/*
* Sylpheed -- a GTK+ based, lightweight, and fast e-mail client
- * Copyright (C) 1999-2003 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
# include <locale.h>
#endif
-#if HAVE_ICONV
-# include <iconv.h>
-#endif
+#include <iconv.h>
#include "intl.h"
#include "codeconv.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_eucjp(gchar *str);
+static void conv_unreadable_8bit(gchar *str);
+static void conv_unreadable_latin(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 void conv_jistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
{
const guchar *in = inbuf;
guchar *out = outbuf;
*out = '\0';
}
-void conv_euctojis(gchar *outbuf, gint outlen, const gchar *inbuf)
+#define JIS_HWDAKUTEN 0x5e
+#define JIS_HWHANDAKUTEN 0x5f
+
+static gint conv_jis_hantozen(guchar *outbuf, guchar jis_code, guchar sound_sym)
+{
+ static guint16 h2z_tbl[] = {
+ /* 0x20 - 0x2f */
+ 0x0000, 0x2123, 0x2156, 0x2157, 0x2122, 0x2126, 0x2572, 0x2521,
+ 0x2523, 0x2525, 0x2527, 0x2529, 0x2563, 0x2565, 0x2567, 0x2543,
+ /* 0x30 - 0x3f */
+ 0x213c, 0x2522, 0x2524, 0x2526, 0x2528, 0x252a, 0x252b, 0x252d,
+ 0x252f, 0x2531, 0x2533, 0x2535, 0x2537, 0x2539, 0x253b, 0x253d,
+ /* 0x40 - 0x4f */
+ 0x253f, 0x2541, 0x2544, 0x2546, 0x2548, 0x254a, 0x254b, 0x254c,
+ 0x254d, 0x254e, 0x254f, 0x2552, 0x2555, 0x2558, 0x255b, 0x255e,
+ /* 0x50 - 0x5f */
+ 0x255f, 0x2560, 0x2561, 0x2562, 0x2564, 0x2566, 0x2568, 0x2569,
+ 0x256a, 0x256b, 0x256c, 0x256d, 0x256f, 0x2573, 0x212b, 0x212c
+ };
+
+ static guint16 dakuten_tbl[] = {
+ /* 0x30 - 0x3f */
+ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x252c, 0x252e,
+ 0x2530, 0x2532, 0x2534, 0x2536, 0x2538, 0x253a, 0x253c, 0x253e,
+ /* 0x40 - 0x4f */
+ 0x2540, 0x2542, 0x2545, 0x2547, 0x2549, 0x0000, 0x0000, 0x0000,
+ 0x0000, 0x0000, 0x2550, 0x2553, 0x2556, 0x2559, 0x255c, 0x0000
+ };
+
+ static guint16 handakuten_tbl[] = {
+ /* 0x4a - 0x4e */
+ 0x2551, 0x2554, 0x2557, 0x255a, 0x255d
+ };
+
+ guint16 out_code;
+
+ jis_code &= 0x7f;
+ sound_sym &= 0x7f;
+
+ if (jis_code < 0x21 || jis_code > 0x5f)
+ return 0;
+
+ if (sound_sym == JIS_HWDAKUTEN &&
+ jis_code >= 0x36 && jis_code <= 0x4e) {
+ out_code = dakuten_tbl[jis_code - 0x30];
+ if (out_code != 0) {
+ *outbuf = out_code >> 8;
+ *(outbuf + 1) = out_code & 0xff;
+ return 2;
+ }
+ }
+
+ if (sound_sym == JIS_HWHANDAKUTEN &&
+ jis_code >= 0x4a && jis_code <= 0x4e) {
+ out_code = handakuten_tbl[jis_code - 0x4a];
+ *outbuf = out_code >> 8;
+ *(outbuf + 1) = out_code & 0xff;
+ return 2;
+ }
+
+ out_code = h2z_tbl[jis_code - 0x20];
+ *outbuf = out_code >> 8;
+ *(outbuf + 1) = out_code & 0xff;
+ return 1;
+}
+
+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 if (iseuchwkana1(*in)) {
- in++;
- if (iseuchwkana2(*in)) {
- HW_IN();
- *out++ = *in++ & 0x7f;
+ if (iseuchwkana2(*(in + 1))) {
+ if (prefs_common.allow_jisx0201_kana) {
+ HW_IN();
+ in++;
+ *out++ = *in++ & 0x7f;
+ } else {
+ guchar jis_ch[2];
+ gint len;
+
+ if (iseuchwkana1(*(in + 2)) &&
+ iseuchwkana2(*(in + 3)))
+ len = conv_jis_hantozen
+ (jis_ch,
+ *(in + 1), *(in + 3));
+ else
+ len = conv_jis_hantozen
+ (jis_ch,
+ *(in + 1), '\0');
+ if (len == 0)
+ in += 2;
+ else {
+ K_IN();
+ in += len * 2;
+ *out++ = jis_ch[0];
+ *out++ = jis_ch[1];
+ }
+ }
} else {
K_OUT();
- if (*in != '\0' && !isascii(*in)) {
+ 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:
+ gchar *eucstr;
+
+ Xalloca(eucstr, outlen, return);
+
+ conv_jistoeuc(eucstr, outlen, inbuf);
+ conv_euctoutf8(outbuf, outlen, eucstr);
+}
+
+static void conv_sjistoutf8(gchar *outbuf, gint outlen, const gchar *inbuf)
+{
+ gchar *tmpstr;
+
+ tmpstr = conv_iconv_strdup(inbuf, CS_SHIFT_JIS, CS_UTF_8);
+ if (tmpstr) {
+ strncpy2(outbuf, tmpstr, outlen);
+ g_free(tmpstr);
+ } else
strncpy2(outbuf, inbuf, outlen);
- break;
- }
}
-void conv_anytoutf8(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_euctoutf8(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- gchar *tmpstr = NULL;
+ 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;
+ }
+ }
+ }
- switch (conv_guess_ja_encoding(inbuf)) {
- case C_ISO_2022_JP:
- tmpstr = conv_codeset_strdup(inbuf, CS_ISO_2022_JP, CS_UTF_8);
+ tmpstr = conv_iconv_strdup_with_cd(inbuf, cd);
+ if (tmpstr) {
strncpy2(outbuf, tmpstr, outlen);
g_free(tmpstr);
+ } else
+ strncpy2(outbuf, inbuf, outlen);
+}
+
+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:
- tmpstr = conv_codeset_strdup(inbuf, CS_SHIFT_JIS, CS_UTF_8);
- strncpy2(outbuf, tmpstr, outlen);
- g_free(tmpstr);
+ conv_sjistoutf8(outbuf, outlen, inbuf);
break;
case C_EUC_JP:
- tmpstr = conv_codeset_strdup(inbuf, CS_EUC_JP, CS_UTF_8);
- strncpy2(outbuf, tmpstr, outlen);
- g_free(tmpstr);
+ conv_euctoutf8(outbuf, outlen, inbuf);
break;
default:
strncpy2(outbuf, inbuf, outlen);
}
}
-void conv_anytojis(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_utf8toeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- switch (conv_guess_ja_encoding(inbuf)) {
- case C_EUC_JP:
- conv_euctojis(outbuf, outlen, inbuf);
- break;
- default:
- strncpy2(outbuf, inbuf, outlen);
- break;
+ 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;
+ }
+ }
}
+
+ tmpstr = conv_iconv_strdup_with_cd(inbuf, cd);
+ if (tmpstr) {
+ strncpy2(outbuf, tmpstr, outlen);
+ g_free(tmpstr);
+ } else
+ strncpy2(outbuf, inbuf, outlen);
+}
+
+static void conv_utf8tojis(gchar *outbuf, gint outlen, const gchar *inbuf)
+{
+ gchar *eucstr;
+
+ Xalloca(eucstr, outlen, return);
+
+ conv_utf8toeuc(eucstr, outlen, inbuf);
+ conv_euctojis(outbuf, outlen, eucstr);
}
static gchar valid_eucjp_tbl[][96] = {
return TRUE;
}
-void conv_unreadable_eucjp(gchar *str)
+static void conv_unreadable_eucjp(gchar *str)
{
register guchar *p = str;
while (*p != '\0') {
- if (isascii(*p)) {
+ if (IS_ASCII(*p)) {
/* convert CR+LF -> LF */
if (*p == '\r' && *(p + 1) == '\n')
memmove(p, p + 1, strlen(p));
/* substitute unprintable code */
*p++ = SUBST_CHAR;
if (*p != '\0') {
- if (isascii(*p))
+ if (IS_ASCII(*p))
p++;
else
*p++ = SUBST_CHAR;
}
}
-void conv_unreadable_8bit(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 (!isascii(*p)) *p = SUBST_CHAR;
+ else if (!IS_ASCII(*p)) *p = SUBST_CHAR;
p++;
}
}
-void conv_unreadable_latin(gchar *str)
+static void conv_unreadable_latin(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)
+ else if ((*p & 0xff) >= 0x7f)
*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);
+ strncpy2(outbuf, inbuf, outlen);
}
-#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);
}
-#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)
-{
- strncpy2(outbuf, inbuf, outlen);
- conv_unreadable_latin(outbuf);
-}
-
void conv_localetodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- strncpy2(outbuf, inbuf, outlen);
- conv_unreadable_locale(outbuf);
+ gchar *tmpstr;
+
+ tmpstr = conv_iconv_strdup(inbuf, conv_get_locale_charset_str(),
+ CS_INTERNAL);
+ if (tmpstr) {
+ strncpy2(outbuf, tmpstr, outlen);
+ g_free(tmpstr);
+ } else
+ strncpy2(outbuf, inbuf, outlen);
}
-void conv_noconv(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_noconv(gchar *outbuf, gint outlen, const gchar *inbuf)
{
strncpy2(outbuf, inbuf, outlen);
}
-CodeConverter *conv_code_converter_new(const gchar *charset)
+CodeConverter *conv_code_converter_new(const gchar *src_charset)
{
CodeConverter *conv;
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;
}
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);
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:
+ 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_latintodisp;
+ 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;
- gint in_size;
- gint in_left;
- gint out_size;
- gint out_left;
- gint n_conv;
- gint 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)) < 0) {
+ &outbuf_p, &out_left)) == (size_t)-1) {
if (EILSEQ == errno) {
+ //g_print("iconv(): at %d: %s\n", in_size - in_left, g_strerror(errno));
inbuf_p++;
in_left--;
if (out_left == 0) {
}
}
- while ((n_conv = iconv(cd, NULL, NULL, &outbuf_p, &out_left)) < 0) {
+ while ((n_conv = iconv(cd, NULL, NULL, &outbuf_p, &out_left)) ==
+ (size_t)-1) {
if (E2BIG == errno) {
EXPAND_BUF();
} else {
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},
{"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 &&
- !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_US_ASCII;
+ return codeset ? codeset : CS_INTERNAL;
+}
+
+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)
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;
}
const gchar *str;
if (prefs_common.outgoing_charset) {
- if (!isalpha(prefs_common.outgoing_charset[0])) {
+ 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)
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:
gchar *buf;
gint buflen;
CharSet cur_charset;
- const gchar *locale;
- cur_charset = conv_get_current_charset();
+ cur_charset = conv_get_locale_charset();
-#warning FIXME_GTK2
-/* Should we always ensure to convert? */
- locale = conv_get_current_locale();
-
- if (locale && !strncasecmp(locale, "ja", 2)) {
+ if (cur_charset == C_EUC_JP) {
buflen = strlen(str) * 2 + 1;
Xalloca(buf, buflen, return);
conv_anytodisp(buf, buflen, str);
const gchar *charset)
{
CharSet cur_charset;
- const gchar *locale;
-
- cur_charset = conv_get_current_charset();
-#warning FIXME_GTK2
-/* Should we always ensure to convert? */
- locale = conv_get_current_locale();
+ cur_charset = conv_get_locale_charset();
- if (locale && !strncasecmp(locale, "ja", 2)) {
+ if (cur_charset == C_EUC_JP) {
gchar *buf;
gint buflen;
Xalloca(buf, buflen, return);
conv_anytodisp(buf, buflen, str);
unmime_header(outbuf, buf);
- } else
+ } else {
+ gchar *tmp = NULL;
unmime_header(outbuf, str);
+
+ if (outbuf && !g_utf8_validate(outbuf, -1, NULL)) {
+ if (conv_get_locale_charset() != C_INTERNAL)
+ tmp = conv_codeset_strdup(outbuf,
+ conv_get_locale_charset_str(),
+ CS_INTERNAL);
+
+ if (tmp) {
+ strncpy(outbuf, tmp, outlen-1);
+ g_free(tmp);
+ } else {
+ conv_unreadable_8bit(outbuf);
+ }
+ }
+ }
+
}
#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 - dest) < MAX_LINELEN + 2) { \
+ if (len - (destp - (guchar *)dest) < MAX_LINELEN + 2) { \
*destp = '\0'; \
return; \
} \
\
if ((cond) && *srcp) { \
- if (destp > dest && left < MAX_LINELEN - 1) { \
+ if (destp > (guchar *)dest && left < MAX_LINELEN - 1) { \
if (isspace(*(destp - 1))) \
destp--; \
else if (is_plain_text && isspace(*srcp)) \
gint mimestr_len;
gchar *mimesep_enc;
gint left;
- const gchar *srcp = src;
- gchar *destp = dest;
+ const guchar *srcp = src;
+ 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;
+ cur_encoding = CS_INTERNAL;
out_encoding = conv_get_outgoing_charset_str();
if (!strcmp(out_encoding, CS_US_ASCII))
out_encoding = CS_ISO_8859_1;
gchar *part_str;
gchar *out_str;
gchar *enc_str;
- const gchar *p = srcp;
+ const guchar *p = srcp;
gint out_str_len;
gint out_enc_str_len;
gint mime_block_len;
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
projects / claws.git / blobdiff