Google Code Downloader

gcd

#!/usr/bin/perl
#
# GCD - Google Code Downloader
# Download source files and rebuild project structure from Google Code Search cache.
#
# usage: ./gcd url path
#
#  Version: 0.3
#  Licence: GPL
#
# Copyright (c) 2007-2008 Marcin 'Malcom' Malich, <me@malcom.pl>
#
# http://projects.malcom.pl/tools/gcd.xhtml
#

use strict;
use warnings;
use LWP::Simple;
use HTML::Entities;

$\ = "\n";

print "\n".'GCD - GoogleCode Downloader v0.3';
print '(c) 2008 MalCom <me@malcom.pl>'."\n";

if (!$ARGV[0] || !$ARGV[1]) {
	print 'Fault of required arguments!';
	print 'usage: ./gcd url path';
	exit;
}

my $c = 0;
&fun($ARGV[0]);
print 'Downloaded '.$c.' files.';

sub fun {
	my($path, $kat);

	if ($_[1]) {
		$path = $ARGV[1].'/'.$_[1];
		$kat = $_[1].'/';
	} else {
		$path = $ARGV[1];
		$kat = '';
	}

	mkdir $path || die "Couldn't create folder $path";

	my $p = get($_[0]);
print $p; exit;
	if (!($p =~ /<div id=filelist><pre class=dirlistpre>(.*?)<\/pre><\/div>/s)) {
		return;
	}

	foreach ($1 =~ /^(.*)/mg) {
		my $fn;

		if (/<a href="(.*)" class="direlem">(.*)<\/a>(.?)/) {
			# klase direlem maja wszystkie linki do katalogow i plikow w danym katalogu,
			# procz .. a on nam i tak nie jest potrzebny ;)
			$fn = $2;
			my $k = $3;

			my $url = 'http://www.google.com'.$1;
			$url =~ s/&amp;/&/g;
				
			if ($k eq '/') {			# jesli to katalog
				&fun($url, $kat.$fn);
				next;

			} else {
				$_ = get($url);
			}

		} elsif (/<b>(.*)<\/b>/) {
			# aktualny plik zawiera source danego pliku
			$fn = $1;
			$_ = $p;

		} else {
			next;
		}

		print 'Add file: '.$kat.$fn;

		if (open FILE, '>'.$path.'/'.$fn) {
			if (/<div id="code"><pre>(.*)<\/pre><\/div>/s) {
				$_ = $1;
				s/<.*?>//gs;			# usuwanie tagow htmla
				s/[\n\r]+$//;			# znakow nowej lini z konca
				print FILE decode_entities($_);
			}
			close FILE;
		} else {
			print "Couldn't create file $path/$fn";
		}
		
		$c++;
	}
	
	$kat = '';
}