Craigslist RSS to email

rssToEmailCraigslist.php

<?php 

/****
* Little PHP script to hit Craigslist RSS feeds, check new posts, and shoot you an email with them.
* Good for finding things ;)
* REQUIRES http://code.google.com/p/phpquery/
****/ 

function resultsSort($a, $b) {
    if ( count($a["posts"]) == count($b["posts"]) ) { return 0; }
    return (count($a["posts"]) < count($b["posts"])) ? 1 : -1;
}

require_once 'phpQuery-onefile.php';

if( !is_file( dirname(__FILE__) .  "/seenPosts.json" ) ){
	touch( dirname(__FILE__) .  "/seenPosts.json" );
	$seenPosts = array( );
}else{
	$seenPosts = json_decode( file_get_contents( dirname(__FILE__) .  "/seenPosts.json" ), true );
}

$keywords = array( "symfony" );
$cities = array( 
			"atlanta",
			"austin",
			"boston",
			"chicago",
			"dallas",
			"denver",
			"detroit",
			"houston",
			"lasvegas",
			"losangeles",
			"miami",
			"minneapolis",
			"newyork",
			"orangecounty",
			"philadelphia",
			"phoenix",
			"portland",
			"raleigh",
			"sacramento",
			"sandiego",
			"seattle",
			"sfbay",
			"washingtondc"
);

$baseUrls = array(
	"http://%s.craigslist.org/search/ggg?query=%s&catAbb=jjj&srchType=A&addThree=&format=rss",
	"http://%s.craigslist.org/search/?areaID=4&subAreaID=&query=%s&catAbb=ggg&format=rss",
);

$results = array( );

$totalPosts = 0;

foreach( $keywords as $kw ){
	
	foreach( $cities as $ct ){
		
		foreach( $baseUrls as $bu ){
			
			$posts = array( );
			$url = sprintf( $bu, $ct, $kw );
			echo $url . "\n";

			$ch = curl_init();
			curl_setopt($ch, CURLOPT_URL, $url);
			curl_setopt($ch, CURLOPT_FAILONERROR, false);
			curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
			curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
			curl_setopt($ch, CURLOPT_TIMEOUT, 30);
			curl_setopt($ch, CURLOPT_HTTPHEADER, Array("User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1.15) Gecko/20080623 Firefox/2.0.0.15") );
			$html = curl_exec($ch);
			curl_close($ch);
			
			$doc = phpQuery::newDocumentXML( $html );
			phpQuery::selectDocument($doc);
			
			foreach( pq("item") as $it ){
				
				$link = pq($it)->children("link:first")->text();
				$body = strip_tags( str_replace( "<br>", "\n", pq($it)->children("description:first")->text() ) );
				
				if( !in_array($link, $seenPosts) ){
				
					$posts[] = array(
						"title" => pq($it)->children("title:first")->text(),
						"body" => $body,
						"link" => $link
					);

					$seenPosts[] = $link;
				}
				
			}
			
			$totalPosts += count( $posts );
			$results[] = array( "url" => $url, "posts" => $posts );
			
		}
		
	}
	
}

usort($results, 'resultsSort');

$emailBody = "Total Posts: " . $totalPosts . "\n\n";
foreach( $results as $obj ){
	
	$emailBody .= $obj["url"] . " ( " . count($obj["posts"]) . " )" . "\n\n";
	
	foreach( $obj["posts"] as $p ){
		$emailBody .= $p["link"] . "\n" . $p["title"] . "\n" . $p["body"] . "\n\n";
	}
	
	$emailBody .= "\n\n\n\n";
}

$headers = 'From: [email protected]' . "\r\n" .
    	   'Reply-To: [email protected]' . "\r\n" .
    	   'X-Mailer: PHP/' . phpversion();
@mail( "[put in your email]", "CL Bot " . date("c"), $emailBody, $headers );

file_put_contents( dirname(__FILE__) .  "/seenPosts.json", json_encode( $seenPosts ) );