Hi guys, thanks for the speedy reply. The problem i guess is that i only want the subroutine, or array sorted once, and it isnt defined until the END of the loop.. lemme attach the code, and hopefully it will help

#!/usr/local/bin/perl

# Author: Dipul Patel


#  This Perl Script takes a home site chosen by the user, then calcula
+tes
#  the distance from that home site to every site in the system and pr
+ints 
#  in a HTML table.



#---------------------------------------------------------------------
# Process information for main site
#---------------------------------------------------------------------
$lat_long = "/tmp/sitesLL";
$filepath = "/data1/PHL/neighbors";


$query_string = $ENV{'QUERY_STRING'};
@search      = split(/&/,$query_string);
@info      = split(/=/, $search[0]);
@siteinfo      = split(/-/, $info[1]);
print "Content-type: text/html\n\n";



#read in lat and long file
open(LL, $lat_long);
@lat_long = <LL>;
close(LL);

#Get long,lat,g,h,j for main site
@dataline  = grep /^$siteinfo[0]/, @lat_long;
@dataparse = split(/\|/, $dataline[0]);
($sitelat, $sitelong) = split(/,/, $dataparse[3]);
    $g = $dataparse[4];
    $h = $dataparse[5];
    $j = $dataparse[6];

#---------------------------------------------------------------------
# Process information for every other site, one at a time
#---------------------------------------------------------------------

#open directory and get sites names
opendir DIR, $filepath;

@allfiles = grep !/^\.\.?$/, readdir DIR;
closedir DIR;
@allfiles = sort @allfiles;
$totalsite=@allfiles;
@allfiles2=@allfiles;
  foreach $site(@allfiles)  {
    chomp $site;
          @frst  = split(/[a-z][a-z]/, $site);
          @cell  = split(/[A-Z]/, $frst[1]);
          push(@othersiteinfo, $cell[0]);
  }


#Open assiociated neighbor list file
open(IN, "$filepath/$siteinfo[1]");
@data=<IN>;
close(IN);

#get sector numbers for neighbors
  foreach $line(@data){
    #chomp $line;
    @information = split (/\|/, $line);
    push (@neighbor,$information[1]);
  }

$n=-1;
$temphigh = 0;
while($n<$totalsite){
    $n++;
    @dataline2  = grep /^$othersiteinfo[$n]/, @lat_long;
    @dataparse2 = split(/\|/, $dataline2[0]);
    ($sitelat2, $sitelong2) = split(/,/, $dataparse2[3]);
        $g2 = $dataparse2[4];
        $h2 = $dataparse2[5];
        $j2 = $dataparse2[6];
#---------------------------------------------------------------------
# calculate distance from main site to other sites
#---------------------------------------------------------------------
    $a=($g2-$g)*($g2-$g);
    $b=($h2-$h)*($h2-$h);
    $c=($j2-$j)*($j2-$j);
    $abc=($a+$b+$c);
    $sqabc=sqrt($abc);
    $distance=(3960*$sqabc);
        $distance = sprintf "%1.2f", $distance;
        $distance2=$distance;
        $distance3=$distance;
        $distance4=$distance;


if ($temphighval >= $temphigh){
    $temphigh=$temphighval;
}


    if ( $distance > 10){
        next;
        }
    
#    if ( $distance == 0){
#        next;
#        }

 # The following code, first compares the sector number of the current
+ site to that of the neighbor list,
 # then it complies 2 arrays, one with all the neighbors, the other ar
+ray with the other sites.

     $element = $othersiteinfo[$n];
    if (grep {$_ eq $element} @neighbor) {
        @array3=($allfiles[$n],$distance2);        
        push (@test1, @array3);    
        push (@highval, $distance2);
        }
    
    else{
        @array2=($allfiles[$n],$distance);
        push (@test2, @array2);
    }


}#end whileloop



my $done_sort=0; # or some more meaningful name
         while( $cond)
           { # probably some code here
             unless( $done_sort)
               { sort_the_array( @array);
                 $done_sort=1;
               }
             # more code here
           }



#create hash and sorted array for non-neighbors
    %distance=@test2;
    @sorted = sort { $distance{$a} <=> $distance{$b} } keys %distance;
    
#extract sector numbers from sitename for use in hyperlinks.
    $k=0;
        @nonneighsect=@sorted;
    for ($h=0; $h<=$#nonneighsect; $h++) {
               $nonneighsect[$h] =~ s/\w{2}(\d{4})\w+/$1/;
           }
    

#create hash and sorted array for neighbors
    %distance2=@test1;
    @sorted2 = sort { $distance2{$a} <=> $distance2{$b} } keys %distan
+ce2;
    
#extract sector numbers from sitename for use in hyperlinks.
        $j=0;
        @neighsect=@sorted2;
    for ($i=0; $i<=$#neighsect; $i++) {
               $neighsect[$i] =~ s/\w{2}(\d{4})\w+/$1/;
       }
       
       
#--------------------
# the following code prints a table, with all the sites sorted first b
+y neighbor, then distance. It also provides links 
# to automattically calculate distances for other sites, as well as ma
+p them.
#--------------------

print "<HTML>\n";
print "<HEAD><TITLE>Distance Tool</TITLE></HEAD>\n";
    print "<TABLE BORDER=1 cellspacing=1 cellpadding=1 WIDTH =65%>\n";
    print "<TR><TH Colspan = 4> Distance From $siteinfo[1] To:<TH>";
#        print " <FONT SIZE=\-5><A HREF=\"http://www-mat.nextel.com/cg
+i-bin/rfn-cgi/neighbormap2.cgi?frames=frames&site=$siteinfo[0]-$sitei
+nfo[1]&rpt=Daily-Total&Lfactor=150000&sector=1\" target=_top>1</a></f
+ont>\n";
#        print " <FONT SIZE=\-5><A HREF=\"http://www-mat.nextel.com/cg
+i-bin/rfn-cgi/neighbormap2.cgi?frames=frames&site=$siteinfo[0]-$sitei
+nfo[1]&rpt=Daily-Total&Lfactor=150000&sector=2\" target=_top>2</a></f
+ont>\n";
#        print " <FONT SIZE=\-5><A HREF=\"http://www-mat.nextel.com/cg
+i-bin/rfn-cgi/neighbormap2.cgi?frames=frames&site=$siteinfo[0]-$sitei
+nfo[1]&rpt=Daily-Total&Lfactor=150000&sector=3\" target=_top>3</a></f
+ont>\n";
    print "<FONT SIZE =\-5>";
    print "<TR><TH>Site<TH>Distance<TH>Neighbor<TH>Map(sector)\n";
        foreach $site3 (@sorted2) {
            print " <TR><TD><FONT SIZE=\-5><A HREF=\"http://www-mat.ne
+xtel.com/cgi-bin/rfn-cgi/distance.cgi?site=$neighsect[$j]-$site3\" >$
+site3<TD>$distance2{$site3}<TD align=center>YES</font>\n";
            print " <TD align=center><FONT SIZE=\-5><A HREF=\"http://w
+ww-mat.nextel.com/cgi-bin/rfn-cgi/neighbormap2.cgi?frames=frames&site
+=$neighsect[$j]-$site3&rpt=Daily-Total&Lfactor=150000&sector=1\" targ
+et=_top>1</a></font>\n";
            print " <FONT SIZE=\-5><A HREF=\"http://www-mat.nextel.com
+/cgi-bin/rfn-cgi/neighbormap2.cgi?frames=frames&site=$neighsect[$j]-$
+site3&rpt=Daily-Total&Lfactor=150000&sector=2\" target=_top>2</a></fo
+nt>\n";
            print " <FONT SIZE=\-5><A HREF=\"http://www-mat.nextel.com
+/cgi-bin/rfn-cgi/neighbormap2.cgi?frames=frames&site=$neighsect[$j]-$
+site3&rpt=Daily-Total&Lfactor=150000&sector=3\" target=_top>3</a></fo
+nt>\n";
            $j++;
        }
        foreach $site2 (@sorted) {
            print " <TR><TD><FONT SIZE=\-5><A HREF=\"http://www-mat.ne
+xtel.com/cgi-bin/rfn-cgi/distance.cgi?site=$nonneighsect[$k]-$site2\"
+>$site2<TD>$distance{$site2}<TD align=center>NO</font>\n";
            print " <TD align=center><FONT SIZE=\-5><A HREF=\"http://w
+ww-mat.nextel.com/cgi-bin/rfn-cgi/neighbormap2.cgi?frames=frames&site
+=$nonneighsect[$k]-$site2&rpt=Daily-Total&Lfactor=150000&sector=1\" t
+arget=_top>1</a></font>\n";
            print " <FONT SIZE=\-5><A HREF=\"http://www-mat.nextel.com
+/cgi-bin/rfn-cgi/neighbormap2.cgi?frames=frames&site=$nonneighsect[$k
+]-$site2&rpt=Daily-Total&Lfactor=150000&sector=2\" target=_top>2</a><
+/font>\n";
            print " <FONT SIZE=\-5><A HREF=\"http://www-mat.nextel.com
+/cgi-bin/rfn-cgi/neighbormap2.cgi?frames=frames&site=$nonneighsect[$k
+]-$site2&rpt=Daily-Total&Lfactor=150000&sector=3\" target=_top>3</a><
+/font>\n";
            $k++;
        }
    print "</FONT>";
print "</TABLE>\n";
print "</HTML>\n";
[download]

now, what i want, is where it says

    if ( $distance > 10){
        next;
        }
[download]

i want the 10 to be replaced with the highest value of the array defined at the end of the loop:

    if (grep {$_ eq $element} @neighbor) {
        @array3=($allfiles[$n],$distance2);        
        push (@test1, @array3);    
        push (@highval, $distance2);
        }
[download]

But.. that array is added to everytime the loop runs, and i need the highest value from it when it is complete, so i can specify the largest distance to map. I hope this helps.
Dipul Patel

So you don't want to sort the array, you want to find the max value in it. So write a subroutine to find the max value in your array and call that (just like chromatic suggested!)

use strict; # ALWAYS!!!!!!

# MaxElementInArray( @array )
sub MaxElementInArray
{
    my $max = shift;
    my $next;
    $max = $next > $max ? $next : $max while $next = shift;
    return $max;
}

# MaxElementInArrayRef( \@array )
sub MaxElementInArrayRef
{
    my $array_ref = shift; 
    my $index = $#$array_ref; 
    my $max = $$array_ref[$index];
    $max = $$array_ref[$index] > $max ?
           $$array_ref[$index] : $max while --$index >= $[;
    return $max;
}
[download]

Update: While examining your code in response to your statement below, I've noticed quite a few style issues: You don't use strict, you are not using CGI.pm, and you are not using subroutines to improve the readability (and maintainability) of your code. I strongly suggest you look into these things. As for your issue of trying to use a number before you have calculated it... this reveals a failure to plan before coding. Now I suspect you will need to re-write much of this code just to get things in a causal order.

Oh yeah, i forgot to mention, this is my second week of learning perl hehe :). Adam, where exactly should i place the code you gave me? I dont understand where to put it, and where/which one to call. Thanks
Dipul Patel

More likely than not there is a problem with the way your code is written, the code you want to run just once should probably be outside of the loop.

Anyway, as a duct-tape kinda fix you can always do:

my $done_sort=0; # or some more meaningful name
while( $cond)
  { # probably some code here
    unless( $done_sort)
      { sort_the_array( @array);
        $done_sort=1;
      }
    # more code here
  }
[download]

But there is probably a better way, which I can't find unless I see the code.

-- Randal L. Schwartz, Perl hacker