Abs directory turned into links

[einerwitzen]

On one perl page that prints out links to itself with ?variable=value on them so it prints out new info.
This value is a directory name. The info is the contents of the directory.
So I can go to this file and click links to view the contents of other directories.
At the top of this multi-tasking perl page is a printed out line of what directory on the server I am viewing, this is printed out from a hard set variable and the sent value.
I want to be able to use this printed out line to have each directory name turned into a link that links again to this same file but with the new value.
its a bit like this:

#!/usr/bin/perl
use CGI;
$query = new CGI;
$dire = $query->param("dire");
print $query->header ( );

$live = "/home/client/www/site";

print "$live<a href=\"./index.cgi?dire=$dire\">$dire</a>";

print "<br><br>"

#then my readdir which reads $live$dire and prints out the contents

So when I click on the linked part it takes me to view the contents of that directory. Simple.

This works fine when I am in "/html" and $dire is "/html" But not when I am in "/html/code" and $dire is "/html/code".
When it is this case and it prints out
home/client/www/site/html/code

I want to be able to click on the "html" part of that line and have it take me to "./index.cgi?dire=/html"
or to click the "code" part of that line and have it take me to "./index.cgi?dire=/html/code"

You get the idea, i want to be able to move back directories. How can I split up $dire and print it out as separate links?

Thank you for any help or pointers. I tried my best to describe the problem but if it's not clear let me know!

[grep]

I would recommend File::Basename:

#!/usr/bin/perl -w
use strict;
use File::Basename;

my $dir = "/usr/bin/mighty/long/path/perl";
my $end = basename($dir);

while ($dir ne "/" and $dir ne "") {
    print "$dir $end\n";
    $dir = dirname($dir);
    $end = basename($dir);
}



grep

grep> chown linux:users /world

[Zaxo]

Apache will do the file listing for you if there is no DirectoryIndex in a publicly available directory.

If you want more control, I'd suggest glob. or File::Find to get the lists and File::Basename to guard against malicious input and dress up the text, as ++grep suggests. Here are some fragments to get you started:

# Specify what directory is meant by a tag.
# Bogus tags will fail.to see anything undesirable
my %dirs = (
    foo => '/foo',
    bar => '/foo/bar',
    baz => '/quux',
);
# restrict to html
my @fles = glob $ENV{DOCUMENT_ROOT}.$dirs{$query->param('dire')}.'/*.h
+tml';

# print GET urls for this script
print join " / ", map
          {$query->a( { href=>$query->url() . "?dire=$_"}, $_)} keys %
+dirs;

# print file list
use File::Basename;
print map {$query->a({ href => $base_uri
                              . substr( ($_, 
                                      length( $ENV{DOCUMENT_ROOT} ))
                     }, basename( $_, '.html')
                    )
           . $query->br() . "\n"
           } @fles;

Warning, this is off the top of my head. Mostly untested.

After Compline,
Zaxo