2005-03-04 [paul] 1.0.1cvs21
[claws.git] / src / codeconv.c
index 01bcf6fe52706d59ebc0a3e39d132b8d7d649d37..d177f743c3d65aae811451ff1d7b8607a658b808 100644 (file)
@@ -1,6 +1,6 @@
 /*
  * Sylpheed -- a GTK+ based, lightweight, and fast e-mail client
- * Copyright (C) 1999-2003 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
@@ -173,6 +173,72 @@ void conv_jistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
        *out = '\0';
 }
 
+#define JIS_HWDAKUTEN          0x5e
+#define JIS_HWHANDAKUTEN       0x5f
+
+static gint conv_jis_hantozen(guchar *outbuf, guchar jis_code, guchar sound_sym)
+{
+       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;
+               }
+       }
+
+       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_euctojis(gchar *outbuf, gint outlen, const gchar *inbuf)
 {
        const guchar *in = inbuf;
@@ -180,7 +246,7 @@ void conv_euctojis(gchar *outbuf, gint outlen, const gchar *inbuf)
        JISState state = JIS_ASCII;
 
        while (*in != '\0') {
-               if (isascii(*in)) {
+               if (IS_ASCII(*in)) {
                        K_OUT();
                        *out++ = *in++;
                } else if (iseuckanji(*in)) {
@@ -192,19 +258,43 @@ void conv_euctojis(gchar *outbuf, gint outlen, const gchar *inbuf)
                                K_OUT();
                                *out++ = SUBST_CHAR;
                                in++;
-                               if (*in != '\0' && !isascii(*in)) {
+                               if (*in != '\0' && !IS_ASCII(*in)) {
                                        *out++ = SUBST_CHAR;
                                        in++;
                                }
                        }
                } else if (iseuchwkana1(*in)) {
-                       in++;
-                       if (iseuchwkana2(*in)) {
-                               HW_IN();
-                               *out++ = *in++ & 0x7f;
+                       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();
-                               if (*in != '\0' && !isascii(*in)) {
+                               in++;
+                               if (*in != '\0' && !IS_ASCII(*in)) {
                                        *out++ = SUBST_CHAR;
                                        in++;
                                }
@@ -217,10 +307,10 @@ void conv_euctojis(gchar *outbuf, gint outlen, const gchar *inbuf)
                                *out++ = *in++ & 0x7f;
                        } else {
                                K_OUT();
-                               if (*in != '\0' && !isascii(*in)) {
+                               if (*in != '\0' && !IS_ASCII(*in)) {
                                        *out++ = SUBST_CHAR;
                                        in++;
-                                       if (*in != '\0' && !isascii(*in)) {
+                                       if (*in != '\0' && !IS_ASCII(*in)) {
                                                *out++ = SUBST_CHAR;
                                                in++;
                                        }
@@ -243,7 +333,7 @@ void conv_sjistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
        guchar *out = outbuf;
 
        while (*in != '\0') {
-               if (isascii(*in)) {
+               if (IS_ASCII(*in)) {
                        *out++ = *in++;
                } else if (issjiskanji1(*in)) {
                        if (issjiskanji2(*(in + 1))) {
@@ -266,7 +356,7 @@ void conv_sjistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
                        } else {
                                *out++ = SUBST_CHAR;
                                in++;
-                               if (*in != '\0' && !isascii(*in)) {
+                               if (*in != '\0' && !IS_ASCII(*in)) {
                                        *out++ = SUBST_CHAR;
                                        in++;
                                }
@@ -382,10 +472,10 @@ static gboolean isprintableeuckanji(guchar c1, guchar c2)
                return (gboolean)valid_eucjp_tbl[c1 - 0xa2][c2 - 0xa0];
 
        if (c1 == 0xcf) {
-               if (c2 >= 0xd4 && c2 <= 0xff)
+               if (c2 >= 0xd4 && c2 <= 0xfe)
                        return FALSE;
        } else if (c1 == 0xf4) {
-               if (c2 >= 0xa7 && c2 <= 0xff)
+               if (c2 >= 0xa7 && c2 <= 0xfe)
                        return FALSE;
        }
 
@@ -397,7 +487,7 @@ void conv_unreadable_eucjp(gchar *str)
        register guchar *p = str;
 
        while (*p != '\0') {
-               if (isascii(*p)) {
+               if (IS_ASCII(*p)) {
                        /* convert CR+LF -> LF */
                        if (*p == '\r' && *(p + 1) == '\n')
                                memmove(p, p + 1, strlen(p));
@@ -411,7 +501,7 @@ void conv_unreadable_eucjp(gchar *str)
                                /* substitute unprintable code */
                                *p++ = SUBST_CHAR;
                                if (*p != '\0') {
-                                       if (isascii(*p))
+                                       if (IS_ASCII(*p))
                                                p++;
                                        else
                                                *p++ = SUBST_CHAR;
@@ -443,7 +533,7 @@ void conv_unreadable_8bit(gchar *str)
                /* convert CR+LF -> LF */
                if (*p == '\r' && *(p + 1) == '\n')
                        memmove(p, p + 1, strlen(p));
-               else if (!isascii(*p)) *p = SUBST_CHAR;
+               else if (!IS_ASCII(*p)) *p = SUBST_CHAR;
                p++;
        }
 }
@@ -566,7 +656,7 @@ CharSet conv_guess_ja_encoding(const gchar *str)
                        if (guessed == C_US_ASCII)
                                return C_ISO_2022_JP;
                        p += 2;
-               } else if (isascii(*p)) {
+               } else if (IS_ASCII(*p)) {
                        p++;
                } else if (iseuckanji(*p) && iseuckanji(*(p + 1))) {
                        if (*p >= 0xfd && *p <= 0xfe)
@@ -743,6 +833,7 @@ CodeConvFunc conv_get_code_conv_func(const gchar *src_charset_str,
        switch (src_charset) {
        case C_ISO_2022_JP:
        case C_ISO_2022_JP_2:
+       case C_ISO_2022_JP_3:
                if (dest_charset == C_AUTO &&
                    conv_get_current_charset() == C_EUC_JP)
                        code_conv = conv_jistodisp;
@@ -767,7 +858,9 @@ CodeConvFunc conv_get_code_conv_func(const gchar *src_charset_str,
        case C_ISO_8859_13:
        case C_ISO_8859_14:
        case C_ISO_8859_15:
-               if (dest_charset == C_AUTO)
+               if (dest_charset == C_AUTO &&
+                   (conv_get_current_charset() == src_charset ||
+                    MB_CUR_MAX > 1))
                        code_conv = conv_latintodisp;
                break;
        case C_SHIFT_JIS:
@@ -781,8 +874,9 @@ CodeConvFunc conv_get_code_conv_func(const gchar *src_charset_str,
                if (dest_charset == C_AUTO &&
                    conv_get_current_charset() == C_EUC_JP)
                        code_conv = conv_euctodisp;
-               else if (dest_charset == C_ISO_2022_JP ||
-                        dest_charset == C_ISO_2022_JP_2)
+               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;
                break;
        default:
@@ -800,12 +894,12 @@ gchar *conv_iconv_strdup(const gchar *inbuf,
        const gchar *inbuf_p;
        gchar *outbuf;
        gchar *outbuf_p;
-       gint in_size;
-       gint in_left;
-       gint out_size;
-       gint out_left;
-       gint n_conv;
-       gint len;
+       size_t in_size;
+       size_t in_left;
+       size_t out_size;
+       size_t out_left;
+       size_t n_conv;
+       size_t len;
 
        if (!src_code)
                src_code = conv_get_outgoing_charset_str();
@@ -813,11 +907,11 @@ gchar *conv_iconv_strdup(const gchar *inbuf,
                dest_code = conv_get_current_charset_str();
 
        /* don't convert if current codeset is US-ASCII */
-       if (!strcasecmp(dest_code, CS_US_ASCII))
+       if (!g_strcasecmp(dest_code, CS_US_ASCII))
                return g_strdup(inbuf);
 
        /* don't convert if src and dest codeset are identical */
-       if (!strcasecmp(src_code, dest_code))
+       if (!g_strcasecmp(src_code, dest_code))
                return g_strdup(inbuf);
 
        cd = iconv_open(dest_code, src_code);
@@ -842,7 +936,7 @@ gchar *conv_iconv_strdup(const gchar *inbuf,
 }
 
        while ((n_conv = iconv(cd, (ICONV_CONST gchar **)&inbuf_p, &in_left,
-                              &outbuf_p, &out_left)) < 0) {
+                              &outbuf_p, &out_left)) == (size_t)-1) {
                if (EILSEQ == errno) {
                        inbuf_p++;
                        in_left--;
@@ -862,7 +956,8 @@ gchar *conv_iconv_strdup(const gchar *inbuf,
                }
        }
 
-       while ((n_conv = iconv(cd, NULL, NULL, &outbuf_p, &out_left)) < 0) {
+       while ((n_conv = iconv(cd, NULL, NULL, &outbuf_p, &out_left)) ==
+              (size_t)-1) {
                if (E2BIG == errno) {
                        EXPAND_BUF();
                } else {
@@ -930,6 +1025,7 @@ static const struct {
        {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_SHIFT_JIS,           CS_SHIFT_JIS},
@@ -979,6 +1075,7 @@ static const struct {
        {"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},
@@ -1218,14 +1315,14 @@ CharSet conv_get_current_charset(void)
 
                /* "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,
+               if (!g_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_strncasecmp(cur_locale, locale_table[i].locale, 2)) {
                                cur_charset = locale_table[i].charset;
                                return cur_charset;
                        }
@@ -1270,14 +1367,14 @@ CharSet conv_get_outgoing_charset(void)
        for (i = 0; i < sizeof(locale_table) / sizeof(locale_table[0]); i++) {
                const gchar *p;
 
-               if (!strncasecmp(cur_locale, locale_table[i].locale,
+               if (!g_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)) {
+                           !g_strncasecmp(cur_locale, locale_table[i].locale, 2)) {
                                out_charset = locale_table[i].out_charset;
                                break;
                        }
@@ -1302,7 +1399,7 @@ const gchar *conv_get_outgoing_charset_str(void)
        const gchar *str;
 
        if (prefs_common.outgoing_charset) {
-               if (!isalpha(prefs_common.outgoing_charset[0])) {
+               if (!isalpha((guchar)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)
@@ -1324,6 +1421,7 @@ gboolean conv_is_multibyte_encoding(CharSet encoding)
        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:
@@ -1342,9 +1440,12 @@ const gchar *conv_get_current_locale(void)
        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);
+       if (!cur_locale || !strlen(cur_locale)) 
+               cur_locale = g_getenv("LC_CTYPE");
+       if (!cur_locale || !strlen(cur_locale)) 
+               cur_locale = g_getenv("LANG");
+       if (!cur_locale || !strlen(cur_locale)) 
+               cur_locale = setlocale(LC_CTYPE, NULL);
 
        debug_print("current locale: %s\n",
                    cur_locale ? cur_locale : "(none)");
@@ -1352,27 +1453,6 @@ const gchar *conv_get_current_locale(void)
        return cur_locale;
 }
 
-void conv_unmime_header_overwrite(gchar *str)
-{
-       gchar *buf;
-       gint buflen;
-       CharSet cur_charset;
-
-       cur_charset = conv_get_current_charset();
-
-       if (cur_charset == C_EUC_JP) {
-               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)
 {
@@ -1397,17 +1477,15 @@ void conv_unmime_header(gchar *outbuf, gint outlen, const gchar *str,
 #define MIMESEP_BEGIN          "=?"
 #define MIMESEP_END            "?="
 
-#define B64LEN(len)    ((len) / 3 * 4 + ((len) % 3 ? 4 : 0))
-
 #define LBREAK_IF_REQUIRED(cond, is_plain_text)                                \
 {                                                                      \
-       if (len - (destp - dest) < MAX_LINELEN + 2) {                   \
+       if (len - (destp - (guchar *)dest) < MAX_LINELEN + 2) {         \
                *destp = '\0';                                          \
                return;                                                 \
        }                                                               \
                                                                        \
        if ((cond) && *srcp) {                                          \
-               if (destp > dest && left < MAX_LINELEN - 1) {           \
+               if (destp > (guchar *)dest && left < MAX_LINELEN - 1) { \
                        if (isspace(*(destp - 1)))                      \
                                destp--;                                \
                        else if (is_plain_text && isspace(*srcp))       \
@@ -1429,8 +1507,8 @@ void conv_encode_header(gchar *dest, gint len, const gchar *src,
        gint mimestr_len;
        gchar *mimesep_enc;
        gint left;
-       const gchar *srcp = src;
-       gchar *destp = dest;
+       const guchar *srcp = src;
+       guchar *destp = dest;
        gboolean use_base64;
 
        if (MB_CUR_MAX > 1) {
@@ -1491,7 +1569,7 @@ void conv_encode_header(gchar *dest, gint len, const gchar *src,
                        gchar *part_str;
                        gchar *out_str;
                        gchar *enc_str;
-                       const gchar *p = srcp;
+                       const guchar *p = srcp;
                        gint out_str_len;
                        gint out_enc_str_len;
                        gint mime_block_len;