zoomgirls_downloader.pl

use strict;
use warnings;

sub norm($) {
    my $n = shift;
    $n = 0 unless $n;
    $n =~ s/x/\./g;
    return $n;
}

our @main;
our $siteroot = "http://zoomgirls.net";
our $get_only_biggest = 1;

our %rdb = (
        '1280x1024' => 1.25,

        '800x600'   => 1.33,
        '1024x768'  => 1.33,
        '1152x864'  => 1.33,
        '1280x960'  => 1.33,
        '1600x1200' => 1.33,
        '1920x1440' => 1.33,

        '1280x800'  => 1.6,
        '1440x900'  => 1.6,
        '1680x1050' => 1.6,
        '1920x1200' => 1.6,
        '1920x1200' => 1.6,
        '2560x1600' => 1.6,

        '1280x720'  => 1.77,
        '1366x768'  => 1.77,
        '1600x900'  => 1.77,
        '1920x1080' => 1.77,
       );

push @main, "${_}_girls" for ('a'..'z');
push @main, $_ for (
    "anime_and_drawings",
    "babes_and_bikes",
    "babes_and_cars",
    "sexy_girls"
);

mkdir("root"); mkdir("pages"); mkdir("walls"); mkdir("pics");

for my $main_cat (@main) {
    my $corpus = "girls";
    $corpus = $main_cat unless $main_cat =~ /^[a-z]_girls$/;
    mkdir("pics/$corpus");

    my ($url, $file);
    $file = "$main_cat-desktop-wallpapers.html";

    unless (-e "root/$file") {
        $url = "$siteroot/$file";
        system("wget $url -O root/$file") == 0 or die;
    }

    open F, "root/$file" or die;
    my $maxpage = 1;
    while (<F>) {
        while (/desktop-wallpapers\/page\/(\d+)"/g) {
            $maxpage = $1 if $1 > $maxpage;
        }
    }
    close F;

    for my $i (1..$maxpage) {
        $file = "pages/$main_cat-desktop-wallpapers_$i.html";
        unless (-e $file) {
            $url = "$siteroot/$main_cat-desktop-wallpapers/page/$i";
            system("wget $url -O $file") == 0 or die;
        }

        open F, "$file" or die;
        while (<F>) {
            next unless /<div class="thumb"><a href="\/(.+?)-wallpapers.html"><em>\d+ views<\/em>/;

            my $id = $1;
            $file = "$id-wallpapers.html";

            unless (-e "walls/$file") {
                $url  = "$siteroot/$file";
                system("wget $url -O walls/$file") == 0 or die;
            }

            open G, "walls/$file" or die;
            print "$file\n";

            my $found = 0;
            my $dlink = undef;
            while (<G>) {
                if (/<a href="\/view\/(.*?).html"/) {
                    $dlink = "$1.jpg";
                }
                next unless /and it is available in (.*?)resolutions/;
                my $res = $1;
                $res =~ s/,?\s*$//;
                unless ($res) {
                    unless ($dlink) {
                        warn "No IMAGE links found!" unless $dlink;
                        $found = 0;
                        last;
                    }
                    $found = 1;
                    $file = $dlink;
                    unless (-e "pics/$corpus/$file") {
                        system("wget $siteroot/download/$file -O pics/$corpus/$file") == 0 or die;
                    }
                    last;
                }

                my %maxr;

                for my $raw (split /, /, $res) {
                    die "raw=[$raw/$res]" unless $raw =~ /^(\d+) x (\d+)$/;
                    my $r = "$1x$2";
                    die "$r not in DB!" unless exists $rdb{$r};
                    my $aspect = $rdb{$r};
                    $maxr{$aspect} = $r if norm($maxr{$aspect}) < norm($r);
                }

                if ($get_only_biggest) {
                    my $mx = '';
                    for my $curr (sort values %maxr) {
                        $mx = $curr if norm($curr) > norm($mx);
                    }
                    %maxr = (1 => $mx);
                }

                for my $curr (sort values %maxr) {
                    $found = 1;
                    $file = "$id-$curr.jpg";
                    unless (-e "pics/$corpus/$file") {
                        system("wget $siteroot/download/$file -O pics/$corpus/$file") == 0 or die;
                    }
                }
            }
            warn "No images found in $file..." unless $found;
            close G;
        }
        close F;
    }

}