ansaurus

Question

Answer 1

+1 A:

You're on the right track; Just use a lookbehind for the next delimiter:

if ( sc.useDelimiter("(?=:28)").hasNext() )
{
  System.out.printf("%n%s%n",  sc.next() );
}
if ( sc.useDelimiter("(?<=</n0:message>)").hasNext() )
{
  System.out.printf("%n%s%n",  sc.next() );
}
if ( sc.useDelimiter("\\z").hasNext() )
{
  System.out.printf("%n%s%n",  sc.next() );
}

But if you've already read the text into a String, it would probably be easier use Matcher.find() or String.split(), or even indexOf() and substring(). I'll elaborate if you're interested.

Alan Moore 2010-08-02 05:23:30

Thanks Alan, that worked perfectly. Can you explain a bit further the look behind and the regular expressions? I understand the first delimiter, not too include :28, so that gives me the first token. So for the second, i dont understand that reg exp or the last one.

shane lee 2010-08-02 23:50:19

Actually i understand the look behind and \\z which means the end of the string but why do i use ?= for first delimiter and ?<= for the second?Thanks,Shane.

shane lee 2010-08-03 00:44:30

If you were to use `(?=</n0:message>)` as the second delimiter, it would match the position *before* the `</n0:message>`, which would become part of the third token. You said you wanted it to stay in the second token, so I used a lookbehind to match the position right *after* it instead.

Alan Moore 2010-08-03 01:58:58

Yes your correct Alan, i just did not understand that expression ?<=. So this value means it matches the position after it, whereas ?= matches position before?

shane lee 2010-08-03 03:25:07

Yep. This might interest you: http://www.regular-expressions.info/lookaround.html

Alan Moore 2010-08-03 03:35:07

Thanks Alan for your help mate.

shane lee 2010-08-03 05:02:46

ansaurus

tags:

views:

answers:

How to use Scanner to break up string

related questions