Kickstarter new project notifier

kickstarter.php

<?php
date_default_timezone_set("Europe/Lisbon");

##############################################################
#                   KICKSTARTER CRAWLER
#  Author: David Gouveia
#  Email:  david.gouveia [_AT_] gmail.com
#
#  This is a very simple script which allows you to crawl
#  kickstarter projects and trigger some alerts. I find
#  this useful to be alerted of new projects based on the
#  keywords fed to the script.
##############################################################

$CACHE_FILE     = "/tmp/kickstarter_db";
$PAGES_TO_CRAWL = 5;
$EMAIL          = "[email protected]";

if ( sizeof( $argv ) < 2 ) die( "You need to specify at least one keyword\n" );
array_shift( $argv );

print "Looking for new projects using the following keywords: " . trim( implode( ", ", $argv) ) . "\n";

$dataset = ( is_file( $CACHE_FILE ) ? unserialize( file_get_contents( $CACHE_FILE ) ) : Array() );
$curPage = loadpage( 1 );

preg_match_all( '/<li class="project grid_4">(.*?)<\/li>/ims', $curPage, $matches );

foreach ( $matches[1] as $match )
{
  $tmp = getDetails( $match );
  if ( $tmp && !checkArray( $dataset, $tmp[4] ) )
  {
    $dataset[] = $tmp;
    foreach( $argv as $keyword )
    {
      if ( stripos( $tmp[0], trim($keyword) ) !== false || stripos( $tmp[2], trim($keyword) ) !== false)
      {
        print "Matching project was found -> $tmp[2]: https://kickstarter.com$tmp[1]\n";
        mail( $EMAIL, "New project found: '$tmp[2]'", "Hello!\nIt seems that there is a new project matching your keywords. For more details go to https://kickstarter.com$tmp[1]", "From: $EMAIL");
      }  
    }
  }
}


file_put_contents( $CACHE_FILE, serialize( $dataset ) );

function getDetails( $match )
{
  preg_match( '/bbcard_blurb">(.*?)<\/p>/ims', $match, $description);
  preg_match( '/<strong>.*?<a href="(.*?)".*?>(.*?)<\/a>.*?<\/strong>/ims', $match, $linkdesc);
  preg_match( '/<div class="project-thumbnail">.*?img.*?src="(.*?)"/ims', $match, $image );
  if ( $description[1] == null || $linkdesc[1] == null || $linkdesc[2] == null) return null;

  return ( Array( trim( $description[1] ),
                  trim( $linkdesc[1] ),
                  trim( $linkdesc[2] ),
                  trim( $image[1] ),
                  md5( trim( $description[1] ) .  trim( $linkdesc[1] ) .  trim( $linkdesc[2] ) . trim( $image[1] ) ),
                  mktime() ) );
}

function checkArray( $mArray, $hash )
{
  if ( sizeof( $mArray ) == 0 ) return false;
  foreach( $mArray as $project )
  {
   if ( $project[4] == $hash ) return true;
  }
  return false;
}

function loadpage ( $page=1 )
{
  return ( file_get_contents( "https://www.kickstarter.com/discover/advanced?page=$page&category_id=16&sort=launch_date&seed=" ) );
}