/*
* Sylpheed -- a GTK+ based, lightweight, and fast e-mail client
- * Copyright (C) 1999-2001 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 <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 '_'
+#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++;
+ }
+ }
+ }
+ } else {
+ K_OUT();
+ *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);
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)
- 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;
- 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;
- }
+#if HAVE_ICONV
+ if (!src_code)
+ src_code = conv_get_outgoing_charset_str();
+ if (!dest_code)
+ dest_code = conv_get_current_charset_str();
- 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 current codeset is US-ASCII */
+ if (!strcasecmp(dest_code, CS_US_ASCII))
+ return g_strdup(inbuf);
- return buf;
-#endif /* !HAVE_LIBJCONV */
+ /* don't convert if src and dest codeset are identical */
+ if (!strcasecmp(src_code, dest_code))
+ return g_strdup(inbuf);
+
+ 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;
+ CodeConvFunc code_conv = conv_noconv;
+ CharSet src_charset;
+ CharSet dest_charset;
- if (!charset) {
- if (conv_get_outgoing_charset() == C_ISO_2022_JP)
- return conv_jistodisp;
+ 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);
+
+ 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, "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, (gchar **)&inbuf_p, &in_left,
+ &outbuf_p, &out_left)) < 0) {
+ if (EILSEQ == errno) {
+ *outbuf_p = '\0';
+ break;
+ } else if (EINVAL == errno) {
+ *outbuf_p = '\0';
+ 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;
+ }
+ }
+
+ iconv_close(cd);
+
+ return outbuf;
+}
+#endif /* HAVE_ICONV */
+
static const struct {
CharSet charset;
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_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_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
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.ujis" , C_EUC_JP , C_ISO_2022_JP},
+ {"ja_JP.EUC" , 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},
+ {"ja_JP" , C_EUC_JP , C_ISO_2022_JP},
+ {"ko_KR" , C_EUC_KR , C_EUC_KR},
+ {"zh_CN.GB2312" , C_GB2312 , C_GB2312},
+ {"zh_CN" , C_GB2312 , C_GB2312},
+ {"zh_TW.eucTW" , 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_ISO_8859_5},
+ {"ru_RU.CP1251" , C_WINDOWS_1251, C_ISO_8859_5},
+
+ {"bg_BG" , C_WINDOWS_1251, C_WINDOWS_1251},
+
+ {"en_US" , 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_DE" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"nl_NL" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"et_EE" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"fi_FI" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"fr_FR" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"is_IS" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"it_IT" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"no_NO" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"pt_PT" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"pt_BR" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"es_ES" , C_ISO_8859_1 , C_ISO_8859_1},
+ {"sv_SE" , C_ISO_8859_1 , C_ISO_8859_1},
+
+ {"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},
+ {"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},
+
+ {"th_TH" , C_TIS_620 , C_TIS_620},
+ /* {"th_TH" , C_WINDOWS_874}, */
+ /* {"th_TH" , C_ISO_8859_11}, */
+
+ {"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},
+ {"lv_LV" , C_ISO_8859_13 , C_ISO_8859_13},
+
+ {"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)
{
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
+ 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;
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;
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)
{
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;
+ 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;
+ for (i = 0; i < sizeof(locale_table) / sizeof(locale_table[0]); i++) {
+ const gchar *p;
+
+ if (!strncasecmp(cur_locale, locale_table[i].locale,
+ strlen(locale_table[i].locale))) {
+ out_charset = locale_table[i].out_charset;
+ break;
+ } else if ((p = strchr(locale_table[i].locale, '_')) &&
+ !strchr(p + 1, '.')) {
+ if (strlen(cur_locale) == 2 &&
+ !strncasecmp(cur_locale, locale_table[i].locale, 2)) {
+ out_charset = locale_table[i].out_charset;
+ break;
+ }
}
}
- 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 !HAVE_ICONV
+ /* encoding conversion without iconv() is only supported
+ on Japanese locale for now */
+ if (out_charset == C_ISO_2022_JP)
+ return out_charset;
+ else
+ return conv_get_current_charset();
#endif
return out_charset;
gchar *cur_locale;
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);
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) {
- gchar *tmp;
- gint len;
-
- len = strlen(str) * 2 + 1;
- Xalloca(tmp, len, return);
- conv_jistodisp(tmp, len, str);
- strncpy2(str, tmp, outlen);
+ 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);
}
-#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)
{
- gchar *buf;
CharSet cur_charset;
cur_charset = conv_get_current_charset();
- Xstrdup_a(buf, str, return);
-#if HAVE_LIBJCONV
- UnMimeHeaderConv(buf, outbuf, outlen);
-#else
- UnMimeHeader(buf);
- strncpy2(outbuf, buf, outlen);
-#endif
if (cur_charset == C_EUC_JP) {
- gint len;
+ gchar *buf;
+ gint buflen;
- len = strlen(outbuf) * 2 + 1;
- Xalloca(buf, len, return);
- conv_jistodisp(buf, len, outbuf);
- strncpy2(outbuf, buf, outlen);
- }
+ buflen = strlen(str) * 2 + 1;
+ Xalloca(buf, buflen, return);
+ conv_anytodisp(buf, buflen, str);
+ unmime_header(outbuf, buf);
+ } else
+ unmime_header(outbuf, str);
}
-#define MAX_ENCLEN 75
#define MAX_LINELEN 76
+#define MIMESEP_BEGIN "=?"
+#define MIMESEP_END "?="
#define B64LEN(len) ((len) / 3 * 4 + ((len) % 3 ? 4 : 0))
-#if HAVE_LIBJCONV
-void conv_encode_header(gchar *dest, gint len, const gchar *src,
- gint header_len)
-{
- 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;
-
- 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 (line_len > 1 && destp > dest) {
- *destp++ = ' ';
- *destp = '\0';
- line_len++;
- }
-
- 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;
-
- to64frombits(destp, raw, raw_len);
- line_len += strlen(destp);
- destp += strlen(destp);
-
- strcpy(destp, mimehdr_end);
- destp += strlen(mimehdr_end);
- line_len += strlen(mimehdr_end);
- } else {
- strcpy(destp, tmp);
- line_len += strlen(destp);
- destp += strlen(destp);
- }
-
- g_free(tmp);
- g_free(raw);
- /* g_print("line_len = %d\n\n", line_len); */
- } while (*wtmpp != (wchar_t)0);
-
- while (iswspace(*wsrcp)) {
- gint mbl;
-
- mbl = wctomb(destp, *wsrcp++);
- if (mbl != -1) {
- destp += mbl;
- line_len += mbl;
- }
- }
- *destp = '\0';
-
- g_free(wtmp);
- }
-
- g_free(wsrc);
-
- /* g_print("dest = %s\n", dest); */
+#define LBREAK_IF_REQUIRED(cond) \
+{ \
+ if (len - (destp - dest) < MAX_LINELEN + 2) { \
+ *destp = '\0'; \
+ return; \
+ } \
+ \
+ if ((cond) && *srcp) { \
+ if (destp > dest && isspace(*(destp - 1))) \
+ destp--; \
+ else if (isspace(*srcp)) \
+ srcp++; \
+ if (*srcp) { \
+ *destp++ = '\n'; \
+ *destp++ = ' '; \
+ left = MAX_LINELEN - 1; \
+ } \
+ } \
}
-#else /* !HAVE_LIBJCONV */
-
-#define JIS_SEQLEN 3
void conv_encode_header(gchar *dest, gint len, const gchar *src,
gint header_len)
{
- 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 gchar *srcp = src;
+ gchar *destp = dest;
+ gboolean use_base64;
+
+ 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);
+ cur_encoding = conv_get_current_charset_str();
+ if (!strcmp(cur_encoding, "US-ASCII"))
+ cur_encoding = "ISO-8859-1";
+ out_encoding = conv_get_outgoing_charset_str();
+ if (!strcmp(out_encoding, "US-ASCII"))
+ out_encoding = "ISO-8859-1";
- if (wp != NULL) {
- wtmp = wcsndup(wsrcp, wp - wsrcp);
- wsrcp = wp + 1;
- while (iswspace(wsrcp[nspc])) nspc++;
- } else {
- wtmp = wcsdup(wsrcp);
- wsrcp += wcslen(wsrcp);
- }
+ mimestr_len = strlen(MIMESEP_BEGIN) + strlen(out_encoding) +
+ strlen(mimesep_enc) + strlen(MIMESEP_END);
- wtmpp = wtmp;
+ left = MAX_LINELEN - header_len;
- do {
- gint prev_mbl = 1, tlen = 0, mb_seqlen = 0;
- gchar *tmp;
- register gchar *tmpp;
+ while (*srcp) {
+ LBREAK_IF_REQUIRED(left <= 0);
- tmpp = tmp = g_malloc(wcslen(wtmpp) * MB_CUR_MAX + 1);
- *tmp = '\0';
+ while (isspace(*srcp)) {
+ *destp++ = *srcp++;
+ left--;
+ LBREAK_IF_REQUIRED(left <= 0);
+ }
- while (*wtmpp != (wchar_t)0) {
- gint mbl;
+ /* 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);
+ while (word_len > 0) {
+ LBREAK_IF_REQUIRED(left <= 0);
+ *destp++ = *srcp++;
+ left--;
+ word_len--;
+ }
- mbl = wctomb(tmpp, *wtmpp);
- if (mbl == -1) {
- g_warning("invalid wide character\n");
- wtmpp++;
- continue;
- }
+ continue;
+ }
- /* 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;
- }
+ while (1) {
+ gint mb_len = 0;
+ gint cur_len = 0;
+ gchar *part_str;
+ gchar *out_str;
+ gchar *enc_str;
+ const gchar *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;
+
+ 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 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
+ 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");
+ out_str = g_strdup(out_str);
}
+ out_str_len = strlen(out_str);
- tmpp += mbl;
- *tmpp = '\0';
+ if (use_base64)
+ out_enc_str_len = B64LEN(out_str_len);
+ else
+ out_enc_str_len =
+ qp_get_q_encoding_len(out_str);
- tlen += mbl;
- prev_mbl = mbl;
-
- wtmpp++;
- }
- /* g_print("tmp = %s, tlen = %d, mb_seqlen = %d\n",
- tmp, tlen, mb_seqlen);
- */
+ g_free(out_str);
- if (tlen == 0) {
- g_free(tmp);
- continue;
- }
-
- if (line_len > 1 && destp > dest) {
- *destp++ = ' ';
- *destp = '\0';
- line_len++;
+ if (mimestr_len + out_enc_str_len <= left) {
+ cur_len += mb_len;
+ p += mb_len;
+ } else if (cur_len == 0) {
+ LBREAK_IF_REQUIRED(1);
+ continue;
+ } else {
+ cont = TRUE;
+ break;
+ }
}
- if (mb_seqlen) {
- gchar *tmp_jis;
-
- 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;
-
- to64frombits(destp, tmp_jis, strlen(tmp_jis));
- line_len += strlen(destp);
- destp += strlen(destp);
-
- strcpy(destp, mimehdr_end);
- destp += strlen(mimehdr_end);
- line_len += strlen(mimehdr_end);
-
- g_free(tmp_jis);
- } else {
- strcpy(destp, tmp);
- line_len += strlen(destp);
- destp += strlen(destp);
+ 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");
+ out_str = g_strdup(out_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;
}
- g_free(tmp);
- /* g_print("line_len = %d\n\n", line_len); */
- } while (*wtmpp != (wchar_t)0);
+ LBREAK_IF_REQUIRED(cont);
- while (iswspace(*wsrcp)) {
- gint mbl;
-
- mbl = wctomb(destp, *wsrcp++);
- if (mbl != -1) {
- destp += mbl;
- line_len += mbl;
- }
+ if (cur_len == 0)
+ break;
}
- *destp = '\0';
-
- g_free(wtmp);
}
- g_free(wsrc);
-
- /* g_print("dest = %s\n", dest); */
+ *destp = '\0';
}
-#endif /* HAVE_LIBJCONV */
+
+#undef LBREAK_IF_REQUIRED
projects / claws.git / blobdiff