Marcus Povey <marcus@dushka.co.uk>

* XML Import processor

git-svn-id: https://code.elgg.org/elgg/trunk@445 36083f99-b078-4883-b0ff-0f9b5a30f544

2 files changed, 101 insertions, 77 deletions

diff --git a/engine/lib/export.php b/engine/lib/export.php
index 4247ae348..e997a1c5a 100644
--- a/engine/lib/export.php
+++ b/engine/lib/export.php
@@ -323,6 +323,40 @@
 		return create_metadata($guid, "import_uuid", $uuid);
 	}
 	
+	
+	$IMPORTED_DATA = array();
+	$IMPORTED_OBJECT_COUNTER = 0;
+	
+	/**
+	 * This function processes an element, passing elements to the plugin stack to see if someone will
+	 * process it.
+	 * 
+	 * If nobody processes the top level element, the sub level elements are processed.
+	 * 
+	 * @param array $element The dom tree.
+	 */
+	function __process_element($element)
+	{
+		global $IMPORTED_DATA, $IMPORTED_OBJECT_COUNTER;
+		
+		// See if anyone handles this element, return true if it is.
+		$handled = trigger_plugin_hook("import", "all", array("name" => $element->name, "element" => $element), $to_be_serialised);
+
+		// If not, then see if any of its sub elements are handled
+		if (!$handled) 
+		{
+			if (isset($element->children)) 
+				foreach ($element->children as $c)
+					__process_element($c);
+		}
+		else
+		{
+			$IMPORTED_OBJECT_COUNTER ++; // Increment validation counter
+			$IMPORTED_DATA[] = $handled; // Return the constructed object
+		}
+
+	}
+	
 	/**
 	 * Export a GUID.
 	 * 
@@ -361,42 +395,6 @@
 		return $output;
 	}
 	
-	
-	
-
-	
-
-	$IMPORTED_DATA = array();
-	$IMPORTED_OBJECT_COUNTER = 0;
-	
-	/**
-	 * This function processes an element, passing elements to the plugin stack to see if someone will
-	 * process it.
-	 * If nobody processes the top level element, the sub level elements are processed.
-	 */
-/*	function __process_element(array $dom)
-	{
-		global $IMPORTED_DATA, $IMPORTED_OBJECT_COUNTER;
-		
-		foreach ($dom as $element)
-		{
-			// See if anyone handles this element, return true if it is.
-			$handled = trigger_plugin_hook("import", "all", array("name" => $element['name'], "element" => $element), $to_be_serialised);
-		
-			// If not, then see if any of its sub elements are handled
-			if (!$handled) 
-			{
-				if (isset($element['elements'])) 
-					__process_element($element['elements']);
-			}
-			else
-			{
-				$IMPORTED_OBJECT_COUNTER ++; // Increment validation counter
-				$IMPORTED_DATA[] = $handled; // Return the constructed object
-			}
-		}
-	}
-	
 	/**
 	 * Import an XML serialisation of an object.
 	 * This will make a best attempt at importing a given xml doc.
@@ -405,24 +403,19 @@
 	 * @return array An array of imported objects (these have already been saved).
 	 * @throws Exception if there was a problem importing the data.
 	 */
-/*	function import($xml)
+	function import($xml)
 	{
 		global $IMPORTED_DATA, $IMPORTED_OBJECT_COUNTER;
 		
 		$IMPORTED_DATA = array();
 		$IMPORTED_OBJECT_COUNTER = 0;
-		$IMPORTED_GUID_MAP = array();
 		
-		$dom = __xml2array($xml);
-		
-		__process_element($dom);
+		__process_element(xml_2_object($xml));
 		
 		if ($IMPORTED_OBJECT_COUNTER!= count($IMPORTED_DATA))
 			throw new ImportException("Not all elements were imported.");
 		
 		return $IMPORTED_DATA;
 	}
-
 	
-	*/
 ?>
\ No newline at end of file
diff --git a/engine/lib/xml.php b/engine/lib/xml.php
index 208763239..7b1c7da7b 100644
--- a/engine/lib/xml.php
+++ b/engine/lib/xml.php
@@ -12,6 +12,25 @@
 	 */
 
 	/**
+	 * @class XMLElement
+	 * A class representing an XML element for import.
+	 */
+	class XmlElement 
+	{
+		/** The name of the element */
+		public $name;
+		
+		/** The attributes */
+		public $attributes;
+		
+		/** CData */
+		public $content;
+		
+		/** Child elements */
+		public $children;
+	};
+	
+	/**
 	 * This function serialises an object recursively into an XML representation.
 	 * The function attempts to call $data->export() which expects a stdClass in return, otherwise it will attempt to
 	 * get the object variables using get_object_vars (which will only return public variables!)
@@ -89,41 +108,53 @@
 	}
 	
 	/**
-	 * XML 2 Array function.
-	 * Taken from http://www.bytemycode.com/snippets/snippet/445/
-	 * @license UNKNOWN - Please contact if you are the original author of this code.
-	 * @author UNKNOWN
+	 * Parse an XML file into an object.
+	 * Based on code from http://de.php.net/manual/en/function.xml-parse-into-struct.php by
+	 * efredricksen at gmail dot com
+	 * 
+	 * @param string $xml The XML.
 	 */
-	function xml2array($xml) 
+	function xml_2_object($xml)
 	{
-        $xmlary = array();
-               
-        $reels = '/<(\w+)\s*([^\/>]*)\s*(?:\/>|>(.*)<\/\s*\\1\s*>)/s';
-        $reattrs = '/(\w+)=(?:"|\')([^"\']*)(:?"|\')/';
-
-        preg_match_all($reels, $xml, $elements);
-
-        foreach ($elements[1] as $ie => $xx) {
-	        $xmlary[$ie]["name"] = $elements[1][$ie];
-	       
-	        if ($attributes = trim($elements[2][$ie])) {
-	                preg_match_all($reattrs, $attributes, $att);
-	                foreach ($att[1] as $ia => $xx)
-	                        $xmlary[$ie]["attributes"][$att[1][$ia]] = $att[2][$ia];
-	        }
-	
-	        $cdend = strpos($elements[3][$ie], "<");
-	        if ($cdend > 0) {
-	                $xmlary[$ie]["text"] = substr($elements[3][$ie], 0, $cdend - 1);
-	        }
-	
-	        if (preg_match($reels, $elements[3][$ie]))
-	                $xmlary[$ie]["elements"] = __xml2array($elements[3][$ie]);
-	        else if ($elements[3][$ie]) {
-	                $xmlary[$ie]["text"] = $elements[3][$ie];
-	        }
-        }
-
-        return $xmlary;
+		$parser = xml_parser_create();
+		
+		// Parse $xml into a structure
+		xml_parser_set_option($parser, XML_OPTION_SKIP_WHITE, 1);
+		xml_parser_set_option($parser, XML_OPTION_CASE_FOLDING, 0);	
+		xml_parse_into_struct($parser, $xml, $tags);
+		
+ 		xml_parser_free($parser);
+ 		
+ 		$elements = array(); 
+  		$stack = array();
+  		
+  		foreach ($tags as $tag)
+  		{
+			$index = count($elements);
+			
+  			if ($tag['type'] == "complete" || $tag['type'] == "open") 
+  			{
+      			$elements[$index] = new XmlElement;
+      			$elements[$index]->name = $tag['tag'];
+      			$elements[$index]->attributes = $tag['attributes'];
+      			$elements[$index]->content = $tag['value'];
+      			
+      			if ($tag['type'] == "open") 
+      			{ 
+        			$elements[$index]->children = array();
+        			$stack[count($stack)] = &$elements;
+        			$elements = &$elements[$index]->children;
+      			}
+    		}
+    		
+    		if ($tag['type'] == "close") 
+    		{
+      			$elements = &$stack[count($stack) - 1];
+      			unset($stack[count($stack) - 1]);
+    		}
+  		}
+  		
+  		return $elements[0];
 	}
+	
 ?>
\ No newline at end of file