PHP file to combine RSS feeds etc, used on gyford.com. Requires many other files to work...

combine_feeds.php

<?php
/**
 * @package	Phil 
 * Combine recent photos, writing, links, etc into one feed and one HTML/PHP include.
 *
 * NOTE: Required the dom.so extension to be enabled in php.ini
 *		 (It wasn't, by default, on Textdrive, 2013-03-14.)
 *
 * v1.4 2015-09-04
 *
 * Version history
 * v1.0 2006-02-17
 * v1.01 2006-03-20 
 * 		Checked for photo tags before cycling through them.
 * v1.1 2006-04-22
 * 		Look for featured photo tag in description, not tags.
 * v1.11 2006-04-30
 *		Fixed problems with putting the wrong item descriptions/content-encodeds in.
 * v1.12 2006-06-18
 *		Stopped including init.php, now just config.php.
 * v1.13 2006-10-18
 * 		Fixed a bug that was doing wrong descriptions for 'Writing' entries.
 * v1.14 2007-05-03
 * 		Added links for wfw:comments in Links and Writing.
 * v1.15 2007-05-07
 *		Changed the featured photo tag stuff slightly.
 * v1.2 2008-09-26
 *		Added the ability to fetch remote feeds, eg PhilipGyford.com.
 * v1.21 2008-10-17
 *		Corrected the total number of photos when some aren't to be used.
 * v1.22 2008-10-21
 * 		Stopped photos from showing when they're marked not to be. 
 * v1.23 2008-10-27
 * 		Added Comments blog to local feeds fetched.
 * v1.24 2008-12-26
 * 		Some fixes to make it more compatible with the rest of the site.
 * v1.25 2008-12-30
 * 		Stopped too-old items from remote feeds being displayed.
 * v1.3 2009-02-09
 *		Added more feeds, support for YouTube feeds, generates two aggregated feeds, plus outputs HTML includes, rather than PHP arrays.
 * v1.31 2009-02-12
 *		Added 'Posted in...' intros to RSS feed items. Fixed number of RSS items displaying.
 * v1.32 2013-03-12
 *		Upgraded SimplePie.
 * v1.33 2013-03-14
 *		Changed 'posted_url' to 'flickr_url' for multiple Flickr images when writing RSS feed.
 * v1.34 2014-06-17
 *		Fixed errors from missing photo descriptions/titles.
 * v1.4 2015-09-04
 *		Updated YouTube to use API v3.
 */

error_reporting(E_ALL);
ini_set('display_errors', '1');

require_once '/home/philgyford/webapps_symlinked/gyfordphil/includes/init.php';
//require_once '/Users/phil/Sites/webfaction/webapps_symlinked/gyfordphil/includes/init.php';

require_once 'Flickr.php';
require_once 'RSSWriter.php';
require_once 'simplepie/autoloader.php';
require_once 'HTML.php';

/**
 ************************************************
 * Configuration
 */

/**
 * Number of days of entries to list on the website front page.
 * It assumes there are that many days' worth of photos/entries available in feeds/includes.
 * Now only used for the Photos page.
 */
define('DAYS_TO_LIST_HTML', 60);

/**
 * Number of entries to list in the RSS feed.
 * We'll fetch enough entries to fill the front page (see above).
 * These RSS entries come out of that, so it's possible the ENTRIES_TO_LIST_RSS number won't be reached.
 */
define('ENTRIES_TO_LIST_RSS', 12);

/**
 * The RSS files we're going to generate.
 */
define('RSS_FILE_PATH_PERSONAL', DOCS_DIR . '/syndication/index-fb.rdf');
define('RSS_FILE_PATH_EVERYTHING', DOCS_DIR . '/syndication/everything-fb.rdf');

/**
 * The directory where the MT-generated files to include are.
 * These contain arrays of data about recent entries in the writing and links blogs.
 */
define('MT_INCLUDE_DIR', INCLUDES_DIR . '/caches/misc/');

/**
 * The directory where the saved HTML/PHP files will be saved.
 * This will feature arrays of data about the latest photos, writing and links for inclusion on the front page.
 * Another file will only feature the photos.
 */
define('SAVED_FILE_DIR', INCLUDES_DIR . '/caches/misc/');


/**
 * The directoy SimplePie will cache remote feeds in.
 */
define('SIMPLEPIE_CACHE_DIR', INCLUDES_DIR . '/caches/simplepie/');

/*************************************************/


/**
 * If you add anything to this, you may want to add stuff in web/includes/Phil/classes/HTML.php (the entriesColumns() method) for the new source.
 */
$sources = array (
	'writing' => array (
		'rss_intro'	=> 'Posted in <a href="http://www.gyford.com/phil/writing/">Phil Gyford\'s Writing</a>'
	),
	'links' => array (
	),
	'comments' => array (
		'rss_intro'	=> 'Posted in <a href="http://www.gyford.com/phil/comments/">Comments posted elsewhere</a>'
	),

	//'philipgyford' => array (
	//	'feed_url'	=> 'http://feeds.feedburner.com/PhilipGyford',
	//	'rss_intro' => 'Posted at <a href="http://www.philipgyford.com/">PhilipGyford.com</a>'
	//),
	// 'overmorgen' => array (
	// 	'feed_url'	=> 'http://feeds.feedburner.com/Overmorgen',
	// 	'rss_intro' => 'Posted at <a href="http://www.overmorgen.com/">Overmorgen</a>'
	// ),
	// 'whitstillman' => array (
	// 	'feed_url'	=> 'http://feeds.feedburner.com/WhitStillman',
	// 	'rss_intro' => 'Posted at <a href="http://www.whitstillman.org/">Whit Stillman</a>'
	// ),

	'septivium' => array (
		'feed_url'	=> 'http://feeds2.feedburner.com/Septivium',
		'rss_intro' => 'Posted at <a href="http://www.septivium.com/">Septivium</a>'
	),
	'pepysdiarynews' => array (
		'feed_url'	=> 'http://feeds.feedburner.com/PepysDiary-SiteNews',
		'rss_intro' => 'Posted at <a href="http://www.pepysdiary.com/about/news/">Pepys\' Diary Site News</a>'
	),
	'youtube' => array (
		//'feed_url'	=> 'https://www.youtube.com/feeds/videos.xml?playlist_id=FLgycUrFUa7ABvcs_vHY0l2Q',
		'feed_url' => 'https://www.googleapis.com/youtube/v3/playlistItems?part=snippet%2CcontentDetails%2Cstatus&playlistId=FLgycUrFUa7ABvcs_vHY0l2Q&key=' . YOUTUBE_API_KEY,
		'rss_intro' => 'Favourited <a href="https://www.youtube.com/playlist?list=FLgycUrFUa7ABvcs_vHY0l2Q">on YouTube</a>'
	)
);

/**
 * The array in which we'll put all the entries to display.
 * This will have keys like '1102167975_flickr'.
 * The values are an array of entry data.
 */
$ITEMS = array();

/**
 * Start doing stuff...
 */
getFlickr();

getYouTube($sources);

getLocalWeblogs($sources);

getRemoteWeblogs($sources);


// Sort in reverse chronological order.
krsort($ITEMS);

writeHTML(array_values($ITEMS));

writeRSS('personal', $sources, array_values($ITEMS));
writeRSS('everything', $sources, array_values($ITEMS));



/**
 * Puts all the photos for the time period into $items.
 * The list of photos for a day is in an array provided by phpFlickr.
 * There are also two extra arrays, featured_indexes and 
 * thumbnail_indexes which indicate which photos for that day
 * are to be featured on the website.
 */
function getFlickr() {
	global $ITEMS;
	
	$from_day = gmdate('Y-m-d', time() - (86400 * (DAYS_TO_LIST_HTML - 1)));

	$days_of_photos = Flickr::getDayCounts($from_day, DAYS_TO_LIST_HTML);

	// Cycle through each day and get photos for any day on which there are some.
	foreach ($days_of_photos as $day) {
		if ($day['count'] > 0) {

			$photo_data = Flickr::dayPhotosForDisplay($day['fromdate'], 'posted');
			
			// It could be that every photo that day was marked not to be displayed so we
			// end up with none to show.
			if ($photo_data['photos']['total'] > 0) {
				$ITEMS[$day['fromdate'] . '_flickr'] = $photo_data;
			}
		}
	}
}


function getLocalWeblogs($sources) {
	global $ITEMS;
		
	foreach ($sources as $blog_key => $source_data) {
		if (!isset($source_data['feed_url'])) {
			// If there's no RSS feed, we assume it's a local file.
			include MT_INCLUDE_DIR . 'mt_recent_entries_' . $blog_key . '.php';

			$ITEMS = array_merge($ITEMS, $entries);
		}
	}
}

function getYouTube($sources) {
	global $ITEMS;

	$contents = file_get_contents($sources['youtube']['feed_url']);
	$json = json_decode($contents, TRUE);

	if (isset($json['items'])) {
		foreach($json['items'] as $item) {
			$snippet = $item['snippet'];
			$video_url = 'https://www.youtube.com/watch?v=' . $snippet['resourceId']['videoId'];
			$item_datetime = date_format(date_create($snippet['publishedAt']), 'Y-m-d H:i:s');

			$entry = array();
			$entry['blog_key'] = 'youtube';
			$entry['title'] = $snippet['title'];
			$entry['permalink'] = $video_url;
			$entry['author'] = '';
			$entry['date'] = $item_datetime;
			$entry['body'] = '<a href="' . $video_url . '"><img src="' . $snippet['thumbnails']['default']['url'] . '" alt="YouTube thumbnail" /></a>';
			$entry['extended'] = true;
			$entry['more'] = '<p>' . $snippet['description'] . '</p>';
		
			$entry['excerpt'] = strip_tags($entry['body']);

			$ITEMS[strtotime($item_datetime) . '_' . $blog_key] = $entry;
		}
	}
}

function getRemoteWeblogs($sources) {
	global $ITEMS;
	
	foreach ($sources as $blog_key => $source_data) {
		if (isset($source_data['feed_url'])) {
			$feed = new SimplePie();
			// Feedburner wasn't letting us see the feed with the default useragent.
			$feed->set_useragent('Gyford.com');
			$feed->enable_cache(true);
			$feed->set_cache_location(SIMPLEPIE_CACHE_DIR);
			$feed->set_feed_url($source_data['feed_url']);
		
			// Assuming we know that all included HTML tags are safe.
			$feed->strip_htmltags(false);
			// We also know we want to keep attributes (particularly 'class'):
			$feed->strip_attributes(false);
		
			// Otherwise we at least need to do this:
			// $strip_htmltags = $feed->strip_htmltags;
			// unset($strip_htmltags[array_search('object', $strip_htmltags)]);
			// unset($strip_htmltags[array_search('param', $strip_htmltags)]);
			// unset($strip_htmltags[array_search('embed', $strip_htmltags)]);
			// $strip_htmltags = array_values($strip_htmltags);
			// $feed->strip_htmltags($strip_htmltags);
			// And something to keep the 'class' attribute.
			$feed->init();
		
			if ($feed->data) {	
				//$from_time =  time() - (86400 * (DAYS_TO_LIST_HTML - 1));
				foreach ($feed->get_items() as $item) {
					// if ($item->get_date('U') < $from_time) {
					// 					// This item is older than we want for the front page, so stop here.
					// 					break;
					// 				}
					$entry = array();
					$entry['blog_key'] = $blog_key;
					$entry['title'] = $item->get_title();
					$entry['permalink'] = $item->get_permalink();
					$entry['author'] = $item->get_author()->get_name();
					$entry['date'] = $item->get_date('Y-m-d H:i:s');

					if ($blog_key == 'youtube') {
						// We need to add an image to the YouTube data.
					
						if ($enclosure = $item->get_enclosure()) {
							$entry['body'] = '<a href="' . $item->get_permalink() . '" title="' . $item->get_title() . '"><img src="' . $enclosure->get_thumbnail() . '" alt="YouTube thumbnail" /></a>';
							$entry['extended'] = true;
							$entry['more'] = '<p>' . $enclosure->get_description() . '</p>';
						}
					
					} else {
						// All the standard text-based feeds, probably generated by me.
						preg_match("/^(.*?<p>.*?<\/p>)(.*)?$/s", $item->get_content(), $matches);
						$entry['body'] = $matches[1];
						if (isset($matches[2]) && $matches[2] != '') {
							$entry['extended'] = true;
							$entry['more'] = trim($matches[2]);
						}
					}
				
					$entry['excerpt'] = strip_tags($entry['body']);

					$ITEMS[strtotime($item->get_date('Y-m-d H:i:s')) . '_' . $blog_key] = $entry;
				}
			}
		}
	}
}


/**
 * Write the files that will be included in the site.
 * @param array $entries An array of hashes - each hash info about an entry.
 */
function writeHTML($entries) {

	// Create the old front page list view. Might as well keep it going.
	if ($fh = @fopen(SAVED_FILE_DIR . 'home.html', 'w')) {
		
		$html = HTML::entries($entries, array('return_html'=>true));
		if (@fwrite($fh, $html)) {
			@fclose($fh);
		}
	}
	
	// Create the columned-view for the front page.
	if ($fh = @fopen(SAVED_FILE_DIR . 'home_columns.html', 'w')) {
		$html = HTML::entries($entries, array('return_html'=>true, 'format' => 'columns'));
		if (@fwrite($fh, $html)) {
			@fclose($fh);
		}
	}

	// Create the list view for the Photos page.
	$photo_entries = array();
	foreach ($entries as $entry) {
		if ($entry['blog_key'] == 'photos') {
			$photo_entries[] = $entry;
		}
	}
	if ($fh = @fopen(SAVED_FILE_DIR . 'photos.html', 'w')) {
		$html = HTML::entries($photo_entries, array('return_html'=>true));
		if (@fwrite($fh, $html)) {
			@fclose($fh);
		}
	}
}


/**
 * @param string $type Either 'personal' or 'everything'.
 * @param array $entries The big array of all entries. 
 */
function writeRSS($type, $sources, $entries) {
	
	if ($type == 'personal') {
		$feed_title = 'Phil Gyford (personal)';
		$feed_description = 'Writing, photos and links by Phil Gyford';
		$blogs_to_use = array('writing', 'photos', 'links', 'comments', 'philipgyford');
	} else {
		$feed_title = 'Phil Gyford (everything)';
		$feed_description = 'Most things created or liked by Phil Gyford across the web';
		$blogs_to_use = array(); // ALL!
	}
	$rss = new RSSWriter(
		"http://www.gyford.com/", 
		$feed_title,
		$feed_description,
		array (
			"dc:date"		=> date(DATE_ATOM),
			"dc:creator"	=> "Phil Gyford <[email protected]>",
			"dc:publisher"	=> "Phil Gyford <[email protected]>",
			"dc:language"	=> "en-gb"
		)
	);
	
	$rss->setImage("http://www.gyford.com/phil/images/rss.jpg", "Phil Gyford");
	
	$rss->useModule("content", "http://purl.org/rss/1.0/modules/content/");
	$rss->useModule("wfw", "http://wellformedweb.org/CommentAPI/");
	
	foreach ($entries as $n => $entry) {
		
		if ($n+1 > ENTRIES_TO_LIST_RSS) {
			break;
		}
		
		// Only add this entry if we're using ALL blogs ($blogs_to_use is empty)
		// or this entry's blog is in the $blogs_to_use list.
		if (count($blogs_to_use) == 0 || in_array($entry['blog_key'], $blogs_to_use)) {

			$comments = '';
			
			if ($entry['blog_key'] == 'photos') {
				if ($entry['photos']['total'] > 0) {
					$link = $entry['photos']['flickr_url'];
					$title = 'Photos for ' . date("j F Y", strtotime($entry['date']));
					$description = '';
					$content_encoded = '';
					foreach ($entry['photos']['photo'] as $photo) {
				
						if ($photo['photo']['description'] == '') {
							$description_text = '';
						} else {
							$description_text = "<br />\n" . preg_replace("/\n/", "<br />\n", $photo['photo']['description']);
						}
						$description .= '<table border="0" cellspacing="10"><tbody><tr>
			<td valign="top" align="center" width="100" style="vertical-align: top;"><a href="' . $photo['urls']['photopage'] . '" title="See a larger version"><img src="' . $photo['urls']['thumbnail'] . '" alt="' . $photo['photo']['title']. '" border="0" /></a></td>
			<td valign="top" style="vertical-align: top;"><strong><a href="' . $photo['urls']['photopage'] . '">' . $photo['photo']['title'] . "</a></strong>
		$description_text</td>
		</tr></tbody></table>\n";
					}
				}

			} elseif ($entry['blog_key'] == 'links') {
				$link = $entry['permalink'];
				$title = 'Links for ' . date("j F Y", strtotime($entry['date']));
				$description = $entry['body'];
				$content_encoded = '';

				if (isset($entry['comments_active']) && $entry['comments_active']) {
					$comments = $entry['permalink'] . '#comments';
				}
			
			} else {
				// writing, philipgyford, etc.
			
				if (isset($entry['remoteurl'])) {
					// Comments feed.
					$link = $entry['remoteurl'];
				} else {
					$link = $entry['permalink'];
				}
				$title = $entry['title'];
				if (isset($entry['extended']) && $entry['extended']) {
					$description = $entry['body'] . "\n<p><a href=\"$link#more\">Read more&#8230</a></p>";
					$content_encoded = $entry['body'] . "\n\n" . $entry['more'];
				} else {
					$description = $entry['body'];
					$content_encoded = $entry['body'];
				}
				if (isset($entry['comments_active']) && $entry['comments_active']) {
					$comments = $entry['permalink'] . '#comments';
				}
			}
		
			$item_details = array (
				'description'	=> $description
			);
			if (isset($content_encoded) && $content_encoded != '') {
				$item_details['content:encoded'] = $content_encoded;
			}
			if (isset($comments) && $comments != '') {
				$item_details['wfw:comments'] = $comments;
			}
			
			if (isset($sources[$entry['blog_key']]) && isset($sources[$entry['blog_key']]['rss_intro'])) {
				$item_details['description'] = '<p>' . $sources[$entry['blog_key']]['rss_intro'] . "</p>\n" . $item_details['description'];
				$item_details['content:encoded'] = '<p>' . $sources[$entry['blog_key']]['rss_intro'] . "</p>\n" . $item_details['content:encoded'];
			}
			
			$item_details['dc:creator'] = 'phil';
			$item_details['dc:date'] = date('c', strtotime($entry['date']));

			$rss->addItem($link, $title, $item_details);
		}
	}

	// RSSWriter prints its stuff to screen.
	// So we use output buffering to write it to a file instead.
	
	$file_path = $type == 'personal' ? RSS_FILE_PATH_PERSONAL : RSS_FILE_PATH_EVERYTHING;

	ob_start();
	$rss->serialize();
	$text = ob_get_contents();
	ob_end_clean();
	if ($fh = @fopen($file_path, 'w')) {
		if (@fputs($fh, $text, strlen($text))) {
			@fclose($fh);
		}
	}
}


/*
 * Replacement for file_get_contents for when URL file-access is disabled.
 * http://uk2.php.net/manual/en/function.fopen.php#55922
 * @param string $url
 * @return string The contents of the file or an empty string on failure.
 */
function fetchFileContents($url)
{
	$ch = curl_init();

	curl_setopt ($ch, CURLOPT_URL, $url);
	curl_setopt ($ch, CURLOPT_HEADER, 0);
	
	// Without this we were getting Internal Server Errors
	// when Last.fm was unavailable and returning nothing after
	// a long delay.
	curl_setopt ($ch, CURLOPT_TIMEOUT, 5);
	
	// Return the result instead of printing it.
	curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
	
	$string = curl_exec ($ch);

	if (curl_errno($ch)) {
		// For debugging.
	   //print curl_error($ch);
	}
	
	curl_close($ch);
	
	if (!is_string($string) || !strlen($string)) {
		// Something went wrong with the curl_exec and it returned false.
		// Use the curl_error() bit above to find out what went wrong.
		return '';
	} else {
		return $string;
	}

}


?>