PERL Beginners - how to get in depth Directory statistics

Author

how to get in depth Directory statistics

Jeevs

2008-03-21, 8:01 am

Hello Forum,

I am in need to write a script which will count the number of files
and directories in a given directory. It will also record the
statistics of the directory within the main directory.
I have thought of a hash structure like
%hash = {
filecnt => value,
dircnt => value,
dir1 => {
filecnt => value,
dircnt => value,
dir1 =>{ ...}

},
dir2 => {
filecnt => value,
dircnt => value,
dir1 =>{ ...}

},

}

+++++++++++++++++++++++++++++

My question is.. Is there already a perl module which will be able to
parse the directory structure and give me the required output. If not
then following is the code I am working on, however I am getting

in using the recursive functionality. Can someone guide me on
this.

My code looks like

use strict;
use warnings;
use Data::Dumper;

my %hash_dir; # this structure will hold the actual directory
structure
my ($file_cnt,$dir_cnt);
my $root = 'f:/dashboard1';
diropen ($root);

#### fmethod to open the directory
sub diropen{
my $dir_name = shift;
opendir (DH, $dir_name) or die "can't opent the directory $@";
my @files_arr = readdir (DH);
dir_browse(\@files_arr);
}

### method to get the statistics of the directory.

sub dir_browse{
my $ref_files_arr = shift;
my @files_arr = @$ref_files_arr;
foreach (@files_arr) {
unless ($_ =~ /\.|\.\./) {
print "$_\n";
if (-f $_) {
$file_cnt++;
}
else{
$dir_cnt++;
}
}
$hash_dir{dir_cnt} = $dir_cnt if (defined $dir_cnt);
$hash_dir{file_cnt} = $file_cnt if (defined $file_cnt);
}
}

print Dumper %hash_dir;

+++++++++++++++++++

J. Peng

2008-03-21, 8:01 am

On Fri, Mar 21, 2008 at 5:01 PM, jeevs <jeevan.ingale@gmail.com> wrote:

> My question is.. Is there already a perl module which will be able to
> parse the directory structure and give me the required output.

sure.like Filesys::Tree module on cpan.

For example, the code below,

use Filesys::Tree qw/tree/;

my $tree = tree({ 'full' => 1,'max-depth' => 3,'pattern' => qr/\.pl$/ } ,'.');
files($tree);

sub files
{
my %tree = %{+shift};
for (keys %tree) {
if ($tree{$_}->{type} eq 'f'){
print $_,"\n"
}elsif ($tree{$_}->{type} eq 'd') {
files($tree{$_}->{contents});
}
}
}

should get the same results as this unix shell command do,

$ find . -type f -name "*.pl" -maxdepth 2

Rob Dixon

2008-03-21, 7:04 pm

jeevs wrote:
>
> I am in need to write a script which will count the number of files
> and directories in a given directory. It will also record the
> statistics of the directory within the main directory.
> I have thought of a hash structure like
> %hash = {
> filecnt => value,
> dircnt => value,
> dir1 => {
> filecnt => value,
> dircnt => value,
> dir1 =>{ ...}
>
> },
> dir2 => {
> filecnt => value,
> dircnt => value,
> dir1 =>{ ...}
>
> },
>
> }

[snip]

My first thought is that your structure is very hard to get anything out
of once it's been built. What is your intention once you've gathered all
of this data?

Also, I assume 'dir1', 'dir2' etc. are the names of the directories?
What happens if a directory happens to be named 'filecnt' or 'dircnt'?

I recommend you think through this a little more before committing
yourself. You are likely to be more successful if your internal data
structure is a basis for the solution to the overall problem rather than
simply echoing the form of the source data.

HTH,

Rob

Sponsored Links