_p;
+ size_t in_size;
+ size_t in_left;
+ size_t out_size;
+ size_t out_left;
+ size_t n_conv;
+ size_t len;
+
inbuf_p = inbuf;
in_size = strlen(inbuf);
in_left = in_size;
while ((n_conv = iconv(cd, (ICONV_CONST gchar **)&inbuf_p, &in_left,
&outbuf_p, &out_left)) == (size_t)-1) {
if (EILSEQ == errno) {
+ if (strict_mode) {
+ g_free(outbuf);
+ return NULL;
+ }
+ //g_print("iconv(): at %d: %s\n", in_size - in_left, g_strerror(errno));
inbuf_p++;
in_left--;
if (out_left == 0) {
outbuf = g_realloc(outbuf, len + 1);
outbuf[len] = '\0';
- iconv_close(cd);
-
return outbuf;
}
-#endif /* HAVE_ICONV */
static const struct {
CharSet charset;
{C_KOI8_U, CS_KOI8_U},
{C_ISO_2022_JP, CS_ISO_2022_JP},
{C_ISO_2022_JP_2, CS_ISO_2022_JP_2},
+ {C_ISO_2022_JP_3, CS_ISO_2022_JP_3},
{C_EUC_JP, CS_EUC_JP},
{C_EUC_JP, CS_EUCJP},
+ {C_EUC_JP_MS, CS_EUC_JP_MS},
{C_SHIFT_JIS, CS_SHIFT_JIS},
{C_SHIFT_JIS, CS_SHIFT__JIS},
{C_SHIFT_JIS, CS_SJIS},
{"ja_JP.ujis" , C_EUC_JP , C_ISO_2022_JP},
{"ja_JP.SJIS" , C_SHIFT_JIS , C_ISO_2022_JP},
{"ja_JP.JIS" , C_ISO_2022_JP , C_ISO_2022_JP},
+#ifdef G_OS_WIN32
+ {"ja_JP" , C_SHIFT_JIS , C_ISO_2022_JP},
+#else
{"ja_JP" , C_EUC_JP , C_ISO_2022_JP},
+#endif
{"ko_KR.EUC-KR" , C_EUC_KR , C_EUC_KR},
{"ko_KR" , C_EUC_KR , C_EUC_KR},
{"zh_CN.GB2312" , C_GB2312 , C_GB2312},
- {"zh_CN.GBK" , C_GBK , C_GB2312},
+ {"zh_CN.GBK" , C_GBK , C_GBK},
{"zh_CN" , C_GB2312 , C_GB2312},
{"zh_HK" , C_BIG5_HKSCS , C_BIG5_HKSCS},
{"zh_TW.eucTW" , C_EUC_TW , C_BIG5},
{"ru_RU" , C_ISO_8859_5 , C_KOI8_R},
{"tg_TJ" , C_KOI8_T , C_KOI8_T},
{"ru_UA" , C_KOI8_U , C_KOI8_U},
+ {"uk_UA.CP1251" , C_WINDOWS_1251, C_KOI8_U},
{"uk_UA" , C_KOI8_U , C_KOI8_U},
{"be_BY" , C_WINDOWS_1251, C_WINDOWS_1251},
return GPOINTER_TO_UINT(g_hash_table_lookup(table, charset));
}
-CharSet conv_get_current_charset(void)
+CharSet conv_get_locale_charset(void)
{
static CharSet cur_charset = -1;
const gchar *cur_locale;
/* "ja_JP.EUC" matches with "ja_JP.eucJP", "ja_JP.EUC" and
"ja_JP". "ja_JP" matches with "ja_JP.xxxx" and "ja" */
- if (!strncasecmp(cur_locale, locale_table[i].locale,
+ if (!g_ascii_strncasecmp(cur_locale, locale_table[i].locale,
+ strlen(locale_table[i].locale))) {
+ cur_charset = locale_table[i].charset;
+ return cur_charset;
+ } else if ((p = strchr(locale_table[i].locale, '_')) &&
+ !strchr(p + 1, '.')) {
+ if (strlen(cur_locale) == 2 &&
+ !g_ascii_strncasecmp(cur_locale, locale_table[i].locale, 2)) {
+ cur_charset = locale_table[i].charset;
+ return cur_charset;
+ }
+ }
+ }
+
+ cur_charset = C_AUTO;
+ return cur_charset;
+}
+
+static CharSet conv_get_locale_charset_no_utf8(void)
+{
+ static CharSet cur_charset = -1;
+ const gchar *cur_locale;
+ const gchar *p;
+ gchar *tmp;
+ gint i;
+
+ if (prefs_common.broken_are_utf8)
+ return conv_get_locale_charset();
+
+ if (cur_charset != -1)
+ return cur_charset;
+
+ cur_locale = conv_get_current_locale();
+ if (!cur_locale) {
+ cur_charset = C_US_ASCII;
+ return cur_charset;
+ }
+
+ if (strcasestr(cur_locale, "UTF-8")) {
+ tmp = g_strdup(cur_locale);
+ *(strcasestr(tmp, ".UTF-8")) = '\0';
+ cur_locale = tmp;
+ }
+
+ if ((p = strcasestr(cur_locale, "@euro")) && p[5] == '\0') {
+ cur_charset = C_ISO_8859_15;
+ return cur_charset;
+ }
+
+ for (i = 0; i < sizeof(locale_table) / sizeof(locale_table[0]); i++) {
+ const gchar *p;
+
+ /* "ja_JP.EUC" matches with "ja_JP.eucJP", "ja_JP.EUC" and
+ "ja_JP". "ja_JP" matches with "ja_JP.xxxx" and "ja" */
+ if (!g_ascii_strncasecmp(cur_locale, locale_table[i].locale,
strlen(locale_table[i].locale))) {
cur_charset = locale_table[i].charset;
return cur_charset;
} else if ((p = strchr(locale_table[i].locale, '_')) &&
!strchr(p + 1, '.')) {
if (strlen(cur_locale) == 2 &&
- !strncasecmp(cur_locale, locale_table[i].locale, 2)) {
+ !g_ascii_strncasecmp(cur_locale, locale_table[i].locale, 2)) {
cur_charset = locale_table[i].charset;
return cur_charset;
}
return cur_charset;
}
-const gchar *conv_get_current_charset_str(void)
+const gchar *conv_get_locale_charset_str(void)
{
static const gchar *codeset = NULL;
if (!codeset)
- codeset = conv_get_charset_str(conv_get_current_charset());
+ codeset = conv_get_charset_str(conv_get_locale_charset());
+
+ return codeset ? codeset : CS_INTERNAL;
+}
+
+const gchar *conv_get_locale_charset_str_no_utf8(void)
+{
+ static const gchar *codeset = NULL;
+
+ if (!codeset)
+ codeset = conv_get_charset_str(conv_get_locale_charset_no_utf8());
+
+ return codeset ? codeset : CS_INTERNAL;
+}
- return codeset ? codeset : CS_US_ASCII;
+CharSet conv_get_internal_charset(void)
+{
+ return C_INTERNAL;
+}
+
+const gchar *conv_get_internal_charset_str(void)
+{
+ return CS_INTERNAL;
}
CharSet conv_get_outgoing_charset(void)
return out_charset;
}
+ if (strcasestr(cur_locale, "UTF-8")) {
+ out_charset = C_UTF_8;
+ return out_charset;
+ }
+
if ((p = strcasestr(cur_locale, "@euro")) && p[5] == '\0') {
out_charset = C_ISO_8859_15;
return out_charset;
for (i = 0; i < sizeof(locale_table) / sizeof(locale_table[0]); i++) {
const gchar *p;
- if (!strncasecmp(cur_locale, locale_table[i].locale,
+ if (!g_ascii_strncasecmp(cur_locale, locale_table[i].locale,
strlen(locale_table[i].locale))) {
out_charset = locale_table[i].out_charset;
break;
} else if ((p = strchr(locale_table[i].locale, '_')) &&
!strchr(p + 1, '.')) {
if (strlen(cur_locale) == 2 &&
- !strncasecmp(cur_locale, locale_table[i].locale, 2)) {
+ !g_ascii_strncasecmp(cur_locale, locale_table[i].locale, 2)) {
out_charset = locale_table[i].out_charset;
break;
}
}
}
-#if !HAVE_ICONV
- /* encoding conversion without iconv() is only supported
- on Japanese locale for now */
- if (out_charset == C_ISO_2022_JP)
- return out_charset;
- else
- return conv_get_current_charset();
-#endif
-
return out_charset;
}
CharSet out_charset;
const gchar *str;
- if (prefs_common.outgoing_charset) {
- if (!isalpha((guchar)prefs_common.outgoing_charset[0])) {
- g_free(prefs_common.outgoing_charset);
- prefs_common.outgoing_charset = g_strdup(CS_AUTO);
- } else if (strcmp(prefs_common.outgoing_charset, CS_AUTO) != 0)
- return prefs_common.outgoing_charset;
- }
-
out_charset = conv_get_outgoing_charset();
str = conv_get_charset_str(out_charset);
- return str ? str : CS_US_ASCII;
+ return str ? str : CS_UTF_8;
}
gboolean conv_is_multibyte_encoding(CharSet encoding)
{
switch (encoding) {
case C_EUC_JP:
+ case C_EUC_JP_MS:
case C_EUC_KR:
case C_EUC_TW:
case C_EUC_CN:
case C_ISO_2022_JP:
case C_ISO_2022_JP_2:
+ case C_ISO_2022_JP_3:
case C_ISO_2022_KR:
case C_ISO_2022_CN:
case C_SHIFT_JIS:
case C_GB2312:
+ case C_GBK:
case C_BIG5:
case C_UTF_8:
case C_UTF_7:
{
const gchar *cur_locale;
+#ifdef G_OS_WIN32
+ cur_locale = g_win32_getlocale();
+#else
cur_locale = g_getenv("LC_ALL");
if (!cur_locale) cur_locale = g_getenv("LC_CTYPE");
if (!cur_locale) cur_locale = g_getenv("LANG");
if (!cur_locale) cur_locale = setlocale(LC_CTYPE, NULL);
+#endif /* G_OS_WIN32 */
debug_print("current locale: %s\n",
cur_locale ? cur_locale : "(none)");
return cur_locale;
}
-void conv_unmime_header_overwrite(gchar *str)
+gboolean conv_is_ja_locale(void)
{
- gchar *buf;
- gint buflen;
- CharSet cur_charset;
- const gchar *locale;
-
- cur_charset = conv_get_current_charset();
+ static gint is_ja_locale = -1;
+ const gchar *cur_locale;
-#warning FIXME_GTK2
-/* Should we always ensure to convert? */
- locale = conv_get_current_locale();
+ if (is_ja_locale != -1)
+ return is_ja_locale != 0;
- if (locale && !strncasecmp(locale, "ja", 2)) {
- buflen = strlen(str) * 2 + 1;
- Xalloca(buf, buflen, return);
- conv_anytodisp(buf, buflen, str);
- unmime_header(str, buf);
- } else {
- buflen = strlen(str) + 1;
- Xalloca(buf, buflen, return);
- unmime_header(buf, str);
- strncpy2(str, buf, buflen);
+ is_ja_locale = 0;
+ cur_locale = conv_get_current_locale();
+ if (cur_locale) {
+ if (g_ascii_strncasecmp(cur_locale, "ja", 2) == 0)
+ is_ja_locale = 1;
}
+
+ return is_ja_locale != 0;
}
-void conv_unmime_header(gchar *outbuf, gint outlen, const gchar *str,
- const gchar *charset)
+gchar *conv_unmime_header(const gchar *str, const gchar *default_encoding)
{
- CharSet cur_charset;
- const gchar *locale;
+ gchar buf[BUFFSIZE];
- cur_charset = conv_get_current_charset();
+ if (is_ascii_str(str))
+ return unmime_header(str);
-#warning FIXME_GTK2
-/* Should we always ensure to convert? */
- locale = conv_get_current_locale();
+ if (default_encoding) {
+ gchar *utf8_buf;
- if (locale && !strncasecmp(locale, "ja", 2)) {
- gchar *buf;
- gint buflen;
+ utf8_buf = conv_codeset_strdup
+ (str, default_encoding, CS_INTERNAL);
+ if (utf8_buf) {
+ gchar *decoded_str;
- buflen = strlen(str) * 2 + 1;
- Xalloca(buf, buflen, return);
- conv_anytodisp(buf, buflen, str);
- unmime_header(outbuf, buf);
- } else
- unmime_header(outbuf, str);
+ decoded_str = unmime_header(utf8_buf);
+ g_free(utf8_buf);
+ return decoded_str;
+ }
+ }
+
+ if (conv_is_ja_locale())
+ conv_anytodisp(buf, sizeof(buf), str);
+ else
+ conv_localetodisp(buf, sizeof(buf), str);
+
+ return unmime_header(buf);
}
#define MAX_LINELEN 76
#define MIMESEP_BEGIN "=?"
#define MIMESEP_END "?="
-#define B64LEN(len) ((len) / 3 * 4 + ((len) % 3 ? 4 : 0))
-
#define LBREAK_IF_REQUIRED(cond, is_plain_text) \
{ \
if (len - (destp - (guchar *)dest) < MAX_LINELEN + 2) { \
} \
}
-void conv_encode_header(gchar *dest, gint len, const gchar *src,
- gint header_len, gboolean addr_field)
+void conv_encode_header_full(gchar *dest, gint len, const gchar *src,
+ gint header_len, gboolean addr_field,
+ const gchar *out_encoding_)
{
const gchar *cur_encoding;
const gchar *out_encoding;
guchar *destp = dest;
gboolean use_base64;
+ g_return_if_fail(g_utf8_validate(src, -1, NULL) == TRUE);
+
if (MB_CUR_MAX > 1) {
use_base64 = TRUE;
mimesep_enc = "?B?";
mimesep_enc = "?Q?";
}
- cur_encoding = conv_get_current_charset_str();
- if (!strcmp(cur_encoding, CS_US_ASCII))
- cur_encoding = CS_ISO_8859_1;
- out_encoding = conv_get_outgoing_charset_str();
+ cur_encoding = CS_INTERNAL;
+
+ if (out_encoding_)
+ out_encoding = out_encoding_;
+ else
+ out_encoding = conv_get_outgoing_charset_str();
+
if (!strcmp(out_encoding, CS_US_ASCII))
out_encoding = CS_ISO_8859_1;
if (addr_field && (*p == '(' || *p == ')'))
break;
- if (MB_CUR_MAX > 1) {
- mb_len = mblen(p, MB_CUR_MAX);
- if (mb_len < 0) {
- g_warning("conv_encode_header(): invalid multibyte character encountered\n");
- mb_len = 1;
- }
- } else
- mb_len = 1;
+ mb_len = g_utf8_skip[*p];
Xstrndup_a(part_str, srcp, cur_len + mb_len, );
out_str = conv_codeset_strdup
(part_str, cur_encoding, out_encoding);
if (!out_str) {
- g_warning("conv_encode_header(): code conversion failed\n");
- conv_unreadable_8bit(part_str);
- out_str = g_strdup(part_str);
+ if (strict_mode) {
+ *dest = '\0';
+ return;
+ } else {
+ g_warning("conv_encode_header(): code conversion failed\n");
+ conv_unreadable_8bit(part_str);
+ out_str = g_strdup(part_str);
+ }
}
out_str_len = strlen(out_str);
*destp = '\0';
}
+void conv_encode_header(gchar *dest, gint len, const gchar *src,
+ gint header_len, gboolean addr_field)
+{
+ conv_encode_header_full(dest,len,src,header_len,addr_field,NULL);
+}
+
#undef LBREAK_IF_REQUIRED
+gchar *conv_filename_from_utf8(const gchar *utf8_file)
+{
+ gchar *fs_file;
+ GError *error = NULL;
+
+ fs_file = g_filename_from_utf8(utf8_file, -1, NULL, NULL, &error);
+ if (error) {
+ g_warning("failed to convert encoding of file name: %s\n",
+ error->message);
+ g_error_free(error);
+ }
+ if (!fs_file)
+ fs_file = g_strdup(utf8_file);
+
+ return fs_file;
+}
+
+gchar *conv_filename_to_utf8(const gchar *fs_file)
+{
+ gchar *utf8_file = NULL;
+ GError *error = NULL;
+
+ utf8_file = g_filename_to_utf8(fs_file, -1, NULL, NULL, &error);
+ if (error) {
+ g_warning("failed to convert encoding of file name: %s\n",
+ error->message);
+ g_error_free(error);
+ }
+
+ if (!utf8_file || !g_utf8_validate(utf8_file, -1, NULL)) {
+ g_free(utf8_file);
+ utf8_file = g_strdup(fs_file);
+ conv_unreadable_8bit(utf8_file);
+ }
+
+ return utf8_file;
+}