I think the warning should be parsing complex domains with regex libraries, not a restriction to regular syntaxes. There are simple cases where backtracking (or some other extension) plus regex makes sense. Something like
(\w+) \d+ \1
is simpler as a regex than any other common tool available.
> Alternative title: you can use regex derivatives...
This is particularly confusing because you mean "derivative" in the sense of "etymologically/conceptually related", but there is actually a (formal) notion of the derivative of a regular expression.
Alternative title: you can use regex derivatives (PCRE, Ruby ::Regexp, Boost.Regex, ...) to match non-regular syntax but you shouldn't.
The advice still holds. In the case of HTML you should reach for an XML parser.