/*
* Sylpheed -- a GTK+ based, lightweight, and fast e-mail client
- * Copyright (C) 1999,2000 Hiroyuki Yamamoto
+ * Copyright (C) 1999-2002 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 <jconv.h>
#endif
-#include <kcc.h>
-
#include "intl.h"
#include "codeconv.h"
#include "unmime.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 '_'
+#define ESC '\033'
+
+#define iseuckanji(c) \
(((c) & 0xff) >= 0xa1 && ((c) & 0xff) <= 0xfe)
-#define iseucss(c) \
- (((c) & 0xff) == 0x8e || ((c) & 0xff) == 0x8f)
-#define isunprintablekanji(c) \
+#define iseuchwkana1(c) \
+ (((c) & 0xff) == 0x8e)
+#define iseuchwkana2(c) \
+ (((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))
+#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++ = '$'; \
+ *out++ = 'B'; \
+ state = JIS_KANJI; \
+ }
+
+#define K_OUT() \
+ if (state != JIS_ASCII) { \
+ *out++ = ESC; \
+ *out++ = '('; \
+ *out++ = 'B'; \
+ state = JIS_ASCII; \
+ }
+
+#define HW_IN() \
+ if (state != JIS_HWKANA) { \
+ *out++ = ESC; \
+ *out++ = '('; \
+ *out++ = 'I'; \
+ state = JIS_HWKANA; \
+ }
+
+#define AUX_IN() \
+ if (state != JIS_AUXKANJI) { \
+ *out++ = ESC; \
+ *out++ = '$'; \
+ *out++ = '('; \
+ *out++ = 'D'; \
+ state = JIS_AUXKANJI; \
+ }
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)
{
- 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;
+ const guchar *in = inbuf;
+ guchar *out = outbuf;
+ JISState state = JIS_ASCII;
+
+ while (*in != '\0') {
+ if (isascii(*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' && !isascii(*in)) {
+ *out++ = SUBST_CHAR;
+ in++;
+ }
+ }
+ } else if (iseuchwkana1(*in)) {
+ in++;
+ if (iseuchwkana2(*in)) {
+ HW_IN();
+ *out++ = *in++ & 0x7f;
+ } else {
+ K_OUT();
+ if (*in != '\0' && !isascii(*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' && !isascii(*in)) {
+ *out++ = SUBST_CHAR;
+ in++;
+ if (*in != '\0' && !isascii(*in)) {
+ *out++ = SUBST_CHAR;
+ in++;
+ }
+ }
+ }
}
+ }
- 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);
+ K_OUT();
+ *out = '\0';
}
void conv_sjistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- KCC_filter(outbuf, "EUC", (gchar *)inbuf, "SJIS", 0, 0, 0);
+ const guchar *in = inbuf;
+ guchar *out = outbuf;
+
+ while (*in != '\0') {
+ if (isascii(*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' && !isascii(*in)) {
+ *out++ = SUBST_CHAR;
+ in++;
+ }
+ }
+ } else if (issjishwkana(*in)) {
+ *out++ = 0x8e;
+ *out++ = *in++;
+ } else {
+ *out++ = SUBST_CHAR;
+ in++;
+ }
+ }
+
+ *out = '\0';
}
void conv_anytoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- KCC_filter(outbuf, "EUC", (gchar *)inbuf, "AUTO", 0, 0, 0);
+ switch (conv_guess_encoding(inbuf)) {
+ case C_ISO_2022_JP:
+ conv_jistoeuc(outbuf, outlen, inbuf);
+ break;
+ case C_SHIFT_JIS:
+ conv_sjistoeuc(outbuf, outlen, inbuf);
+ break;
+ default:
+ strncpy2(outbuf, inbuf, outlen);
+ break;
+ }
}
void conv_anytojis(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- KCC_filter(outbuf, "JISBB", (gchar *)inbuf, "AUTO", 0, 0, 0);
+ switch (conv_guess_encoding(inbuf)) {
+ case C_EUC_JP:
+ conv_euctojis(outbuf, outlen, inbuf);
+ break;
+ default:
+ strncpy2(outbuf, inbuf, outlen);
+ break;
+ }
}
-#define SUBST_CHAR '_'
-
void conv_unreadable_eucjp(gchar *str)
{
register guchar *p = str;
memmove(p, p + 1, strlen(p));
/* printable 7 bit code */
p++;
- } else if (iskanji(*p)) {
- if (iskanji(*(p + 1)) && !isunprintablekanji(*p))
+ } else if (iseuckanji(*p)) {
+ if (iseuckanji(*(p + 1)) && !isunprintableeuckanji(*p))
/* printable euc-jp code */
p += 2;
else {
*p++ = SUBST_CHAR;
}
}
- } else if (iseucss(*p)) {
- if ((*(p + 1) & 0x80) != 0)
+ } else if (iseuchwkana1(*p)) {
+ if (iseuchwkana2(*(p + 1)))
/* euc-jp hankaku kana */
p += 2;
else
*p++ = SUBST_CHAR;
+ } else if (iseucaux(*p)) {
+ if (iseuckanji(*(p + 1)) && iseuckanji(*(p + 2))) {
+ /* auxiliary kanji */
+ p += 3;
+ } else
+ *p++ = SUBST_CHAR;
} else
/* substitute unprintable 1 byte code */
*p++ = SUBST_CHAR;
p += 2;
len -= 2;
}
- } else if (iskanji(*p)) {
+ } else if (iseuckanji(*p)) {
p += 2;
len -= 2;
} else {
}
}
+CharSet conv_guess_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 (isascii(*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;
+}
+
void conv_jistodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
{
conv_jistoeuc(outbuf, outlen, inbuf);
conv_unreadable_eucjp(outbuf);
}
+void conv_anytodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
+{
+ conv_anytoeuc(outbuf, outlen, inbuf);
+ conv_unreadable_eucjp(outbuf);
+}
+
void conv_ustodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
{
strncpy2(outbuf, inbuf, outlen);
func = conv_get_code_conv_func(src_codeset);
if (func != conv_noconv) {
- if (func == conv_jistodisp || func == conv_sjistodisp)
+ if (func == conv_jistodisp ||
+ func == conv_sjistodisp ||
+ func == conv_anytodisp)
len = strlen(inbuf) * 2 + 1;
else
len = strlen(inbuf) + 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)
+ 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 (jconv_alloc_conv(inbuf, strlen(inbuf), &buf, &len,
codesets, n_codesets,
&actual_codeset, dest_codeset)
== 0)
return buf;
- else
+ else {
+ g_warning("code conversion from %s to %s failed\n",
+ codesets && codesets[0] ? codesets[0] : "(unknown)",
+ dest_codeset);
return NULL;
+ }
#else /* !HAVE_LIBJCONV */
if (src_codeset) {
if (!strcasecmp(src_codeset, CS_EUC_JP) ||
- !strcasecmp(src_codeset, "EUCJP"))
+ !strcasecmp(src_codeset, CS_EUCJP))
src_charset = C_EUC_JP;
else if (!strcasecmp(src_codeset, CS_SHIFT_JIS) ||
!strcasecmp(src_codeset, "SHIFT-JIS") ||
CodeConvFunc conv_get_code_conv_func(const gchar *charset)
{
CodeConvFunc code_conv;
+ CharSet cur_charset;
if (!charset) {
- if (conv_get_outgoing_charset() == C_ISO_2022_JP)
- return conv_jistodisp;
+ cur_charset = conv_get_current_charset();
+ if (cur_charset == C_EUC_JP || cur_charset == C_SHIFT_JIS)
+ return conv_anytodisp;
else
return conv_noconv;
}
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"))
+ !strcasecmp(charset, CS_EUCJP))
code_conv = conv_euctodisp;
else
code_conv = conv_noconv;
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_ISO_8859_1, CS_ISO_8859_1},
{C_ISO_8859_2, CS_ISO_8859_2},
{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_11, CS_ISO_8859_11},
{C_ISO_8859_13, CS_ISO_8859_13},
{C_ISO_8859_15, CS_ISO_8859_15},
{C_BALTIC, CS_BALTIC},
{C_ISO_2022_JP, CS_ISO_2022_JP},
{C_ISO_2022_JP_2, CS_ISO_2022_JP_2},
{C_EUC_JP, CS_EUC_JP},
+ {C_EUC_JP, CS_EUCJP},
{C_SHIFT_JIS, CS_SHIFT_JIS},
{C_ISO_2022_KR, CS_ISO_2022_KR},
{C_EUC_KR, CS_EUC_KR},
{C_GB2312, CS_GB2312},
{C_EUC_TW, CS_EUC_TW},
{C_BIG5, CS_BIG5},
+ {C_TIS_620, CS_TIS_620},
+ {C_WINDOWS_874, CS_WINDOWS_874},
};
#if !HAVE_LIBJCONV
{"zh_TW" , C_BIG5},
{"ru_RU.KOI8-R" , C_KOI8_R},
- {"ru_RU.CP1251" , C_CP1251},
+ {"ru_RU.CP1251" , C_WINDOWS_1251},
+
+ {"bg_BG" , C_WINDOWS_1251},
{"en_US" , C_ISO_8859_1},
{"ca_ES" , C_ISO_8859_1},
{"iw_IL" , C_ISO_8859_8},
{"tr_TR" , C_ISO_8859_9},
+ {"th_TH" , C_TIS_620},
+ /* {"th_TH" , C_WINDOWS_874}, */
+ /* {"th_TH" , C_ISO_8859_11}, */
+
{"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},
+ {"C" , C_US_ASCII},
+ {"POSIX" , C_US_ASCII},
+ {"ANSI_X3.4-1968" , C_US_ASCII},
};
#endif /* !HAVE_LIBJCONV */
const gchar *conv_get_current_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());
return codeset ? codeset : "US-ASCII";
-#endif
}
CharSet conv_get_outgoing_charset(void)
/* 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++) {
+ 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;
return str ? str : "US-ASCII";
}
+const gchar *conv_get_current_locale(void)
+{
+ gchar *cur_locale;
+
+ cur_locale = g_getenv("LC_ALL");
+ 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)");
+
+ return cur_locale;
+}
+
void conv_unmime_header_overwrite(gchar *str)
{
gchar *buf;
- gint outlen;
+ gint buflen;
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)
- conv_unreadable_eucjp(str);
-#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
+ buflen = strlen(str) * 2 + 1;
+ Xalloca(buf, buflen, return);
+ conv_anytodisp(buf, buflen, str);
+ unmime_header(str, buf);
+ } else {
+ buflen = strlen(str) + 1;
+ Xalloca(buf, buflen, return);
+ unmime_header(buf, str);
+ strncpy2(str, buf, buflen);
+ }
}
void conv_unmime_header(gchar *outbuf, gint outlen, const gchar *str,
const gchar *charset)
{
- 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)
- conv_unreadable_eucjp(outbuf);
-#else
- UnMimeHeader(buf);
if (cur_charset == C_EUC_JP) {
- gchar *tmp;
- gint len;
+ gchar *buf;
+ gint buflen;
- len = strlen(buf) * 2 + 1;
- Xalloca(tmp, len, {strncpy2(outbuf, buf, outlen); return;});
- conv_jistodisp(tmp, len, buf);
- strncpy2(outbuf, tmp, outlen);
+ buflen = strlen(str) * 2 + 1;
+ Xalloca(buf, buflen, return);
+ conv_anytodisp(buf, buflen, str);
+ unmime_header(outbuf, buf);
} else
- strncpy2(outbuf, buf, outlen);
-#endif
+ unmime_header(outbuf, str);
}
#define MAX_ENCLEN 75
gchar *mimehdr_enctype = "?B?";
const gchar *mimehdr_charset;
- //g_print("src = %s\n", src);
+ /* g_print("src = %s\n", src); */
mimehdr_charset = conv_get_outgoing_charset_str();
/* convert to wide-character string */
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';
while (*wsrcp) {
wchar_t *wp, *wtmp, *wtmpp;
gint nspc = 0;
+ gboolean str_is_non_ascii;
/* irresponsible buffer overrun check */
if ((len - (destp - dest)) < (MAX_LINELEN + 1) * 2) break;
wp = wsrcp;
while ((wp = find_wspace(wp)) != NULL)
if (!is_next_nonascii(wp)) break;
- } else
+ str_is_non_ascii = TRUE;
+ } else {
wp = find_wspace(wsrcp);
+ str_is_non_ascii = FALSE;
+ }
if (wp != NULL) {
wtmp = wcsndup(wsrcp, wp - wsrcp);
wtmpp = wtmp;
do {
- gint tlen = 0, str_ascii = 1;
+ gint tlen = 0;
gchar *tmp; /* internal codeset */
gchar *raw; /* converted, but not base64 encoded */
register gchar *tmpp;
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");
}
/* g_free(raw); */
src_codeset = conv_get_current_charset_str();
- //printf ("tmp = %s, tlen = %d, mbl\n",
- // tmp, tlen, mbl);
+ /* 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,
!= 0) {
g_warning("can't convert\n");
tmpp[0] = '\0';
+ wtmpp++;
continue;
}
- if (!str_ascii) {
+ if (str_is_non_ascii) {
gint dlen = mimehdr_len +
B64LEN(raw_len);
if ((line_len + dlen +
*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;
+ } 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;
wtmpp++;
}
- //g_print("tmp = %s, tlen = %d, mb_seqlen = %d\n",
- // tmp, tlen, mb_seqlen);
+ /* g_print("tmp = %s, tlen = %d, mb_seqlen = %d\n",
+ tmp, tlen, mb_seqlen); */
if (tlen == 0 || raw_len == 0) {
g_free(tmp);
line_len++;
}
- if (!str_ascii) {
+ if (str_is_non_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;
- to64frombits(destp, raw, raw_len);
+ base64_encode(destp, raw, raw_len);
line_len += strlen(destp);
destp += strlen(destp);
g_free(tmp);
g_free(raw);
- //g_print("line_len = %d\n\n", line_len);
+ /* g_print("line_len = %d\n\n", line_len); */
} while (*wtmpp != (wchar_t)0);
while (iswspace(*wsrcp)) {
g_free(wsrc);
- //g_print("dest = %s\n", dest);
+ /* g_print("dest = %s\n", dest); */
}
#else /* !HAVE_LIBJCONV */
gchar *mimehdr_enctype = "?B?";
const gchar *mimehdr_charset;
- //g_print("src = %s\n", src);
+ /* 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 */
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';
while (*wsrcp) {
wchar_t *wp, *wtmp, *wtmpp;
gint nspc = 0;
+ gboolean str_is_non_ascii;
/* irresponsible buffer overrun check */
if ((len - (destp - dest)) < (MAX_LINELEN + 1) * 2) break;
wp = wsrcp;
while ((wp = find_wspace(wp)) != NULL)
if (!is_next_nonascii(wp)) break;
- } else
+ str_is_non_ascii = TRUE;
+ } else {
wp = find_wspace(wsrcp);
+ str_is_non_ascii = FALSE;
+ }
if (wp != NULL) {
wtmp = wcsndup(wsrcp, wp - wsrcp);
if (prev_mbl == 1 && mbl == 2)
mb_seqlen += JIS_SEQLEN * 2;
- if (mb_seqlen) {
+ if (str_is_non_ascii) {
gint dlen = mimehdr_len +
B64LEN(tlen + mb_seqlen + mbl);
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;
+ } 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;
}
tmpp += mbl;
wtmpp++;
}
- //g_print("tmp = %s, tlen = %d, mb_seqlen = %d\n",
- // tmp, tlen, mb_seqlen);
+ /* g_print("tmp = %s, tlen = %d, mb_seqlen = %d\n",
+ tmp, tlen, mb_seqlen); */
if (tlen == 0) {
g_free(tmp);
line_len++;
}
- if (mb_seqlen) {
+ if (str_is_non_ascii) {
gchar *tmp_jis;
tmp_jis = g_new(gchar, tlen + mb_seqlen + 1);
destp += mimehdr_begin_len;
line_len += mimehdr_begin_len;
- to64frombits(destp, tmp_jis, strlen(tmp_jis));
+ base64_encode(destp, tmp_jis, strlen(tmp_jis));
line_len += strlen(destp);
destp += strlen(destp);
}
g_free(tmp);
- //g_print("line_len = %d\n\n", line_len);
+ /* g_print("line_len = %d\n\n", line_len); */
} while (*wtmpp != (wchar_t)0);
while (iswspace(*wsrcp)) {
g_free(wsrc);
- //g_print("dest = %s\n", dest);
+ /* g_print("dest = %s\n", dest); */
}
#endif /* HAVE_LIBJCONV */