That's better (I think). Your input data look like this:

* /vobs/cs_test_scripts /usr/add-on/puccase_vob01/ccvob01/cs_test_scri
+pts.vbs public (replicated)
* /scm /usr/add-on/puccase_vob01/ccvob01/scm.test.vbs
[download]

And your output should be:

/usr/add-on/puccase_vob01/ccvob01/cs_test_scripts.vbs
/usr/add-on/puccase_vob01/ccvob01/scm.test.vbs
[download]

I wouldn't do this with regular expressions, I'd use split:

my @data = (
  "* /vobs/cs_test_scripts /usr/add-on/puccase_vob01/ccvob01/cs_test_s
+cripts.vbs public (replicated)",
  "* /scm /usr/add-on/puccase_vob01/ccvob01/scm.test.vbs"
);

for ( @data ) {
  my $path = (split)[2];
  print "$path\n";
}
[download]

Outputs:

/usr/add-on/puccase_vob01/ccvob01/cs_test_scripts.vbs
/usr/add-on/puccase_vob01/ccvob01/scm.test.vbs
[download]

Is that what you're after?