You can block based on Regex but you can't do the same in severity mapping.
This means you either:
1. Have to block the object with Regex.
2. Have to severity map the object using it's name only or basic wildcard functionality.
There are a lot of cases where (1) is not appropriate. Equally there are a lot of cases such as when the object name changes where (2) cannot be used due to a lack of Regex support.
The best fix would be to overhaul the 3 features (block, ignore & map) as the UX for them is horrendous and they behave in different ways. It would be nicer for a user to be able to give an inbound URL, add regex rules for this URL and then choose an action since they are either going to want to Severity Map it OR Block it, then choose the conditions such as on Certain HTTP Responses or other errors (NET_INTERRUPT) etc, then finally choose between blocking, mapping or ignoring. So one interface could surely do both jobs?
Update: This was released in 8.4