diff libpurple/protocols/oscar/oscar.c @ 25497:67ad619bd97f

propagate from branch 'im.pidgin.pidgin' (head 3d1e6b8637ff9133c8f17e9aeaeea1284c898d86) to branch 'im.pidgin.pidgin.yaz' (head cf1ed778db33093adccaec9aa0399ac72a0b712f)

--- a/libpurple/protocols/oscar/oscar.c	Mon Jun 04 06:57:25 2007 +0000
+++ b/libpurple/protocols/oscar/oscar.c	Mon Jun 04 07:22:38 2007 +0000
@@ -303,7 +303,25 @@
 	gchar *utf8 = NULL;
 
 	if ((encoding == NULL) || encoding[0] == '\0') {
-		purple_debug_info("oscar", "Empty encoding, assuming UTF-8\n");
+		purple_debug_info("yaz oscar", "Empty encoding, validate as UTF-8\n");
+		if(g_utf8_validate(text, textlen, NULL)){
+			size_t newlen;
+			utf8 = sanitize_utf(text, textlen, &newlen);
+			goto done;
+		}
+		// not UTF-8
+		purple_debug_info("yaz oscar", "Empty encoding, assuming UCS-2BE\n");
+		sanitize_ucs(text, textlen);
+		utf8 = g_convert(text, textlen, "UTF-8", "UCS-2BE", NULL, NULL, NULL);
+		if(utf8){
+			if(!g_utf8_validate(utf8, strlen(utf8), NULL)){
+				purple_debug_info("yaz oscar", "Invalid conversion\n");
+				g_free(utf8);
+				utf8 = NULL;
+			}
+		} else {
+			purple_debug_info("yaz oscar", "Conversion failed\n");
+		}
 	} else if (!g_ascii_strcasecmp(encoding, "iso-8859-1")) {
 		utf8 = g_convert(text, textlen, "UTF-8", "iso-8859-1", NULL, NULL, NULL);
 	} else if (!g_ascii_strcasecmp(encoding, "ISO-8859-1-Windows-3.1-Latin-1") ||
@@ -311,6 +329,7 @@
 	{
 		utf8 = g_convert(text, textlen, "UTF-8", "Windows-1252", NULL, NULL, NULL);
 	} else if (!g_ascii_strcasecmp(encoding, "unicode-2-0")) {
+		sanitize_ucs(text, textlen);
 		utf8 = g_convert(text, textlen, "UTF-8", "UCS-2BE", NULL, NULL, NULL);
 	} else if (g_ascii_strcasecmp(encoding, "utf-8")) {
 		purple_debug_warning("oscar", "Unrecognized character encoding \"%s\", "
@@ -331,7 +350,7 @@
 		else
 			utf8 = g_strndup(text, textlen);
 	}
-
+done:
 	return utf8;
 }
 
@@ -488,6 +507,7 @@
 		if ((b != NULL) && (PURPLE_BUDDY_IS_ONLINE(b)))
 		{
 			*msg = g_convert(from, strlen(from), "UCS-2BE", "UTF-8", NULL, &msglen, NULL);
+			botch_ucs(*msg, msglen);
 			if (*msg != NULL)
 			{
 				*charset = AIM_CHARSET_UNICODE;
@@ -522,6 +542,7 @@
 	 * Nothing else worked, so send as UCS-2BE.
 	 */
 	*msg = g_convert(from, strlen(from), "UCS-2BE", "UTF-8", NULL, &msglen, &err);
+	botch_ucs(*msg, msglen);
 	if (*msg != NULL) {
 		*charset = AIM_CHARSET_UNICODE;
 		*charsubset = 0x0000;
@@ -1961,6 +1982,8 @@
 		tmp = purple_plugin_oscar_decode_im_part(account, userinfo->sn, curpart->charset,
 				curpart->charsubset, curpart->data, curpart->datalen);
 		if (tmp != NULL) {
+			purple_str_strip_char(tmp, 0x0d); // yaz: strip CR
+//			purple_debug_info("yaz oscar", "tmp=%s",tmp);
 			g_string_append(message, tmp);
 			g_free(tmp);
 		}
@@ -2061,6 +2084,7 @@
 		char *encoding, *utf8name, *tmp;
 		GHashTable *components;
 
+//		purple_debug_info("yaz oscar", "chat request %s\n", args->msg);
 		if (!args->info.chat.roominfo.name || !args->info.chat.roominfo.exchange) {
 			g_free(message);
 			return 1;
@@ -2083,6 +2107,8 @@
 		g_hash_table_replace(components, g_strdup("room"), utf8name);
 		g_hash_table_replace(components, g_strdup("exchange"),
 				g_strdup_printf("%d", args->info.chat.roominfo.exchange));
+		purple_debug_info("yaz oscar", "about to call serv_got_chat_invite\n");
+//		purple_debug_info("yaz oscar", "name=%s message=%s\n", name ? name : args->info.chat.roominfo.name, message);
 		serv_got_chat_invite(gc,
 				     utf8name,
 				     userinfo->sn,
@@ -2317,7 +2343,7 @@
 	 * for this suck-ass part of the protocol by splitting the string into at
 	 * most 1 baby string.
 	 */
-	msg1 = g_strsplit(args->msg, "\376", (args->type == 0x01 ? 1 : 0));
+	msg1 = g_strsplit(args->msg, "\376", (args->type == 0x01 ? 1 : 0)); // \376 is 0xfe
 	for (numtoks=0; msg1[numtoks]; numtoks++);
 	msg2 = (gchar **)g_malloc((numtoks+1)*sizeof(gchar *));
 	for (i=0; msg1[i]; i++) {
@@ -4344,6 +4370,7 @@
 	charset = oscar_charset_check(str);
 	if (charset == AIM_CHARSET_UNICODE) {
 		encoded = g_convert(str, strlen(str), "UCS-2BE", "UTF-8", NULL, ret_len, NULL);
+		botch_ucs(encoded, *ret_len);
 		*encoding = "unicode-2-0";
 	} else if (charset == AIM_CHARSET_CUSTOM) {
 		encoded = g_convert(str, strlen(str), "ISO-8859-1", "UTF-8", NULL, ret_len, NULL);
@@ -5359,7 +5386,7 @@
 		charsetstr = "unicode-2-0";
 	else if (charset == AIM_CHARSET_CUSTOM)
 		charsetstr = "iso-8859-1";
-	aim_chat_send_im(od, c->conn, 0, buf2, len, charsetstr, "en");
+	aim_chat_send_im(od, c->conn, 0, buf2, len, charsetstr, "JA");
 	g_free(buf2);
 
 	return 0;