File: moreover_api.php

<?php

	/**************************************************************************
	      Title: Moreover PHP API
		  Author: Michael Bell (mikeb@map.com)
		 		    Based upon news_parser written by Carter Comunale (carter@brasscity.com)
		 Version: 1.0
		    Date: July 10th, 2001
   Description: This collection of classes are my attempt to create a complete
   				 API in PHP to the Moreover.com content syndication service. It
   				 has the ability to retrieve a listing of all current content
   				 channels and categories as well as retrieve all articles for
   				 those categories.
		 Classes: Moreover - Main API class
		 			 moreoverChannel - Channel class
		 			 moreoverCategory - Category class
		 			 moreoverArticle - Article class
	**************************************************************************/

	class moreoverArticle {
		var $ArticleID;
		var $URL;
		var $HeadlineText;
		var $Source;
		var $MediaType;
		var $Cluster;
		var $Tagline;
		var $DocumentURL;
		var $HarvestTime;
		var $AccessRegistration;
		var $AccessStatus;

		function moreoverArticle () {
			// Constructor - Nothing needs to be initilized in this class.
		}
	}

	class moreoverCategory {
		var $CategoryName;
		var $FeedName;
		var $ArticleObjects;

		function moreover_category () {
			// Constructor - Just empty out the ArticleObjects array
			$this->ArticleObjects = array ();
		}
	}

	class moreoverChannel {
		var $ChannelName;
		var $CategoryObjects;

		function moreoverChannel () {
			// Constructor - Just empty out the CategoryObjects array
			$this->CategoryObjects = array ();
		}
	}

	class Moreover {
		var $getArticleURL;
		var $getChannelURL;
		var $getSearchURL;
		var $ArticleObjects;
		var $ChannelObjects;
		var $_XMLParser;
		var $_CurrentTag;
		var $_CurrentArticle;
		var $_CurrentChannel;
		var $_CurrentCategory;

		function Moreover () {
			// Constructor
			// Empty out ArticleObjects array
			$this->ArticleObjects = array ();

			// Empty out ChannelObjects array
			$this->ChannelObjects = array ();

			// Setup the proper URL strings
			$this->getArticleURL = "http://www.moreover.com/cgi-local/page?o=xml&c=";
			$this->getSearchURL  = "http://p.moreover.com/cgi-local/page?o=xml&k=";
			$this->getChannelURL = "http://w.moreover.com/categories/nested_category_list.xml";
		}

		function getArticles () {
			$this->_clearArrays ();

			foreach (func_get_args () as $argument) {
				if (is_array ($argument)) {
					foreach ($argument as $arg) {
						$this->_interactAndParse ($this->getArticleURL . urlencode ($arg));
					}
				}
				else {
					$this->_interactAndParse ($this->getArticleURL . urlencode ($argument));
				}
			}

			return $this->ArticleObjects;
		}

		function getChannel ($channel) {
			$output = array ();
			$this->getChannels ();

			foreach ($this->ChannelObjects as $channelObj) {
				if ($channelObj->ChannelName == $channel) {
					$output = $channelObj->CategoryObjects;
					break;
				}
			}

			return $output;
		}

		function getChannels () {
			$this->_clearArrays ();

			$this->_interactAndParse ($this->getChannelURL);

			return $this->ChannelObjects;
		}

		function searchArticles () {
			$this->_clearArrays ();

			foreach (func_get_args () as $argument) {
				if (is_array ($argument)) {
					foreach ($argument as $arg) {
						$this->_interactAndParse ($this->getSearchURL . urlencode ($arg));
					}
				}
				else {
					$this->_interactAndParse ($this->getSearchURL . urlencode ($argument));
				}
			}

			return $this->ArticleObjects;
		}

		function _clearArrays () {
			$this->ArticleObjects = array ();
			$this->ChannelObjects = array ();
		}

		function _interactAndParse ($XMLFile) {
			// Initilize the XML parser
			$this->_XMLParser = xml_parser_create ('UTF-8');
			xml_parser_set_option ($this->_XMLParser, XML_OPTION_CASE_FOLDING, TRUE);
			xml_parser_set_option ($this->_XMLParser, XML_OPTION_TARGET_ENCODING, 'UTF-8');
			xml_set_element_handler ($this->_XMLParser, "_tagOpen", "_tagClose");
			xml_set_character_data_handler ($this->_XMLParser, "_cdata");

			xml_set_object ($this->_XMLParser, &$this);
			if (!($fp = fopen ($XMLFile, 'r'))) {
				echo "Could not open $XMLFile for parsing!\n";
			}
			while ($data = fread ($fp, 4096)) {
				if (!($data = utf8_encode ($data))) {
					echo 'ERROR'."\n";
				}
				if (!xml_parse ($this->_XMLParser, $data, feof ($fp))) {
					die (sprintf ( "XML error: %s at line %d\n\n",
						xml_error_string (xml_get_error_code ($this->_XMLParser)),
						xml_get_current_line_number ($this->_XMLParser)));
				}
			}
			fclose ($fp);

			$this->_cleanUp ();
		}

		function _tagOpen ($parser, $tag, $attributes) {
			$this->_CurrentTag = $tag;
			switch ($tag) {
				case "ARTICLE": 		$this->_CurrentArticle = new moreoverArticle ();
											break;
				case "CATEGORY": 		$this->_CurrentCategory = new moreoverCategory();
											break;
				case "CHANNEL": 		$this->_CurrentChannel = new moreoverChannel();
											break;
			}
		}

		function _tagClose ($parser, $tag) {
			switch ($tag) {
				case "ARTICLE": 	array_push($this->ArticleObjects, $this->_CurrentArticle);
										break;
				case "CATEGORY": 	array_push($this->_CurrentChannel->CategoryObjects, $this->_CurrentCategory);
										break;
				case "CHANNEL":	array_push($this->ChannelObjects, $this->_CurrentChannel);
										break;
			}
		}

		function _cdata($parser, $cdata) {
			switch ($this->_CurrentTag) {
				case "URL":
						if (!$this->_CurrentArticle->URL) {
							$this->_CurrentArticle->URL = $cdata;
						}
						break;
				case "HEADLINE_TEXT":
						if (!$this->_CurrentArticle->HeadlineText) {
							$this->_CurrentArticle->HeadlineText = $cdata;
						}
						break;
				case "SOURCE":
						if (!$this->_CurrentArticle->Source) {
							$this->_CurrentArticle->Source = $cdata;
						}
						break;
				case "MEDIA_TYPE":
						if (!$this->_CurrentArticle->MediaType) {
							$this->_CurrentArticle->MediaType = $cdata;
						}
						break;
				case "CLUSTER":
						if (!$this->_CurrentArticle->Cluster) {
							$this->_CurrentArticle->Cluster = $cdata;
						}
						break;
				case "TAGLINE":
						if (!$this->_CurrentArticle->Tagline) {
							$this->_CurrentArticle->Tagline = $cdata;
						}
						break;
				case "DOCUMENT_URL":
						if (!$this->_CurrentArticle->DocumentURL) {
							$this->_CurrentArticle->DocumentURL = $cdata;
						}
						break;
				case "HARVEST_TIME":
						if (!$this->_CurrentArticle->HarvestTime) {
							$this->_CurrentArticle->HarvestTime = $cdata;
						}
						break;
				case "ACCESS_REGISTRATION":
						if (!$this->_CurrentArticle->AccessRegistration) {
							$this->_CurrentArticle->AccessRegistration = $cdata;
						}
						break;
				case "ACCESS_STATUS":
						if (!$this->_CurrentArticle->AccessStatus) {
							$this->_CurrentArticle->AccessStatus = $cdata;
						}
						break;
				case "CHANNEL_NAME":
						if (!$this->_CurrentChannel->ChannelName) {
							$this->_CurrentChannel->ChannelName = $cdata;
						}
						break;
				case "CATEGORY_NAME":
						if (!$this->_CurrentCategory->CategoryName) {
							$this->_CurrentCategory->CategoryName = $cdata;
						}
						break;
				case "FEED_NAME":
						if (!$this->_CurrentCategory->FeedName) {
							$this->_CurrentCategory->FeedName = $cdata;
						}
						break;
			}
		}

		function _cleanUp () {
			xml_parser_free ($this->_XMLParser);
		}
	}
?>


Example use...<PRE>

&lt;?php

	include ("moreover_api.php");

	$mo_api = new Moreover ();

	// Retrieve headlines in the "Broadcasting industry news" category.
	print_r ($mo_api->getArticles ("Broadcasting industry news"));

	// Retrieve headlines with the keyword "Linux".
	print_r ($mo_api->searchArticles ("Linux"));

	// Retrieve headlines with the keyword "Linux" then retrieve headlines with
	// "MP3" and combine them into one responce.
	print_r ($mo_api->searchArticles ("Linux", "MP3"));

	// Retrieve headlines with the keyword "Linux" then retrieve headlines with
	// "MP3" and combine them into one responce. (Same as above, different call)
	print_r ($mo_api->searchArticles (array ("Linux", "MP3")));

	// Rerieve all channels and categories that Moreover.com has available.
	print_r ($mo_api->getChannels ());

	// Rerieve all categories within the specified channel.
	print_r ($mo_api->getChannel ("Business: general"));

?&gt;

</PRE>