Dominus on the non-greedy version of the ? quantifier

[Aristotle]

So I was just looking at Dominus’ Regular Expression Mastery talk, since I hadn’t seen it before. I’ve been around the block enough times to know the subject matter cold, but of course understanding and mastery are separate issues and it’s always good to take another look – in complex subjects, other people will often bring up angles to look at some of their aspects that you never thought of before. Anyway, I was skimming the slides, but it didn’t look like I’d be taking home anything from this talk.

Until I came to the absolute end, the very last bullet of the very last slide:

• Also there’s a non-greedy version X??
• I used to pay US$60 for a live sighting of ?? in the wild
• But one day I thought of

  if ( $option =~ /^-f(i(e(ld??)??)??)??$/ ) {
      ...
  }
  

I’ve been staring at that example for 10 minutes and I really can’t see how it would ever match differently from the greedy version:

/^-f(i(e(ld?)?)?)?$/

From what I can tell, since the entire pattern is anchored at both ends and the captures are nested, it must always match the exact same things, whether the quantifiers are greedy or not.

Am I missing something?

[rinceWind]

My take on regexp greediness, is that this affects the order in which matches are tried.

The greedy form of * and + try the longest match first, and successively decreasing the candidate match. Non-greedy forms *? and +? start with the shortest match and gradually increase the length.

The ? operator matches 0 or 1 occurrence. 1 occurrence is tried first. The ?? operator tries 0 occurrences first.

In the example you have given, it does not matter which is tried first, as you will get the same result. If there were another quantifier downstream of the ??, this could make a difference, as it would see the string to the left of the clause matching the ?? first, wherease a greedy ? would present the remainder of the string to the right of the match, first.

[Aristotle]

I know what the quantifiers mean and how greediness works – which is exactly why I’m asking. I cannot see what Dominus is hinting at by bringing up this particular example, because greediness does not affect its result in any way.

I was hoping someone would have an idea as to what he might have been thinking.

[jhourcle]

...because greediness does not affect its result in any way.

I was hoping someone would have an idea as to what he might have been thinking.

You're only thinking about the final outcome. Sometimes, changes are not made for the outcome, but because it changes how it performs, and one method may be better for the normally expected input.

Unfortunately, in this case, I'm not seeing that on my system the non-greedy being significantly better at matching the short version, and it's significantly worse at matching the long version:

Read more... (868 Bytes)

Oh -- and it _does_ change the values of $1, $2, $3, but we have no reason from the code snippet given that they're important.,/strike>

Update: Removed extra comma, resulting in sloppy english grammer, and stressed the 'not'. Was :

Unfortunately, in this case, I'm not seeing that on my system, the non-greedy being significantly better at matching the short version ...

Update: tye is correct in the lack of difference. Why a non-greedy expression would match 1 when it has the option of matching 0 makes no sense to me, though.

[tilly]

Yes, you're missing something.

The point is not what the code would do when run. It is what it could do to Dominus' wallet.

[Aristotle]

Oh, d’oh. Thanks for the pointer. :-)