How this regex works?

Question

System.out.println("du hast mich".replaceAll("(?<=^(.*)) ", ", $1 "));
// prints "du, du hast, du hast mich"

what is the mean of ^ symbol after the look behind ? (I know standard mean of this symbol is start of the line) and why dot symbol matches up to du then du hast then du hast mich.In briesf why the dot symbol didn't match the whole string?

Please give me an explanation how this regex works properly.I am wondering.Thanks for your interest.

Depending on the context, ^ can mean "beginning of the line" or "not (expression)" (but here, it is the beginning of the line) — Arnaud Denoyelle
– Arnaud Denoyelle, Commented Oct 24, 2013 at 14:19

Kendall Frey · Accepted Answer · 2013-10-24 14:19:21Z

3

(?<= ) is the syntax for lookbehind. The ^ is just the "start of string" anchor. Essentially what the regex is saying is:

"Match a space which is preceded by the start of the string and any number of characters. The characters preceding the space are the first captured group."

answered Oct 24, 2013 at 14:19

Kendall Frey

44.6k21 gold badges113 silver badges151 bronze badges

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

2 Comments

Kendall Frey Over a year ago

@MelihAltıntaş It only matches up until the end of the lookbehind, which stops just before the space (or it wouldn't be lookbehind anymore). And each space matches separately, so you get several matches.

Melih Altıntaş Over a year ago

thanks for your interest :) I didn't see the last space and i try to understand so far :))

Sotirios Delimanolis · Accepted Answer · 2013-10-24 14:30:54Z

2

Kendall has the explanation. Here's the step by step.

du hast mich
 ^ regex hasn't matched anything so no replacement

writes

du

du hast mich
  ^ regex matches

replaces the match with a comma and everything before the space

, du

du hast mich
      ^ no match

writes

hast

du hast mich
       ^ regex matches

replaces that match with a comma and everything before the space

, du hast

du hast mich
           ^ no match

leaves it as is

mich

combine all that and you get

du, du hast, du hast mich

answered Oct 24, 2013 at 14:30

Sotirios Delimanolis

281k62 gold badges718 silver badges744 bronze badges

2 Comments

Sotirios Delimanolis Over a year ago

@MelihAltıntaş Note the space after the last ) in your regex.

Melih Altıntaş Over a year ago

thanks for your interest :) I didn't see the last space and i try to understand so far :))

Alan Moore · Accepted Answer · 2013-10-24 16:08:46Z

0

That regex shouldn't work at all. What it should do is throw an exception because of the open-ended quantifier (.*) in the lookbehind. You seem to have discovered a glitch that lets you bypass that rule. But don't use it! It's definitely a bug, not a feature.

Java's lookbehinds have always been a little twitchy, which I attribute to its complicated known maximum length requirement for lookbehind subexpressions. I've come to feel that feature was a mistake; it's just not useful to enough to justify the hassles it brought with it. This is why I try to avoid using any quantifiers in my lookbehinds.

answered Oct 24, 2013 at 16:08

Alan Moore

75.6k13 gold badges110 silver badges161 bronze badges

Collectives™ on Stack Overflow

How this regex works?

3 Answers 3

2 Comments

2 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

2 Comments

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related