in reply to Re^2: Perl regexp matching is slow??
in thread Perl regexp matching is slow??

I think you could rework this to only include [A-Z] in the opcode, if there is a number following the [A-Z] its not important. In other words just look for OPEN not OPEN1 etc.

Also i think that youll need to add MINMOD to the list of non DFA'able constructs, and maybe also should put OPEN in there as well since capturing with a DFA is distinctly non-trivial (actually im not even sure you can do it with a pure-DFA, maybe with NFA based DFA simulation). And why is NOTHING in both lists?

---
$world=~s/war/peace/g

Replies are listed 'Best First'.
Re^4: Perl regexp matching is slow??
by diotalevi (Canon) on Mar 01, 2007 at 02:28 UTC

    NOTHING is in the lists because it showed up as an op in the word extraction. You oughta notice that what I wrote was really, really basic and was just doing existence tests for the named ops.

    ⠤⠤ ⠙⠊⠕⠞⠁⠇⠑⠧⠊

[reply]

In Section Meditations