Scans http://brettterpstra.com/otherstuff/systematic-linkage/ and attempts to output the "Top Picks" in markdown format producing HTML like this: http://cloud.zoooot.com/Q8Wr

BTPicks.php

<?php

$curl_error_str = "";
$curl_error_no = 0;
$httpCode = 0;
$curl_info = "";

$page="";
$html="";

#echo mb_internal_encoding() . "\n";
#mb_internal_encoding("UTF-8");
#echo mb_internal_encoding() . "\n";

$url = 'http://brettterpstra.com/otherstuff/systematic-linkage/';

if (!$page = CURL($url)){

  $str = "\nScript: " . $_SERVER["SCRIPT_FILENAME"] . "\nFile: " . __FILE__ . "\nFunction: "
	. __FUNCTION__ . "() - line: " . __LINE__ .  "\nERROR($curl_error_no) - $curl_error_str\nHTTP Response Code: $httpCode\nCURL INFO:\n$curl_info";

	die("ERROR getting $url: $str");
}
else{
	// strip out all the header stuff
	$start = '<h3 id="introductions-all-around">';
	$end = '<footer>';

	$html = GetBetween($page,$start,$end );

	//assume no error from GetBetween - TODO

	// IMP - add in the encoding = UTF-8, otherwise DOMDocument returns nodevalues in ASCII even though all chars may not be valid
	$html = '<?xml version="1.0" encoding="utf-8" ?><h3 id="introductions-all-around">' . $html;
}

// create DOM
$xml = new DOMDocument();
$xml->loadHTML($html);

foreach ( $xml->getElementsByTagName( 'a' ) as $link ) {
	
	if ( $link->getAttribute( 'href' )){

		
		if(isMaybeATopPick($link->getAttribute( 'href' ), $link->nodeValue ) == true){


			// web page is encoded as UTF-8 but  $link->nodeValue was coming back as ASCII
			// until I added the xml enc in the html
			// hence this faffing about

			// echo mb_detect_encoding($link->nodeValue);
			// echo "  ";
			// echo mb_detect_encoding($link->getAttribute( 'href' ));
			// echo "  ";
			// echo($link->getAttribute( 'href' ) . " = ");
			// print($link->nodeValue);
			 //			echo "\n";

			//$string = iconv('ASCII', 'UTF-8//IGNORE', $link->nodeValue);
			//$string= preg_replace ('/([\x80-\xff])/se', "pack (\"C*\", (ord ($1) >> 6) | 0xc0, (ord ($1) & 0x3f) | 0x80)", $link->nodeValue);
			//echo($link->getAttribute( 'href' ) . " = " . $link->nodeValue . "\n");

			echo "* [$link->nodeValue](" . $link->getAttribute( 'href' ) . ")\n";
		}
	}
}

exit;


function isMaybeATopPick($url, $text){

	$picks = array( "app", "mac", "itunes.apple.com", "OS X", "iOS", "iTunes", "iPhone", "iPad", "Apple" );

	foreach ($picks as $needle) {

		if( stripos($url, $needle) !== false || stripos($text, $needle) !== false){

			if( andIsNotOneOfThese($url, $text) == true){
				return true;
			}
		}
	}

	return false;
}

function andIsNotOneOfThese($url, $text){

	$nonPicks = array( "alpha.app.net", "join.app.net", "minimalmac", "http://macsparky.com", "macdrifter.com" , "https://twitter.com", "Frakes", "MacStories", "5by5.tv/mpu", "www.theverge.com");

	foreach ($nonPicks as $needle) {

		if( stripos($text, $needle) !==  false || stripos($url, $needle) !==  false){

			return false;
		}
	}

	return true;
}

function GetBetween($content,$start,$end){
	$r = explode($start, $content);

	if (isset($r[1])){
		$r = explode($end, $r[1]);
		return $r[0];
	}
	return '';
}

function CURL($url, $retries = 3)
{

	global $httpCode;
	global $curl_error_str;
	global $curl_error_no;
	global $curl_info;
	
	$timeout = 5;
	$connectionTimeout = 2;

	$curl = curl_init($url);

	if (is_resource($curl) === true)
	{
		curl_setopt($curl, CURLOPT_FAILONERROR, true);
		curl_setopt($curl, CURLOPT_TIMEOUT, $timeout);
		curl_setopt($curl, CURLOPT_RETURNTRANSFER, true);
		curl_setopt($curl, CURLOPT_SSL_VERIFYHOST, false);
		curl_setopt($curl, CURLOPT_SSL_VERIFYPEER, false);
		curl_setopt($curl, CURLOPT_CONNECTTIMEOUT, $connectionTimeout);
        curl_setopt($curl, CURLOPT_NOSIGNAL, 1);
       	//curl_setopt($curl, CURLOPT_VERBOSE, true);

		$result = false;

		while (($result === false) && (--$retries > 0))
		{
			//echo "CURL retries = $retries\n";
			$result = curl_exec($curl);
		}

		if(!$result){

			$httpCode = curl_getinfo($curl, CURLINFO_HTTP_CODE);

			$curl_info_temp = curl_getinfo($curl);

			if(is_array($curl_info_temp) == TRUE){

				$curl_info = var_export($curl_info_temp, 1);
			}

		}

		// Check if any error occured
		if(curl_errno($curl))
		{

			$curl_error_str = curl_error($curl);
			$curl_error_no = curl_errno($curl);

			//echo 'Curl error: ' . curl_error($curl);
			//echo 'Curl error no: ' . curl_errno($curl);
		}


		curl_close($curl);
	}
	else{

		$curl_error_str = "curl_init() failed for $url";
	}

	return $result;
}

?>