Re: Our regex vs. POSIX on "longest match"

hubert depesz lubaczewski <depesz(at)depesz(dot)com> writes:
> I stand on position that mixing greedy and non-greedy operators should
> be possible, and that it should work according to POLA - i.e. greedines
> of given atom shouldn't be influenced by other atoms.

[ shrug... ] That sounds good, but it's pretty much vacuous as far as
defining a principled alternative behavior goes. It's easy to
demonstrate cases where atoms *must* be influenced by other ones.
A trivial example is
(.*)(.*)
It doesn't matter whether the second atom is greedy or not: it's not
going to get to eat anything because the first one does instead.
IOW this is just the same as
(.*)(.*?)
--- they are both overall-greedy.

regards, tom lane