perl, match one of multiple regular expressions

Question

I've currently got the following code:

if ($str =~ m{^[A-Z]+-\d+$} || $str =~ m{^\d+$}){
    # do stuff
}

Is it possible to combine the 2 regular expressions into a single expression? And would that improve performance at all?

What you ask, as answered by [Wiktor Stribiżew](https://stackoverflow.com/users/3832970/wiktor-stribi%C5%BCew), in the first place improves how the intent is conveyed -- match `\d+` optionally preceded by that other. It's just clearer that way, than when written with `||`. Which is faster depends on which case happens more often, but I'd say that _on average_ the single regex should be faster since it never starts the engine twice. — zdim, Jul 14 '17 at 20:58
As a side note, there is an interesting post where [regex optimization in Perl](https://stackoverflow.com/questions/36420517/is-it-faster-to-use-alternation-than-subsequent-replacements-in-regular-expressi/36611087#36611087) is discussed. If the patterns could not be "joined" with an optional non-capuring group, I think it would be more appropriate to just use the `||` with two separate regexps as *"alternations hinder the optimizer"*. — Wiktor Stribiżew, Jul 14 '17 at 21:14

score 3 · Accepted Answer · answered Jul 14 '17 at 20:46

3

I would use an optional non-capturing group and combine these two into

if ($str =~ m{^(?:[A-Z]+-)?\d+$}) {
    # do stuff
}

Details

^ - start of string
(?:[A-Z]+-)? - an optional non-capturing group (? quantifier makes it match 1 or 0 times)
\d+ - 1 or more digits
$ - end of string.

answered Jul 14 '17 at 20:46

Wiktor Stribiżew

For more complex regexps, consider the `/x` modifier. – choroba Jul 14 '17 at 20:57

1 Answers1