#!/usr/bin/perl

# script written by Zbigniew Koziol, softquake@gmail.com

# This script will download all archival pages day by day, till today.
# "archival" means these pages that are accessible by using the following type of URL:
# http://marucha.wordpress.com/yyyy/mm/dd/
# As an example: 
# http://marucha.wordpress.com/2006/09/06/ -- seems to be the first date of posting
# that date will approximately be 1157553223 seconds since the epoche.

	$start_time = 1157553223; 	# 2006/09/06 in seconds since the epoche
	$today = time(); 		# now in seconds since the epoche
	$oneday = 24*3600; 		# seconds of 24-hours

	$my_start_time = $start_time;
	
	while ($my_start_time < $today) {
		($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) = gmtime($my_start_time);
		$year = $year + 1900;
		$mon  = $mon + 1;
		if ($mon < 10) { $mon = "0" . $mon; }
		if ($mday < 10) { $mday = "0" . $mday; }
		$my_download_url = "http://marucha.wordpress.com/" . $year . "/" . $mon . "/" . $mday . "/";
		$my_save_file = $year . "_" . $mon . "_" . $mday . ".html";
#		print "$my_download_url\t$my_save_file\n";
		system("/usr/bin/wget -c $my_download_url -O one_day/$my_save_file");		
		$my_start_time = $my_start_time + $oneday;
	}

# After that, remove all files with size==0 bytes from directory one_day, and run from
# that directory: ls -1 * >> ../postings.dat