Properly split long IRC messages

This commit adds handling of long IRC messages to the core. In contrast
to the `splitlong.pl' plugin, multi-byte encoded and recoded messages
are properly split.

To allow for this, a new function has been added to the server struct:
`split_message'. `split_message' returns a string array with the message
splitted to substrings of a length that the server can handle. If a
protocol module doesn't have any limit, it can simply return a singleton
array with a copy of the message.

The `MSG' chat command now calls `split_message' before `send_message',
and emits `message own_public' / `message own_private' with each
substring, so that the string splitting will be visible in the UI.

`split_message' in the IRC module uses `recode_split' which in turn uses
iconv to properly split multi-byte encoded (and recoded) messages.

src/core/chat-commands.c

@@ -378,12 +378,23 @@ static void cmd_msg(const char *data, SERVER_REC *server, WI_ITEM_REC *item)
 		}
 	}
 	if (target != NULL) {
-		signal_emit("server sendmsg", 4, server, target, msg,
+		char **splitmsgs = server->split_message(server, target, msg);
+		char *m;
+		int n = 0;
+
+		while ((m = splitmsgs[n++])) {
+			signal_emit("server sendmsg", 4, server, target, m,
 				    GINT_TO_POINTER(target_type));
+			signal_emit(target_type == SEND_TARGET_CHANNEL ?
+			            "message own_public" :
+				    "message own_private", 4, server, m,
+				    target, origtarget);
+		}
+		g_strfreev(splitmsgs);
+	} else {
+		signal_emit("message own_private", 4, server, msg, target,
+			    origtarget);
 	}
-	signal_emit(target != NULL && target_type == SEND_TARGET_CHANNEL ?
-		    "message own_public" : "message own_private", 4,
-		    server, msg, target, origtarget);
 
 	if (free_ret && target != NULL) g_free(target);
 	cmd_params_free(free_arg);

src/core/misc.c

@@ -966,3 +966,20 @@ char *ascii_strdown(char *str)
 		*s = g_ascii_tolower (*s);
 	return str;
 }
+
+char **strsplit_len(const char *str, int len)
+{
+	char **ret;
+	int n = strlen(str) / len;
+	int i;
+
+	if (strlen(str) % len)
+		n++;
+
+	ret = g_new(char *, n + 1);
+	for (i = 0; i < n; i++, str += len)
+		ret[i] = g_strndup(str, len);
+	ret[n] = NULL;
+
+	return ret;
+}

src/core/misc.h

@@ -115,4 +115,7 @@ uoff_t str_to_uofft(const char *str);
 /* find `item' from a space separated `list' */
 int find_substr(const char *list, const char *item);
 
+/* split `str' into `len' sized substrings */
+char **strsplit_len(const char *str, int len);
+
 #endif

src/core/recode.c

@@ -182,6 +182,87 @@ char *recode_out(const SERVER_REC *server, const char *str, const char *target)
 	return recoded;
 }
 
+char **recode_split(const SERVER_REC *server, const char *str,
+		    const char *target, int len)
+{
+	GIConv cd = (GIConv)-1;
+	const char *from = translit_charset;
+	const char *to = translit_charset;
+	char *translit_to = NULL;
+	const char *inbuf = str;
+	const char *previnbuf = inbuf;
+	char *tmp = NULL;
+	char *outbuf;
+	gsize inbytesleft = strlen(inbuf);
+	gsize outbytesleft = len;
+	int n = 0;
+	char **ret;
+
+	if (!str)
+		return NULL;
+
+	if (settings_get_bool("recode")) {
+		to = find_conversion(server, target);
+		if (to == NULL)
+			/* default outgoing charset if set */
+			to = settings_get_str("recode_out_default_charset");
+		if (to && *to != '\0') {
+			if (settings_get_bool("recode_transliterate") &&
+			    !is_translit(to))
+				to = translit_to = g_strconcat(to,
+							       "//TRANSLIT",
+							       NULL);
+		} else {
+			to = from;
+		}
+	}
+
+	cd = g_iconv_open(to, from);
+	if (cd == (GIConv)-1) {
+		/* Fall back to splitting by byte. */
+		ret = strsplit_len(str, len);
+		goto out;
+	}
+
+	tmp = g_malloc(outbytesleft);
+	outbuf = tmp;
+	ret = g_new(char *, 1);
+	while (g_iconv(cd, (char **)&inbuf, &inbytesleft, &outbuf,
+		       &outbytesleft) == -1) {
+		if (errno != E2BIG) {
+			/*
+			 * Conversion failed. Fall back to splitting
+			 * by byte.
+			 */
+			ret[n] = NULL;
+			g_strfreev(ret);
+			ret = strsplit_len(str, len);
+			goto out;
+		}
+
+		/* Outbuf overflowed, split the input string. */
+		ret[n++] = g_strndup(previnbuf, inbuf - previnbuf);
+		ret = g_renew(char *, ret, n + 1);
+		previnbuf = inbuf;
+
+		/* Reset outbuf for the next substring. */
+		outbuf = tmp;
+		outbytesleft = len;
+	}
+	/* Copy the last substring into the array. */
+	ret[n++] = g_strndup(previnbuf, inbuf - previnbuf);
+	ret = g_renew(char *, ret, n + 1);
+	ret[n] = NULL;
+
+out:
+	if (cd != (GIConv)-1)
+		g_iconv_close(cd);
+	g_free(translit_to);
+	g_free(tmp);
+
+	return ret;
+}
+
 void recode_update_charset(void)
 {
 	const char *charset = settings_get_str("term_charset");

src/core/recode.h

@@ -3,6 +3,8 @@
 
 char *recode_in (const SERVER_REC *server, const char *str, const char *target);
 char *recode_out (const SERVER_REC *server, const char *str, const char *target);
+char **recode_split(const SERVER_REC *server, const char *str,
+		    const char *target, int len);
 gboolean is_valid_charset(const char *charset);
 gboolean is_utf8(void);
 void recode_update_charset(void);

src/core/server-rec.h

@@ -61,6 +61,9 @@ const char *(*get_nick_flags)(SERVER_REC *server);
 /* send public or private message to server */
 void (*send_message)(SERVER_REC *server, const char *target,
 		     const char *msg, int target_type);
+/* split message in case it is too long for the server to receive */
+char **(*split_message)(SERVER_REC *server, const char *target,
+			const char *msg);
 
 /* -- Default implementations are used if NULL -- */
 CHANNEL_REC *(*channel_find_func)(SERVER_REC *server, const char *name);

src/irc/core/irc-servers.c

@@ -102,6 +102,30 @@ static void send_message(SERVER_REC *server, const char *target,
 	g_free(recoded);
 }
 
+static char **split_message(SERVER_REC *server, const char *target,
+			    const char *msg)
+{
+	IRC_SERVER_REC *ircserver = IRC_SERVER(server);
+	int userhostlen = 63;  /* Maximum length defined by protocol. */
+
+	g_return_val_if_fail(ircserver != NULL, NULL);
+	g_return_val_if_fail(target != NULL, NULL);
+	g_return_val_if_fail(msg != NULL, NULL);
+
+	/*
+	 * If we have joined a channel, userhost will be set, so we can
+	 * calculate the exact maximum length.
+	 */
+	if (ircserver->userhost != NULL)
+		userhostlen = strlen(ircserver->userhost);
+
+	/* length calculation shamelessly stolen from splitlong.pl */
+	return recode_split(SERVER(server), msg, target,
+			    510 - strlen(":! PRIVMSG  :") -
+			    strlen(ircserver->nick) - userhostlen -
+			    strlen(target));
+}
+
 static void server_init(IRC_SERVER_REC *server)
 {
 	IRC_SERVER_CONNECT_REC *conn;
@@ -288,6 +312,7 @@ static void sig_connected(IRC_SERVER_REC *server)
 
 	server->isnickflag = isnickflag_func;
 	server->ischannel = ischannel_func;
+	server->split_message = split_message;
 	server->send_message = send_message;
 	server->query_find_func =
 		(QUERY_REC *(*)(SERVER_REC *, const char *)) irc_query_find;