I have a hard time understanding your question. Why would you want to "extract" a known substring from another: you already have it! Maybe you meant "remove"? If that's the case, then the simplest thing IMO is to use a regular expression:

$string =~ s/\Q$substring//;
[download]

my $offset = index( $string, $substring );
substr( $string, $offset, length $substring, '' ) if $offset > -1;
[download]

The way that I understood the question was that if you consider the string to be broken up into `words' by the spaces, the OP requires all words apart from the first two and the last two.

You mean like using the regular expression thingies as described in perlre ?

/J\

use strict;
use warnings;

...
my $source_string;
# set the value of $source_string to the original string

$source_string =~ /^\S+\s+\S+\s+(.+)\s+\S+\s+\S+$/;
my $wanted_string = $1;
[download]

while(<DATA>) {
   my @pieces = split;
   my $extracted = join ' ', @pieces[ 2 .. @pieces - 3 ];
   print $extracted, "\n";
}

__DATA__
WX 200 A string of interest 86.34 0906
WTTS 320 Peer here my peer 17.34 1001
XGR 400 Please take me with you 19.87 1201
[download]

Do you want to remove the substring from the larger string or pull it out for processing? I'm getting hung up on your choice of "extract."

If your data is consistent with:

WX 200 A string of interest 86.34 0906
WTTS 320 Peer here my peer 17.34 1001
XGR 400 Please take me with you 19.87 1201

You could:

while(<STRING_SRC>) {
  /[A-Za-z]+ [0-9]+ (.*)?[0-9.]+ [0-9]+/;
  $dotoit = $1;
  do_func $dotoit;
}


The regular expression assigns your string to $1.
If you need to extract, as in pull out, the value then
while(<STRING_SRC>) {
  /[A-Za-z]+ [0-9]+ (.*)?[0-9.]+ [0-9]+/;
  $dotoit = $1;
  s/$dotoit//;
  fund_do $dotoit;
}



Be Appropriate && Follow Your Curiosity