parserxml.php

来自「php 开发的内容管理系统」· PHP 代码 · 共 644 行 · 第 1/2 页

				$parts = $this->sub_makeXHTML($parser);
				$parts = explode('|', $parts);
				$title = array_shift($parts);
				$ret .= $this->getTemplateXHTML($title, $parts, & $parser);
				break;

			case 'TEMPLATEVAR':
				$x = $this->sub_makeXHTML($parser);
				if (isset ($parser->mCurrentTemplateOptions["{$x}"]))
					$ret .= $parser->mCurrentTemplateOptions["{$x}"];
				break;

			# Internal use, not generated by wiki2xml parser
			case 'IGNORE':
				$ret .= $this->sub_makeXHTML($parser);

			case 'NOWIKI':
				$parser->nowiki++;
				$ret .= $this->sub_makeXHTML($parser, '');
				$parser->nowiki--;


			# Unknown HTML extension
			case 'EXTENSION': # This is currently a dummy!!!
				$ext = $this->attrs['NAME'];

				$ret .= '<'.$ext.'>';
				$ret .= $this->sub_makeXHTML($parser);
				$ret .= '</'.$ext.'> ';
				break;


			# Table stuff

			case 'TABLE':
				$ret .= $this->sub_makeXHTML($parser, 'table');
				break;
			case 'TABLEROW':
				$ret .= $this->sub_makeXHTML($parser, 'tr');
				break;
			case 'TABLECELL':
				$ret .= $this->sub_makeXHTML($parser, 'td');
				break;
			case 'TABLEHEAD':
				$ret .= $this->sub_makeXHTML($parser, 'th');
				break;
			case 'CAPTION':
				$ret .= $this->sub_makeXHTML($parser, 'caption');
				break;
			case 'ATTRS': # SPECIAL CASE : returning attributes
				return $this->getTheseAttrs();


			# Lists stuff
			case 'LISTITEM':
				if ($parser->mListType == 'dl')
					$ret .= $this->sub_makeXHTML($parser, 'dd');
				else
					$ret .= $this->sub_makeXHTML($parser, 'li');
				break;
			case 'LIST':
					$type = 'ol'; # Default
					if ($this->attrs['TYPE'] == 'bullet')
						$type = 'ul';
					else
						if ($this->attrs['TYPE'] == 'indent')
							$type = 'dl';
					$oldtype = $parser->mListType;
					$parser->mListType = $type;
					$ret .= $this->sub_makeXHTML($parser, $type);
					$parser->mListType = $oldtype;
				break;

			# Something else entirely
			default:
				$ret .= '<'.$n.'>';
				$ret .= $this->sub_makeXHTML($parser);
				$ret .= '</'.$n.'> ';
			} // switch($n)

		$ret = "\n{$ret}\n";
		$ret = str_replace("\n\n", "\n", $ret);
		return $ret;
	}

	/**
	 * A function for additional debugging output
	 */
	function myPrint() {
		$ret = "<ul>\n";
		$ret .= "<li> <b> Name: </b> $this->name </li>\n";
		// print attributes
		$ret .= '<li> <b> Attributes: </b>';
		foreach ($this->attrs as $name => $value) {
			$ret .= "$name => $value; ";
		}
		$ret .= " </li>\n";
		// print children
		foreach ($this->children as $child) {
			if (is_string($child)) {
				$ret .= "<li> $child </li>\n";
			} else {
				$ret .= $child->myPrint();
			}
		}
		$ret .= "</ul>\n";
		return $ret;
	}
}

$ancStack = array (); // the stack with ancestral elements

// START Three global functions needed for parsing, sorry guys
/** @todo document */
function wgXMLstartElement($parser, $name, $attrs) {
	global $ancStack;

	$newElem = new element;
	$newElem->name = $name;
	$newElem->attrs = $attrs;

	array_push($ancStack, $newElem);
}

/** @todo document */
function wgXMLendElement($parser, $name) {
	global $ancStack, $rootElem;
	// pop element off stack
	$elem = array_pop($ancStack);
	if (count($ancStack) == 0)
		$rootElem = $elem;
	else
		// add it to its parent
		array_push($ancStack[count($ancStack) - 1]->children, $elem);
}

/** @todo document */
function wgXMLcharacterData($parser, $data) {
	global $ancStack;
	$data = trim($data); // Don't add blank lines, they're no use...
	// add to parent if parent exists
	if ($ancStack && $data != "") {
		array_push($ancStack[count($ancStack) - 1]->children, $data);
	}
}
// END Three global functions needed for parsing, sorry guys

/**
 * Here's the class that generates a nice tree
 * @package MediaWiki
 * @subpackage Experimental
 */
class xml2php {

	/** @todo document */
	function & scanFile($filename) {
		global $ancStack, $rootElem;
		$ancStack = array ();

		$xml_parser = xml_parser_create();
		xml_set_element_handler($xml_parser, 'wgXMLstartElement', 'wgXMLendElement');
		xml_set_character_data_handler($xml_parser, 'wgXMLcharacterData');
		if (!($fp = fopen($filename, 'r'))) {
			die('could not open XML input');
		}
		while ($data = fread($fp, 4096)) {
			if (!xml_parse($xml_parser, $data, feof($fp))) {
				die(sprintf("XML error: %s at line %d", xml_error_string(xml_get_error_code($xml_parser)), xml_get_current_line_number($xml_parser)));
			}
		}
		xml_parser_free($xml_parser);

		// return the remaining root element we copied in the beginning
		return $rootElem;
	}

	/** @todo document */
	function scanString($input) {
		global $ancStack, $rootElem;
		$ancStack = array ();

		$xml_parser = xml_parser_create();
		xml_set_element_handler($xml_parser, 'wgXMLstartElement', 'wgXMLendElement');
		xml_set_character_data_handler($xml_parser, 'wgXMLcharacterData');

		if (!xml_parse($xml_parser, $input, true)) {
			die(sprintf("XML error: %s at line %d", xml_error_string(xml_get_error_code($xml_parser)), xml_get_current_line_number($xml_parser)));
		}
		xml_parser_free($xml_parser);

		// return the remaining root element we copied in the beginning
		return $rootElem;
	}

}

/**
 * @todo document
 * @package MediaWiki
 * @subpackage Experimental
 */
class ParserXML extends Parser {
	/**#@+
	 * @private
	 */
	# Persistent:
	var $mTagHooks, $mListType;

	# Cleared with clearState():
	var $mOutput, $mAutonumber, $mDTopen, $mStripState = array ();
	var $mVariables, $mIncludeCount, $mArgStack, $mLastSection, $mInPre;

	# Temporary:
	var $mOptions, $mTitle, $mOutputType, $mTemplates, // cache of already loaded templates, avoids
	// multiple SQL queries for the same string
	$mTemplatePath; // stores an unsorted hash of all the templates already loaded
	// in this path. Used for loop detection.

	var $nowikicount, $mCurrentTemplateOptions;

	/**#@-*/

	/**
	 * Constructor
	 *
	 * @public
	 */
	function ParserXML() {
		$this->mTemplates = array ();
		$this->mTemplatePath = array ();
		$this->mTagHooks = array ();
		$this->clearState();
	}

	/**
	 * Clear Parser state
	 *
	 * @private
	 */
	function clearState() {
		$this->mOutput = new ParserOutput;
		$this->mAutonumber = 0;
		$this->mLastSection = "";
		$this->mDTopen = false;
		$this->mVariables = false;
		$this->mIncludeCount = array ();
		$this->mStripState = array ();
		$this->mArgStack = array ();
		$this->mInPre = false;
	}

	/**
	* Turns the wikitext into XML by calling the external parser
	*
	*/
	function html2xml(& $text) {
		global $wgWiki2xml;

		# generating html2xml command path
		$a = $wgWiki2xml;
		$a = explode('/', $a);
		array_pop($a);
		$a[] = 'html2xml';
		$html2xml = implode('/', $a);
		$a = array ();

		$tmpfname = tempnam( wfTempDir(), 'FOO' );
		$handle = fopen($tmpfname, 'w');
		fwrite($handle, utf8_encode($text));
		fclose($handle);
		exec($html2xml.' < '.$tmpfname, $a);
		$text = utf8_decode(implode("\n", $a));
		unlink($tmpfname);
	}

	/** @todo document */
	function runXMLparser(& $text) {
		global $wgWiki2xml;

		$this->html2xml($text);

		$tmpfname = tempnam( wfTempDir(), 'FOO');
		$handle = fopen($tmpfname, 'w');
		fwrite($handle, $text);
		fclose($handle);
		exec($wgWiki2xml.' < '.$tmpfname, $a);
		$text = utf8_decode(implode("\n", $a));
		unlink($tmpfname);
	}

	/** @todo document */
	function plain_parse(& $text, $inline = false, $templateOptions = array ()) {
		$this->runXMLparser($text);
		$nowikicount = 0;
		$w = new xml2php;
		$result = $w->scanString($text);

		$oldTemplateOptions = $this->mCurrentTemplateOptions;
		$this->mCurrentTemplateOptions = $templateOptions;

		if ($inline) { # Inline rendering off for templates
			if (count($result->children) == 1)
				$result->children[0]->name = 'IGNORE';
		}

		if (1)
			$text = $result->makeXHTML($this); # No debugging info
		else
			$text = $result->makeXHTML($this).'<hr>'.$text.'<hr>'.$result->myPrint();
		$this->mCurrentTemplateOptions = $oldTemplateOptions;
	}

	/** @todo document */
	function parse($text, & $title, $options, $linestart = true, $clearState = true) {
		$this->plain_parse($text);
		$this->mOutput->setText($text);
		return $this->mOutput;
	}

}
?>