If there's no SRV record, maybe we should default to the input domain, as suggested in the RFC?
* If an SRV record is not found, the client will need to prompt
the user to enter the FQDN and port number information
directly or use some other heuristic, for example, using the
extracted "domain" as the FQDN and default HTTPS or HTTP port
numbers.
Or at least return a fixed error value which can be handled by the user.
If there's no SRV record, maybe we should default to the input domain, as suggested in the RFC?
Or at least return a fixed error value which can be handled by the user.