Juliusz Gonera wrote:
I understand that thanks to this patch we can be 100% sure that the error comes from AbuseFilter even if the `error-msg` has no "abusefilter-" prefix. However, we should also include some information which would allow us to figure out which action from "Actions taken when matched" box in filter config a given API response is referring to. That would be much more reliable than looking for "abusefilter-warning-" prefix which can be omitted by admins setting up the warning.
I'll take a look at adding this.
— Andrew Garrett