/*
* Sylpheed -- a GTK+ based, lightweight, and fast e-mail client
- * Copyright (C) 1999-2001 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 "config.h"
#endif
+#include "defs.h"
+
#include <glib.h>
+#include <glib/gi18n.h>
#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>
-#endif
+#include <iconv.h>
-#include <kcc.h>
-
-#include "intl.h"
#include "codeconv.h"
#include "unmime.h"
#include "base64.h"
+#include "quoted-printable.h"
#include "utils.h"
#include "prefs_common.h"
-#define iskanji(c) \
+typedef enum
+{
+ JIS_ASCII,
+ JIS_KANJI,
+ JIS_HWKANA,
+ JIS_AUXKANJI
+} JISState;
+
+#define SUBST_CHAR 0x5f;
+#define ESC '\033'
+
+#define iseuckanji(c) \
(((c) & 0xff) >= 0xa1 && ((c) & 0xff) <= 0xfe)
-#define iseucss(c) \
- (((c) & 0xff) == 0x8e || ((c) & 0xff) == 0x8f)
-#define isunprintablekanji(c) \
- (((c) & 0xff) >= 0xa9 && ((c) & 0xff) <= 0xaf)
+#define iseuchwkana1(c) \
+ (((c) & 0xff) == 0x8e)
+#define iseuchwkana2(c) \
+ (((c) & 0xff) >= 0xa1 && ((c) & 0xff) <= 0xdf)
+#define iseucaux(c) \
+ (((c) & 0xff) == 0x8f)
+#define issjiskanji1(c) \
+ ((((c) & 0xff) >= 0x81 && ((c) & 0xff) <= 0x9f) || \
+ (((c) & 0xff) >= 0xe0 && ((c) & 0xff) <= 0xfc))
+#define issjiskanji2(c) \
+ ((((c) & 0xff) >= 0x40 && ((c) & 0xff) <= 0x7e) || \
+ (((c) & 0xff) >= 0x80 && ((c) & 0xff) <= 0xfc))
+#define issjishwkana(c) \
+ (((c) & 0xff) >= 0xa1 && ((c) & 0xff) <= 0xdf)
+
+#define K_IN() \
+ if (state != JIS_KANJI) { \
+ *out++ = ESC; \
+ *out++ = '$'; \
+ *out++ = 'B'; \
+ state = JIS_KANJI; \
+ }
+
+#define K_OUT() \
+ if (state != JIS_ASCII) { \
+ *out++ = ESC; \
+ *out++ = '('; \
+ *out++ = 'B'; \
+ state = JIS_ASCII; \
+ }
+
+#define HW_IN() \
+ if (state != JIS_HWKANA) { \
+ *out++ = ESC; \
+ *out++ = '('; \
+ *out++ = 'I'; \
+ state = JIS_HWKANA; \
+ }
-void conv_jistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
+#define AUX_IN() \
+ if (state != JIS_AUXKANJI) { \
+ *out++ = ESC; \
+ *out++ = '$'; \
+ *out++ = '('; \
+ *out++ = 'D'; \
+ state = JIS_AUXKANJI; \
+ }
+
+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 void conv_jistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- KCC_filter(outbuf, "EUC", (gchar *)inbuf, "JISBB", 0, 0, 0);
+ const guchar *in = inbuf;
+ guchar *out = outbuf;
+ JISState state = JIS_ASCII;
+
+ while (*in != '\0') {
+ if (*in == ESC) {
+ in++;
+ if (*in == '$') {
+ if (*(in + 1) == '@' || *(in + 1) == 'B') {
+ state = JIS_KANJI;
+ in += 2;
+ } else if (*(in + 1) == '(' &&
+ *(in + 2) == 'D') {
+ state = JIS_AUXKANJI;
+ in += 3;
+ } else {
+ /* unknown escape sequence */
+ state = JIS_ASCII;
+ }
+ } else if (*in == '(') {
+ if (*(in + 1) == 'B' || *(in + 1) == 'J') {
+ state = JIS_ASCII;
+ in += 2;
+ } else if (*(in + 1) == 'I') {
+ state = JIS_HWKANA;
+ in += 2;
+ } else {
+ /* unknown escape sequence */
+ state = JIS_ASCII;
+ }
+ } else {
+ /* unknown escape sequence */
+ state = JIS_ASCII;
+ }
+ } else if (*in == 0x0e) {
+ state = JIS_HWKANA;
+ in++;
+ } else if (*in == 0x0f) {
+ state = JIS_ASCII;
+ in++;
+ } else {
+ switch (state) {
+ case JIS_ASCII:
+ *out++ = *in++;
+ break;
+ case JIS_KANJI:
+ *out++ = *in++ | 0x80;
+ if (*in == '\0') break;
+ *out++ = *in++ | 0x80;
+ break;
+ case JIS_HWKANA:
+ *out++ = 0x8e;
+ *out++ = *in++ | 0x80;
+ break;
+ case JIS_AUXKANJI:
+ *out++ = 0x8f;
+ *out++ = *in++ | 0x80;
+ if (*in == '\0') break;
+ *out++ = *in++ | 0x80;
+ break;
+ }
+ }
+ }
+
+ *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)
{
- size_t inlen, len;
-
- inlen = strlen(inbuf);
- if (iskanji(inbuf[inlen - 1]) || iseucss(inbuf[inlen - 1])) {
- /* if tail end of the string is not ended with ascii,
- add dummy return code. */
- gchar *tmpin, *tmpout;
-
- /* length of original string + '\n' + '\0' */
- tmpin = alloca(inlen + 2);
- if (tmpin == NULL) {
- g_warning(_("can't allocate memory\n"));
- KCC_filter(outbuf, "JISBB", (gchar *)inbuf, "EUC",
- 0, 0, 0);
- return;
- }
- strcpy(tmpin, inbuf);
- tmpin[inlen] = '\n';
- tmpin[inlen + 1] = '\0';
-
- tmpout = alloca(outlen + 1);
- if (tmpout == NULL) {
- g_warning(_("can't allocate memory\n"));
- KCC_filter(outbuf, "JISBB", (gchar *)inbuf, "EUC",
- 0, 0, 0);
- return;
+ 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;
}
+ }
- KCC_filter(tmpout, "JISBB", tmpin, "EUC", 0, 0, 0);
- len = strlen(tmpout);
- if (tmpout[len - 1] == '\n')
- tmpout[len - 1] = '\0';
- strncpy2(outbuf, tmpout, outlen);
- } else
- KCC_filter(outbuf, "JISBB", (gchar *)inbuf, "EUC", 0, 0, 0);
+ 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;
}
-void conv_sjistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_euctojis(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- KCC_filter(outbuf, "EUC", (gchar *)inbuf, "SJIS", 0, 0, 0);
+ const guchar *in = inbuf;
+ guchar *out = outbuf;
+ JISState state = JIS_ASCII;
+
+ while (*in != '\0') {
+ if (IS_ASCII(*in)) {
+ K_OUT();
+ *out++ = *in++;
+ } else if (iseuckanji(*in)) {
+ if (iseuckanji(*(in + 1))) {
+ K_IN();
+ *out++ = *in++ & 0x7f;
+ *out++ = *in++ & 0x7f;
+ } else {
+ K_OUT();
+ *out++ = SUBST_CHAR;
+ in++;
+ if (*in != '\0' && !IS_ASCII(*in)) {
+ *out++ = SUBST_CHAR;
+ in++;
+ }
+ }
+ } else if (iseuchwkana1(*in)) {
+ 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();
+ in++;
+ if (*in != '\0' && !IS_ASCII(*in)) {
+ *out++ = SUBST_CHAR;
+ in++;
+ }
+ }
+ } else if (iseucaux(*in)) {
+ in++;
+ if (iseuckanji(*in) && iseuckanji(*(in + 1))) {
+ AUX_IN();
+ *out++ = *in++ & 0x7f;
+ *out++ = *in++ & 0x7f;
+ } else {
+ K_OUT();
+ if (*in != '\0' && !IS_ASCII(*in)) {
+ *out++ = SUBST_CHAR;
+ in++;
+ if (*in != '\0' && !IS_ASCII(*in)) {
+ *out++ = SUBST_CHAR;
+ in++;
+ }
+ }
+ }
+ } else {
+ K_OUT();
+ *out++ = SUBST_CHAR;
+ in++;
+ }
+ }
+
+ K_OUT();
+ *out = '\0';
}
-void conv_anytoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_sjistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- KCC_filter(outbuf, "EUC", (gchar *)inbuf, "AUTO", 0, 0, 0);
+ const guchar *in = inbuf;
+ guchar *out = outbuf;
+
+ while (*in != '\0') {
+ if (IS_ASCII(*in)) {
+ *out++ = *in++;
+ } else if (issjiskanji1(*in)) {
+ if (issjiskanji2(*(in + 1))) {
+ guchar out1 = *in;
+ guchar out2 = *(in + 1);
+ guchar row;
+
+ row = out1 < 0xa0 ? 0x70 : 0xb0;
+ if (out2 < 0x9f) {
+ out1 = (out1 - row) * 2 - 1;
+ out2 -= out2 > 0x7f ? 0x20 : 0x1f;
+ } else {
+ out1 = (out1 - row) * 2;
+ out2 -= 0x7e;
+ }
+
+ *out++ = out1 | 0x80;
+ *out++ = out2 | 0x80;
+ in += 2;
+ } else {
+ *out++ = SUBST_CHAR;
+ in++;
+ if (*in != '\0' && !IS_ASCII(*in)) {
+ *out++ = SUBST_CHAR;
+ in++;
+ }
+ }
+ } else if (issjishwkana(*in)) {
+ *out++ = 0x8e;
+ *out++ = *in++;
+ } else {
+ *out++ = SUBST_CHAR;
+ in++;
+ }
+ }
+
+ *out = '\0';
}
-void conv_anytojis(gchar *outbuf, gint outlen, const gchar *inbuf)
+static void conv_jistoutf8(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- KCC_filter(outbuf, "JISBB", (gchar *)inbuf, "AUTO", 0, 0, 0);
+ gchar *eucstr;
+
+ Xalloca(eucstr, outlen, return);
+
+ conv_jistoeuc(eucstr, outlen, inbuf);
+ conv_euctoutf8(outbuf, outlen, eucstr);
}
-#define SUBST_CHAR '_'
+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);
+}
-void conv_unreadable_eucjp(gchar *str)
+static void conv_euctoutf8(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- register guchar *p = str;
+ static iconv_t cd = (iconv_t)-1;
+ static gboolean iconv_ok = TRUE;
+ gchar *tmpstr;
- 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 (iskanji(*p)) {
- if (iskanji(*(p + 1)) && !isunprintablekanji(*p))
- /* printable euc-jp code */
- p += 2;
- else {
- /* substitute unprintable code */
- *p++ = SUBST_CHAR;
- if (*p != '\0') {
- if (isascii(*p))
- p++;
- else
- *p++ = SUBST_CHAR;
- }
+ 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;
}
- } else if (iseucss(*p)) {
- if ((*(p + 1) & 0x80) != 0)
- /* euc-jp hankaku kana */
- p += 2;
- else
- *p++ = SUBST_CHAR;
- } else
- /* substitute unprintable 1 byte code */
- *p++ = SUBST_CHAR;
+ }
}
+
+ 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_8bit(gchar *str)
+static void conv_anytoutf8(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- register guchar *p = str;
+ 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_euctoutf8(outbuf, outlen, inbuf);
+ break;
+ default:
+ strncpy2(outbuf, inbuf, outlen);
+ break;
+ }
+}
- 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++;
+static void conv_utf8toeuc(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_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);
}
-void conv_unreadable_latin(gchar *str)
+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 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) >= 0x80 && (*p & 0xff) <= 0x9f)
- *p = SUBST_CHAR;
+ else if (!IS_ASCII(*p)) *p = SUBST_CHAR;
p++;
}
}
p += 2;
len -= 2;
}
- } else if (iskanji(*p)) {
+ } else if (iseuckanji(*p)) {
p += 2;
len -= 2;
} else {
}
}
-void conv_jistodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
+CharSet conv_guess_ja_encoding(const gchar *str)
+{
+ const guchar *p = str;
+ CharSet guessed = C_US_ASCII;
+
+ while (*p != '\0') {
+ if (*p == ESC && (*(p + 1) == '$' || *(p + 1) == '(')) {
+ if (guessed == C_US_ASCII)
+ return C_ISO_2022_JP;
+ p += 2;
+ } else if (IS_ASCII(*p)) {
+ p++;
+ } else if (iseuckanji(*p) && iseuckanji(*(p + 1))) {
+ if (*p >= 0xfd && *p <= 0xfe)
+ return C_EUC_JP;
+ else if (guessed == C_SHIFT_JIS) {
+ if ((issjiskanji1(*p) &&
+ issjiskanji2(*(p + 1))) ||
+ issjishwkana(*p))
+ guessed = C_SHIFT_JIS;
+ else
+ guessed = C_EUC_JP;
+ } else
+ guessed = C_EUC_JP;
+ p += 2;
+ } else if (issjiskanji1(*p) && issjiskanji2(*(p + 1))) {
+ if (iseuchwkana1(*p) && iseuchwkana2(*(p + 1)))
+ guessed = C_SHIFT_JIS;
+ else
+ return C_SHIFT_JIS;
+ p += 2;
+ } else if (issjishwkana(*p)) {
+ guessed = C_SHIFT_JIS;
+ p++;
+ } else {
+ p++;
+ }
+ }
+
+ return guessed;
+}
+
+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);
+}
+
+void conv_utf8todisp(gchar *outbuf, gint outlen, const gchar *inbuf)
+{
+ if (g_utf8_validate(inbuf, -1, NULL) == TRUE)
+ strncpy2(outbuf, inbuf, outlen);
+ else
+ conv_ustodisp(outbuf, outlen, 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);
}
-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;
+
+ tmpstr = conv_iconv_strdup(inbuf, conv_get_locale_charset_str(),
+ CS_INTERNAL);
+ if (tmpstr && g_utf8_validate(tmpstr, -1, NULL)) {
+ strncpy2(outbuf, tmpstr, outlen);
+ g_free(tmpstr);
+ } else if (tmpstr && !g_utf8_validate(tmpstr, -1, NULL)) {
+ g_free(tmpstr);
+ conv_utf8todisp(outbuf, outlen, inbuf);
+ } else
+ conv_utf8todisp(outbuf, outlen, inbuf);
}
-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)
+static const gchar *
+conv_get_fallback_for_private_encoding(const gchar *encoding)
+{
+ 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 *src_charset)
{
CodeConverter *conv;
+ src_charset = conv_get_fallback_for_private_encoding(src_charset);
+
conv = g_new0(CodeConverter, 1);
-#if !HAVE_LIBJCONV
- conv->code_conv_func = conv_get_code_conv_func(charset);
-#endif
- 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_LIBJCONV
- gchar *str;
-
- str = conv_codeset_strdup(inbuf, conv->charset_str, NULL);
- if (!str)
- return -1;
+ 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_iconv_strdup(inbuf, conv->charset_str, NULL);
+ if (!str)
+ return -1;
+ else {
+ strncpy2(outbuf, str, outlen);
+ g_free(str);
+ }
}
-#else /* !HAVE_LIBJCONV */
- conv->code_conv_func(outbuf, outlen, inbuf);
-#endif
return 0;
}
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;
+ 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;
+ 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)
- 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);
}
-#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();
-
- if (jconv_alloc_conv(inbuf, strlen(inbuf), &buf, &len,
- codesets, n_codesets,
- &actual_codeset, dest_codeset)
- == 0)
- return buf;
+ return conv_iconv_strdup(inbuf, src_code, dest_code);
+}
+
+CodeConvFunc conv_get_code_conv_func(const gchar *src_charset_str,
+ const gchar *dest_charset_str)
+{
+ CodeConvFunc code_conv = conv_noconv;
+ CharSet src_charset;
+ CharSet dest_charset;
+
+ if (!src_charset_str)
+ src_charset = conv_get_locale_charset();
else
- return NULL;
-#else /* !HAVE_LIBJCONV */
- if (src_codeset) {
- if (!strcasecmp(src_codeset, CS_EUC_JP) ||
- !strcasecmp(src_codeset, "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;
+ src_charset = conv_get_charset_from_str(src_charset_str);
+
+ /* auto detection mode */
+ if (!src_charset_str && !dest_charset_str) {
+ if (conv_is_ja_locale())
+ return conv_anytodisp;
+ else
+ return conv_noconv;
}
- 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);
+ dest_charset = conv_get_charset_from_str(dest_charset_str);
+
+ if (dest_charset == C_US_ASCII)
+ return conv_ustodisp;
+
+ switch (src_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:
+ 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)
+ 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)
+ code_conv = conv_euctodisp;
+ 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 buf;
-#endif /* !HAVE_LIBJCONV */
+ return code_conv;
}
-CodeConvFunc conv_get_code_conv_func(const gchar *charset)
+gchar *conv_iconv_strdup(const gchar *inbuf,
+ const gchar *src_code, const gchar *dest_code)
{
- CodeConvFunc code_conv;
+ iconv_t cd;
+ gchar *outbuf;
- if (!charset) {
- if (conv_get_outgoing_charset() == C_ISO_2022_JP)
- return conv_jistodisp;
- else
- return conv_noconv;
+ if (!src_code)
+ src_code = conv_get_outgoing_charset_str();
+ if (!dest_code)
+ 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;
+ out_size = (in_size + 1) * 2;
+ outbuf = g_malloc(out_size);
+ outbuf_p = outbuf;
+ out_left = out_size;
+
+#define EXPAND_BUF() \
+{ \
+ len = outbuf_p - outbuf; \
+ out_size *= 2; \
+ outbuf = g_realloc(outbuf, out_size); \
+ outbuf_p = outbuf + len; \
+ out_left = out_size - len; \
+}
+
+ while ((n_conv = iconv(cd, (ICONV_CONST gchar **)&inbuf_p, &in_left,
+ &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) {
+ EXPAND_BUF();
+ }
+ *outbuf_p++ = SUBST_CHAR;
+ out_left--;
+ } else if (EINVAL == errno) {
+ break;
+ } else if (E2BIG == errno) {
+ EXPAND_BUF();
+ } else {
+ g_warning("conv_iconv_strdup(): %s\n",
+ g_strerror(errno));
+ break;
+ }
}
- 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;
-#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, "EUCJP"))
- code_conv = conv_euctodisp;
- else
- code_conv = conv_noconv;
+ while ((n_conv = iconv(cd, NULL, NULL, &outbuf_p, &out_left)) ==
+ (size_t)-1) {
+ if (E2BIG == errno) {
+ EXPAND_BUF();
+ } else {
+ g_warning("conv_iconv_strdup(): %s\n",
+ g_strerror(errno));
+ break;
+ }
+ }
- return code_conv;
+#undef EXPAND_BUF
+
+ len = outbuf_p - outbuf;
+ outbuf = g_realloc(outbuf, len + 1);
+ outbuf[len] = '\0';
+
+ return outbuf;
}
static const struct {
gchar *const name;
} charsets[] = {
{C_US_ASCII, CS_US_ASCII},
+ {C_US_ASCII, CS_ANSI_X3_4_1968},
{C_UTF_8, CS_UTF_8},
+ {C_UTF_7, CS_UTF_7},
{C_ISO_8859_1, CS_ISO_8859_1},
{C_ISO_8859_2, CS_ISO_8859_2},
+ {C_ISO_8859_3, CS_ISO_8859_3},
{C_ISO_8859_4, CS_ISO_8859_4},
{C_ISO_8859_5, CS_ISO_8859_5},
+ {C_ISO_8859_6, CS_ISO_8859_6},
{C_ISO_8859_7, CS_ISO_8859_7},
{C_ISO_8859_8, CS_ISO_8859_8},
{C_ISO_8859_9, CS_ISO_8859_9},
+ {C_ISO_8859_10, CS_ISO_8859_10},
+ {C_ISO_8859_11, CS_ISO_8859_11},
{C_ISO_8859_13, CS_ISO_8859_13},
+ {C_ISO_8859_14, CS_ISO_8859_14},
{C_ISO_8859_15, CS_ISO_8859_15},
{C_BALTIC, CS_BALTIC},
+ {C_CP1250, CS_CP1250},
{C_CP1251, CS_CP1251},
+ {C_CP1252, CS_CP1252},
+ {C_CP1253, CS_CP1253},
+ {C_CP1254, CS_CP1254},
+ {C_CP1255, CS_CP1255},
+ {C_CP1256, CS_CP1256},
+ {C_CP1257, CS_CP1257},
+ {C_CP1258, CS_CP1258},
+ {C_WINDOWS_1250, CS_WINDOWS_1250},
{C_WINDOWS_1251, CS_WINDOWS_1251},
+ {C_WINDOWS_1252, CS_WINDOWS_1252},
+ {C_WINDOWS_1253, CS_WINDOWS_1253},
+ {C_WINDOWS_1254, CS_WINDOWS_1254},
+ {C_WINDOWS_1255, CS_WINDOWS_1255},
+ {C_WINDOWS_1256, CS_WINDOWS_1256},
+ {C_WINDOWS_1257, CS_WINDOWS_1257},
+ {C_WINDOWS_1258, CS_WINDOWS_1258},
{C_KOI8_R, CS_KOI8_R},
+ {C_KOI8_T, CS_KOI8_T},
{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},
{C_ISO_2022_KR, CS_ISO_2022_KR},
{C_EUC_KR, CS_EUC_KR},
{C_ISO_2022_CN, CS_ISO_2022_CN},
{C_EUC_CN, CS_EUC_CN},
{C_GB2312, CS_GB2312},
+ {C_GBK, CS_GBK},
{C_EUC_TW, CS_EUC_TW},
{C_BIG5, CS_BIG5},
+ {C_BIG5_HKSCS, CS_BIG5_HKSCS},
+ {C_TIS_620, CS_TIS_620},
+ {C_WINDOWS_874, CS_WINDOWS_874},
+ {C_GEORGIAN_PS, CS_GEORGIAN_PS},
+ {C_TCVN5712_1, CS_TCVN5712_1},
};
-#if !HAVE_LIBJCONV
static const struct {
gchar *const locale;
CharSet charset;
+ CharSet out_charset;
} locale_table[] = {
- {"ja_JP.eucJP" , C_EUC_JP},
- {"ja_JP.ujis" , C_EUC_JP},
- {"ja_JP.EUC" , C_EUC_JP},
- {"ja_JP.SJIS" , C_SHIFT_JIS},
- {"ja_JP.JIS" , C_ISO_2022_JP},
- {"ja_JP" , C_EUC_JP},
- {"ko_KR" , C_EUC_KR},
- {"zh_CN.GB2312" , C_GB2312},
- {"zh_CN" , C_GB2312},
- {"zh_TW.eucTW" , C_EUC_TW},
- {"zh_TW.Big5" , C_BIG5},
- {"zh_TW" , C_BIG5},
-
- {"ru_RU.KOI8-R" , C_KOI8_R},
- {"ru_RU.CP1251" , C_CP1251},
-
- {"en_US" , C_ISO_8859_1},
- {"ca_ES" , C_ISO_8859_1},
- {"da_DK" , C_ISO_8859_1},
- {"de_DE" , C_ISO_8859_1},
- {"nl_NL" , C_ISO_8859_1},
- {"et_EE" , C_ISO_8859_1},
- {"fi_FI" , C_ISO_8859_1},
- {"fr_FR" , C_ISO_8859_1},
- {"is_IS" , C_ISO_8859_1},
- {"it_IT" , C_ISO_8859_1},
- {"no_NO" , C_ISO_8859_1},
- {"pt_PT" , C_ISO_8859_1},
- {"pt_BR" , C_ISO_8859_1},
- {"es_ES" , C_ISO_8859_1},
- {"sv_SE" , C_ISO_8859_1},
-
- {"hr_HR" , C_ISO_8859_2},
- {"hu_HU" , C_ISO_8859_2},
- {"pl_PL" , C_ISO_8859_2},
- {"ro_RO" , C_ISO_8859_2},
- {"sk_SK" , C_ISO_8859_2},
- {"sl_SI" , C_ISO_8859_2},
- {"ru_RU" , C_ISO_8859_5},
- {"el_GR" , C_ISO_8859_7},
- {"iw_IL" , C_ISO_8859_8},
- {"tr_TR" , C_ISO_8859_9},
-
- {"lt_LT.iso88594" , C_ISO_8859_4},
- {"lt_LT.ISO8859-4" , C_ISO_8859_4},
- {"lt_LT.ISO_8859-4" , C_ISO_8859_4},
- {"lt_LT" , C_ISO_8859_13},
- {"lv_LV" , C_ISO_8859_13},
-
- {"C" , C_US_ASCII},
- {"POSIX" , C_US_ASCII},
+ {"ja_JP.eucJP" , C_EUC_JP , C_ISO_2022_JP},
+ {"ja_JP.EUC-JP" , C_EUC_JP , C_ISO_2022_JP},
+ {"ja_JP.EUC" , C_EUC_JP , C_ISO_2022_JP},
+ {"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_GBK},
+ {"zh_CN" , C_GB2312 , C_GB2312},
+ {"zh_HK" , C_BIG5_HKSCS , C_BIG5_HKSCS},
+ {"zh_TW.eucTW" , C_EUC_TW , C_BIG5},
+ {"zh_TW.EUC-TW" , C_EUC_TW , C_BIG5},
+ {"zh_TW.Big5" , C_BIG5 , C_BIG5},
+ {"zh_TW" , C_BIG5 , C_BIG5},
+
+ {"ru_RU.KOI8-R" , C_KOI8_R , C_KOI8_R},
+ {"ru_RU.KOI8R" , C_KOI8_R , C_KOI8_R},
+ {"ru_RU.CP1251" , C_WINDOWS_1251, C_KOI8_R},
+ {"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},
+ {"bg_BG" , C_WINDOWS_1251, C_WINDOWS_1251},
+
+ {"yi_US" , C_WINDOWS_1255, C_WINDOWS_1255},
+
+ {"af_ZA" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"br_FR" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"ca_ES" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"da_DK" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"de_AT" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"de_BE" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"de_CH" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"de_DE" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"de_LU" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"en_AU" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"en_BW" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"en_CA" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"en_DK" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"en_GB" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"en_HK" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"en_IE" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"en_NZ" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"en_PH" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"en_SG" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"en_US" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"en_ZA" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"en_ZW" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_AR" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_BO" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_CL" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_CO" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_CR" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_DO" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_EC" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_ES" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_GT" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_HN" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_MX" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_NI" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_PA" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_PE" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_PR" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_PY" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_SV" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_US" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_UY" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_VE" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"et_EE" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"eu_ES" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"fi_FI" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"fo_FO" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"fr_BE" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"fr_CA" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"fr_CH" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"fr_FR" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"fr_LU" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"ga_IE" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"gl_ES" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"gv_GB" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"id_ID" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"is_IS" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"it_CH" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"it_IT" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"kl_GL" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"kw_GB" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"ms_MY" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"nl_BE" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"nl_NL" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"nn_NO" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"no_NO" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"oc_FR" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"pt_BR" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"pt_PT" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"sq_AL" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"sv_FI" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"sv_SE" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"tl_PH" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"uz_UZ" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"wa_BE" , C_ISO_8859_1 , C_ISO_8859_1},
+
+ {"bs_BA" , C_ISO_8859_2 , C_ISO_8859_2},
+ {"cs_CZ" , C_ISO_8859_2 , C_ISO_8859_2},
+ {"hr_HR" , C_ISO_8859_2 , C_ISO_8859_2},
+ {"hu_HU" , C_ISO_8859_2 , C_ISO_8859_2},
+ {"pl_PL" , C_ISO_8859_2 , C_ISO_8859_2},
+ {"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},
+
+ {"sr_YU@cyrillic" , C_ISO_8859_5 , C_ISO_8859_5},
+ {"sr_YU" , C_ISO_8859_2 , C_ISO_8859_2},
+
+ {"mt_MT" , C_ISO_8859_3 , C_ISO_8859_3},
+
+ {"lt_LT.iso88594" , C_ISO_8859_4 , C_ISO_8859_4},
+ {"lt_LT.ISO8859-4" , C_ISO_8859_4 , C_ISO_8859_4},
+ {"lt_LT.ISO_8859-4" , C_ISO_8859_4 , C_ISO_8859_4},
+ {"lt_LT" , C_ISO_8859_13 , C_ISO_8859_13},
+
+ {"mk_MK" , C_ISO_8859_5 , C_ISO_8859_5},
+
+ {"ar_AE" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_BH" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_DZ" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_EG" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_IQ" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_JO" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_KW" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_LB" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_LY" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_MA" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_OM" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_QA" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_SA" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_SD" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_SY" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_TN" , C_ISO_8859_6 , C_ISO_8859_6},
+ {"ar_YE" , C_ISO_8859_6 , C_ISO_8859_6},
+
+ {"el_GR" , C_ISO_8859_7 , C_ISO_8859_7},
+ {"he_IL" , C_ISO_8859_8 , C_ISO_8859_8},
+ {"iw_IL" , C_ISO_8859_8 , C_ISO_8859_8},
+ {"tr_TR" , C_ISO_8859_9 , C_ISO_8859_9},
+
+ {"lv_LV" , C_ISO_8859_13 , C_ISO_8859_13},
+ {"mi_NZ" , C_ISO_8859_13 , C_ISO_8859_13},
+
+ {"cy_GB" , C_ISO_8859_14 , C_ISO_8859_14},
+
+ {"ar_IN" , C_UTF_8 , C_UTF_8},
+ {"en_IN" , C_UTF_8 , C_UTF_8},
+ {"se_NO" , C_UTF_8 , C_UTF_8},
+ {"ta_IN" , C_UTF_8 , C_UTF_8},
+ {"te_IN" , C_UTF_8 , C_UTF_8},
+ {"ur_PK" , C_UTF_8 , C_UTF_8},
+
+ {"th_TH" , C_TIS_620 , C_TIS_620},
+ /* {"th_TH" , C_WINDOWS_874}, */
+ /* {"th_TH" , C_ISO_8859_11}, */
+
+ {"ka_GE" , C_GEORGIAN_PS , C_GEORGIAN_PS},
+ {"vi_VN.TCVN" , C_TCVN5712_1 , C_TCVN5712_1},
+
+ {"C" , C_US_ASCII , C_US_ASCII},
+ {"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 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;
}
-CharSet conv_get_current_charset(void)
+const gchar *conv_get_charset_str(CharSet charset)
{
- static CharSet cur_charset = -1;
- gint i;
+ GHashTable *table;
-#if HAVE_LIBJCONV
- const gchar *cur_codeset;
-#else
+ 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_locale_charset(void)
+{
+ static CharSet cur_charset = -1;
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 = 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);
-
- debug_print("current locale: %s\n",
- cur_locale ? cur_locale : "(none)");
-
+ cur_locale = conv_get_current_locale();
if (!cur_locale) {
cur_charset = C_US_ASCII;
return cur_charset;
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" */
- if (!strncasecmp(cur_locale, locale_table[i].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 (!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;
}
}
}
-#endif
cur_charset = C_AUTO;
return cur_charset;
}
-const gchar *conv_get_current_charset_str(void)
+const gchar *conv_get_locale_charset_str(void)
{
-#if HAVE_LIBJCONV
- return jconv_info_get_current_codeset();
-#else
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 : "US-ASCII";
-#endif
+ 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)
{
static CharSet out_charset = -1;
-
-#if HAVE_LIBJCONV
- gint i, j, n_pref_codesets;
- const gchar *const *pref_codesets;
-#else
- CharSet cur_charset;
-#endif
+ const gchar *cur_locale;
+ 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 = 2; j < sizeof(charsets) / sizeof(charsets[0]); j++) {
- if (!strcasecmp(pref_codesets[i], charsets[j].name)) {
- out_charset = charsets[j].charset;
- return out_charset;
- }
- }
+ cur_locale = conv_get_current_locale();
+ if (!cur_locale) {
+ out_charset = C_AUTO;
+ 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;
- }
+ if (strcasestr(cur_locale, "UTF-8")) {
+ out_charset = C_UTF_8;
+ return out_charset;
}
- out_charset = C_AUTO;
-#else
- cur_charset = conv_get_current_charset();
- switch (cur_charset) {
- case C_EUC_JP:
- case C_SHIFT_JIS:
- out_charset = C_ISO_2022_JP;
- break;
- default:
- out_charset = cur_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 (!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 &&
+ !g_ascii_strncasecmp(cur_locale, locale_table[i].locale, 2)) {
+ out_charset = locale_table[i].out_charset;
+ break;
+ }
+ }
}
-#endif
return out_charset;
}
CharSet out_charset;
const gchar *str;
- if (prefs_common.outgoing_charset) {
- if (!isalpha(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 : "US-ASCII";
+ return str ? str : CS_UTF_8;
}
-void conv_unmime_header_overwrite(gchar *str)
+gboolean conv_is_multibyte_encoding(CharSet encoding)
{
- gchar *buf;
- gint outlen;
- CharSet cur_charset;
-
- cur_charset = conv_get_current_charset();
-
-#if HAVE_LIBJCONV
- Xstrdup_a(buf, str, return);
- outlen = strlen(str) + 1;
- UnMimeHeaderConv(buf, str, outlen);
- if (cur_charset == C_EUC_JP) {
- gchar *tmp;
- gint len;
-
- len = strlen(str) * 2 + 1;
- Xalloca(tmp, len, return);
- conv_jistodisp(tmp, len, str);
- strncpy2(str, tmp, outlen);
+ 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:
+ return TRUE;
+ default:
+ return FALSE;
}
-#else
- if (cur_charset == C_EUC_JP) {
- gchar *tmp;
- gint len;
-
- Xstrdup_a(buf, str, {return;});
- outlen = strlen(str) + 1;
- UnMimeHeader(buf);
- len = strlen(buf) * 2 + 1;
- Xalloca(tmp, len, {strncpy2(str, buf, outlen); return;});
- conv_jistodisp(tmp, len, buf);
- strncpy2(str, tmp, outlen);
- } else
- UnMimeHeader(str);
-#endif
}
-void conv_unmime_header(gchar *outbuf, gint outlen, const gchar *str,
- const gchar *charset)
+const gchar *conv_get_current_locale(void)
{
- gchar *buf;
- CharSet cur_charset;
-
- Xstrdup_a(buf, str, return);
- cur_charset = conv_get_current_charset();
-
-#if HAVE_LIBJCONV
- UnMimeHeaderConv(buf, outbuf, outlen);
- if (cur_charset == C_EUC_JP) {
- gchar *tmp;
- gint len;
+ const gchar *cur_locale;
- len = strlen(outbuf) * 2 + 1;
- Xalloca(tmp, len, return);
- conv_jistodisp(tmp, len, outbuf);
- strncpy2(outbuf, tmp, outlen);
- }
+#ifdef G_OS_WIN32
+ cur_locale = g_win32_getlocale();
#else
- UnMimeHeader(buf);
- if (cur_charset == C_EUC_JP) {
- gchar *tmp;
- gint len;
-
- len = strlen(buf) * 2 + 1;
- Xalloca(tmp, len, {strncpy2(outbuf, buf, outlen); return;});
- conv_jistodisp(tmp, len, buf);
- strncpy2(outbuf, tmp, outlen);
- } else
- strncpy2(outbuf, buf, outlen);
-#endif
-}
+ 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 */
-#define MAX_ENCLEN 75
-#define MAX_LINELEN 76
+ debug_print("current locale: %s\n",
+ cur_locale ? cur_locale : "(none)");
-#define B64LEN(len) ((len) / 3 * 4 + ((len) % 3 ? 4 : 0))
+ return cur_locale;
+}
-#if HAVE_LIBJCONV
-void conv_encode_header(gchar *dest, gint len, const gchar *src,
- gint header_len)
+gboolean conv_is_ja_locale(void)
{
- wchar_t *wsrc;
- wchar_t *wsrcp;
- gchar *destp;
- size_t line_len, mimehdr_len, mimehdr_begin_len;
- gchar *mimehdr_init = "=?";
- gchar *mimehdr_end = "?=";
- gchar *mimehdr_enctype = "?B?";
- const gchar *mimehdr_charset;
-
- /* g_print("src = %s\n", src); */
- mimehdr_charset = conv_get_outgoing_charset_str();
-
- /* convert to wide-character string */
- wsrcp = wsrc = strdup_mbstowcs(src);
-
- mimehdr_len = strlen(mimehdr_init) + strlen(mimehdr_end) +
- strlen(mimehdr_charset) + strlen(mimehdr_enctype);
- mimehdr_begin_len = strlen(mimehdr_init) +
- strlen(mimehdr_charset) + strlen(mimehdr_enctype);
- /* line_len = 1; */
- line_len = header_len;
- destp = dest;
- *dest = '\0';
-
- g_return_if_fail(wsrc != NULL);
-
- while (*wsrcp) {
- wchar_t *wp, *wtmp, *wtmpp;
- gint nspc = 0;
-
- /* irresponsible buffer overrun check */
- if ((len - (destp - dest)) < (MAX_LINELEN + 1) * 2) break;
-
- /* encode string including space
- if non-ASCII string follows */
- if (is_next_nonascii(wsrcp)) {
- wp = wsrcp;
- while ((wp = find_wspace(wp)) != NULL)
- if (!is_next_nonascii(wp)) break;
- } else
- wp = find_wspace(wsrcp);
-
- if (wp != NULL) {
- wtmp = wcsndup(wsrcp, wp - wsrcp);
- wsrcp = wp + 1;
- while (iswspace(wsrcp[nspc])) nspc++;
- } else {
- wtmp = wcsdup(wsrcp);
- wsrcp += wcslen(wsrcp);
- }
-
- wtmpp = wtmp;
-
- do {
- gint tlen = 0, str_ascii = 1;
- gchar *tmp; /* internal codeset */
- gchar *raw; /* converted, but not base64 encoded */
- register gchar *tmpp;
- gint raw_len;
-
- tmpp = tmp = g_malloc(wcslen(wtmpp) * MB_CUR_MAX + 1);
- *tmp = '\0';
- raw = g_strdup("");
- raw_len = 0;
-
- while (*wtmpp != (wchar_t)0) {
- gint mbl;
- gint dummy;
- gchar *raw_new = NULL;
- int raw_new_len = 0;
- const gchar *src_codeset;
-
- if (*wtmpp < 32 || *wtmpp >= 127)
- str_ascii = 0;
- mbl = wctomb(tmpp, *wtmpp);
- if (mbl == -1) {
- g_warning("invalid wide character\n");
- wtmpp++;
- continue;
- }
- /* g_free(raw); */
- src_codeset = conv_get_current_charset_str();
- /* printf ("tmp = %s, tlen = %d, mbl\n",
- tmp, tlen, mbl); */
- if (jconv_alloc_conv(tmp, tlen + mbl,
- &raw_new, &raw_new_len,
- &src_codeset, 1,
- &dummy, mimehdr_charset)
- != 0) {
- g_warning("can't convert\n");
- tmpp[0] = '\0';
- continue;
- }
- if (!str_ascii) {
- gint dlen = mimehdr_len +
- B64LEN(raw_len);
- if ((line_len + dlen +
- (*(wtmpp + 1) ? 0 : nspc) +
- (line_len > 1 ? 1 : 0))
- > MAX_LINELEN) {
- g_free(raw_new);
- if (tlen == 0) {
- *destp++ = '\n';
- *destp++ = ' ';
- line_len = 1;
- str_ascii = 1;
- continue;
- } else {
- *tmpp = '\0';
- break;
- }
- }
- } else if ((line_len + tlen + mbl +
- (*(wtmpp + 1) ? 0 : nspc) +
- (line_len > 1 ? 1 : 0))
- > MAX_LINELEN) {
- g_free(raw_new);
- if (1 + tlen + mbl +
- (*(wtmpp + 1) ? 0 : nspc)
- >= MAX_LINELEN) {
- *tmpp = '\0';
- break;
- }
- *destp++ = '\n';
- *destp++ = ' ';
- line_len = 1;
- continue;
- }
-
- tmpp += mbl;
- *tmpp = '\0';
-
- tlen += mbl;
-
- g_free(raw);
- raw = raw_new;
- raw_len = raw_new_len;
+ static gint is_ja_locale = -1;
+ const gchar *cur_locale;
- wtmpp++;
- }
- /*
- g_print("tmp = %s, tlen = %d, mb_seqlen = %d\n",
- tmp, tlen, mb_seqlen);
- */
-
- if (tlen == 0 || raw_len == 0) {
- g_free(tmp);
- g_free(raw);
- continue;
- }
+ if (is_ja_locale != -1)
+ return is_ja_locale != 0;
- if (line_len > 1 && destp > dest) {
- *destp++ = ' ';
- *destp = '\0';
- line_len++;
- }
+ 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;
+ }
- if (!str_ascii) {
- g_snprintf(destp, len - strlen(dest), "%s%s%s",
- mimehdr_init, mimehdr_charset,
- mimehdr_enctype);
- destp += mimehdr_begin_len;
- line_len += mimehdr_begin_len;
+ return is_ja_locale != 0;
+}
- to64frombits(destp, raw, raw_len);
- line_len += strlen(destp);
- destp += strlen(destp);
+gchar *conv_unmime_header(const gchar *str, const gchar *default_encoding)
+{
+ gchar buf[BUFFSIZE];
- strcpy(destp, mimehdr_end);
- destp += strlen(mimehdr_end);
- line_len += strlen(mimehdr_end);
- } else {
- strcpy(destp, tmp);
- line_len += strlen(destp);
- destp += strlen(destp);
- }
+ if (is_ascii_str(str))
+ return unmime_header(str);
- g_free(tmp);
- g_free(raw);
- /* g_print("line_len = %d\n\n", line_len); */
- } while (*wtmpp != (wchar_t)0);
+ if (default_encoding) {
+ gchar *utf8_buf;
- while (iswspace(*wsrcp)) {
- gint mbl;
+ utf8_buf = conv_codeset_strdup
+ (str, default_encoding, CS_INTERNAL);
+ if (utf8_buf) {
+ gchar *decoded_str;
- mbl = wctomb(destp, *wsrcp++);
- if (mbl != -1) {
- destp += mbl;
- line_len += mbl;
- }
+ decoded_str = unmime_header(utf8_buf);
+ g_free(utf8_buf);
+ return decoded_str;
}
- *destp = '\0';
-
- g_free(wtmp);
}
- g_free(wsrc);
+ if (conv_is_ja_locale())
+ conv_anytodisp(buf, sizeof(buf), str);
+ else
+ conv_localetodisp(buf, sizeof(buf), str);
- /* g_print("dest = %s\n", dest); */
+ return unmime_header(buf);
}
-#else /* !HAVE_LIBJCONV */
-#define JIS_SEQLEN 3
+#define MAX_LINELEN 76
+#define MAX_HARD_LINELEN 996
+#define MIMESEP_BEGIN "=?"
+#define MIMESEP_END "?="
+
+#define LBREAK_IF_REQUIRED(cond, is_plain_text) \
+{ \
+ if (len - (destp - (guchar *)dest) < MAX_LINELEN + 2) { \
+ *destp = '\0'; \
+ return; \
+ } \
+ \
+ if ((cond) && *srcp) { \
+ if (destp > (guchar *)dest && left < MAX_LINELEN - 1) { \
+ if (isspace(*(destp - 1))) \
+ destp--; \
+ else if (is_plain_text && isspace(*srcp)) \
+ srcp++; \
+ if (*srcp) { \
+ *destp++ = '\n'; \
+ *destp++ = ' '; \
+ left = MAX_LINELEN - 1; \
+ } \
+ } \
+ } \
+}
-void conv_encode_header(gchar *dest, gint len, const gchar *src,
- gint header_len)
+void conv_encode_header_full(gchar *dest, gint len, const gchar *src,
+ gint header_len, gboolean addr_field,
+ const gchar *out_encoding_)
{
- wchar_t *wsrc;
- wchar_t *wsrcp;
- gchar *destp;
- size_t line_len, mimehdr_len, mimehdr_begin_len;
- gchar *mimehdr_init = "=?";
- gchar *mimehdr_end = "?=";
- gchar *mimehdr_enctype = "?B?";
- const gchar *mimehdr_charset;
-
- /* g_print("src = %s\n", src); */
- mimehdr_charset = conv_get_outgoing_charset_str();
- if (strcmp(mimehdr_charset, "ISO-2022-JP") != 0) {
- /* currently only supports Japanese */
- strncpy2(dest, src, len);
- return;
+ const gchar *cur_encoding;
+ const gchar *out_encoding;
+ gint mimestr_len;
+ gchar *mimesep_enc;
+ gint left;
+ 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?";
+ } else {
+ use_base64 = FALSE;
+ mimesep_enc = "?Q?";
}
- /* convert to wide-character string */
- wsrcp = wsrc = strdup_mbstowcs(src);
-
- mimehdr_len = strlen(mimehdr_init) + strlen(mimehdr_end) +
- strlen(mimehdr_charset) + strlen(mimehdr_enctype);
- mimehdr_begin_len = strlen(mimehdr_init) +
- strlen(mimehdr_charset) + strlen(mimehdr_enctype);
- /* line_len = 1; */
- line_len = header_len;
- destp = dest;
- *dest = '\0';
-
- g_return_if_fail(wsrc != NULL);
-
- while (*wsrcp) {
- wchar_t *wp, *wtmp, *wtmpp;
- gint nspc = 0;
-
- /* irresponsible buffer overrun check */
- if ((len - (destp - dest)) < (MAX_LINELEN + 1) * 2) break;
-
- /* encode string including space
- if non-ASCII string follows */
- if (is_next_nonascii(wsrcp)) {
- wp = wsrcp;
- while ((wp = find_wspace(wp)) != NULL)
- if (!is_next_nonascii(wp)) break;
- } else
- wp = find_wspace(wsrcp);
-
- if (wp != NULL) {
- wtmp = wcsndup(wsrcp, wp - wsrcp);
- wsrcp = wp + 1;
- while (iswspace(wsrcp[nspc])) nspc++;
- } else {
- wtmp = wcsdup(wsrcp);
- wsrcp += wcslen(wsrcp);
- }
-
- wtmpp = wtmp;
-
- do {
- gint prev_mbl = 1, tlen = 0, mb_seqlen = 0;
- gchar *tmp;
- register gchar *tmpp;
-
- tmpp = tmp = g_malloc(wcslen(wtmpp) * MB_CUR_MAX + 1);
- *tmp = '\0';
+ cur_encoding = CS_INTERNAL;
- while (*wtmpp != (wchar_t)0) {
- gint mbl;
+ if (out_encoding_)
+ out_encoding = out_encoding_;
+ else
+ out_encoding = conv_get_outgoing_charset_str();
- mbl = wctomb(tmpp, *wtmpp);
- if (mbl == -1) {
- g_warning("invalid wide character\n");
- wtmpp++;
- continue;
- }
+ if (!strcmp(out_encoding, CS_US_ASCII))
+ out_encoding = CS_ISO_8859_1;
- /* length of KI + KO */
- if (prev_mbl == 1 && mbl == 2)
- mb_seqlen += JIS_SEQLEN * 2;
-
- if (mb_seqlen) {
- gint dlen = mimehdr_len +
- B64LEN(tlen + mb_seqlen + mbl);
-
- if ((line_len + dlen +
- (*(wtmpp + 1) ? 0 : nspc) +
- (line_len > 1 ? 1 : 0))
- > MAX_LINELEN) {
- if (tlen == 0) {
- *destp++ = '\n';
- *destp++ = ' ';
- line_len = 1;
- mb_seqlen = 0;
- continue;
- } else {
- *tmpp = '\0';
- break;
- }
- }
- } else if ((line_len + tlen + mbl +
- (*(wtmpp + 1) ? 0 : nspc) +
- (line_len > 1 ? 1 : 0))
- > MAX_LINELEN) {
- if (1 + tlen + mbl +
- (*(wtmpp + 1) ? 0 : nspc)
- >= MAX_LINELEN) {
- *tmpp = '\0';
- break;
- }
- *destp++ = '\n';
- *destp++ = ' ';
- line_len = 1;
- continue;
- }
+ mimestr_len = strlen(MIMESEP_BEGIN) + strlen(out_encoding) +
+ strlen(mimesep_enc) + strlen(MIMESEP_END);
- tmpp += mbl;
- *tmpp = '\0';
+ left = MAX_LINELEN - header_len;
- tlen += mbl;
- prev_mbl = mbl;
+ while (*srcp) {
+ LBREAK_IF_REQUIRED(left <= 0, TRUE);
- wtmpp++;
- }
- /* g_print("tmp = %s, tlen = %d, mb_seqlen = %d\n",
- tmp, tlen, mb_seqlen);
- */
+ while (isspace(*srcp)) {
+ *destp++ = *srcp++;
+ left--;
+ LBREAK_IF_REQUIRED(left <= 0, TRUE);
+ }
- if (tlen == 0) {
- g_free(tmp);
- continue;
+ /* output as it is if the next word is ASCII string */
+ if (!is_next_nonascii(srcp)) {
+ gint word_len;
+
+ word_len = get_next_word_len(srcp);
+ LBREAK_IF_REQUIRED(left < word_len, TRUE);
+ while (word_len > 0) {
+ LBREAK_IF_REQUIRED(left + (MAX_HARD_LINELEN - MAX_LINELEN) <= 0, TRUE)
+ *destp++ = *srcp++;
+ left--;
+ word_len--;
}
- if (line_len > 1 && destp > dest) {
- *destp++ = ' ';
- *destp = '\0';
- line_len++;
- }
+ continue;
+ }
- if (mb_seqlen) {
- gchar *tmp_jis;
+ /* don't include parentheses in encoded strings */
+ if (addr_field && (*srcp == '(' || *srcp == ')')) {
+ LBREAK_IF_REQUIRED(left < 2, FALSE);
+ *destp++ = *srcp++;
+ left--;
+ }
- tmp_jis = g_new(gchar, tlen + mb_seqlen + 1);
- conv_euctojis(tmp_jis,
- tlen + mb_seqlen + 1, tmp);
- g_snprintf(destp, len - strlen(dest), "%s%s%s",
- mimehdr_init, mimehdr_charset,
- mimehdr_enctype);
- destp += mimehdr_begin_len;
- line_len += mimehdr_begin_len;
+ while (1) {
+ gint mb_len = 0;
+ gint cur_len = 0;
+ gchar *part_str;
+ gchar *out_str;
+ gchar *enc_str;
+ const guchar *p = srcp;
+ gint out_str_len;
+ gint out_enc_str_len;
+ gint mime_block_len;
+ gboolean cont = FALSE;
+
+ while (*p != '\0') {
+ if (isspace(*p) && !is_next_nonascii(p + 1))
+ break;
+ /* don't include parentheses in encoded
+ strings */
+ if (addr_field && (*p == '(' || *p == ')'))
+ break;
+
+ 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);
+ }
+ out_str_len = strlen(out_str);
- to64frombits(destp, tmp_jis, strlen(tmp_jis));
- line_len += strlen(destp);
- destp += strlen(destp);
+ if (use_base64)
+ out_enc_str_len = B64LEN(out_str_len);
+ else
+ out_enc_str_len =
+ qp_get_q_encoding_len(out_str);
- strcpy(destp, mimehdr_end);
- destp += strlen(mimehdr_end);
- line_len += strlen(mimehdr_end);
+ g_free(out_str);
- g_free(tmp_jis);
- } else {
- strcpy(destp, tmp);
- line_len += strlen(destp);
- destp += strlen(destp);
+ if (mimestr_len + out_enc_str_len <= left) {
+ cur_len += mb_len;
+ p += mb_len;
+ } else if (cur_len == 0) {
+ LBREAK_IF_REQUIRED(1, FALSE);
+ continue;
+ } else {
+ cont = TRUE;
+ break;
+ }
}
- g_free(tmp);
- /* g_print("line_len = %d\n\n", line_len); */
- } while (*wtmpp != (wchar_t)0);
+ if (cur_len > 0) {
+ 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)
+ out_enc_str_len = B64LEN(out_str_len);
+ else
+ out_enc_str_len =
+ qp_get_q_encoding_len(out_str);
+
+ Xalloca(enc_str, out_enc_str_len + 1, );
+ if (use_base64)
+ base64_encode(enc_str, out_str, out_str_len);
+ else
+ qp_q_encode(enc_str, out_str);
+
+ g_free(out_str);
+
+ /* output MIME-encoded string block */
+ mime_block_len = mimestr_len + strlen(enc_str);
+ g_snprintf(destp, mime_block_len + 1,
+ MIMESEP_BEGIN "%s%s%s" MIMESEP_END,
+ out_encoding, mimesep_enc, enc_str);
+ destp += mime_block_len;
+ srcp += cur_len;
+
+ left -= mime_block_len;
+ }
- while (iswspace(*wsrcp)) {
- gint mbl;
+ LBREAK_IF_REQUIRED(cont, FALSE);
- mbl = wctomb(destp, *wsrcp++);
- if (mbl != -1) {
- destp += mbl;
- line_len += mbl;
- }
+ if (cur_len == 0)
+ break;
}
- *destp = '\0';
+ }
+
+ *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;
+}
- g_free(wtmp);
+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);
}
- g_free(wsrc);
+ 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);
+ }
- /* g_print("dest = %s\n", dest); */
+ return utf8_file;
}
-#endif /* HAVE_LIBJCONV */