Custom Source Handler to deal with multiple eZ content-objects that have been exported into a single XML source file.

Note 1: This requires a small modification to ImportOperator.php (below) so that you can pass the name of the file to the command line command:

#php extension/data_import/scripts/run.php -s <siteaccess> -i ImportOperator -d MyCustomSourceXMLHandler -f <name-of-source-XML-file-without.xml-suffix>

Note 2: This also assumes that your source XML is in the same format as those exports found in the data_import 'examples' folder (or has been transformed using XSL into this format from another XML format, say from the ezxmlexport extension)

Note 3: This only works for 'folder' and 'article' content-classes and doesn't seem to want to import inline images, but I'm working on it.

Note 4: You will of course need to add this class to the ezp_extension.php array in order to eZ to recognise your new class.

ImportOperator.php Modification

To come...

Custom Source Handler

<?php


/*


*       @decription:    Generic exported eZ content-class import class. Relies on the


'n' export method from ezxmlexport.


* Heavily modified version of the example script: XMLFolders.php


*       @author:                R.Michell April 2010


*       @package:               data_import


*/


 

class MyCustomSourceXMLHandler extends XmlHandlerPHP


{


        var $handlerTitle = 'Generic Handler';                                                                          // Default handler name.


        var $logfile = 'data_import.log';                                                                                       // Log file-name.               


        const REMOTE_IDENTIFIER = 'xmlimport_';                                                                         // Default. Is appended-to later..


        var $root_node = 'all';                                                                                                         // Source XML root node element.


        var $xml_source_path = 'extension/data_import/dataSource/exports';                      // Path to parent dir od source XML file(s) for import.


        var $xml_source_file;                                                                                                           // ezxmlexport uses an export name for the export's parent dir and XML filename.


        var $parent_id_fallback = 2;                                                                                            // Fallback to root node ('Main') if a parent_id cannot be found for an imported object


 

        /* 

        * Constructor


        */


        public function MyCustomSourceXMLHandler()


        {


                $message = 'Starting data_import process';


                $logfile = $this->logfile;


                $this->writeLog($message,$logfile);


        }


 

        /*


        * @desc                 Log stuff to ezpublish log using ezcomponents


        * @return               void


        * @args                 string $message, string $logfile


        * @author               Russell Michell April 2010


        * @ToDo                 Relies on presence of ezcomponents - run a check for its existence.


        *                               Find native eZ function to get ezpublish intall path and pass to $writer var


        */


        private function logger($message,$logfile)


        {


                // Get the one and only instance of the ezcLog.


                $log = ezcLog::getInstance();


 

                // Get an instance to the default log mapper.


                $mapper = $log->getMapper();


 

                // Create a new Unix file writer, that writes to the file: "default.log".


                $writer = new ezcLogUnixFileWriter('/var/www/htdocs/ez-GeoNet/var/log/',$logfile);


 

                // Create a filter that accepts every message (default behavior).


                $filter = new ezcLogFilter;


 

                // Combine the filter with the writer in a filter rule.


                $rule = new ezcLogFilterRule( $filter, $writer, true );


 

                // And finally assign the rule to the mapper.


                $mapper->appendRule( $rule );


 

                // Write a message to the log.


                $log->log($message,ezcLog::WARNING); 

        }


 

        private function writeLog( $message, $newlogfile = '')


        {


                if($newlogfile)


                {


                        $logfile = $newlogfile;


                }


                else


                {


                        $logfile = $this->logfile;


                }


                $this->logger($message,$logfile);


        }


        


        // Mapping for source XML field name to an eZ attribute name:


        function geteZAttributeIdentifierFromField()


        {


                $field_name = $this->current_field->getAttribute('name');


                if($this->getTargetContentClass() == 'folder')


                {


                        switch ($field_name)


                        {


                                case 'shortname':


                                        return 'short_name';


                                break;


                                case 'showsubitems':


                                        return 'show_children';


                                break;


                                case 'publishdate':


                                        return 'publish_date';


                                break;


                                case 'shortdescription':


                                        return 'short_description';


                                break;


                                default:


                                        return $field_name;


                                break;


                        }


                }


                if($this->getTargetContentClass() == 'article')


                {


                        switch ($field_name)


                        {


                                case 'name':


                                        return 'title';


                                break;


                                case 'shortname':


                                        return 'short_title';


                                break;


                                case 'description':


                                        return 'body';


                                break;


                                case 'publishdate':


                                        return 'publish_date';


                                break;


                                case 'shortdescription':


                                        return 'intro';


                                break;


                                default:


                                        return $field_name;


                                break;


                        }


                }


                else


                {


                        switch ($field_name)


                        {


                                case 'shortname':


                                        return 'short_name';


                                break;


                                case 'showsubitems':


                                        return 'show_children';


                                break;


                                case 'publishdate':


                                        return 'publish_date';


                                break;


                                default:


                                        return $field_name;


                                break;


                        }


                }


        }


 

        /*


        * Handles 'stringy' fields (Text,XML etc) before storing them in ez publish


        */


        function getValueFromField()


        {


                switch( $this->current_field->getAttribute('name') )


                {


                        case 'publishdate':


                        {


                                $return_unix_ts = time();


                                $us_formated_date = $this->current_field->nodeValue;


                                $parts = explode('/', $us_formated_date );


                                if( count( $parts ) == 3)


                                {


                                        $return_unix_ts = mktime( 0,0,0, $parts[0], $parts[1] , $parts[2] );


                                }


                                return $return_unix_ts;


                        }


                        break;


                        case 'shortdescription':


                        case 'description':


                        {


                                $xml_text_parser = new XmlTextParser();


                                // Having probs with the in-page anchors: <a name="eztoc27594_0_1" id="eztoc27594_0_1"></a> - they don't get imported. Parse them out:


                                if(preg_match("#<a name=\"eztoc.*\" id=\"eztoc.*\"></a>#e",$this->current_field->nodeValue))


                                {


                                        $this->current_field->nodeValue = preg_replace("#<a name=\"eztoc.*\" id=\"eztoc.*\"></a>#",'',$this->current_field->nodeValue);


                                }


                                $xmltext = $xml_text_parser->Html2XmlText( $this->current_field->nodeValue );


                               echo $xmltext;


                                if($xmltext !== false)


                                {


                                        return $xmltext;


                                }


                                else


                                {


                                        $message = 'Failed to parse XML for attribute: '.$this->current_field->getAttribute('name');


                                        $this->writeLog(self::REMOTE_IDENTIFIER.$this->current_row->getAttribute('id').': '.$message,$logfile);


                                        return false;


                                }


                        }


                        break;


                        default:


                        {


                                return $this->current_field->nodeValue;


                        }


                        break;


                }


        }


 

        /*


        * @desc         Logic where to place the current content node into the content tree


        *


        */


        function getParentNodeId()


        {


                $parent_id = $this->parent_id_fallback;


 

                // Conventional method taken right out of XMLFolders.php example


                if($this->current_row->getAttribute('parent_id'))


                {


                        $parent_remote_id = $this->current_row->getAttribute('parent_id');


                        $eZ_object = eZContentObject::fetchByRemoteID( self::REMOTE_IDENTIFIER.$parent_remote_id );


                        if($eZ_object)


                        {


                                $parent_id = $eZ_object->attribute('main_node_id');


                        }


                        else


                        {


                                // If a parent_id can't be found using self::REMOTE_IDENTIFIER.$parent_remote_id as the key, lose the REMOTE_IDENTIFIER and try that with eZContentObject::fetch()


                                $eZ_object = eZContentObject::fetch( $parent_remote_id );


                                if($eZ_object)


                                {


                                        $parent_id = $eZ_object->attribute('main_node_id');


                                }


                        }


                }


                return $parent_id;


        }


 

        function getDataRowId()


        {


                return self::REMOTE_IDENTIFIER.$this->current_row->getAttribute('id');


        }


 

        /*


        * - Allow the flexibility to extract data from multiple content-classes a single source XML file from ezxmlexport:


        * - See comments by Joachim Karl at: http://ez.no/developer/contribs/import_export/data_import


        */                              


        function getTargetContentClass()


        {       


                if($this->current_row->getAttribute('type'))


                {


                        return $this->current_row->getAttribute('type');


                }


                else


                {


                        $message = 'eZ content-class not found. Given class name was: '.$this->current_row->getAttribute('type');


                        $this->writeLog(self::REMOTE_IDENTIFIER.$this->current_row->getAttribute('id').': '.$message,$logfile);


                        return false;


                }


        }       


 

        function readData()


        {


                $filename = $this->xml_source_path.'/'.$this->xml_source_file.'/'.$this->xml_source_file.'.transformed.xml';


                if(isset($this->xml_source_path) && isset($this->xml_source_file))


                {


                        $filename = $this->xml_source_path.'/'.$this->xml_source_file.'/'.$this->xml_source_file.'.transformed.xml';


                        if(!is_file($filename))


                        {


                                $message = 'Cannot open '.$filename.' for reading. Please check files/dirs exist and permissions are set correctly'."\n";


                                $this->writeLog(self::REMOTE_IDENTIFIER.$this->current_row->getAttribute('id').': '.$message,$logfile);


                                return false;


                        }


                        else


                        {


                                return $this->parse_xml_document($filename,$this->root_node);


                        }


                }


                else


                {


                                $message = 'Source export file cannot be found or is not set. Please check files/dirs exist and permissions are set correctly'."\n";


                                $this->writeLog(self::REMOTE_IDENTIFIER.$this->current_row->getAttribute('id').': '.$message,$logfile);


                                return false;


                }


        }


 

        function post_publish_handling( $eZ_object, $force_exit )


        {


                $force_exit = false;


                return true;


        }


}


?>