regex lazy before capturing group

Question

Using the following regex

<w:p.*?\$\{test\}.*?\/w:p>

I'm trying to match the first

<w:p>

before the "${test}" and the first

</w:p>

after. The after worked just fine, using the ? quantifier, but it refuses to stop at the first

<w:body><w:p w:rsidRDefault="00271ADB"/><w:p w:rsidR="00C15291"><w:pPr><w:p w:rsidR="0093632F" w:rsidRDefault="0093632F"><w:pPr><w:rPr></w:rPr></w:pPr><w:r><w:rPr></w:rPr><w:br/><w:t>${test}</w:t></w:r></w:p></w:body>

This is what I expected the result to be:

<w:p w:rsidR="0093632F" w:rsidRDefault="0093632F"><w:pPr><w:rPr></w:rPr></w:pPr><w:r><w:rPr></w:rPr><w:br/><w:t>${test}</w:t></w:r></w:p>

but instead this is what being returned

<w:p w:rsidRDefault="00271ADB"/><w:p w:rsidR="00C15291"><w:pPr><w:p w:rsidR="0093632F" w:rsidRDefault="0093632F"><w:pPr><w:rPr></w:rPr></w:pPr><w:r><w:rPr></w:rPr><w:br/><w:t>${test}</w:t></w:r></w:p>

This is the result in the editor: https://i.sstatic.net/4ri4C.png

And this is the result I'm expecting: https://i.sstatic.net/W87K9.png

Sam · Accepted Answer · 2014-06-05 14:38:08Z

3

You'll have to change the first .*? into a repeated group with a negative lookahead. You should also notice that I added a \s after <w:p, this is so <w:pPr doesn't get matched. If you have some <w:p> instances, you may need to change this to <w:p(?:\s|>).

<w:p\s(?:(?!<w:p\s).)*?\$\{test\}.*?\/w:p>

Demo

RegEx matches from left to right, so there is no real way to say "lazy before". Instead of .*? I used (?:(?!<w:p\s).)*?. Lets break that down:

(?:         (?# begin non-capturing group for grouping/repetition)
  (?!       (?# begin negative lookahead)
    <w:p\s  (?# no <w:p ahead)
  )         (?# end negative lookahead)
  .         (?# match any character)
)*?         (?# lazy repetition)

How this works is as soon as we match <w:p\s, we enter the non-capturing/repeated group. It does a zero-length assertion to make sure <w:p\s doesn't exist ahead of that point, and then matches a character. This lazily repeats until we hit ${test}. If the expression sees a <w:p\s in the lookahead, it will fail..and a new match will start back up, matching that <w:p\s in the beginning (and starting to do more lookaheads).

answered Jun 5, 2014 at 14:38

Sam

20.5k2 gold badges48 silver badges71 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Diego Vieira Over a year ago

Very clever! Thank you Sam for the explanation as well :)

zx81 Over a year ago

Nice detailed answer delving into laziness. :) +1

Collectives™ on Stack Overflow

regex lazy before capturing group

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related