(((c) & 0xff) >= 0xa1 && ((c) & 0xff) <= 0xdf)
#define iseucaux(c) \
(((c) & 0xff) == 0x8f)
-#define isunprintableeuckanji(c) \
- (((c) & 0xff) >= 0xa9 && ((c) & 0xff) <= 0xaf)
#define issjiskanji1(c) \
((((c) & 0xff) >= 0x81 && ((c) & 0xff) <= 0x9f) || \
(((c) & 0xff) >= 0xe0 && ((c) & 0xff) <= 0xfc))
void conv_anytoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- switch (conv_guess_encoding(inbuf)) {
+ switch (conv_guess_ja_encoding(inbuf)) {
case C_ISO_2022_JP:
conv_jistoeuc(outbuf, outlen, inbuf);
break;
void conv_anytojis(gchar *outbuf, gint outlen, const gchar *inbuf)
{
- switch (conv_guess_encoding(inbuf)) {
+ switch (conv_guess_ja_encoding(inbuf)) {
case C_EUC_JP:
conv_euctojis(outbuf, outlen, inbuf);
break;
}
}
+static gchar valid_eucjp_tbl[][96] = {
+ /* 0xa2a0 - 0xa2ff */
+ { 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0,
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1,
+ 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1,
+ 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0,
+ 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, 0 },
+
+ /* 0xa3a0 - 0xa3ff */
+ { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0,
+ 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0,
+ 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0 },
+
+ /* 0xa4a0 - 0xa4ff */
+ { 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 },
+
+ /* 0xa5a0 - 0xa5ff */
+ { 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0 },
+
+ /* 0xa6a0 - 0xa6ff */
+ { 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0,
+ 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0,
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 },
+
+ /* 0xa7a0 - 0xa7ff */
+ { 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+ 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 },
+
+ /* 0xa8a0 - 0xa8ff */
+ { 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 }
+};
+
+static gboolean isprintableeuckanji(guchar c1, guchar c2)
+{
+ if (c1 <= 0xa0 || c1 >= 0xf5)
+ return FALSE;
+ if (c2 <= 0xa0 || c2 == 0xff)
+ return FALSE;
+
+ if (c1 >= 0xa9 && c1 <= 0xaf)
+ return FALSE;
+
+ if (c1 >= 0xa2 && c1 <= 0xa8)
+ return (gboolean)valid_eucjp_tbl[c1 - 0xa2][c2 - 0xa0];
+
+ if (c1 == 0xcf) {
+ if (c2 >= 0xd4 && c2 <= 0xff)
+ return FALSE;
+ } else if (c1 == 0xf4) {
+ if (c2 >= 0xa7 && c2 <= 0xff)
+ return FALSE;
+ }
+
+ return TRUE;
+}
+
void conv_unreadable_eucjp(gchar *str)
{
register guchar *p = str;
/* printable 7 bit code */
p++;
} else if (iseuckanji(*p)) {
- if (iseuckanji(*(p + 1)) && !isunprintableeuckanji(*p))
+ if (isprintableeuckanji(*p, *(p + 1))) {
/* printable euc-jp code */
p += 2;
- else {
+ } else {
/* substitute unprintable code */
*p++ = SUBST_CHAR;
if (*p != '\0') {
/* convert CR+LF -> LF */
if (*p == '\r' && *(p + 1) == '\n')
memmove(p, p + 1, strlen(p));
- else if ((*p & 0xff) >= 0x80 && (*p & 0xff) <= 0x9f)
+ else if ((*p & 0xff) >= 0x7f && (*p & 0xff) <= 0x9f)
*p = SUBST_CHAR;
p++;
}
}
}
-CharSet conv_guess_encoding(const gchar *str)
+CharSet conv_guess_ja_encoding(const gchar *str)
{
const guchar *p = str;
CharSet guessed = C_US_ASCII;
strncpy2(outbuf, inbuf, outlen);
}
+void conv_localetodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
+{
+ strncpy2(outbuf, inbuf, outlen);
+
+ switch (conv_get_current_charset()) {
+ case C_US_ASCII:
+ case C_ISO_8859_1:
+ case C_ISO_8859_2:
+ case C_ISO_8859_4:
+ case C_ISO_8859_5:
+ case C_ISO_8859_7:
+ case C_ISO_8859_8:
+ case C_ISO_8859_9:
+ case C_ISO_8859_11:
+ case C_ISO_8859_13:
+ case C_ISO_8859_15:
+ conv_unreadable_latin(outbuf);
+ break;
+ case C_EUC_JP:
+ conv_unreadable_eucjp(outbuf);
+ break;
+ default:
+ break;
+ }
+}
+
CodeConverter *conv_code_converter_new(const gchar *charset)
{
CodeConverter *conv;
#if HAVE_ICONV
if (!src_code)
src_code = conv_get_outgoing_charset_str();
- if (!dest_code)
+ if (!dest_code) {
dest_code = conv_get_current_charset_str();
+ if (!strcasecmp(dest_code, CS_US_ASCII))
+ dest_code = CS_ISO_8859_1;
+ }
/* don't convert if current codeset is US-ASCII */
if (!strcasecmp(dest_code, CS_US_ASCII))
dest_charset = conv_get_charset_from_str(dest_charset_str);
+ if (dest_charset == C_US_ASCII)
+ return conv_ustodisp;
+
switch (src_charset) {
case C_ISO_2022_JP:
case C_ISO_2022_JP_2:
outbuf_p = outbuf;
out_left = out_size;
- while ((n_conv = iconv(cd, (gchar **)&inbuf_p, &in_left,
+ while ((n_conv = iconv(cd, (ICONV_CONST gchar **)&inbuf_p, &in_left,
&outbuf_p, &out_left)) < 0) {
if (EILSEQ == errno) {
- *outbuf_p = '\0';
+ g_free(outbuf);
+ outbuf = NULL;
break;
} else if (EINVAL == errno) {
- *outbuf_p = '\0';
+ g_free(outbuf);
+ outbuf = NULL;
break;
} else if (E2BIG == errno) {
out_size *= 2;
}
}
+ if (outbuf) {
+ iconv(cd, NULL, NULL, &outbuf_p, &out_left);
+ outbuf = g_realloc(outbuf, strlen(outbuf) + 1);
+ }
+
iconv_close(cd);
return outbuf;
{"ANSI_X3.4-1968" , C_US_ASCII , C_US_ASCII},
};
-const gchar *conv_get_charset_str(CharSet charset)
+static GHashTable *conv_get_charset_to_str_table(void)
{
+ static GHashTable *table;
gint i;
+ if (table)
+ return table;
+
+ table = g_hash_table_new(NULL, g_direct_equal);
+
for (i = 0; i < sizeof(charsets) / sizeof(charsets[0]); i++) {
- if (charsets[i].charset == charset)
- return charsets[i].name;
+ if (g_hash_table_lookup(table, GUINT_TO_POINTER(charsets[i].charset))
+ == NULL) {
+ g_hash_table_insert
+ (table, GUINT_TO_POINTER(charsets[i].charset),
+ charsets[i].name);
+ }
}
- return NULL;
+ return table;
}
-CharSet conv_get_charset_from_str(const gchar *charset)
+static gint str_case_equal(gconstpointer v, gconstpointer v2)
+{
+ return strcasecmp((const gchar *)v, (const gchar *)v2) == 0;
+}
+
+static guint str_case_hash(gconstpointer key)
{
+ const gchar *p = key;
+ guint h = *p;
+
+ if (h) {
+ h = tolower(h);
+ for (p += 1; *p != '\0'; p++)
+ h = (h << 5) - h + tolower(*p);
+ }
+
+ return h;
+}
+
+static GHashTable *conv_get_charset_from_str_table(void)
+{
+ static GHashTable *table;
gint i;
- if (!charset) return C_AUTO;
+ if (table)
+ return table;
+
+ table = g_hash_table_new(str_case_hash, str_case_equal);
for (i = 0; i < sizeof(charsets) / sizeof(charsets[0]); i++) {
- if (!strcasecmp(charsets[i].name, charset))
- return charsets[i].charset;
+ g_hash_table_insert(table, charsets[i].name,
+ GUINT_TO_POINTER(charsets[i].charset));
}
- return C_AUTO;
+ return table;
+}
+
+const gchar *conv_get_charset_str(CharSet charset)
+{
+ GHashTable *table;
+
+ table = conv_get_charset_to_str_table();
+ return g_hash_table_lookup(table, GUINT_TO_POINTER(charset));
+}
+
+CharSet conv_get_charset_from_str(const gchar *charset)
+{
+ GHashTable *table;
+
+ if (!charset) return C_AUTO;
+
+ table = conv_get_charset_from_str_table();
+ return GPOINTER_TO_UINT(g_hash_table_lookup(table, charset));
}
CharSet conv_get_current_charset(void)
{
static CharSet cur_charset = -1;
const gchar *cur_locale;
+ const gchar *p;
gint i;
if (cur_charset != -1)
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;
if (!codeset)
codeset = conv_get_charset_str(conv_get_current_charset());
- return codeset ? codeset : "US-ASCII";
+ return codeset ? codeset : CS_US_ASCII;
}
CharSet conv_get_outgoing_charset(void)
{
static CharSet out_charset = -1;
const gchar *cur_locale;
+ const gchar *p;
gint i;
if (out_charset != -1)
return out_charset;
}
+ if ((p = strcasestr(cur_locale, "@euro")) && p[5] == '\0') {
+ out_charset = C_ISO_8859_15;
+ return out_charset;
+ }
+
for (i = 0; i < sizeof(locale_table) / sizeof(locale_table[0]); i++) {
const gchar *p;
out_charset = conv_get_outgoing_charset();
str = conv_get_charset_str(out_charset);
- return str ? str : "US-ASCII";
+ return str ? str : CS_US_ASCII;
}
const gchar *conv_get_current_locale(void)
#define B64LEN(len) ((len) / 3 * 4 + ((len) % 3 ? 4 : 0))
-#define LBREAK_IF_REQUIRED(cond) \
+#define LBREAK_IF_REQUIRED(cond, plaintext) \
{ \
if (len - (destp - dest) < MAX_LINELEN + 2) { \
*destp = '\0'; \
} \
\
if ((cond) && *srcp) { \
- if (destp > dest && isspace(*(destp - 1))) \
- destp--; \
- else if (isspace(*srcp)) \
- srcp++; \
- if (*srcp) { \
- *destp++ = '\n'; \
- *destp++ = ' '; \
- left = MAX_LINELEN - 1; \
+ if (destp > dest && left < MAX_LINELEN - 1) { \
+ if (isspace(*(destp - 1))) \
+ destp--; \
+ else if (plaintext && isspace(*srcp)) \
+ srcp++; \
+ if (*srcp) { \
+ *destp++ = '\n'; \
+ *destp++ = ' '; \
+ left = MAX_LINELEN - 1; \
+ } \
} \
} \
}
}
cur_encoding = conv_get_current_charset_str();
- if (!strcmp(cur_encoding, "US-ASCII"))
- cur_encoding = "ISO-8859-1";
+ if (!strcmp(cur_encoding, CS_US_ASCII))
+ cur_encoding = CS_ISO_8859_1;
out_encoding = conv_get_outgoing_charset_str();
- if (!strcmp(out_encoding, "US-ASCII"))
- out_encoding = "ISO-8859-1";
+ if (!strcmp(out_encoding, CS_US_ASCII))
+ out_encoding = CS_ISO_8859_1;
mimestr_len = strlen(MIMESEP_BEGIN) + strlen(out_encoding) +
strlen(mimesep_enc) + strlen(MIMESEP_END);
left = MAX_LINELEN - header_len;
while (*srcp) {
- LBREAK_IF_REQUIRED(left <= 0);
+ LBREAK_IF_REQUIRED(left <= 0, TRUE);
while (isspace(*srcp)) {
*destp++ = *srcp++;
left--;
- LBREAK_IF_REQUIRED(left <= 0);
+ LBREAK_IF_REQUIRED(left <= 0, TRUE);
}
/* output as it is if the next word is ASCII string */
gint word_len;
word_len = get_next_word_len(srcp);
- LBREAK_IF_REQUIRED(left < word_len);
+ LBREAK_IF_REQUIRED(left < word_len, TRUE);
while (word_len > 0) {
- LBREAK_IF_REQUIRED(left <= 0);
+ LBREAK_IF_REQUIRED(left + 22 <= 0, TRUE);
*destp++ = *srcp++;
left--;
word_len--;
(part_str, cur_encoding, out_encoding);
if (!out_str) {
g_warning("conv_encode_header(): code conversion failed\n");
- out_str = g_strdup(out_str);
+ conv_unreadable_8bit(part_str);
+ out_str = g_strdup(part_str);
}
out_str_len = strlen(out_str);
cur_len += mb_len;
p += mb_len;
} else if (cur_len == 0) {
- LBREAK_IF_REQUIRED(1);
+ LBREAK_IF_REQUIRED(1, FALSE);
continue;
} else {
cont = TRUE;
(part_str, cur_encoding, out_encoding);
if (!out_str) {
g_warning("conv_encode_header(): code conversion failed\n");
- out_str = g_strdup(out_str);
+ conv_unreadable_8bit(part_str);
+ out_str = g_strdup(part_str);
}
out_str_len = strlen(out_str);
left -= mime_block_len;
}
- LBREAK_IF_REQUIRED(cont);
+ LBREAK_IF_REQUIRED(cont, FALSE);
if (cur_len == 0)
break;