Re: Fixing ill-conditioned regular expressions. Proof of concept.

[Stefan Monnier]

> I really did mean R*\(\)R*, with R being the same on both sides of the
> \(\), but the *s possibly being +s.

Ah, well R*<E>R* where <E> can match the empty string is definitely
nasty given the current implementation technique of our regexp
matcher, yes.
It's not specific to \(\), OTOH.

> _That_ is nasty and undefined.

Well, it's actually defined (by the "leftmost longest" rule), but
I agree it's a nasty case.


        Stefan