/*
* Sylpheed -- a GTK+ based, lightweight, and fast e-mail client
- * Copyright (C) 1999-2002 Hiroyuki Yamamoto
+ * Copyright (C) 1999-2003 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 <string.h>
#include <ctype.h>
#include <stdlib.h>
-
-#if (HAVE_WCTYPE_H && HAVE_WCHAR_H)
-# include <wchar.h>
-# include <wctype.h>
-#endif
+#include <errno.h>
#if HAVE_LOCALE_H
# include <locale.h>
#endif
-#if HAVE_LIBJCONV
-# include <jconv.h>
+#if HAVE_ICONV
+# include <iconv.h>
#endif
#include "intl.h"
(((c) & 0xff) >= 0xa1 && ((c) & 0xff) <= 0xdf)
#define iseucaux(c) \
(((c) & 0xff) == 0x8f)
-#define isunprintableeuckanji(c) \
- (((c) & 0xff) >= 0xa9 && ((c) & 0xff) <= 0xaf)
#define issjiskanji1(c) \
((((c) & 0xff) >= 0x81 && ((c) & 0xff) <= 0x9f) || \
(((c) & 0xff) >= 0xe0 && ((c) & 0xff) <= 0xfc))
void conv_anytoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- switch (conv_guess_encoding(inbuf)) {
+ switch (conv_guess_ja_encoding(inbuf)) {
case C_ISO_2022_JP:
conv_jistoeuc(outbuf, outlen, inbuf);
break;
void conv_anytojis(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- switch (conv_guess_encoding(inbuf)) {
+ switch (conv_guess_ja_encoding(inbuf)) {
case C_EUC_JP:
conv_euctojis(outbuf, outlen, inbuf);
break;
}
}
+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)
+{
+ 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 <= 0xff)
+ return FALSE;
+ } else if (c1 == 0xf4) {
+ if (c2 >= 0xa7 && c2 <= 0xff)
+ return FALSE;
+ }
+
+ return TRUE;
+}
+
void conv_unreadable_eucjp(gchar *str)
{
register guchar *p = str;
/* printable 7 bit code */
p++;
} else if (iseuckanji(*p)) {
- if (iseuckanji(*(p + 1)) && !isunprintableeuckanji(*p))
+ if (isprintableeuckanji(*p, *(p + 1))) {
/* printable euc-jp code */
p += 2;
- else {
+ } else {
/* substitute unprintable code */
*p++ = SUBST_CHAR;
if (*p != '\0') {
/* convert CR+LF -> LF */
if (*p == '\r' && *(p + 1) == '\n')
memmove(p, p + 1, strlen(p));
- else if ((*p & 0xff) >= 0x80 && (*p & 0xff) <= 0x9f)
+ else if ((*p & 0xff) >= 0x7f && (*p & 0xff) <= 0x9f)
*p = SUBST_CHAR;
p++;
}
}
}
-CharSet conv_guess_encoding(const gchar *str)
+CharSet conv_guess_ja_encoding(const gchar *str)
{
const guchar *p = str;
CharSet guessed = C_US_ASCII;
strncpy2(outbuf, inbuf, outlen);
}
+void conv_localetodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
+{
+ strncpy2(outbuf, inbuf, outlen);
+
+ switch (conv_get_current_charset()) {
+ case C_US_ASCII:
+ case C_ISO_8859_1:
+ case C_ISO_8859_2:
+ case C_ISO_8859_4:
+ case C_ISO_8859_5:
+ case C_ISO_8859_7:
+ case C_ISO_8859_8:
+ case C_ISO_8859_9:
+ case C_ISO_8859_11:
+ case C_ISO_8859_13:
+ case C_ISO_8859_15:
+ conv_unreadable_latin(outbuf);
+ break;
+ case C_EUC_JP:
+ conv_unreadable_eucjp(outbuf);
+ break;
+ default:
+ break;
+ }
+}
+
CodeConverter *conv_code_converter_new(const gchar *charset)
{
CodeConverter *conv;
conv = g_new0(CodeConverter, 1);
-#if !HAVE_LIBJCONV
- conv->code_conv_func = conv_get_code_conv_func(charset);
-#endif
+ conv->code_conv_func = conv_get_code_conv_func(charset, NULL);
conv->charset_str = g_strdup(charset);
conv->charset = conv_get_charset_from_str(charset);
gint conv_convert(CodeConverter *conv, gchar *outbuf, gint outlen,
const gchar *inbuf)
{
-#if HAVE_LIBJCONV
- gchar *str;
-
- str = conv_codeset_strdup(inbuf, conv->charset_str, NULL);
- if (!str)
- return -1;
+#if HAVE_ICONV
+ if (conv->code_conv_func != conv_noconv)
+ conv->code_conv_func(outbuf, outlen, inbuf);
else {
- strncpy2(outbuf, str, outlen);
- g_free(str);
+ gchar *str;
+
+ str = conv_codeset_strdup(inbuf, conv->charset_str, NULL);
+ if (!str)
+ return -1;
+ else {
+ strncpy2(outbuf, str, outlen);
+ g_free(str);
+ }
}
-#else /* !HAVE_LIBJCONV */
+#else /* !HAVE_ICONV */
conv->code_conv_func(outbuf, outlen, inbuf);
#endif
}
gchar *conv_codeset_strdup(const gchar *inbuf,
- const gchar *src_codeset, const gchar *dest_codeset)
+ const gchar *src_code, const gchar *dest_code)
{
gchar *buf;
size_t len;
-#if HAVE_LIBJCONV
- gint actual_codeset;
- const gchar *const *codesets;
- gint n_codesets;
-#else /* !HAVE_LIBJCONV */
- CharSet src_charset = C_AUTO, dest_charset = C_AUTO;
-#endif
+ CodeConvFunc conv_func;
- if (!dest_codeset) {
- CodeConvFunc func;
+ conv_func = conv_get_code_conv_func(src_code, dest_code);
+ if (conv_func != conv_noconv) {
+ len = (strlen(inbuf) + 1) * 3;
+ buf = g_malloc(len);
+ if (!buf) return NULL;
- func = conv_get_code_conv_func(src_codeset);
- if (func != conv_noconv) {
- if (func == conv_jistodisp ||
- func == conv_sjistodisp ||
- func == conv_anytodisp)
- len = strlen(inbuf) * 2 + 1;
- else
- len = strlen(inbuf) + 1;
- buf = g_malloc(len);
- if (!buf) return NULL;
- func(buf, len, inbuf);
- buf = g_realloc(buf, strlen(buf) + 1);
- return buf;
- }
+ conv_func(buf, len, inbuf);
+ return g_realloc(buf, strlen(buf) + 1);
}
- /* don't convert if src and dest codeset are identical */
- if (src_codeset && dest_codeset &&
- !strcasecmp(src_codeset, dest_codeset))
- return g_strdup(inbuf);
-
-#if HAVE_LIBJCONV
- if (src_codeset) {
- codesets = &src_codeset;
- n_codesets = 1;
- } else
- codesets = jconv_info_get_pref_codesets(&n_codesets);
- if (!dest_codeset) {
- dest_codeset = conv_get_current_charset_str();
- /* don't convert if current codeset is US-ASCII */
- if (!strcasecmp(dest_codeset, CS_US_ASCII))
- return g_strdup(inbuf);
+#if HAVE_ICONV
+ if (!src_code)
+ src_code = conv_get_outgoing_charset_str();
+ if (!dest_code) {
+ dest_code = conv_get_current_charset_str();
+ if (!strcasecmp(dest_code, CS_US_ASCII))
+ dest_code = CS_ISO_8859_1;
}
- if (jconv_alloc_conv(inbuf, strlen(inbuf), &buf, &len,
- codesets, n_codesets,
- &actual_codeset, dest_codeset)
- == 0)
- return buf;
- else {
-#if 0
- g_warning("code conversion from %s to %s failed\n",
- codesets && codesets[0] ? codesets[0] : "(unknown)",
- dest_codeset);
-#endif /* 0 */
- return NULL;
- }
-#else /* !HAVE_LIBJCONV */
- if (src_codeset) {
- if (!strcasecmp(src_codeset, CS_EUC_JP) ||
- !strcasecmp(src_codeset, CS_EUCJP))
- src_charset = C_EUC_JP;
- else if (!strcasecmp(src_codeset, CS_SHIFT_JIS) ||
- !strcasecmp(src_codeset, "SHIFT-JIS") ||
- !strcasecmp(src_codeset, "SJIS"))
- src_charset = C_SHIFT_JIS;
- if (dest_codeset && !strcasecmp(dest_codeset, CS_ISO_2022_JP))
- dest_charset = C_ISO_2022_JP;
- }
+ /* don't convert if current codeset is US-ASCII */
+ if (!strcasecmp(dest_code, CS_US_ASCII))
+ return g_strdup(inbuf);
- if ((src_charset == C_EUC_JP || src_charset == C_SHIFT_JIS) &&
- dest_charset == C_ISO_2022_JP) {
- len = (strlen(inbuf) + 1) * 3;
- buf = g_malloc(len);
- if (buf) {
- if (src_charset == C_EUC_JP)
- conv_euctojis(buf, len, inbuf);
- else
- conv_anytojis(buf, len, inbuf);
- buf = g_realloc(buf, strlen(buf) + 1);
- }
- } else
- buf = g_strdup(inbuf);
+ /* don't convert if src and dest codeset are identical */
+ if (!strcasecmp(src_code, dest_code))
+ return g_strdup(inbuf);
- return buf;
-#endif /* !HAVE_LIBJCONV */
+ 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 *charset)
+CodeConvFunc conv_get_code_conv_func(const gchar *src_charset_str,
+ const gchar *dest_charset_str)
{
- CodeConvFunc code_conv;
- CharSet cur_charset;
+ CodeConvFunc code_conv = conv_noconv;
+ CharSet src_charset;
+ CharSet dest_charset;
- if (!charset) {
- cur_charset = conv_get_current_charset();
- if (cur_charset == C_EUC_JP || cur_charset == C_SHIFT_JIS)
+ if (!src_charset_str)
+ src_charset = conv_get_current_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)
return conv_anytodisp;
else
return conv_noconv;
}
- if (!strcasecmp(charset, CS_ISO_2022_JP) ||
- !strcasecmp(charset, CS_ISO_2022_JP_2))
- code_conv = conv_jistodisp;
- else if (!strcasecmp(charset, CS_US_ASCII))
- code_conv = conv_ustodisp;
- else if (!strncasecmp(charset, CS_ISO_8859_1, 10))
- code_conv = conv_latintodisp;
-#if !HAVE_LIBJCONV
- else if (!strncasecmp(charset, "ISO-8859-", 9))
- code_conv = conv_latintodisp;
+ dest_charset = conv_get_charset_from_str(dest_charset_str);
+
+ if (dest_charset == C_US_ASCII)
+ return conv_ustodisp;
+
+ switch (src_charset) {
+ case C_ISO_2022_JP:
+ case C_ISO_2022_JP_2:
+ if (dest_charset == C_AUTO)
+ 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:
+#if !HAVE_ICONV
+ case C_ISO_8859_2:
+ case C_ISO_8859_4:
+ case C_ISO_8859_5:
+ case C_ISO_8859_7:
+ case C_ISO_8859_8:
+ case C_ISO_8859_9:
+ case C_ISO_8859_11:
+ case C_ISO_8859_13:
+ case C_ISO_8859_15:
#endif
- else if (!strcasecmp(charset, CS_SHIFT_JIS) ||
- !strcasecmp(charset, "SHIFT-JIS") ||
- !strcasecmp(charset, "SJIS") ||
- !strcasecmp(charset, "X-SJIS"))
- code_conv = conv_sjistodisp;
- else if (!strcasecmp(charset, CS_EUC_JP) ||
- !strcasecmp(charset, CS_EUCJP))
- code_conv = conv_euctodisp;
- else
- code_conv = conv_noconv;
+ if (dest_charset == C_AUTO)
+ code_conv = conv_latintodisp;
+ break;
+ case C_SHIFT_JIS:
+ if (dest_charset == C_AUTO)
+ code_conv = conv_sjistodisp;
+ else if (dest_charset == C_EUC_JP)
+ code_conv = conv_sjistoeuc;
+ break;
+ case 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)
+ code_conv = conv_euctojis;
+ 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;
+
+ cd = iconv_open(dest_code, src_code);
+ if (cd == (iconv_t)-1)
+ return NULL;
+
+ inbuf_p = inbuf;
+ in_size = strlen(inbuf) + 1;
+ in_left = in_size;
+ out_size = in_size * 2;
+ outbuf = g_malloc(out_size);
+ outbuf_p = outbuf;
+ out_left = out_size;
+
+ while ((n_conv = iconv(cd, (ICONV_CONST gchar **)&inbuf_p, &in_left,
+ &outbuf_p, &out_left)) < 0) {
+ if (EILSEQ == errno) {
+ g_free(outbuf);
+ outbuf = NULL;
+ break;
+ } else if (EINVAL == errno) {
+ g_free(outbuf);
+ outbuf = NULL;
+ break;
+ } else if (E2BIG == errno) {
+ out_size *= 2;
+ outbuf = g_realloc(outbuf, out_size);
+ inbuf_p = inbuf;
+ in_left = in_size;
+ outbuf_p = outbuf;
+ out_left = out_size;
+ } else {
+ g_warning("conv_iconv_strdup(): %s\n",
+ g_strerror(errno));
+ *outbuf_p = '\0';
+ break;
+ }
+ }
+
+ if (outbuf) {
+ iconv(cd, NULL, NULL, &outbuf_p, &out_left);
+ outbuf = g_realloc(outbuf, strlen(outbuf) + 1);
+ }
+
+ iconv_close(cd);
+
+ return outbuf;
+}
+#endif /* HAVE_ICONV */
+
static const struct {
CharSet charset;
gchar *const name;
{C_EUC_JP, CS_EUC_JP},
{C_EUC_JP, CS_EUCJP},
{C_SHIFT_JIS, CS_SHIFT_JIS},
+ {C_SHIFT_JIS, CS_SHIFT__JIS},
+ {C_SHIFT_JIS, CS_SJIS},
{C_ISO_2022_KR, CS_ISO_2022_KR},
{C_EUC_KR, CS_EUC_KR},
{C_ISO_2022_CN, CS_ISO_2022_CN},
{C_WINDOWS_874, CS_WINDOWS_874},
};
-#if !HAVE_LIBJCONV
static const struct {
gchar *const locale;
CharSet charset;
{"zh_TW.Big5" , C_BIG5 , C_BIG5},
{"zh_TW" , C_BIG5 , C_BIG5},
- {"ru_RU.KOI8-R" , C_KOI8_R , C_ISO_8859_5},
- {"ru_RU.CP1251" , C_WINDOWS_1251, C_ISO_8859_5},
-
+ {"ru_RU.KOI8-R" , C_KOI8_R , C_KOI8_R},
+ {"ru_RU.CP1251" , C_WINDOWS_1251, C_KOI8_R},
+ {"ru_RU" , C_ISO_8859_5 , C_KOI8_R},
+ {"ru_UA" , C_KOI8_U , C_KOI8_U},
+ {"uk_UA" , C_KOI8_U , C_KOI8_U},
+ {"be_BY" , C_WINDOWS_1251, C_WINDOWS_1251},
{"bg_BG" , C_WINDOWS_1251, C_WINDOWS_1251},
{"en_US" , C_ISO_8859_1 , C_ISO_8859_1},
{"ro_RO" , C_ISO_8859_2 , C_ISO_8859_2},
{"sk_SK" , C_ISO_8859_2 , C_ISO_8859_2},
{"sl_SI" , C_ISO_8859_2 , C_ISO_8859_2},
- {"ru_RU" , C_ISO_8859_5 , C_ISO_8859_5},
{"el_GR" , C_ISO_8859_7 , C_ISO_8859_7},
{"iw_IL" , C_ISO_8859_8 , C_ISO_8859_8},
{"tr_TR" , C_ISO_8859_9 , C_ISO_8859_9},
{"POSIX" , C_US_ASCII , C_US_ASCII},
{"ANSI_X3.4-1968" , C_US_ASCII , C_US_ASCII},
};
-#endif /* !HAVE_LIBJCONV */
-const gchar *conv_get_charset_str(CharSet charset)
+static GHashTable *conv_get_charset_to_str_table(void)
{
+ static GHashTable *table;
gint i;
+ if (table)
+ return table;
+
+ table = g_hash_table_new(NULL, g_direct_equal);
+
for (i = 0; i < sizeof(charsets) / sizeof(charsets[0]); i++) {
- if (charsets[i].charset == charset)
- return charsets[i].name;
+ if (g_hash_table_lookup(table, GUINT_TO_POINTER(charsets[i].charset))
+ == NULL) {
+ g_hash_table_insert
+ (table, GUINT_TO_POINTER(charsets[i].charset),
+ charsets[i].name);
+ }
}
- return NULL;
+ return table;
}
-CharSet conv_get_charset_from_str(const gchar *charset)
+static gint str_case_equal(gconstpointer v, gconstpointer v2)
+{
+ return strcasecmp((const gchar *)v, (const gchar *)v2) == 0;
+}
+
+static guint str_case_hash(gconstpointer key)
+{
+ const gchar *p = key;
+ guint h = *p;
+
+ if (h) {
+ h = tolower(h);
+ for (p += 1; *p != '\0'; p++)
+ h = (h << 5) - h + tolower(*p);
+ }
+
+ return h;
+}
+
+static GHashTable *conv_get_charset_from_str_table(void)
{
+ static GHashTable *table;
gint i;
- if (!charset) return C_AUTO;
+ if (table)
+ return table;
+
+ table = g_hash_table_new(str_case_hash, str_case_equal);
for (i = 0; i < sizeof(charsets) / sizeof(charsets[0]); i++) {
- if (!strcasecmp(charsets[i].name, charset))
- return charsets[i].charset;
+ g_hash_table_insert(table, charsets[i].name,
+ GUINT_TO_POINTER(charsets[i].charset));
}
- return C_AUTO;
+ return table;
+}
+
+const gchar *conv_get_charset_str(CharSet charset)
+{
+ GHashTable *table;
+
+ table = conv_get_charset_to_str_table();
+ return g_hash_table_lookup(table, GUINT_TO_POINTER(charset));
+}
+
+CharSet conv_get_charset_from_str(const gchar *charset)
+{
+ GHashTable *table;
+
+ if (!charset) return C_AUTO;
+
+ table = conv_get_charset_from_str_table();
+ return GPOINTER_TO_UINT(g_hash_table_lookup(table, charset));
}
CharSet conv_get_current_charset(void)
{
static CharSet cur_charset = -1;
- gint i;
-
-#if HAVE_LIBJCONV
- const gchar *cur_codeset;
-#else
const gchar *cur_locale;
-#endif
+ const gchar *p;
+ gint i;
if (cur_charset != -1)
return cur_charset;
-#if HAVE_LIBJCONV
- cur_codeset = jconv_info_get_current_codeset();
- for (i = 0; i < sizeof(charsets) / sizeof(charsets[0]); i++) {
- if (!strcasecmp(cur_codeset, charsets[i].name)) {
- cur_charset = charsets[i].charset;
- return cur_charset;
- }
- }
-#else
cur_locale = conv_get_current_locale();
if (!cur_locale) {
cur_charset = C_US_ASCII;
return cur_charset;
}
+ 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" and "ja_JP.EUC" */
- /* "ja_JP" matches with "ja_JP.xxxx" and "ja" */
+ /* "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,
strlen(locale_table[i].locale))) {
cur_charset = locale_table[i].charset;
}
}
}
-#endif
cur_charset = C_AUTO;
return cur_charset;
if (!codeset)
codeset = conv_get_charset_str(conv_get_current_charset());
- return codeset ? codeset : "US-ASCII";
+ return codeset ? codeset : CS_US_ASCII;
}
CharSet conv_get_outgoing_charset(void)
{
static CharSet out_charset = -1;
- gint i;
-
-#if HAVE_LIBJCONV
- gint j, n_pref_codesets;
- const gchar *const *pref_codesets;
-#else
const gchar *cur_locale;
-#endif
+ const gchar *p;
+ gint i;
if (out_charset != -1)
return out_charset;
-#if HAVE_LIBJCONV
- /* skip US-ASCII and UTF-8 */
- pref_codesets = jconv_info_get_pref_codesets(&n_pref_codesets);
- for (i = 0; i < n_pref_codesets; i++) {
- for (j = 3; j < sizeof(charsets) / sizeof(charsets[0]); j++) {
- if (!strcasecmp(pref_codesets[i], charsets[j].name)) {
- out_charset = charsets[j].charset;
- return out_charset;
- }
- }
- }
-
- for (i = 0; i < n_pref_codesets; i++) {
- if (!strcasecmp(pref_codesets[i], "UTF-8")) {
- out_charset = C_UTF_8;
- return out_charset;
- }
- }
-
- out_charset = C_AUTO;
-#else
cur_locale = conv_get_current_locale();
if (!cur_locale) {
out_charset = C_AUTO;
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;
}
}
- /* encoding conversion without libjconv is only supported
+#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;
-
- out_charset = conv_get_current_charset();
+ else
+ return conv_get_current_charset();
#endif
return out_charset;
out_charset = conv_get_outgoing_charset();
str = conv_get_charset_str(out_charset);
- return str ? str : "US-ASCII";
+ return str ? str : CS_US_ASCII;
}
const gchar *conv_get_current_locale(void)
#define B64LEN(len) ((len) / 3 * 4 + ((len) % 3 ? 4 : 0))
-#define LBREAK_IF_REQUIRED(cond) \
+#define LBREAK_IF_REQUIRED(cond, plaintext) \
{ \
- if (len - (destp - dest) < MAX_LINELEN) { \
+ if (len - (destp - dest) < MAX_LINELEN + 2) { \
*destp = '\0'; \
return; \
} \
\
- if (cond) { \
- if (destp > dest && isspace(*(destp - 1))) \
- destp--; \
- *destp++ = '\n'; \
- *destp++ = ' '; \
- left = MAX_LINELEN - 1; \
+ if ((cond) && *srcp) { \
+ if (destp > dest && left < MAX_LINELEN - 1) { \
+ if (isspace(*(destp - 1))) \
+ destp--; \
+ else if (plaintext && isspace(*srcp)) \
+ srcp++; \
+ if (*srcp) { \
+ *destp++ = '\n'; \
+ *destp++ = ' '; \
+ left = MAX_LINELEN - 1; \
+ } \
+ } \
} \
}
}
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();
- if (!strcmp(out_encoding, "US-ASCII"))
- out_encoding = "ISO-8859-1";
+ if (!strcmp(out_encoding, CS_US_ASCII))
+ out_encoding = CS_ISO_8859_1;
mimestr_len = strlen(MIMESEP_BEGIN) + strlen(out_encoding) +
strlen(mimesep_enc) + strlen(MIMESEP_END);
left = MAX_LINELEN - header_len;
while (*srcp) {
- LBREAK_IF_REQUIRED(left <= 0);
+ LBREAK_IF_REQUIRED(left <= 0, TRUE);
while (isspace(*srcp)) {
*destp++ = *srcp++;
left--;
- LBREAK_IF_REQUIRED(left <= 0);
+ LBREAK_IF_REQUIRED(left <= 0, TRUE);
}
/* output as it is if the next word is ASCII string */
gint word_len;
word_len = get_next_word_len(srcp);
- LBREAK_IF_REQUIRED(left < word_len);
- while(*srcp && !isspace(*srcp)) {
+ LBREAK_IF_REQUIRED(left < word_len, TRUE);
+ while (word_len > 0) {
+ LBREAK_IF_REQUIRED(left + 22 <= 0, TRUE);
*destp++ = *srcp++;
left--;
- LBREAK_IF_REQUIRED(left <= 0);
+ word_len--;
}
continue;
if (isspace(*p) && !is_next_nonascii(p + 1))
break;
- mb_len = mblen(p, MB_CUR_MAX);
- if (mb_len < 0) {
- g_warning("invalid multibyte character encountered\n");
- 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;
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);
+ }
out_str_len = strlen(out_str);
if (use_base64)
cur_len += mb_len;
p += mb_len;
} else if (cur_len == 0) {
- LBREAK_IF_REQUIRED(1);
+ LBREAK_IF_REQUIRED(1, FALSE);
continue;
} else {
cont = TRUE;
Xstrndup_a(part_str, srcp, cur_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);
+ }
out_str_len = strlen(out_str);
if (use_base64)
left -= mime_block_len;
}
- LBREAK_IF_REQUIRED(cont);
+ LBREAK_IF_REQUIRED(cont, FALSE);
if (cur_len == 0)
break;