BTW, if you’re a Perl programmer and you’ve been jealous of all the cool kids and node.js, AnyEvent is how you do node.js-style programming in Perl.

UPDATE: this is a better example.
Doug Wilson gist of HTTP async with Mojo::IOLoop. On my test system, this now runs slightly faster than the AnyEvent::HTTP solutions.

Below is *a* solution using Mojolicous. I am sure there are other frameworks that can do the same thing. Benchmarks showed that the AnyEvent program is faster anywhere from fractions of a ms to one entire second depending on the iteration. In this simple case, AnyEvent::HTTP may be the correct solution, however I think as part of a larger project you are still better off going with an async web framework like Mojolicious.

I tried to keep the same style as the original author for comparison and testing. Almost entirely copied and pasted from the Mojolicious documentation and the original article; here is what I whipped up:

#!/usr/bin/env perl

use strict;

use warnings;

use Mojo::UserAgent;

use Mojo::IOLoop;

use Time::HiRes qw(time);

my $url = [

        "https://www.google.com",

        "http://www.windley.com/",

        "https://www.bing.com",

        "http://www.example.com",

        "http://www.wetpaint.com",

        "http://www.uh.cu"

        ];

my $ua = Mojo::UserAgent->new;

my $start = time;

foreach my $u (@$url) {

    print "\n";

    my $now = time;

    $ua->get($u, => sub {

        my ($ua, $tx) = @_;

        if (my $res = $tx->success) {

            print "$u has length " . (length $tx->res->body) . " and loaded in " . (time() - $now) . "ms";

        }

        else {

            my ($message, $code) = $tx->error;

            print "Error for $u : ($code) $message"

        }

        Mojo::IOLoop->stop;

    });

    Mojo::IOLoop->start;

}

print "\nTotal elapsed time: " . (time - $start) . "ms\n";

Go ahead now, take your shots.

my $xftp = Net::xFTP->new('Foreign', 'hostname', user => $user, password => $password,);

my $next = Archive::Tar->iter('mytarball.tar.gz');
while (my $file = $next->() )  {
   
  if ($file->type == 5) {    # 5 is the filetype number for directories in this case

    $xftp->mkdir($file->full_path);
    next;
  }

  open FH, '<', $file->get_content_by_ref;

  $xftp->put(*FH, $file->full_path) or die $xftp->message;
}

If for some reason the files you need are expanded on the filesystem, you can use File::Find to get a list of the files you want, then use Archive::Tar to create the archive in memory and loop through that.

find(sub{push @files,$File::Find::name}, "./");

my $archive;
open (my $in, '>', \$archive);
Archive::Tar->create_archive($in,0,@files);

open (my $out, '<', \$archive);
my $tar = Archive::Tar->new($out);

foreach my $file ($tar->get_files) {
  if ($file->type == 5) {
    $xftp->mkdir($file->full_path);
    next;
  }

  open (FH, "<", \$file->get_content);
  $xftp->put(*FH, $file->full_path) or die $xftp->message;
}

A couple of things to note. The filehandle typeglob does not work with the regular SFTP protocol module (but does work for FTP and FTPS and Foreign). Use Foreign as the protocol if you want SFTP; this will use Net::SFTP::Foreign and not Net::SFTP. Also, if using an in memory created archive $file->get_content_by_ref does not work, use \$file->get_content.