MixCloud RSS

index.php

<?php

/***********************************
	   config stuff up here 
************************************/

set_time_limit(0);

date_default_timezone_set('America/New_York');

$feeds = ["medschoolmusic", "evil_concussion", "respectdnb", "lifestylemusic", "Cyantific"];
foreach ($feeds as $feed) {
	getFeed($feed);
}

/***************************************
  nothing to configure below this line
****************************************/ 

function getFeed($my_podcast) {
	include_once('curl.php');

	$my_feed_url = $my_podcast.".rss";
	$language = "en-us";
	$user_info = json_decode(curlGet('http://api.mixcloud.com/'.$my_podcast .'/'));

	$itunes_image = $user_info->pictures->large;
	$my_description = stripslashes($user_info->biog);
	$updated = date(DATE_RSS,strtotime($user_info->updated_time));
	$my_title = $user_info->name;
	$my_link = $user_info->url;
	 
	/* write out the outer shell, channel, globals */ 
	$updated= date("D, d M Y H:i:s T",strtotime("now"));
	$output = "<?xml version=\"1.0\" encoding=\"UTF-8\"?>
		<rss version=\"2.0\" xmlns:itunes=\"http://www.itunes.com/dtds/podcast-1.0.dtd\"
			 xmlns:atom=\"http://www.w3.org/2005/Atom\">	
		<channel>
			<title><![CDATA[$my_title]]></title>
			<link>$my_link</link>
			<description><![CDATA[$my_description]]></description>
			<image>
				<url>$itunes_image</url>
				<link>$my_link</link>
				<description><![CDATA[$my_title]]></description>
				<title><![CDATA[$my_title]]></title>
			</image>
			<language>$language</language>
			<lastBuildDate>$updated</lastBuildDate>
			<pubDate>$updated</pubDate>
			<itunes:explicit>no</itunes:explicit>
			<atom:link href=\"$my_feed_url\" rel=\"self\" type=\"application/rss+xml\" /> 

			";

	$nextURL = null;
	$x = 0;
	do {
		if (++$x > 10) {
			break;
		}
		/* First get the info page for this playlist */
		$url = $nextURL ? $nextURL : $my_podcast;
		$my_podcast_page = file_get_contents('http://www.mixcloud.com/'.$url);

		$my_podcast_page = mb_convert_encoding($my_podcast_page, 'HTML-ENTITIES', "UTF-8");

		$doc = new DOMDocument();
		/* hide warnings - html docs likely won't parse correctly */ 
		libxml_use_internal_errors(true);
		$doc->loadHTML($my_podcast_page); 
		
		$xpath = new DOMXpath($doc);

		if ($xpath->query('//div[@class="infinitescroll-end"]')->length == 0) {
			break;
		}
		$nextURL = $xpath->query('//div[@class="infinitescroll-end"]')->item(0)->getAttribute("m-next-page-url");
		
		$episodes = $xpath->query('//div[@class="card-elements-container cf"]'); 
		//echo '<p>episodes has '. $episodes->length .'</p>';
		if($episodes->length > 0) {
			//echo '<p>List of episodes:</p><ol>';
			foreach ($episodes as $container) {
				$episode_image = $xpath->query('.//div[@class="card-cloudcast-image"]/a/img',$container);
				$large_photo = 'http:' . $episode_image->item(0)->getAttribute("src");
				$episode_info = $xpath->query('.//div[@class="card-cloudcast-image"]/span',$container);
				if ($episode_info->length == 0) { //episodes that are disabled have no title
					continue;
				}
				$e_title = $episode_info->item(0)->getAttribute("m-title");
				$e_url = 'http://www.mixcloud.com'. $episode_info->item(0)->getAttribute("m-url");
				$e_description = json_decode(curlGet('http://api.mixcloud.com'.$episode_info->item(0)->getAttribute("m-url")))->description;
				$e_preview = $episode_info->item(0)->getAttribute("m-preview");
				$length = strpos($e_preview, "preview");
				$e_server = substr($e_preview,0,$length - 1);
				$e_server = str_replace("audiocdn", "stream", $e_server);
				// todo - should not just be 39 magic number, but where 'preview/' is in url
				$e_identifier = substr($e_preview,$length + 9);
				$e_identifier = rtrim($e_identifier,".mp3"); 
				$e_download =  $e_server . '/c/m4a/64/'. $e_identifier .'.m4a'; 
				$e_original = $e_server . '/c/originals/' . $e_identifier . '.mp3';
				$item_size = get_Size($e_download);
				/* if $item_size is 168 this means not found */ 
				if($item_size > 200) {
					$episode_update = $xpath->query('.//div[@class="card-stats cf"]/span[@class="card-date"]/time',$container); 
					if($episode_update) {
						$pubDate = strtotime($episode_update->item(0)->nodeValue);
					} else {
						$pubDate = "false";
					} 
					$output .= "<item>
					<pubDate>". date(DATE_RSS,$pubDate) ."</pubDate>
					<title><![CDATA[$e_title]]></title>
					<link>$e_url</link>
					<description><![CDATA[$e_description]]></description>
					<itunes:image href=\"$large_photo\" />
					<enclosure url=\"$e_download\" length=\"$item_size\" type=\"audio/mp4\" />
					<guid isPermaLink=\"true\">$e_url</guid>
				</item>
					";
				}
			}
		}
	} while ($nextURL);

	/* seems like we're getting the closing footer too early */
	sleep(2); 

	/* and output the closing footer */
	$output .= "
		</channel>
	</rss>";

	header("Content-Type: application/rss+xml; charset=UTF-8");
	file_put_contents($my_podcast.".rss", $output);
	chmod($my_podcast.".rss", 0755);
	header("Location:". $my_podcast.".rss");
}

Thanks Viodet. This is working fine for me now.

Copper-Dale, you probably don't have the whole file. The portion above isn't the complete file. I would download it again.

Do you know a way around the user agent issues? I tried to set the user agent in the curl.php but that didn't work. It looks like iTunes makes a direct call from the feed file generated. Do you think it's possible to set the user agent in the generated file to spoof the user agent?

Apparently all right, spotify approved the link but processing is slow, has anyone gone through it?