Copy the last modified Dir from one location to another using Perl - perl

Question

fairly new to perl so this is most likely is not the best code which is why I am posting. I got this to work but was wondering if there is a better way. I do not have the ability to download modules. I am copying the last modified directory in a build folder from one server to another server. The argument allows me to choose which build directory to choose from.
Thanks
#!C:\strawberry\perl
use warnings;
use strict;
use File::Copy::Recursive;
my $NewFolder = `(dir /o-d/ad/b \\\\myserver1.name.com\\builds\\$ARGV[0] | head -1)`;
chomp($NewFolder);
$dir1 = "\\\\myserver1.name.com\\builds\\$ARGV[0]/$NewFolder";
$dir2 = "\\\\myserver2.name.com\\builds\\$ARGV[0]/Backup/$NewFolder";
File::Copy::Recursive::dircopy $dir1, $dir2 or die "Copy failed: $!";

Answer 1

Use forward slashes. It just makes your code easier to read:
$dir1 = "\\\\myserver1.name.com\\builds\\$ARGV[0]/$NewFolder";
vs.
$dir1 = "//myserver1.name.com/builds/$ARGV[0]/$NewFolder";
Also, don't do system calls where Perl can do it. For example, Perl can see the last modification date of a file via the stat. Even better is the File::stat module that makes the stat command so much easier to use.
Don't use #ARGV in your programs. Instead, read the variables from #ARGV into your own variables. It makes your program easier to understand, and your own variables have limited scope while #ARGV is global.
Use modern conventions. Variable names should be in all lower case, and use underscores to separate out words. That is $new_folder vs. $NewFolder. Is this arbitrary? Yes, but it's a convention followed by most Perl developers. It means not wondering if the variable is $newFolder, $NewFolder, or $newfolder because you know by these rules it is $new_folder.
And finally, use autodie; This will kill your program whenever a file operation fails. This turns perl from a check function for errors programming language into a exception checking language. This way, you don't have to worry whether or not you have to check for a failed IO operation.
Here's a completely untested, error ridden example:
use strict;
use warnings;
use autodie;
use File::Copy::Recursive qw(dircopy); #Optional Module
use File::Stat;
use constants {
ORIG_SERVER => '//myserver1.name.com/builds',
TO_SERVER => '//myserver2.name.com/builds',
};
my $from_directory = shift;
#
# Find newest directory
#
opendir my $dir_fh, ORIG_SERVER . "/$from_directory";
my $newest_directory;
while ( my $sub_directory = readdir $dir_fh ) {
next if $sub_directory eq "." or $sub_directory eq "..";
next unless -d $sub_directory;
if ( not defined $newest_directory ) {
$youngest_directory = $sub_directory;
next;
}
my $youngest_directory_stat = stat ORIG_SERVER . "/$directory/$newest_directory";
my $sub_directory_stat = stat ORIG_SERVER . "/$directory/$sub_directory";
if ( $newest_directory_stat->mtime > $sub_directory_stat->mtime ) {
$newest_directory = $sub_directory;
}
}
dircopy ORIG_SERVER . "/$directory/$youngest_directory",
TO_SERVER . "/$directory/$youngest_directory/backup";
My program is a lot longer than your program because your program depended upon various system operating commands, like dir and head which I don't believe is a standard Windows OS command. Instead, I read each entry under that directory into my loop. Anything that's not a directory, I toss (next if -d $sub_directory) and I toss out the special directories . and ...
After that, I use stat to find the youngest directory which to me means the one with the newest modification time. Note that Unix doesn't store creation time. However, according to perlport ctime is creation time on Win32, so you might prefer that instead of mtime.
If I didn't use File::stat, instead of this:
my $youngest_directory_stat = stat ORIG_SERVER . "/$directory/$newest_directory";
my $sub_directory_stat = stat ORIG_SERVER . "/$directory/$sub_directory";
if ( $newest_directory_stat->mtime > $sub_directory_stat->mtime ) {
$newest_directory = $sub_directory;
}
I could have done this:
my $newest = ORIG_SERVER . "/$directory/$newest_directory";
my $sub_dir = ORIG_SERVER . "/$directory/$sub_directory";
if ( stat( $newest )[9] > stat( $sub_dir )[9] ) {
$newest_directory = $sub_directory;
}
The stat command without File::stat returns an array of values, and I could have simply used the [9] element of that array. However, what is 9? Even though it could of saved me a few lines of code, and including an extra Perl module, it's better to use File::stat.
One thing you notice is that constants don't interpolate which means I have to keep doing things like this:
my $youngest_directory_stat = stat ORIG_SERVER . "/$directory/$newest_directory";
However, you can use this bit of Perlish black magic to interpolate constants inside quotes:
my $youngest_directory_stat = stat "#{[ORIG_SERVER]}/$directory/$newest_directory";
Hope that helps.