First please learn the value of indenting you code properly so you can see the flow. If you don't know how get a copy of perltidy from sourceforge and let it do it for you till it comes naturally. Try something like this:

#! perl -w

use strict;
use File::Copy;

my $infile = "c:\\doclist1.chr";
my @directories;

# first lets build our directory list
open IN, "<$infile" or die "Couldn't open $infile, $!";
while (<IN>) {
    chomp;
    my @fields   = split /,/;
    my $path_str = $fields[6];
    do { warn "Empty field 7"; next } unless $path_str;
    my @path = split /\\/, $path_str;
    # push the directories into an array
    push @directories, join "\\", @path[ 0, 5, 6 ];
}
close IN;

# now we have a directory list in @directories let's process it
for my $dir ( @directories ) {
    process_dir($dir);
}

exit;

sub process_dir {
    my $dir = shift;
    do { warn "$dir does not exist!\n"; return } unless -e $dir;
    opendir DIR, $dir or do { warn "Could not open $dir $!\n" ; return
+ };
    while ( my $file = readdir DIR ) {
        next unless -f "$dir\\$file";
        next unless $file =~ m/\.rtf$/;
        copy( "$dir\\$file", "C:\\testfiles\\$file" )
          or die "Failed to copy $file: $!\n";
    }
}
[download]

Once you have the basic logic working you can add in File::Find if you need to recurse a directory structure in your process dirs sub. There were a lot of problems with your code. Foremost was that you wanted to push a list of directories into @directory but were not pushing and even if you were the my declaration in the loop would erase it every loop.

You actually don't need @directories at all a you can just call process_dirs() in your main loop like this:

#! perl -w

use strict;
use File::Copy;

my $infile = "c:\\doclist1.chr";
open IN, "<$infile" or die "Couldn't open $infile, $!";
while (<IN>) {
    chomp;
    my @fields   = split /,/;
    my $path_str = $fields[6];
    do { warn "Empty field 7"; next } unless $path_str;
    my @path = split /\\/, $path_str;
    my $dir = join "\\", @path[ 0, 5, 6 ];
    process_dir($dir);
}
close IN;

sub process_dir {
    my $dir = shift;
    do { warn "$dir does not exist!\n"; return } unless -e $dir;
    opendir DIR, $dir or do { warn "Could not open $dir $!\n" ; return
+ };
    while ( my $file = readdir DIR ) {
        next unless -f "$dir\\$file";
        next unless $file =~ m/\.rtf$/;
        copy( "$dir\\$file", "C:\\testfiles\\$file" )
          or die "Failed to copy $file: $!\n";
    }
}
[download]

cheers

tachyon

s&&rsenoyhcatreve&&&s&n.+t&"$'$`$\"$\&"&ee&&y&srve&&d&&print

Thanks for your help!!! I didn't think of it, but each file that I'm copying to the new directory has the same name called "word1.rtf" I would need to rename each file because if not they will not copy with the same name. Have you seen anywhere or do you know of a method to rename files dynamicaly while you are copying files to another directory? I'll appreciate any suggestions. Thanks a lot

All you need to do is

copy( "C:\\somename", C:\\some_path\\some_other_name" )
[download]

To generate the new names dynamically you use something like:

# this is a test example
my $name = 'doc.txt';
for ( 1 .. 10 ) {
    my $unique = get_unique_filename($name);
    $names{$unique}++;
    print "$unique\n";
}

sub get_unique_filename {
    my $name = shift;
    $name =~ s/(\.[^\.]+)$//;
    my $ext = $1 ? $1 : '';
    my $tries = '';
    my $max_tries = 100;
    while (  not unique_name($name.$tries.$ext) ) {
        $tries++;
        die "Could not make a unique file name $name\n" if $tries > $m
+ax_tries;
    }
  return $name.$tries.$ext;
}

# this tests if our name exists in the test hash
# you need to test if the file exists here instead
sub unique_name {
    my $name = shift;
  return defined $names{$name} ? 0 : 1;
  # return -e $name ? 0 : 1;
}
[download]

Without the demo code all you need is a single sub that looks like:

sub get_unique_filename {
    my $name = shift;
    $name =~ s/(\.[^\.]+)$//;
    my $ext = $1 ? $1 : '';
    my $tries = '';
    my $max_tries = 100;
    while (  -e $name.$tries.$ext ) {
        $tries++;
        die "Could not make a unique file name $name\n" if $tries > $m
+ax_tries;
    }
  return $name.$tries.$ext;
}
[download]

cheers

tachyon

s&&rsenoyhcatreve&&&s&n.+t&"$'$`$\"$\&"&ee&&y&srve&&d&&print

 my $dir; # my goes here
 foreach $dir (@dir) {  # my does not go here.
    if ($File::Find::dir ne $dir)
    {
        $File::Find::prune = 1;
        return 0;
    }
[download]

Cabrion, I change my code a bit but I still cannot get it to work. I keeps givng me "Global Symbol $dir requires explicit package name in line 32 and 35" I'm really hitting a wall could you help??? I don't know why it keeps giving me the error. I really need to do a recursive directory search. I'll appreciate your help...

#! perl 

use strict;
use File::Copy;
use File::Find;

no warnings 'File::Find';

my $infile = "c:\\doclist1.chr";
my @directories;

# first lets build our directory list
open IN, "<$infile" or die "Couldn't open $infile, $!";
while (<IN>) {
    chomp;
    my @fields   = split /,/;
    my $path_str = $fields[6];
    do { warn "Empty field 7"; next } unless $path_str;
    my @path = split /\\/, $path_str;
    # push the directories into an array
    push @directories, join "\\", @path[ 0, 5, 6 ];
}
close IN;

exit;

# now we have a directory list in @directories let's process it
for my $dir ( @directories ) {
    process_directories($dir);
}

find(\&process_directories, $dir);

sub process_directories {
    if ($File::Find::dir ne $dir)
    {
        $File::Find::prune = 1;
        return 0;
    }

    return 0 if ($_ !~ /\.rtf$/);

    copy($File::Find::name, "C:\\testfiles\\$_") or die "Failed to cop
+y $_: $!\n";
    return 1;
}
[download]