Fakeweb slow

[Brian Cardarella]

We're using Fakeweb in Cucumber Scenario Outlines with 50 examples
(states)

In each outline we are faking out 3 resources. We're using regular
expressions to match the URL Nothing fancy, just a greedy pattern
match at the end of each URL.

Short story long, this is slowing out test suite down by a lot. I was
told that Fakeweb does some brute force pattern recognition stuff. Is
this true? Any tips on speeding this up?

We're on Fakeweb 1.2.8