Right! Except that I wouldn't call them "system variables". How 'bout "numbered match variables" instead?

perl -MYAPE::Regex::Explain -e 'print YAPE::Regex::Explain->new(qr/(aa
+aaa(?:(?:(?!aaaaa).)*))/s)->explain'
[download]

The regular expression:

(?s-imx:(aaaaa(?:(?:(?!aaaaa).)*)))

matches as follows:
  
NODE                     EXPLANATION
----------------------------------------------------------------------
(?s-imx:                 group, but do not capture (with . matching
                         \n) (case-sensitive) (with ^ and $ matching
                         normally) (matching whitespace and #
                         normally):
----------------------------------------------------------------------
  (                        group and capture to \1:
----------------------------------------------------------------------
    aaaaa                    'aaaaa'
----------------------------------------------------------------------
    (?:                      group, but do not capture:
----------------------------------------------------------------------
      (?:                      group, but do not capture (0 or more
                               times (matching the most amount
                               possible)):
----------------------------------------------------------------------
        (?!                      look ahead to see if there is not:
----------------------------------------------------------------------
          aaaaa                    'aaaaa'
----------------------------------------------------------------------
        )                        end of look-ahead
----------------------------------------------------------------------
        .                        any character
----------------------------------------------------------------------
      )*                       end of grouping
----------------------------------------------------------------------
    )                        end of grouping
----------------------------------------------------------------------
  )                        end of \1
----------------------------------------------------------------------
)                        end of grouping
----------------------------------------------------------------------
[download]

hi, do we need 'g' modifier here ?, i tried your pattern matching with this code :

$str = "aaaaa\nbbbbb\nccccc\nddddd\naaaaa\nddddd\neeeee\n";
@a = $str =~ /(aaaaa(?:(?:(?!aaaaa).)*))/s;
print "1 : $a[0]\n 2 : $a[1]\n";
[download]

and this code only match "aaaaa bbbbb ccccc ddddd" ($a[0])

and if we put g in pattern matching, i mean like this :

$str = "aaaaa\nbbbbb\nccccc\nddddd\naaaaa\nddddd\neeeee\n";
@a = $str =~ /(aaaaa(?:(?:(?!aaaaa).)*))/gs;
print "1 : $a[0]\n 2 : $a[1]\n";
[download]

this will print out : "aaaaa bbbbb ccccc ddddd" and "aaaaa ddddd eeeee".

thanks, zak

I sat all day, tried to understand prasadbabu's code, then i asked for help to id-perl.

Then someone named Jacinta Richardson told me about this, and she said :

/
      aaaaa               # Find me aaaaa
           (?:            # Followed by, but do not capture
            (?:           # Group but do not capture
                (?!       # Something which is not
                    aaaaa
                    .    #and any char including newline
                )
            )*             # As many as possible
        )
    /gs        # Repeat the match, dots can include newlines
[download]

The first grouping is unnessary, but not a problem.

Negative look-aheads ask the regular _expression to look at the nextvalue and only include it in the match if it does not match that part of the _expression.

Thus the regular _expression finds: aaaaa\nbbbbb\nccccc\n

in its first run, stopping at the "aaaaa\n" which matches the negative look-ahead and then in its second run finds: aaaaa\nddddd\neeeee\n

That's what she said, and then i realize that Jacinta Richadson known as jarich here.

Thanks for your time Jarich, and hope this help jeanluca too