First sorry for completely forgetting about this patch series.
Am 25.08.21 um 23:02 schrieb [email protected]:
From: Selva Nair <[email protected]>
Selecting the remote host via the management iterface
Typo in interface
+
+static bool
+management_callback_remote_entry_get(void *arg, unsigned int index, char
**remote)
+{
+ assert(arg);
+ assert(remote);
+
+ struct context *c = (struct context *) arg;
+ struct connection_list *l = c->options.connection_list;
+ bool ret = true;
+
+ if (index < l->len)
+ {
+ struct connection_entry *ce = l->array[index];
+ const char *proto = proto2ascii(ce->proto, ce->af, false);
+
+ /* space for output including 2 commas and a nul */
+ int len = strlen(ce->remote) + strlen(ce->remote_port) + strlen(proto)
+ 2 + 1;
+ char *out = malloc(len);
+ check_malloc_return(out);
Not sure that we need to exit the program here instead of of return
false but the error is so seldom that it does not really matter. Using
gc_arena would have been also a bit nicer but not too problematic.
+
+ openvpn_snprintf(out, len, "%s,%s,%s", ce->remote, ce->remote_port,
proto);
+ *remote = out;
+ }
+ else
+ {
+ ret = false;
+ msg(M_WARN, "Out of bounds index in management query for remote entry:
index = %u", index);
+ }
+
+ return ret;
I feel we could eliminate ret variable here and just do return true and
return false from the true branches of the if.
There are minor things that could be improved but the current state is
still acceptable.
Acked-By: Arne Schwabe <[email protected]>
_______________________________________________
Openvpn-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openvpn-devel
