Re: extended rx.el

[Stefan Monnier]

> I think that info is worth including somewhere -- in the rx.el
> commentary if nowhere else, but probably in the doc for shy groups.

I'd rather keep this info close to the source code (i.e. regex.c) since it
could change at any time.  I think people intuitively (and correctly with
the current implementation) know/expect that \(?:foo\) will be slower than
foo but faster than \(foo\) which is all that really matters.

> I'm glad I guessed correctly that I didn't know enough to make a
> useful measurement!

Performance is always too tricky to worry about ;-)


        Stefan