<?php

/**

 * XML parser for ODS files

 *

 * Parses the content.xml file

 *

 * @package CSVImproved

 * @subpackage Helpers

 * @author Roland Dalmulder

 * @link http://www.csvimproved.com

 * @copyright 	Copyright (C) 2006 - 2012 RolandD Cyber Produksi

 * @license 	GNU/GPLv3 http://www.gnu.org/licenses/gpl-3.0.html

 * @version $Id: ods.php 754 2009-02-14 10:53:32Z Suami $

 *

 * @todo add support for zip file

 * @todo add multiple sheet processing

 */

 

/**

 * Parse ODS files

 *

 * @package CSVImproved

 * @subpackage Helpers

 */

class ODSParser {

	

	/* Set private variables */

	/** @var string filename */

	private $_file = null;

	/** @var string tag */

	private $_tag = null;

	/** @var object the xml parser */

	private $_xml_parser = null;

	/** @var bool set if inside a tag */

	private $_insideitem = false;

	/** @var array tags to process */

	private $_tagprocess = array('TABLE:TABLE-CELL', 'TABLE:TABLE-ROW');

	/** @var array holds the data per node */

	private $_data = array();

	/** @var integer counts the positions */

	private $_poscount = 0;

	/** @var integer counts the number of times a string is repeated */

	private $_columns_repeated = 0;

	

	/* Set public variables */

	/** @var array holds the parsed data */

	public $fulldata = array();

	/** @var int holds the number of rows */

	public $rows = null;

	/** @var int holds the number of columns */

	public $cols = null;

	

	/**

	 * Start to process the XML file

	 *

	 * @todo change die to a proper return result

	 * @todo add support for <table:table-cell/>

	 */

	public function read($filename) {

		$this->_file = $filename;

		list($this->_xml_parser, $fp) = $this->new_xml_parser($this->_file);

		if (!$this->_xml_parser) {

			die("could not open XML input");

		}

		$data = '';

		while (!feof($fp)) {

			$data .= fread($fp, 4096);

		}

		fclose($fp);

		

		/* Regex to find data elements */

		$regex = "#<text:p>(.*?)</text:p>#s";

		/* Find the content links */

		$matches = array();

		$result = preg_match_all($regex, $data, $matches);

		

		/* Loop through the links to link them */

		foreach ($matches[1] as $key => $value) {

			if (!stristr($value, '<![CDATA[')) {

				$find = $matches[0][$key];

				$replace = '<text:p><![CDATA['.$value.']]></text:p>';

				$data = str_replace($find, $replace, $data);

			}

		}

		if (!xml_parse($this->_xml_parser, $data, true)) {

			

			die(sprintf("XML error: %s at line %d\n",

						xml_error_string(xml_get_error_code($this->_xml_parser)),

						xml_get_current_line_number($this->_xml_parser)));

		}

		

		xml_parser_free($this->_xml_parser);

		$this->cols = count($this->fulldata[0]);

		$this->rows = count($this->fulldata);

		return true;

	}

	

	/**

	 * Handle the opening element

	 */

	private function startElement($parser, $tagname, $attribs) {

		if ($this->_insideitem) {

			$this->_tag = $tagname;

		} 

		elseif (in_array($tagname, $this->_tagprocess)) {

			$this->_insideitem = true;

			if (isset($attribs['TABLE:NUMBER-COLUMNS-REPEATED'])) {

				$this->_columns_repeated = $attribs['TABLE:NUMBER-COLUMNS-REPEATED'];

				$this->_poscount += $this->_columns_repeated;

			}

			else $this->_poscount++;

		} 

	}

	

	/**

	 * Handle the closing element

	 */

	private function endElement($parser, $name) {

		/* Add empty fields */

		if ($this->_poscount > count($this->_data)) {

			$addfields = $this->_poscount - count($this->_data);

			for ($addempty = 0; $addempty < $addfields; $addempty++) {

				$this->_data[] = '';

			}

		}

		/* Empty the data array otherwise all data gets put into 1 array */

		if ($name == 'TABLE:TABLE-ROW') {

			/* Check for empty sheets */

			if (count($this->_data) <> 1 && !empty($this->_data[1])) {

				$this->fulldata[] = $this->_data;

			}

			$this->_data = array();

			$this->_poscount = 0;

		}

		

		/* Reset everything */

		$this->_tag = null;

		$this->_insideitem = false;

	}

	

	/**

	 * Handle the data

	 *

	 * @todo parse <text:s text:c="2"/>

	 */

	private function characterData($parser, $data) {

		if ($this->_insideitem) {

			switch ($this->_tag) {

				case 'TEXT:P':

					/* Replace single spaces */

					$data = str_replace('<text:s/>', ' ', $data);

					if (stristr($data, '<text:s')) {

						/* Regex to find data elements */

						$regex = "#<text:s text:c=\"(.*?)\"/>#s";

						/* Find the spaces */

						$matches = array();

						$result = preg_match_all($regex, $data, $matches);

						

						/* Replace the spaces */

						foreach ($matches[1] as $key => $value) {

							$find = '<text:s text:c="'.$value.'"/>';

							$replace = str_repeat(" ", $value);

							$data = str_replace($find, $replace, $data);

						}

					}

					if (!isset($this->_data[1])) $this->_data[1] = $data;

					else {

						if ($this->_columns_repeated >= 2) {

							for ($print = 0; $print < $this->_columns_repeated; $print++) {

								$this->_data[] = $data;

							}

							$this->_columns_repeated = 0;

						}

						else $this->_data[] = $data;

					}

					break;

			}

		}

		

	}

	

	private function new_xml_parser($file) {

		$this->_xml_parser = xml_parser_create("UTF-8");

		xml_parser_set_option($this->_xml_parser, XML_OPTION_CASE_FOLDING, 1);

		xml_set_object($this->_xml_parser, $this);

		xml_set_element_handler($this->_xml_parser, "startElement", "endElement");

		xml_set_character_data_handler($this->_xml_parser, "characterData");

		$fp = fopen($file, "rb");

		if (!$fp) return false;

		else return array($this->_xml_parser, $fp);

	}

}

?>